2018 ABoVE Science Team Meeting

Poster Abstracts

Presenters are in italics.

Poster Session A

Poster Session B
1. Fire Disturbance
2. Vegetation Dynamics and Distribution

Poster Session C
1. Permafrost and Hydrology
2. Wildlife and Ecosystem Services

Poster Session A

Tuesday 4:30 PM

Carbon Dynamics

Lake Change in the 13 Boreal and Arctic Study Regions in the recent ~60 years

Prajna Lindgren, University of Alaska Fairbanks, pregmi@alaska.edu
Franz Meyer, University of Alaska Fairbanks, fjmeyer@alaska.edu
Katey Walter Anthony, University of Alaska Fairbanks, kmwalteranthony@alaska.edu
Lisa Wirth, University of Alaska Fairbanks, lisa@gina.alaska.edu
Andrew Herbst, University of Alaska Fairbanks, amherbst@alaska.edu

Permafrost in the northern latitude stores a significant amount of carbon. However, formation and expansion of lakes in ice-rich permafrost region facilitate the release of soil-stored carbon in the form of methane (CH4), an important greenhouse gas. Previous studies have shown that lakes are changing at a greater rate in the last few decades due to warming permafrost and many regions have experienced significant lake expansion consequently increasing the release of old permafrost carbon. Therefore, in this study, we utilized historic aerial images (1 m spatial resolution) acquired in the 1950s and recent SPOT satellite images acquired post-2009 (2.5 spatial resolution) to map historic and modern lake margins using semi-automated object-oriented image analysis in 13 different sites distributed over Alaska to produce fine-scale maps of lake area change. These sites include (1) Barrow Peninsula, (2) North of Teshupuk, (3) Atqasuk, (4) Inigok Fish Creek in the north, (5) Minto, (6) Fairbanks, (7) Standard Dunbar, (8) Fox in the interior, (9) Gerstle, (10) Moon, (11) Mansfield, (12) Northway Boundary in the east, and (13) Northern Seward Peninsula in the west. The preliminary results show that there is both lake expansion and shrinkage. The interior of Alaska shows major lake expansion especially in Fairbanks area, an increase in lake area by approx. 42%, while the eastern study sites show mostly lake area shrinkage, with the highest lake area loss in Moon by approx. 15%. We have also found that the location of methane seeps and their density align very well with expanding part of the lakes in Fairbanks. The mapping on the rest of the study sites is still a work in progress, however for some sites like Barrow in the north, our initial interpretation shows that many lakes are experiencing lake expansion. Our next step, after creating lake change map products, will be to compare the lake expansion in these regions with CH4 seep locations that we have observed in the field as well as mapped in the recent high-resolution aerial and satellite imagery.

Associated Project: Meyer-01

Geologic methane seeps in Arctic Alaska

Katey Walter Anthony, University of Alaska, Fairbanks, kmwalteranthony@alaska.edu
Janelle Sharp, University of Alaska Fairbanks, Janelle.Sharp@nana.com
Melanie Engram, University of Alaska, Fairbanks, melanie.engram@alaska.edu
Prajna Lindgren, University of Alaska, Fairbanks, pregmi@alaska.edu
Peter Anthony, University of Alaska Fairbanks, pmanthony@alaska.edu
Franz Meyer, University of Alaska, Fairbanks, fjmeyer@alaska.edu

Methane (CH4), a potent greenhouse gas, accumulates in subsurface hydrocarbon reservoirs. In the Arctic, impermeable icy permafrost is thought to form a ‘cryosphere cap’ that traps gas leaking from these reservoirs, restricting flow to the atmosphere. We document the release of geologic methane to the atmosphere from large gas seeps in arctic Alaskan lakes using remote sensing and field work. While some seeps occur along boundaries of thawing permafrost (e.g. near large rivers with presumably through-going taliks and along the continuous-discontinuous permafrost border), others occur in inland lakes where continuous permafrost thickness likely exceeds talik depths beneath lakes. This raises questions about gas migration and release mechanisms in the continuous permafrost zone.

Associated Project(s):
Meyer-01
Miller-C-05

Estimating lake methane ebullition with Synthetic Aperture Radar (SAR)

Melanie Engram, University of Alaska, Fairbanks, melanie.engram@alaska.edu
Katey Walter Anthony, University of Alaska, Fairbanks, kmwalteranthony@alaska.edu
Torsten Sachs, GFX German Research Centre for Geoscience, Potsdam, Germany, torsten.sachs@gfz-potsdam.de
Guido Grosse, Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research, guido.grosse@awi.de
Franz Meyer, University of Alaska, Fairbanks, fjmeyer@alaska.edu

Lakes cover vast extents of lowland Arctic landscapes and are recognized as a major source of atmospheric methane, a potent greenhouse gas. Yet estimates of Arctic and sub-Arctic lake-source methane emissions are highly uncertain, due in large part to the spatial and temporal irregularity of ebullition (bubbling), but further compounded by the expense and logistical challenge of obtaining enough widespread field measurements to accurately represent a heterogeneous lake-scape. Previous work in synthetic aperture radar (SAR) remote sensing showed backscatter intensity from early winter lake ice correlated with methane ebullition as measured by field surveys in one region in Alaska. Here, we show this relationship holds true for four additional geographic areas in Alaska and create a regionally robust regression model with L-band SAR intensity and methane measurements. We invert this model and, in conjunction with data from long-term bubble traps, estimate methane ebullition from SAR backscatter intensity for over 5,000 lakes on landscape scales. We observed strong lake-size relationships to ebullition and regional variability in lake emissions related to climate and permafrost carbon stocks. Rather than upscaling from a limited number of field observations, our regional lake methane emissions estimates are based on SAR observations for each lake which show increased L-band radar intensity from methane ebullition bubbles disrupting and warping the highly reflective ice/water interface. This new inexpensive approach to remote-sensing lake ebullition offers a unique opportunity to improve knowledge about greenhouse gas fluxes for seasonally ice-covered lakes across the globe.

Associated Project: Meyer-01

Mapping methane plumes from the ABoVE domain using AVIRIS-NG

Andrew Thorpe, JPL, Andrew.K.Thorpe@jpl.nasa.gov
David Thompson, Jet Propulsion Laboratory / Caltech, David.R.Thompson@jpl.nasa.gov
Charles Miller, NASA JPL, charles.e.miller@jpl.nasa.gov
Riley Duren, NASA JPL, riley.m.duren@jpl.nasa.gov
Michael Bernas, JPL, Michael.Bernas@jpl.nasa.gov
Winston Olson-Duval, JPL, Winston.Olson-Duvall@jpl.nasa.gov

Airborne imaging spectrometers like the next generation Airborne Visible/Infrared Imaging Spectrometer (AVIRIS-NG) are well suited for identifying local methane (CH4) sources by covering large regions with the high spatial resolution necessary to resolve emissions. As part of the ABoVE campaign, AVIRIS-NG surveyed portions of Alaska and Canada that contain potential CH4 emission sources from both anthropogenic and natural sources. While data analysis is ongoing, a number of CH4 plumes have been observed from anthropogenic sources, including gas flaring stacks, coal seams, coal mines, and well pads. Future work will focus on regions with possible natural CH4 sources. Imaging spectrometers permit direct attribution of emissions to individual point sources which is particularly useful given the large uncertainties associated with anthropogenic and natural emissions.

Associated Project: Miller-C-05

The predominance of young carbon in Arctic whole-lake CH4 and CO2 emissions and implications for Boreal yedoma lakes

Clayton Elder, Jet Propulsion Laboratory, cdelder@uci.edu
Xiaomei Xu, University of California, Irvine, xxu@uci.edu
Jennifer Walker, University of California, Irvine, jclehman@uci.edu
Katey Walter Anthony, University of Alaska, Fairbanks, kmwalteranthony@alaska.edu
John Pohlman, USGS, jpohlman@usgs.gov
Chris Arp, University of Alaska, Fairbanks, cdarp@alaska.edu
Benjamin Gaglioti, Lamont-Doherty Earth Observatory, gaglioti@ldeo.columbia.edu
Amy Townsend-Small, University of Cincinnati, townseay@ucmail.uc.edu
Hinkel Kenneth, Michigan Technological University, kmhinkel@mtu.edu
Claudia Czimczik, University Of California, Irvine, czimczik@uci.edu

Lakes in Arctic and Boreal regions are hotspots for atmospheric exchange of the greenhouse gases CO2 and CH4. Thermokarst lakes are a subset of these Northern lakes that may further accelerate climate warming by mobilizing ancient permafrost C (> 11,500 years old) that has been disconnected from the active C cycle for millennia. Northern lakes are thus potentially powerful agents of the permafrost C-climate feedback. While they are critical for projecting the magnitude and timing these feedbacks from the rapidly warming circumpolar region, we lack datasets capturing the diversity of northern lakes, especially regarding their CH4 contributions to whole-lake C emissions and their ability to access and mobilize ancient C. We measured the radiocarbon (14C) ages of CH4 and CO2 emitted from 60 understudied lakes and ponds in Arctic and Boreal Alaska during winter and summer to estimate the ages of the C sources yielding these gases. Integrated mean ages for whole-lake emissions were inferred from the 14C-age of dissolved gases sampled beneath seasonal ice. Additionally, we measured concentrations and 14C values of gases emitted by ebullition and diffusion in summer to apportion C emission pathways. Using a multi-sourced mass balance approach, we found that whole-lake CH4 and CO2 emissions were predominantly sourced from relatively young C in most lakes. In Arctic lakes, CH4 originated from 850 14C-year old C on average, whereas dissolved CO2 was sourced from 400 14C-year old C, and represented 99% of total C flux. Although ancient C had a minimal influence (11% of total emissions), we discovered that lakes in finer-textured aeolian deposits (Yedoma) emitted twice as much ancient C as lakes in sandy regions. In Boreal, yedoma-type lakes, CH4 and CO2 were fueled by significantly older sources, and mass balance results estimated CH4-ebullition to comprise 50-60% of whole-lake CH4 emissions. The mean 14C-age of Boreal emissions was 6,000 14C-years for CH4-C, and 2,400 14C-years for CO2-C. Seasonal differences in dissolved CH4 revealed a clear influence of trapped ebullition dissolving into the water below lake ice in Boreal, but not Arctic lakes. Together, our data demonstrate that regional surficial geology exerts a larger control than climate on C ages and gas emission pathways from lakes.

Associated Project: Miller-C-05

Wildfire Effects on Permafrost and Soil Moisture in Spruce Forests of Interior Alaska

Christopher Potter, NASA ARC, chris.potter@nasa.gov
Charles Hugny, Interior Alaska Wilderness Consultants, chugny@yahoo.com

In the summer of 2015, hundreds of forest fires burned across the state of Alaska. Several uncontrolled wildfires near the town of Tanana on the Yukon River were responsible for the largest portion of the area burned statewide. In July 2017, field measurements were carried out in both unburned and burned forested areas nearly adjacent to one-another, all within 15 miles of the village of Tanana. These surveys were used to first visually verify locations of different burn severity class (low, moderate, or high) estimated in 2016 from Landsat images (collected before and after the 2015 Tanana-area wildfires). Surface and profile measurements to 30 cm depth of soil layers at these same locations were collected for evidence of moss layer and forest biomass burning. Soil temperature and moisture content was measured to 30 cm depth, and depth to permafrost was estimated by soil excavation, wherever necessary. Digital thermal infra-red (TIR) images of the soil profiles were taken at each site location, and root-zone organic layer samples were extracted for further chemical analysis. Results supported the hypothesis that the loss of surface organic layers is a major factor determining post-fire soil water and temperature changes and the depth of permafrost thawing. In the most severely burned forest sites, complete consumption of the living moss organic layer was strongly associated with both warming at the soil surface layer and increases in soil water content, relative to unburned forest sites. Soil temperature at both 10 cm and 30 cm depth at burned forest sites increased by 8 - 10 deg. C compared to unburned forest sites. Below 15 cm soil depth, the temperature of unburned sites dropped gradually to frozen levels by 30 cm depth, while soil temperatures at burned sites remained above 5o C to 30 cm depth. The water content measured at 3 cm depth at burned sites was commonly in excess of 30% by volume, compared to unburned forest sites. The strong correlation between burn index values and depth to permafrost measured across all forest sites sampled in July 2017 showed that the new ice-free profile in severely burned forest areas was commonly 50 cm deeper than in unburned forest soils.

Associated Project: Miller-C-03

Landscape Patterns of Vegetation Burning in Ecosystems of Interior Alaska Derived from Satellite Image Analysis and Field Measurements

Christopher Potter, NASA ARC, chris.potter@nasa.gov

In the summer of 2015, hundreds of forest fires burned across the state of Alaska. Several uncontrolled wildfires near the town of Tanana on the Yukon River were responsible for the largest portion of the area burned statewide. In July 2017, field measurements were carried out in both unburned and burned forested areas nearly adjacent to one-another to visually verify locations of different burn severity classes (low, moderate, or high) estimated in 2016 from MODIS and Landsat satellite images. Spatial analysis of landscape patterns of vegetation burning and severity classes were mapped in association with topography, pre-fire live biomass density, and land cover types. Results showed that the time-course of fire spread was repeated in several cycles, from relatively higher elevations (> 100 m), to lower elevations (< 30 m), and on slopes of greater than 10% to slopes lower than 5%. Burning cycles were also associated with higher-to-lower pre-fire NDVI burning and (pre-fire) land cover classes of predominantly forest and shrub cover at the beginning of a cycle, into herbaceous grasslands and river flats at the end of each cycle. Results from field measurements in unburned and nearby burned forest sites, all within 15 miles of the village of Tanana in 2017, were consistent with the hypothesis that the loss of surface organic layers in boreal ecosystem fires is a major factor determining post-fire soil temperature changes and the depth of thawing. Soil temperature profiles to 30 cm depth at burned forest sites increased by an average of 8o - 10o C compared to unburned forest sites.

Associated Project: Miller-C-03

11:

The 2017 ASCENDS/ABoVE Airborne Campaign and initial look at Pulsed Lidar Measurements of CO2 Column Concentrations

James Abshire, NASA Goddard, james.b.abshire@nasa.gov
Haris Riris, NASA Goddard, haris.riris@nasa.gov
Graham Allan, NASA Goddard/Sigma, graham.r.allan@nasa.gov
Jianping Mao, ESSIC/University of Maryland, jianping.mao@nasa.gov
William Hasselbrack, NASA Goddard/Sigma, wiilliam.e.hasselbrack@nasa.gov
Kenji Numata, NASA Goddard, kenji.numata@nasa.gov
Jeffrey Chen, NASA Goddard, jeffrey.r.chen@nasa.gov
Randy Kawa, NASA Goddard, stephan.r.kawa@nasa.gov
Joshua Digangi, NASA Langley, joshua.p.digangi@nasa.gov
Yonghoon Choi, NASA Langley, yonghoon.choi-1@nasa.gov

The 2017 ASCENDS airborne campaign was flown on the NASA DC-8 in late July and early August. The campaign objectives were to assess the accuracy of airborne IPDA lidar measurements of CO2 column concentrations (XCO2), and to extend these lidar measurements to the ABoVE study area in the Arctic. Eight science flights were conducted with simultaneous XCO2 measurements from NASA Goddard’s CO2 Sounder lidar and the NASA Langley ACES lidar along with in-situ CO2 measurements made at aircraft altitude with the AVOCET and Picarro in-situ sensors. Over 40 spiral-down maneuvers were conducted during the campaign over locations in California, Northwest Territories (NWT) Canada and over Alaska, along with the transit flights from California to Alaska and return. Since each spiral maneuver allows comparing the retrievals of XCO2 from the lidar against those computed from in-situ measured CO2, this campaign allowed an unprecedented opportunity to assess the lidar measurements of XCO2 over a diverse set of conditions, including those in the Arctic. The CO2 Sounder lidar flown on the campaign is a pulsed, multiple-wavelength integrated path differential absorption lidar that measures XCO2 in a nadir path from the aircraft. The lidar measures the range resolved shape of the 1572.33 nm CO2 absorption line to scattering surfaces, including to the ground and to the tops of clouds. The measurements for this campaign used 30 fixed-wavelength samples distributed across the line. Analysis estimates the lidar range and pulse energies at each wavelength 10 times per second. For each second the retrievals solve for the CO2 absorption line shape and the column average CO2 concentrations by using radiative transfer calculations, the aircraft altitude and range to the scattering surface, and the atmospheric conditions. The presentation will describe all the flight paths, and will present the early results from analyzing the retrieved XCO2 from the CO2 Sounder lidar in the campaign over some of the regions. It will also present the initial results of comparing the lidar-retrieved XCO2 to those computed from the in-situ sensors at some of the spiral-down locations and discuss the ongoing analyses.

Associated Project: Abshire-01

12:

DC-8 In Situ Measurements for ASCENDS and ABoVE 2017

Stephan Kawa, NASA GSFC, stephan.r.kawa@nasa.gov
Joshua DiGangi, NASA Langley Research Center, joshua.p.digangi@nasa.gov
Glenn Diskin, NASA Langley Research Center, glenn.s.diskin@nasa.gov
Yonghoon Choi, NASA LaRC/SSAI, yonghoon.choi-1@nasa.gov
John Nowak, NASA Langley Research Center, john.b.nowak@nasa.gov
Jianping Mao, NASA GSFC/ESSIC, Jianping.Mao@nasa.gov
James Abshire, NASA GSFC, James.B.Abshire@nasa.gov
Bing Lin, NASA Langley Research Center, bing.lin@nasa.gov
Edward Browell, NASA Langley Research Center/UNH, edward.v.browell@nasa.gov

During July and August 2017 the NASA DC-8 completed 8 flights from Palmdale, CA and Fairbanks, AK in support of atmospheric CO2 measurements using lidar instruments designed as demonstrators for NASA’s planned ASCENDS mission (Active Sensing of CO2 Emissions over Nights, Days, & Seasons). In addition to these CO2 laser remote sensing instruments, the aircraft carried a set of in situ sensors to provide ancillary data for the lidar XCO2 data validation, as well as to establish the airborne measurement context. In situ data are available for CO2 and CH4 from two different Picarro instruments, H2O from the Diode Laser Hygrometer (DLH), and CO from the Differential Absorption Carbon Monoxide Measurement (DACOM). The airborne campaign covered a wide range of Arctic/Boreal locations in 4 flights from Fairbanks plus the incoming and outgoing transits. Approximately 40 hours of flight data are available at high latitudes including 30 spiral profiles from ~9 km altitude to the surface. These data form a unique observational resource for the ABoVE 2017 intensive campaign. Here we present an initial analysis of the in situ complement of the DC-8 campaign data. We will show an overview of the data along with some systematic features and findings. The overriding first impression from this data set is the high degree of variability in the tracer data with altitude, location, and time. Inferring the influence of local sources, sinks, and transport will be challenging in such a dynamic environment. We look forward to connection with other in situ, ground-based and satellite remote sensing, and model output data together with application of the DC-8 data in syntheses for flux inference and process studies in the ABoVE domain.

Associated Project: Abshire-01

13:

First Results for Active Remote Sensing of Carbon Dioxide During the ABoVE 2017 Airborne Field Campaign using the Active Sensing of CO2 Emissions over Nights, Days, and Seasons (ASCENDS) CarbonHawk Experiment Simulator (ACES) Instrument

Mike Obland, NASA Langley, michael.d.obland@nasa.gov
Bing Lin, NASA Langley, bing.lin@nasa.gov
Byron Meadows, NASA LARC, byron.l.meadows@nasa.gov
Wialliam Carrion, SSAI, william.carrion@nasa.gov
Jonathan Hicks, SSAI, Jonathan.w.hicks@nasa.gov
Joseph Sparrow, NASA Langley, joseph.a.sparrow@nasa.gov
Susan Kooi, SSAI, susan.a.kooi@nasa.gov
Tai-Fang Fan, SSAI, Tai-fang.fan-1@nasa.gov
Edward Browell, STARSS III, edward.v.browell@nasa.gov

The Active Sensing of CO2 Emissions over Nights, Days, and Seasons (ASCENDS) CarbonHawk Experiment Simulator (ACES) is a NASA Langley Research Center instrument funded by NASA’s Science Mission Directorate that seeks to advance technologies critical to measuring atmospheric column carbon dioxide (CO2) mixing ratios in support of the NASA ASCENDS mission. The ACES instrument, an Intensity-Modulated Continuous-Wave (IM-CW) lidar, was designed for high-altitude aircraft operations and can be directly applied to space instrumentation to meet the ASCENDS mission requirements. The ACES design demonstrates advanced technologies critical for developing an airborne simulator and spaceborne instrument with lower platform consumption of size, mass, and power, and with improved performance. ACES recently flew on the NASA DC-8 aircraft during the 2017 ABoVE/ASCENDS airborne measurement campaign to test ASCENDS-related technologies in the challenging Arctic environment. ACES and the NASA Goddard CO2 Sounder collected the first Integrated Path Differential Absorption (IPDA) lidar measurements in the Arctic. Data were collected over a wide variety of surface reflectivities, terrain, and atmospheric conditions during the campaign’s 8 research flights (over 55 hours of science data). Simultaneous data were collected by in situ instrumentation on the aircraft.

Associated Project: Abshire-01

27:

Investigating sensitivity of soil freeze/thaw dynamics and cold-season respiration to snow cover changes in Alaska

Yonghong Yi, University of Montana, yonghong.yi@ntsg.umt.edu
John Kimball, University of Montana, johnk@ntsg.umt.edu
Donatella Zona, San Diego State University, d.zona@sheffield.ac.uk
Kyle Arndt, San Diego State University, karndt-w@sdsu.edu
Richard Chen, University of Southern California, chenrh@usc.edu
Mahta Moghaddam, University of Southern California, mahta@usc.edu
Rolf Reichle, NASA Goddard Space Flight Center, rolf.reichle@nasa.gov
Walter Oechel, San Diego State University, woechel@mail.sdsu.edu

The contribution of cold season respiration to the boreal-arctic carbon cycle and its potential feedback to climate change remain poorly quantified. Here, we developed an integrated modeling framework combining airborne low frequency (L+P-band) radar retrievals and landscape level (≥1km) satellite environmental observations with a detailed permafrost carbon model to investigate underlying processes controlling soil freeze/thaw (FT) dynamics and cold season CO2 respiration in Alaska. The permafrost carbon model simulates snow cover and soil thermal dynamics with soil water phase change included and accounts for soil carbon decomposition up to 3m below surface. Initial comparisons with tower-based measurements show that the model can well capture the soil temperature dynamics and seasonality of cold season respiration in both tundra and boreal forest areas, with large CO2 emissions in late fall and early winter that gradually diminish over the winter. Model outputs include active layer thickness (ALT) and regional carbon fluxes at 1-km resolution spanning the recent satellite era (2001-present) across Alaska. The regional simulations indicate that vegetation productivity plays a major role in controlling the total soil carbon respiration, while snow cover changes are dominant in controlling the relative contribution of different soil depths to total soil carbon emissions. In the future work, radar retrievals of soil moisture and soil organic carbon (SOC) will be used to improve the regional parameterization of the permafrost model; airborne atmospheric CO2 measurements will also be used to evaluate model processes controlling cold season respiration, particularly how a deeper unfrozen active layer with warming may contribute to changes in cold season respiration.

Associated Project(s):
Kimball-04
Moghaddam-03

32:

From Archaea to the Atmosphere: Integrating Microbial, Isotopic and Landscape-Scale Observations to Quantify Methane Emissions from Global High-Latitude Ecosystems

Ruth Varner, University of New Hampshire, ruth.varner@unh.edu
Michael Palace, ESRC-University of New Hampshire, michael.palace@unh.edu
Virginia Rich, Ohio State University, virginia.isabel.rich@gmail.com
Justin Fisk, Applied Geosolutions, Inc., jfisk@appliedgeosolutions.com
Bobby Braswell, Applied GeoSolutions, rbraswell@appliedgeosolutions.com
Nathan Torbick, Applied Geosolutions, ntorbick@appliedgeosolutions.com
Carmody McCalley, Rochester Institute of Technology, ckmsbi@rit.edu
Jia Deng, University of New Hampshire, dengjia85@gmail.com
Joanne Shorter, Aerodyne Research, Inc., shorter@aerodyne.com
Patrick Crill, Stockholm University, patrick.crill@geo.su.se
Christina Herrick, University of New Hampshire, herrick@eos.sr.unh.edu
L. Lamit, Syracuse University, ljlamit@mtu.edu

High latitude peatlands are a significant source of atmospheric methane. These sources are spatially and temporally heterogeneous, resulting in a wide range of global estimates for the atmospheric budget. At these high latitudes, increasing atmospheric temperatures are causing degradation of permafrost, creating changes in surface moisture, hydrology, vegetation and microbial communities resulting in dynamic changes to methane cycling. The temporal and spatial scale of disturbance from permafrost degradation varies depending on the transfer of heat and the hydrological connectivity of an ecosystem. The primary goal of our proposed work is to to combine remote sensing data and biogeochemical modeling to quantify methane emissions and isofluxes at the pan-arctic scale. We will accomplish this goal by addressing the following objectives: 1. Improve the ability of biogeochemical models to reliably estimate emissions of methane and 13CH4 from high latitude ecosystems by linking above and belowground processes through measurements and modeling, 2. Improve the estimate of water table and land cover using remote sensing techniques to be able to scale CH4 and 13CH4 emissions, and 3. Produce multi-scale scale maps of emissions and isofluxes and errors associated by using remote sensing (Landsat, MODIS, PALSAR-2, Sentinel-1, WorldView2, UAS, G-LIHT) to scale to the pan-Arctic region Using this combination of a validated biogeochemical process-based-model with ground verified multi-temporal and spatial remote sensing platforms, we will estimate the spatial distribution of methane emissions and its C isotopes across the high latitude peatland ecosystems. This project will quantitatively reduce uncertainties in the global methane budget related to these ecosystems and will allow us to link below and above ground processes on large spatial scales using cutting edge microbial, isotopic and remote sensing techniques.

34:

Surface-atmosphere Interactions in Canada’s Low Arctic

Elyn Humphreys, Carleton University, elyn_humphreys@carleton.ca
Peter Lafleur, Trent University, plafleur@trentu.ca

The low Arctic tundra in the eastern portion of the ABoVE study domain includes many surface types including lakes, ponds, wetlands, and tundra with varying soil depth and shrub cover. Observations of ecosystem-scale and plot-scale carbon dioxide, methane, energy and water exchanges have been ongoing since 2004 at the Daring Lake Tundra Ecosystem Research Station to characterize the spatial and temporal variability in surface-atmosphere interactions in this landscape. In 2017, we observed large seasonal variations in methane fluxes at a sedge fen and in net ecosystem exchange of CO2 and evapotranspiration at three tundra sites with varying shrub cover. In past growing seasons we found greater net uptake of CO2 with greater shrub cover but no concurrent increase in evapotranspiration. Snow depth increases with greater and taller shrub cover and along with moss cover, influences winter soil temperatures and active layer development. Greater shrub cover does not reduce late-winter albedo at our sites but does increase summer albedo and sensible heating of the atmosphere. Thus, our research reinforces existing knowledge of certain shrub tundra climate feedbacks and refutes others. Recent shoulder and winter season measurements of CO2 fluxes are being analyzed to help assess the annual carbon sink or source strength of these ecosystems.

50:

Mapping Cold Season Soil CO2 Emissions in the Arctic-Boreal Region

Jennifer Watts, Woods Hole Research Center, jwatts@whrc.org
Susan Natali, Woods Hole Research Center, snatali@whrc.org
Christina Minions, Woods Hole Research Center, cminions@whrc.org
Sarah Ludwig, Woods Hole Research Center, ludda.ludwig@gmail.com
Brendan Rogers, Woods Hole Research Center, brogers@whrc.org
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

The impact of ecosystem warming on the Arctic-boreal carbon budget remains highly uncertain. Carbon stored in thawing permafrost soils may be increasingly vulnerable to microbial mineralization and transfer to the atmosphere as CO2 given lessening cold temperature constraints. The magnitude of CO2 loss during winter and shoulder (autumn/spring) seasons could greatly reduce net ecosystem carbon sink in northern latitudes. This project aims to improve understanding of cold season soil CO2 emissions from boreal forests and tundra within Alaska and western Canada by incorporating new information gained from a network of Forced Diffusion (FD) flux sensors and tower eddy covariance observations. This combined database represents 16 FD sites and 23 tower locations across the Arctic Boreal Vulnerability Experiment (ABoVE) domain. We explore various statistical approaches and information from multi-scale (VIS-IR, microwave) satellite remote sensing and ancillary inputs to extrapolate the cold season fluxes to the greater region. Initial results from a generalized additive model (GAM) show the importance of soil carbon stocks, land surface temperature, landscape wetness (e.g. soil moisture, inundation), and permafrost status as predictors of soil CO2 flux. Our analysis also emphasizes the need for a new comprehensive vegetation map specific to the ABoVE domain that better characterizes ecosystem heterogeneity across tundra, boreal wetland and forest communities.

Associated Project: Natali-01

51:

Dissolved organic carbon export and its contribution to the carbon budget in a boreal peat landscape undergoing rapid permafrost thaw

Oliver Sonnentag, Université de Montréal, oliver.sonnentag@umontreal.ca
Julien Fouché, Université de Montréal, julien.fouche@tuta.io
Manuel Helbig, McGill University, mhelbig85@gmail.com
Karoline Lillie, Université de Montréal, karoline.wischnewski@gmail.com
Gabriel Hould Gosselin, Université de Montréal, ghgosselin@gmail.com
William Quinton, Wilfrid Laurier University, wquinton@wlu.ca
Tim Moore, McGill University, tim.moore@mcgill.ca

In boreal lowlands with warm and thin isolated permafrost, increasing temperatures cause a thaw-induced expansion of permafrost-free wetlands at the expense of forested permafrost peat plateaus. Permafrost thaw associated with warming may enhance decomposition of soil organic matter but also modify dissolved organic carbon (DOC) export to aquatic systems, which may play a non-negligible role for the carbon budget. We quantified the DOC export from a boreal peat landscape in the southern Northwest Territories (Canada) and estimate its contribution to the net ecosystem carbon balance. DOC export ranged from 2.5 to 4.0 g C m-2 from April to August, which accounted for ~51% (2014), 25% (2015) and 16% (2016) of the annual net ecosystem exchange. Our findings suggest that thawing boreal peat landscapes along the southern limit of permafrost are currently carbon neutral and increased vegetation productivity may turn the site into a weak sink.

52:

Understanding the resilience of arctic community assembly and soil C to fire disturbance and nutrient addition

Adrian Rocha, Univ of Notre Dame, arocha1@nd.edu
David Medvigy, Univ of Notre Dame, dmedvigy@nd.edu

Fires in the arctic are anticipated to increase in a future warmer world, yet remain poorly understood in terms of their impacts on ecosystem structure and function. The Anaktuvuk River fire was an unprecedented event that occurred in 2008 on the North Slope of Alaska. We present nearly a decade of post-fire recovery of Net Ecosystem Exchange of CO2 fluxes and ecosystem C stocks from Severely- and Un-burned sites located in close proximity to each other. NEE at the Severely burned site recovered within 4 years-aided by the rapid recovery and stimulation of vegetation productivity and a shift to shrub canopies. Soil C and N stocks remain depleted, and as a result, it is hypothesized that the stimulation of vegetation productivity is not sustainable. In 2016, we setup a multifactorial N, P and N+P nutrient addition experiment at the Severely burned and Unburned sites to further understand the role of nutrients in stimulating vegetation productivity and C sequestration. Five blocks of treatments have been fertilized at the beginning of each growing season, and a biomass harvest is planned for summer 2018. We will use these data to parameterize ED2-MEND, which is a vegetation demography model linked to biogeochemistry model of C, N, P. ED2-MEND will help to understand the development and post-fire C stock and flux recovery into the future and decrease the uncertainty regarding the impacts of tundra fires on climate.

Associated Project: Rocha-01

57:

Vetting atmospheric CO2 and CH4 from ecosystem models and satellites by comparison to ground-based observations

Nicole Jacobs, University of Alaska Fairbanks, najacobs@alaska.edu
William Simpson, University of Alaska Fairbanks, wrsimpson@alaska.edu
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Debra Wunch, University of Toronto, dwunch@atmosp.physics.utoronto.ca
Róisín Commane, Harvard University, rcommane@g.harvard.edu
Taylor Jones, Harvard University, taylorjones@g.harvard.edu
Steven Wofsy, Harvard University, wofsy@fas.harvard.edu
Harrison Parker, California Institute of Technology, hparker2@gmail.com
Manvendra Dubey, Los Alamos National Laboratory, dubey@lanl.gov
Thomas Blumenstock, Karlsruhe Institute of Technology, thomas.blumenstock@kit.edu
Frank Hase, Karlsruhe Institute of Technology, frank.hase@kit.edu
Qiansi Tu, Karlsruhe Institute of Technology, qiansi.tu@partner.kit.edu
Gregory Osterman, JPL, gregory.b.osterman@jpl.nasa.gov

The Boreal Forest and Northern wetlands exchange globally significant amounts of greenhouse gases and are undergoing rapid ecosystem change, including thawing of permafrost, motivating the need to study carbon gas exchange in the region. In August of 2016, we began the Arctic Mobile Infrared Greenhouse Gas Observations (AMIGGO) campaign in the region around Fairbanks, Alaska with the goal of satellite validation and measurement of natural ecosystem carbon exchange fluxes in the Boreal Forest. In this campaign, we used the EM27/SUN mobile solar-viewing Fourier-transform infrared spectrometer (EM27/SUN FTS) to retrieve column-averaged dry-air mole fractions of CO2 and CH4 (XCO2 and XCH4) with the GGG2014 algorithm. The EM27/SUN FTS was developed / designed by the Karlsruhe Institute of Technology (KIT) in cooperation with Bruker Optics and has been deployed in urban areas to measure anthropogenic fluxes of CO2, CH4, and CO (Gisi et al., 2012, doi:10.5194/amt-5-2969-2012, Hase et al., 2016, doi: 10.5194/amt-9-2303-2016). In the past year, applications for the EM27/SUN FTS have expanded beyond urban settings and they have been deployed by many groups for the purposes of cost-effective satellite validation and measurement of agricultural emissions across regional scales. These instruments show great potential for use in vetting model estimates of regional CH4 fluxes from Boreal forest and tundra ecosystems in Alaska and Canada and improving satellite observations of atmospheric CO2 and CH4 over the high latitude Boreal forest regions. These measurements can improve our ability to scale up fluxes from kilometer-scale eddy covariance observations to the regional-scale. We present observations of CH4 gradients during the fall of 2016 with two EM27/SUN FTS simultaneously deployed in Nenana and Fairbanks and comparisons of ground-based and satellite-based observations of XCO2 coordinated with OCO-2 target overpasses in 2016 and 2017. A comparison between XCO2 observations around Fairbanks and from the TCCON station in East Trout Lake, Saskatchewan will also be shown. We plan to continue CH4 gradient observations within the Boreal Forest / Northern wetlands in summer of 2018 and are seeking collaboration and support for an expanded project that would observe over larger regions of Alaska. Analysis of CH4 gradient observations also involves inverse modeling and vetting of forward model flux predictions from established ecosystem models such as CLM4.5-BGC. Both inverse and forward modeling interpretations will incorporate eddy covariance observations in Bonanza Creek LTER and eco-regions maps of Alaska constructed from data collected by other ABoVE projects. Once sound methods are developed for interpreting CH4 gradients amongst EM27/SUN FTS on 50-100km scales we intend to evaluate gradients over the entire North American Boreal Forest between Fairbanks and East Trout Lake.

Associated Project: Wunch-01

58:

Atmospheric Tomography (ATom) in the context of ABoVE

Róisín Commane, Harvard University, rcommane@g.harvard.edu
Atmospheric Tomography, Science Team, wofsy@g.harvard.edu

The Atmospheric Tomography Mission (ATom) is a NASA EVS Airborne project sampling profiles of trace gases and aerosols from the surface to the top of the troposphere throughout the Pacific and Atlantic basins. The project also sampled within the ABoVE domain. We will present an overview of the flights and trace gas and aerosol measurements made that can coordinate with ABoVE.

59:

Linking the carbon and water cycles through carbonyl sulfide

Róisín Commane, Harvard University, rcommane@g.harvard.edu
Ian Baker, Colorado State University, baker@atmos.colostate.edu
Stephen Montzka, NOAA, Stephen.A.Montzka@noaa.gov
Colm Sweeney, NOAA/ESRL GMD, Colm.Sweeney@noaa.gov
Debra Wunch, University of Toronto, dwunch@atmosp.physics.utoronto.ca

Fluxes of carbonyl sulfide (OCS) provide a way to link the carbon and water cycles in the terrestrial biosphere. OCS provides a quantitative, independent measure of biospheric activity, especially stomatal conductance, and allows us to partition photosynthesis and respiration on the ecosystem scale in Arctic and Boreal ecosystems. We can use OCS to better quantify the timing of gross photosynthesis in both spring and fall. It may also help us to quantify changes in photosynthetic uptake over time. In the mid-latitudes, vegetative uptake is the dominant sink of OCS (through hydrolysis by carbonic anhydrase), with a small but significant soil sink (Commane et al., 2015, PNAS). In one conifer forest in Finland, night-time uptake of OCS was almost 30% of the peak daytime uptake, with soil uptake contributing less than 15% of the overall flux (Kooijmans et al., 2017, ACP). There is a wealth of regional-scale OCS concentration measurements in the ABoVE domain. Regular NOAA vertical profile flights have measured OCS over key locations within the ABoVE domain, including East Trout Lake in northern Canada, the location of TCCON total column OCS (XOCS) measurements since July 2016. The combination of the diurnally resolved TCCON total column XOCS with the NOAA profiles will be a key measure of the regional OCS/CO2 ratio. However, there is limited information about the fluxes of OCS in arctic and boreal ecosystems with which to interpret these regional-scale observations. For example, field measurements from the 1988 ABLE 3A project suggested a small OCS sources from tundra wetlands. Unfortunately, the techniques used in this study were later found to be flawed (Hines and Durham, 1992, JGR). We can parameterize ecosystem processes based on studies from lower latitudes in a process-based ecosystem model (SiB), but, without on the ground measurements within the ABoVE domain, any interpretation will be subject to large uncertainties.

Associated Project: Wunch-01

74:

Effect of soil moisture on the temperature sensitivity of Northern soils

Christina Minions, Woods Hole Research Center, cminions@whrc.org
Susan Natali, Woods Hole Research Center, snatali@whrc.org
Sarah Ludwig, Woods Hole Research Center, ludda.ludwig@gmail.com
Dave Risk, St. Francis Xavier University, Canada, drisk@stfx.ca

Arctic and Boreal ecosystems are vast reservoirs of carbon, and are particularly sensitive to climate warming. Changes in the temperature and precipitation regimes of these regions could significantly alter soil respiration rates, impacting atmospheric concentrations and affecting climate change feedbacks. Many incubation studies have shown that both temperature and soil moisture are important environmental drivers of soil respiration; this relationship, however, has rarely been demonstrated with in situ data. Here we present the results of a study at six field sites in Alaska from 2016 to 2017. Low-power automated multiplexer soil gas systems were used to measure soil surface CO₂ flux from three forced diffusion chambers and soil profile concentrations from three soil depth chambers at hourly intervals at each site. HOBO Onset dataloggers were used to monitor soil moisture and temperature profiles. Temperature sensitivity (Q10) was determined at each site using inversion analysis applied over different time periods. With highly resolved data sets, we were able to observe the changes in soil respiration in response to changes in temperature and soil moisture. Through regression analysis we confirmed that temperature is the primary driver in soil respiration, but soil moisture becomes dominant beyond a certain threshold, suppressing CO₂ flux in soils with high moisture content. This field study supports the conclusions made from previous soil incubation studies, and provides valuable insights into the impact of both temperature and soil moisture changes on soil respiration.

Associated Project: Natali-01

75:

Increased methane loss due to later soil freezing in the Arctic may explain renewed atmospheric methane growth

Kyle Arndt, San Diego State University, kyleaarndt@gmail.com
Walter Oechel, San Diego State University, woechel@mail.sdsu.edu
Jordan Goodrich, Scripps Institute of Oceanography, jordan.p.goodrich@gmail.com
Aram Kalhori, San Diego State University, akalhori@mail.sdsu.edu
Josh Hashemi, San Diego State University, Joshhashemi@gmail.com
Colm Sweeney, NOAA/ESRL GMD, Colm.Sweeney@noaa.gov
Donatella Zona, San Diego State University (USA), dzona@mail.sdsu.edu

Over the last decade, the methane (CH₄) atmospheric concentration is increasing at the fastest rate since the last twenty years, and isotopic analyses show that this increase is due to increased emissions from ecosystems (Nisbet et al., 2016; Kirschke et al., 2013). Northern high latitudes wetlands have substantial potential in affecting atmospheric CH₄ concentration, given the extensive carbon stored in their soils (Hugelius et al., 2014), and their rate of warming double the global mean (Achberger et al., 2011). However, it remains difficult to predict or attribute changes in atmospheric CH₄ concentration to increased emissions from any specific ecosystems, given the sparse baseline CH₄ flux estimates available from wetlands, particularly from northern high latitudes. Here we show that soil freezing in the fall is occurring 1.4 days year^-1 later over the last decade across a variety of sites in the North Slope of Alaska, resulting in a longer “zero-curtain” (period when the soil temperature hovers around zero, Outcalt et al., 1990, Hinkel et al., 2001)). Methane emissions across 14 site-years from five sites in the North Slope of Alaska remained consistently higher during the zero curtain then after the complete soil freezing. This later soil freezing was associated with the fall atmospheric CH₄ enhancement observed across the North Slope of Alaska over the last 10 years. The consistently higher CH₄ emission rates before soil freeing, suggests that warming, and later soil freezing could explain some of the increased in the fall atmospheric CH₄ concentration.

Associated Project: Oechel-01

79:

Changes in boreal and arctic ecosystems productivity : deriving metrics of photosynthetic activity and limitation from MAIAC data.

Gabriel Hmimina, University of Lincoln Nebraska, hmimina@unl.edu
Rong Yu, University of Nebraska, rong.j.yu@gmail.com
Karl Huemmrich, NASA GSFC/UMBC, karl.f.huemmrich@nasa.gov
Dave Billesbach, University of Nebraska, dbillesbach1@unl.edu
John Gamon, University of Nebraska, jgamon@gmail.com

Large-scale greening and browning trends have been reported in northern terrestrial ecosystems over the last two decades. The greening has been interpreted as an increased productivity in response to increases in temperature. Boreal and arctic ecosystem productivity is expected to increase as the length of growing seasons increases, resulting in a bigger northern carbon sink. While evidences of such greening based on the use of remotely-sensed vegetation indices is compelling, analysis over the network of flux tower sites available in northern latitudes paints a more complex story, and raise some issues as to whether vegetation indices based on NIR reflectance at large spatial scales are suited to the analysis of complex northern landscapes that exhibit strong patterns in snow and standing water cover. In a broader sense, whether “greenness” is a sufficiently good proxy of ecosystem productivity in northern latitudes is unclear. The current work focused on deriving continuous estimates of ecosystem potential productivity and photosynthetic limitation over a network of flux towers, and on using MAIAC reflectance data to extrapolate those metrics over time and space. A novel partitioning method was used to derive ecophysiological parameters from sparse carbon fluxes measurements, and those parameters were then used to train a random forest approach in order to predict them from remote-sensing data.The resulting metrics can be used to estimate potential yearly productivity and loss of productivity due to photosynthesis limitation over a wide range of ecosystems. Trends in productivity were derived from both flux-tower and remote-sensing derived data and were deconvolved into changes in potential productivity and changes in limitation, showing contrasting spatial patterns of productivity increase in arctic ecosystems, decrease, and of imbalance between potential productivity and photosynthesis limitation in boreal ecosystems.

Associated Project: Gamon-01

90:

Quantifying Carbon Fluxes over the Alaska North Slope Using Eddy Covariance Flux Tower Observations and Machine Learning

Stephen Shirley, University of Montana, stephen.shirley@ntsg.umt.edu
Jennifer Watts, Woods Hole Research Center, jwatts@whrc.org
John Kimball, University of Montana, john.kimball@mso.umt.edu
Donatella Zona, San Diego State University (USA), dzona@mail.sdsu.edu
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Walter Oechel, San Diego State University, woechel@mail.sdsu.edu
Susan Natali, Woods Hole Research Center, snatali@whrc.org

Arctic permafrost regions contain over half of the global soil organic carbon pool. Due to climate warming in Arctic-boreal regions, this large store of carbon is vulnerable to microbial decomposition and loss to the atmosphere as greenhouse gases. While a warmer and wetter Arctic may increase vegetation growth and productivity, a drier climate could contribute to greater magnitudes of ecosystem respiration and methane oxidation, heightening the release of stored soil carbon to the atmosphere as carbon dioxide (CO2) and methane (CH4), and shifting the Arctic from a carbon sink to carbon source. Empirical and satellite data-driven models provide a useful method to quantify and monitor Arctic-boreal net ecosystem carbon budgets at coarse spatial scales, and are complimented by in situ eddy covariance flux tower measurements of ecosystem CO2 and CH4 flux at a less than 1 km scale. This presentation provides an update of ongoing efforts to model carbon dioxide and methane fluxes over the Alaska North Slope using machine learning random forest models trained with observations from eight eddy covariance flux towers located across northern Alaskan tundra. Random forest model inputs were determined based on their importance factor and include MERRA2 reanalysis soil moisture and temperature, snow mass, photosynthetic active radiation, and wind speed, as well as MODIS leaf area index (MCD15A2H), normalized difference vegetation index (NDVI; MOD13Q1/MYD13Q1), and daytime land surface temperature (LST; MOD11A2/MYD11A2). Eddy covariance tower sites have been categorized by wetland type using a fine resolution synthetic-aperture radar derived classification of Alaska wetlands. Here we present a site comparison of estimated daily CO2 and CH4 fluxes from the empirical random forest carbon flux model to eight in situ eddy covariance flux tower observations, along with Soil Moisture Active Passive Level 4 Carbon (SMAP L4C) and Terrestrial Carbon Flux (TCF) satellite data driven model outputs. Our results reveal shifts in the importance of different environmental controls influencing CO2 and CH4 fluxes both seasonally and between lowland and upland tundra sites. These results are being used to derive local scale (100-m resolution) carbon flux maps over northern Alaska for linking field, airborne and satellite based assessments and understanding of the regional carbon budget.

Associated Project(s):
Kimball-04
Oechel-01

92:

Arctic carbon flux estimated from the ED2 model: Preliminary results and model development

Erik Larson, Harvard University, erik_larson@fas.harvard.edu
Takeshi Ise, Kyoto University, kais@kyoto-u.ac.jp
J. Munger, Harvard University, jwmunger@seas.harvard.edu
Steven Wofsy, Harvard University, wofsy@fas.harvard.edu
Róisín Commane, Harvard University, rcommane@seas.harvard.edu
Donatella Zona, San Diego State University (USA), dzona@mail.sdsu.edu
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Paul Moorcroft, Harvard University, paul_moorcroft@harvard.edu

The Ecosystem Demography model v2.1 (ED2) is used to simulate several different arctic plant communities located at tower sites in the ABoVE domain. ED2 is a process based mechanistic model of the terrestrial biosphere. The model tracks carbon and nitrogen fluxes between the soil, living tissue and atmosphere. The goal of the study is to model the seasonal cycle of carbon (CO2 and CH4) uptake and emission in the Arctic and how this flux is affected by temperature. The ED2 model had not previously been run in the Arctic, and there are multiple avenues of model development underway to simulate realistic plant growth, including new Arctic specific plant functional types. Furthermore, we are adding dynamic soil carbon to the model that will allow for peat accumulation and reduction. We will discuss the current and planned model developments and present preliminary results including comparisons with observed data at the following tower sites: Barrow, Anaktuvuk River, Atqasuk, Imnavait Creek, Ivotuk, and Poker Flat Research Range.

Associated Project: Munger-04

97:

Monitoring the role of subsurface physical processes in carbon flux from a degraded permafrost site using novel multidisciplinary observations

Stephanie James, U.S. Geological Survey, Denver CO, sjames@usgs.gov
Burke Minsley, U.S. Geological Survey, Denver CO, bminsley@usgs.gov
Mark Waldrop, U.S. Geological Survey, Menlo Park CA, mwaldrop@usgs.gov

Understanding the impact of permafrost thaw and thermokarst formation on the global carbon budget remains incomplete due to the complex relationships between above and below ground processes, dramatic seasonal changes, and limitations in subsurface measurement capabilities. To address some of these uncertainties and better the understanding of the subsurface physical controls on carbon fluxes at the site scale, we will be conducting a multidisciplinary field experiment using novel combinations of geophysical and biogeochemical methods. This work will take place from 04/2018 to 07/2019 at the Alaskan Peatland Experiment (APEX) thermokarst research site near Fairbanks, AK. Passive seismic, nuclear magnetic resonance (NMR), temperature, soil moisture, and both surficial and subsurface carbon gas measurements will be recorded along a transect between forested permafrost and two collapse-scar bogs of different age. The unique combination of complementary geophysical methods (passive seismic + NMR) and in situ biogeochemical data will characterize the relationships between subsurface physical properties and surface carbon fluxes. Of particular interest is the potential role of unfrozen water within permafrost and frozen soils in sustaining microbial activity, which may be significant in enhancing carbon export but has been poorly characterized in situ. An array of seismic stations will continuously monitor subsurface freeze/thaw and water saturation changes along the thermokarst gradient and periodic borehole NMR profiles will provide direct measurements of unfrozen water content. These geophysical recordings will be collocated with in situ measurements of subsurface gas concentrations and temperature. Other key variables such as organic layer thickness, surface gas fluxes, and active layer thickness will also be measured at select locations. Initial work has begun using SUTRA-ICE to generate 1-D numerical simulations of water and energy transport for representative vertical profiles along the planned transect to provide a conceptual framework and help clarify the complex relationships between surface conditions and subsurface properties. We plan to integrate simulations with empirical data from the field experiment to aid interpretations, as well as combine with forward modeling of seismic velocity variations to quantify the sensitivity of the passive seismic technique to subtle changes in ice and water saturation. This multifaceted, interdisciplinary approach will serve as a proof-of-concept for the integration of geophysical and biogeochemical measurements for better understanding the impacts of thermokarst development on ecosystem-scale carbon fluxes in a lowland boreal forest.

Associated Project: James-01

100:

Carbon dioxide and methane flux analysis over ABoVE domain during the 2017 ArctiCAP airborne measurement campaign: methods and early work

Luke Schiferl, Harvard University, schiferl@seas.harvard.edu
Róisín Commane, Harvard University, rcommane@seas.harvard.edu
Stephen Conley, Scientific Aviation, Inc., sconley@scientificaviation.com
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Manuel Helbig, Université de Montréal, manuel.helbig@umontreal.ca
John Henderson, Atmospheric and Environmental Research, Inc., jhenders@aer.com
Elyn Humphreys, Carleton University, elynhumphreys@cunet.carleton.ca
Miriam Hurkuck, Université de Montréal, mhurkuck@wlu.ca
Kristina Luus, Dublin Institute of Technology, niinaluus@gmail.com
Philip Marsh, Wilfrid Laurier University, pmarsh@wlu.ca
Kathryn McKain, NOAA ESRL GMD, kathryn.mckain@noaa.gov
Walter Oechel, San Diego State University, woechel@mail.sdsu.edu
William Quinton, Wilfrid Laurier University, wquinton@wlu.ca
Oliver Sonnentag, Université de Montréal, oliver.sonnentag@umontreal.ca
Colm Sweeney, NOAA/ESRL GMD, colm.sweeney@noaa.gov
Sonja Wolter, NOAA ESRL GMD, sonja.wolter@noaa.gov
Donatella Zona, San Diego State University, dzona@mail.sdsu.edu
Steven Wofsy, Harvard University, swofsy@seas.harvard.edu

The Arctic Carbon Atmospheric Profiles (ArctiCAP) project successfully measured carbon dioxide (CO₂) and methane (CH₄) atmospheric concentration vertical profiles in the ABoVE domain during six flight campaigns throughout April-November 2017. Here we highlight an early look at the measured concentrations during ArctiCAP and present the tools which will be used to calculate regional fluxes of CO₂ and CH₄. Using methods developed during the Carbon in the Arctic Reservoirs Vulnerability Experiment (CARVE) (2012-2014), we will quantify ABoVE-domain CO₂ and CH₄ fluxes seasonally and spatially and identify the sources and dominant processes in this area. A back-trajectory atmospheric transport model (WRF-STILT) initialized from ArctiCAP observation points will allow us to identify source regions for ArctiCAP sampled data. We use meteorological reanalysis (NARR), remote sensing products such as solar-induced fluorescence (SIF) and snow cover area, and flux tower data along with vegetation maps as inputs to an updated Polar Vegetation Photosynthesis and Respiration Model with SIF (PVPRM-SIF), which will be used as the CO₂ flux prior. Finally, we will perform an inversion using the aircraft observations, source footprints, and flux prior to obtain optimized CO₂ and CH₄ fluxes in time and space. This will allow us to develop an annual budget for CO₂ and CH₄ throughout the ABoVE domain, improving upon our understanding and representation of the carbon system in arctic and boreal ecosystems.

Associated Project: Sweeney-01

110:

The Timing and Frequency of Rain on Snow Events in Alaska

Caleb Pan, University of Montana, caleb.pan@mso.umt.edu
Peter Kirchner, National Park Service, peter_kirchner@nps.gov
John Kimball, University of Montana, john.kimball@mso.umt.edu
Youngwook Kim, University of Montana, youngwook.kim@ntsg.umt.edu
Ulrich Kamp, University of Michigan, ulrich.kamp@mso.umt.edu

Rain-on-snow (ROS) affects many of the physical properties of a snow pack including energy content, depth, density and grain size. In northern latitudes ROS, wet snow and the icing events that follow, effect ecosystem processes at multiple scales including, wildlife movement, carbon cycling, human transportation, and hydrology. Here we present a new dataset for Alaska, USA that has been developed from satellite optical and passive microwave (PM) remote sensing. Our detection uses a spectral gradient ratio of calibrated AMSR-E and AMSR-2 36.5 GHz and 19 GHz brightness temperatures mapped to a 6-km resolution on a polar EASE-grid projection. Our classification includes snow cover extent (MOD10A2) and a DEM as additional ancillary inputs. Data were evaluated against in-situ observations of ROS events using a 2-tier validation process. Spatial and temporal trends of wet snow were evaluated over the recent satellite record (2002-2016) in relation to the regional climate divisions. These data demonstrate an accuracy of 85.9 % between observed ROS and wet snow PM returns with strong linkages between regional patterns and periods of above normal temperatures across Alaska.

Associated Project: Kimball-04

Crosscutting

47:

ABoVE Extension of the ArcticDEM

Elizabeth Hoy, NASA GSFC / Global Science and Technology, Inc., elizabeth.hoy@nasa.gov
Julien Peters, NASA GSFC / Patuxent Technology Partners, julien.b.peters@nasa.gov
Bruce Van Aartsen, NASA GSFC, bruce.vanaartsen@nasa.gov
Peter Griffith, NASA GSFC / SSAI, peter.c.griffith@nasa.gov
Claire Porter, Polar Geospatial Center, University of Minnesota, porte254@umn.edu
Paul Morin, University of Minnesota - Polar Geospatial Center, lpaul@umn.edu
Daniel Duffy, NASA GSFC, daniel.q.duffy@nasa.gov

The ArcticDEM is a digital surface model (DSM) of the Arctic being created through a partnership between the National Science Foundation (NSF) and the National Geospatial-Intelligence Agency (NGA). The project will culminate in the production of high-resolution optical stereo imagery DSMs for all land area north of 60°N. However, as the ABoVE domain extends south of 60°N, ABoVE is working in partnership with the NASA Center for Climate Simulation (NCCS) to extend the development of DSMs into the ABoVE domain south of 60°N using the same open-source software and high-performance computing techniques the ArcticDEM creators are developing. Using processing available through NASA’s Discover supercomputer, and petabytes of DigitalGlobe imagery including the panchromatic bands of the WorldView-1, WorldView-2, WorldView-3, and GeoEye-1 satellites, ABoVE is extending DSMs into regions of British Columbia, Alberta, Saskatchewan, and Manitoba. Completed DSM strips are currently available in the ABoVE Science Cloud, and plans are underway to share these strips with the broader community.

Associated Project: Morin-01

89:

GLOBE Observer: citizen science to support ABoVE research

Peder Nelson, Oregon State University, Peder.Nelson@oregonstate.edu
Holli Riebeek, NASA GSFC / SSAI, holli.a.riebeek@nasa.gov

GLOBE Observer is an international citizen science initiative to understand our global environment. Using a mobile app to capture location, photos ,and text, citizen scientists help track changes in clouds, water, plants, and other life. ABoVE research scientists can then use this data to verify satellite data and improve their models. Here, we will introduce the newest data collection protocol, ‘Land Cover’, and share observations from 'Cloud Observer' and 'Mosquito Habitat Mapper'. In addition, students of all ages are able to do real scientific research as part of the NASA-led Global Observations to Benefit the Environment (GLOBE) Program.

107:

Using GIS Services in the ABoVE Science Cloud

James Shute, NASA GSFC / Computer Science Corp, james.k.shute@nasa.gov
Laura Carriere, Computer Science Corp, laura.carriere@nasa.gov
Daniel Duffy, NASA GSFC, daniel.q.duffy@nasa.gov
Elizabeth Hoy, NASA GSFC / Global Science and Technology, Inc., elizabeth.hoy@nasa.gov
Julien Peters, NASA GSFC / Patuxent Technology Partners, julien.b.peters@nasa.gov
Yingshuo Shen, NASA GSFC, yingshuo.shen@nasa.gov

The NASA Center for Climate Simulation (NCCS) at the Goddard Space Flight Center built and is maintaining the ABoVE Science Cloud for its stakeholders, to include ABoVE scientists, industry partners, and the public. This geospatial platform includes three GIS subsystems operating in a highly-available, virtualized environment: 1. Spatial Analytics Platform: the primary NCCS GIS provides users discoverability of the vast DigitalGlobe/NGA raster assets within the NCCS environment; 2. Disaster Mapping Platform: supplies mapping and analytics services to NASA’s Disaster Response Group; and 3. Internal Enterprise GIS (Advanced Data Analytics Platform/ADAPT): includes the full suite of Esri and open source GIS software applications and services. All systems benefit from NCCS’s cutting-edge infrastructure, to include a mixed/heterogeneous environment featuring seamless sharing of information between Linux and Windows subsystems; in-depth system monitoring and warning systems; and the following data assets and applications: • NGA DigitalGlobe data: 2.8 Petabytes (PB) • Multiple data discovery tools for NGA DigitalGlobe data • Landsat data: 150 Terabytes (TB) • NCCS Dataportal CREATE-IP data: 52 Terabytes (TB) • NCCS Dataportal NEX-GDDP data: 11 Terabytes (TB) • Multiple Global and Regional Digital Elevation Model (DEM) datasets • MODIS TERRA and AQUA data: 5 Terabytes (TB) • Global Landslide Catalog Reporter and Viewer applications The vast ABoVE Science Cloud infrastructure is, at times, difficult to access and navigate. This poster and associated presentation will focus on the following key topics: 1. Streamlined access to the environment(s); 2. Locating and utilizing data archives; 3. Consuming existing GIS services and applications; 4. Capitalizing on infrastructure for deploying (and sharing) new GIS-related science products; 5. Known technical challenges; and 6. Methods for maintaining efficiency and effectiveness while utilizing the resources.

108:

The NASA Arctic-Boreal Vulnerability Experiment and the ABoVE Science Cloud: Accelerating Science with Cloud Technologies

Ellen Salmon, NCCS, NASA GSFC Code 606.2, ellen.salmon@nasa.gov
Laura Carriere, NASA GSFC/CSRA, laura.carriere@nasa.gov
James Shute, NASA GSFC/CSRA, james.k.shute@nasa.gov
Scott Sinno, NASA GSFC/PTP, scott.sinno@nasa.gov
Ben Bledsoe, NASA GSFC/PTP, benjamin.c.bledsoe@nasa.gov
Elizabeth Hoy, NASA GSFC / Global Science and Technology, Inc., elizabeth.hoy@nasa.gov
Peter Griffith, NASA GSFC / SSAI, peter.c.griffith@nasa.gov
John Thompson, NASA GSFC/PTP, john.h.thompson@nasa.gov
Garrison Vaughan, NASA GSFC/ITC, garrison.r.vaughan@nasa.gov
Julien Peters, NASA GSFC/PTP, julien.b.peters@nasa.gov
Daniel Duffy, NASA GSFC, daniel.q.duffy@nasa.gov

The NASA Center for Climate Simulation (NCCS – https://www.nccs.nasa.gov/) has partnered with the NASA Carbon Cycle and Ecosystems Office (CCEO) to provide a high performance science cloud to accelerate the pace of science for the ABoVE field campaign. Resources from the Advanced Data Analytics Platform (ADAPT – https://www.nccs.nasa.gov/services/adapt) have been reserved to create the ABoVE Science Cloud (ASC). The ASC combines high performance computing, storage, and purpose built virtual environments to create an environment specifically designed for the analysis of big data. Through this architecture, the ASC provides an agile environment that contributes to data integration, geospatial product generation, modeling, data stewardship and long-term data preservation by aiding researchers through the entire process of the data lifecycle. The ASC can provide one scientist with hundreds of virtual machines custom-configured to accelerate computation and visualization. It also serves as a Geographical Information System proximate to petascale high-resolution imagery. Furthermore, by using the ABoVE Science Cloud as a shared and centralized resource, researchers reduce costs for their proposed work, making proposed research more competitive. This poster will provide an overview of the ASC capabilities, tools and data sets available, and show examples of how the ASC is being used to meet the needs of the ABoVE campaign. In addition, this presentation will provide a demonstration for how to log into the system, access data, and run applications.

109:

Earth Data Analytics Service (EDAS) – Server Side Analytics at the NCCS

Laura Carriere, Computer Science Corp, laura.carriere@nasa.gov
Thomas Maxwell, CSRA, thomas.maxwell@nasa.gov
Daniel Duffy, GSFC, daniel.q.duffy@nasa.gov

As the availability and volume of Earth data grow, researchers spend more time downloading and processing their data than doing science. The NASA Center for Climate Simulation (NCCS) has developed the Earth Data Analytics Service (EDAS), a high-performance big data analytics framework built on Apache Spark, to allow researchers to leverage our compute power to analyze large datasets located at the NCCS through a web-based interface, thereby eliminating the need to download the data. EDAS provides access to a suite of “canonical operations”—min, max, sum, difference, average, root mean square, anomaly, and standard deviation— that researchers can combine to develop various workflows. EDAS uses a dynamic caching architecture, a custom framework, and a streaming parallel in-memory workflow for efficiently processing huge datasets within limited memory spaces at interactive response times. These operations and datasets can be accessed via a Web Processing Service (WPS) API using applications written by the user. This poster will present information on accessing EDAS as well as a variety of use cases for using EDAS to evaluate climate reanalysis data such as MERRA2. Examples provided will utilize Jupyter Notebooks.

DAAC

111:

From the Scientist to the Archive: Data Publication/Archive at the ORNL DAAC

Debjani Deb, Oak Ridge National Laboratory, debd@ornl.gov
Alison Boyer, Oak Ridge National Laboratory, boyerag@ornl.gov
Suresh Vannan, Oak Ridge National Laboratory, santhanavans@ornl.gov
Tammy Beaty, Oak Ridge National Laboratory, beatytw@ornl.gov

The ORNL DAAC archives and publishes data and information relevant to biogeochemical, ecological, and environmental processes primarily produced by NASA's Terrestrial Ecology Program. The data set submission process at the ORNL DAAC is managed by a semi-automated process to provide a consistent data provider experience and to create a uniform data product. The Semi Automated Data Publication System (SAuS) involves a formalized work flow, that makes data set submission easier for data providers and data publication faster and is being used to monitor the efficiency of the data publication workflows, and summarize how various aspects of the data publication process have affected the archive’s ability to expedite data publication.

112:

ORNL DAAC: NASA data archive for ABoVE field campaign

Suresh Vannan, Oak Ridge National Laboratory, santhanavans@ornl.gov
Alison Boyer, Oak Ridge National Laboratory, boyerag@ornl.gov
Debjani Deb, Oak Ridge National Laboratory, debd@ornl.gov

The Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) for biogeochemical dynamics is one of the NASA Earth Observing System Data and Information System (EOSDIS) data centers managed by the Earth Science Data and Information System (ESDIS) Project, which is responsible for providing scientific and other users access to data from NASA's Earth Science Missions. ORNL DAAC is operated by the ORNL Environmental Sciences Division and is responsible for data archival, product development and distribution, and user support for biogeochemical and ecological data and models. ORNL DAAC is the NASA designated archive for the ABoVE field campaign datasets.

Modeling

40:

Soil carbon residence time in the Arctic – model response to key environmental drivers

Deborah Huntzinger, Northern Arizona University, deborah.huntzinger@nau.edu
Joshua Fisher, NASA JPL, jbfisher@jpl.nasa.gov
Christopher Schwalm, Woods Hole Research Center, schwalm.christopher@gmail.com
Daniel Hayes, University of Maine, daniel.j.hayes@maine.edu
Eric Stofferahn, Jet Propulsion Laboratory / Caltech, ericstofferahn@gmail.com
Wouter Hantson, University of Maine, wouter.hantson@maine.edu
Kevin Schaefer, National Snow and Ice Data Center, kevin.schaefer@nsidc.org
Anna Michalak, Carnegie Institution Of Washington, michalak@stanford.edu
Yuanyuan Fang, Carnegie Institution for Science, yyfang@carnegiescience.edu
Yaxing Wei, Oak Ridge National Laboratory, weiy@ornl.gov

Carbon residence time is one of the most important factors controlling carbon cycling in ecosystems. Residence time depends on carbon allocation and conversion among various carbon pools and the rate of organic matter decomposition; all of which rely on environmental conditions, primarily temperature and soil moisture. As a result, residence time is an emergent property of models and a strong determinant of terrestrial carbon storage capacity. However, residence time is poorly constrained in process-based models due, in part, to the lack of data with which to benchmark global-scale models in order to guide model improvements and, ultimately, reduce uncertainty in model projections. Here we focus on improving the understanding of the drivers to observed and simulated carbon residence time in the Arctic-Boreal region (ABR). Carbon-cycling in the ABR represents one of the largest sources of uncertainty in historical and future projections of land-atmosphere carbon dynamics. This uncertainty is depicted in the large spread of terrestrial biospheric model (TBM) estimates of carbon flux and ecosystem carbon pool size in this region. Recent efforts, such as the Arctic-Boreal Vulnerability Experiment (ABoVE), have increased the availability of spatially explicit in-situ and remotely sensed carbon and ecosystem focused data products in the ABR. Together with simulations from Multi-scale Synthesis and Terrestrial Model Intercomparison Project (MsTMIP), we use these observations to evaluate the ability of models to capture soil carbon stocks and changes in the ABR. Specifically, we compare simulated versus observed soil carbon residence times in order to evaluate the functional response and sensitivity of modeled soil carbon stocks to changes in key environmental drivers. Understanding how simulated carbon residence time compares with observations and what drives these differences is critical for improving projections of changing carbon dynamics in the ABR and globally.

Associated Project: Fisher-01

48:

ArcticDEM Validation & Accuracy Assessment

Salvatore Candela, Ohio State University, candela.13@osu.edu
Ian Howat, Ohio State University, howat.4@osu.edu
Paul Morin, University of Minnesota - Polar Geospatial Center, lpaul@umn.edu
Myoung-Jong Noh, Ohio State University, noh.5@osu.edu
Claire Porter, Polar Geospatial Center, University of Minnesota, porte254@umn.edu
Elizabeth Hoy, NASA GSFC / Global Science and Technology, Inc., elizabeth.hoy@nasa.gov

The topography of the Arctic was among the most poorly mapped on Earth. Yet, the terrain of the Arctic is undergoing rapid changes making such data critical for both scientific investigations and infrastructure planning. The objective of ArcticDEM is to produce and openly distribute high resolution (2-5 m) Digital Surface Models (DSMs) of the entire Arctic landmass, including all areas above 60 degrees N, and all of Alaska, Greenland and the Kamchatka Peninsula. While the DSMs have an internal (pixel-to-pixel) accuracy of 0.2 m (Noh and Howat, 2015), the initial geolocation may have systematic offsets of 3-5 m in the vertical and horizontal resulting from sensor model errors. Both the strips and mosaics are registered to seasonally-subsetted and quality controlled ICESat-1 elevations due its density of coverage at high latitudes and high report accuracy (~10cm). ICESat, however, has a relatively coarse measurement footprint (~70 m) which may impact the precision of the registration. Further, the ICESat data predates the ArcticDEM imagery by a decade, so that temporal changes in the surface may also impact the registration. Finally, biases may exist between different the different sensors in the ArcticDEM constellation. Our objective is to use high-accuracy airborne LiDAR surveys conducted close in time to DSM acquisition to constrain the vertical accuracy of ArcticDEM strips over a range of terrains.

[poster]

Associated Project: Morin-01

66:

Does Your Terrestrial Model Capture Key Arctic-Boreal Relationships?: Functional Benchmarks in the ABoVE Model Benchmarking System

Eric Stofferahn, Jet Propulsion Laboratory / Caltech, ericstofferahn@gmail.com
Joshua Fisher, NASA JPL, jbfisher@jpl.nasa.gov
Daniel Hayes, University of Maine, daniel.j.hayes@maine.edu
Christopher Schwalm, Woods Hole Research Center, schwalm.christopher@gmail.com
Deborah Huntzinger, Northern Arizona University, deborah.huntzinger@nau.edu
Wouter Hantson, University of Maine, wouter.hantson@maine.edu

The Arctic-Boreal Region (ABR) is a major source of uncertainties for terrestrial biosphere model (TBM) simulations. These uncertainties are precipitated by a lack of observational data from the region, affecting the parameterizations of cold environment processes in the models. Addressing these uncertainties requires a coordinated effort of data collection and integration of the following key indicators of the ABR ecosystem: disturbance, vegetation / ecosystem structure and function, carbon pools and biogeochemistry, permafrost, and hydrology. We are continuing to develop the model-data integration framework for NASA’s Arctic Boreal Vulnerability Experiment (ABoVE), wherein data collection is driven by matching observations and model outputs to the ABoVE indicators via the ABoVE Grid and Projection. The data are used as reference datasets for a benchmarking system which evaluates TBM performance with respect to ABR processes. The benchmarking system utilizes two types of performance metrics to identify model strengths and weaknesses: standard metrics, based on the International Land Model Benchmarking (ILaMB) system, which relate a single observed variable to a single model output variable, and functional benchmarks, wherein the relationship of one variable to one or more variables (e.g. the dependence of vegetation structure on snow cover, the dependence of active layer thickness (ALT) on air temperature and snow cover) is ascertained in both observations and model outputs. This in turn provides guidance to model development teams for reducing uncertainties in TBM simulations of the ABR.

Associated Project: Fisher-01

82:

A Biogeochemical Forecasting System for Arctic Wetland Methane Emissions

Zhen Zhang, University of Maryland, yuisheng@gmail.com
Abhishek Chatterjee, NASA GSFC / USRA GESTAR, abhishek.chatterjee@nasa.gov
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Jeffrey Masek, NASA GSFC, Jeffrey.G.Masek@nasa.gov
Charles Miller, NASA JPL, charles.e.miller@jpl.nasa.gov
Lesley Ott, NASA GSFC, lesley.e.ott@nasa.gov
Colm Sweeney, NOAA/ESRL GMD, Colm.Sweeney@noaa.gov
Benjamin Poulter, NASA GSFC, benjamin.poulter@nasa.gov

Wetlands are the largest source of methane in the Arctic region, accounting for more than two-thirds of the total budget. Measurements of wetland methane fluxes in the Arctic are challenging given the large spatial extent and heterogeneity of sources and also because temporal dynamics are dependent on pulses of emissions, i.e., in the shoulder-seasons, that are difficult to capture. Here, we present a framework for providing short-term (one week to 3 months) forecasts of wetland methane emissions using LPJ as a prognostic land-surface model coupled within the Goddard Earth Observing System Model (GEOS-5) Earth system model. The LPJ biogeochemical model estimates daily wetland methane emissions by combining wetland area with soil microbial activity, accounting for local topography, soil moisture, soil temperature, and vertically discretized freeze-thaw dynamics. Meteorological forecasts are provided by the NASA Global Modeling and Assimilation Office and with a three-month projection and at 12.5 km spatial resolution. We demonstrate LPJ forecasts for wetland methane emissions for the ABOVE domain during the 2017 campaign and give examples on how the forecasts can be operationalized as an additional component of targeting in-situ field and airborne measurements (e.g., AVIRIS-NG and ArctiCAP), and to potentially reduce bias by covering a wider range of low to high-flux areas.

Associated Project: Chatterjee-01

Poster Session B

Wednesday 4:45 PM

Fire Disturbance

41:

Assessing Wildfire Effects in North American Boreal Peatlands through Field and Remote Sensing Analysis

Laura Bourgeau-Chavez, Michigan Tech Research Institute, lchavez@mtu.edu
Sarah Endres, Michigan Tech Research Institute (MTRI), slendres@mtu.edu
Jeremy Graham, Michigan Tech Research Institute, jeremyg@mtu.edu
Nancy French, Michigan Tech Research Institute (MTRI), nhfrench@mtu.edu
Michael Battaglia, Michigan Technological University, mjbattag@mtu.edu
Chelene Hanes, Canadian Forest Service, chelene.hanes@canada.ca
William de Groot, Canadian Forest Service, bill.degroot@nrcan-rncan.gc.ca
Jennifer Baltzer, Wilfrid Laurier University, jbaltzer@wlu.ca
Mike Flannigan, University of Alberta, mike.flannigan@ualberta.ca

Higher temperatures and reduced precipitation is leading to widespread seasonal drying in some arctic and boreal landscapes, thereby increasing wildfire frequency and severity. Research studies are needed to better understand the vulnerability of boreal ecosystems to a changing climate. In 2014, Northwest Territories (NWT) Canada had a record breaking year of wildfire, burning over 3.4 million hectares of upland forests, peatlands, and even emergent wetlands. Fire activity occurred across seasons (spring, summer, and fall) in the Taiga Shield and Taiga & Boreal Plains ecozones. We are using field and remote sensing studies to understand the vulnerability and resiliency of boreal ecosystems (with a focus on peatlands) to wildfire in NWT and Alberta. We are using field data to understand the relationship between burn severity, soil moisture and coniferous and deciduous tree recruitment at a fine scale and using remote sensing to understand landscape scale fire effects. Landsat and radar satellite imagery are being used to develop remote sensing algorithms specific to peatlands to map and monitor not only burn severity but also organic soil moisture, and peatland type (e.g. bog vs. fen). Field data analysis of tree recruitment, in situ moisture, burn severity, fuel loading and other biophysical parameters are currently being synthesized. Peatland maps, burn severity maps and initial analysis of tree recruitment data will be presented. The field and remote sensing data are being prepared for integration into CanFIRE (a carbon emissions and fire effects model). This spatial information allows for better quantification of the landscape heterogeneity of peatlands and fire effects, thus providing new insights to landscape scale changes and allowing improved understanding of the implications of increasing wildfire in boreal ecosystems.

Associated Project: Bourgeau-Chavez-03

54:

Workshop Outcomes: Opportunities to Apply Remote Sensing in Boreal/Arctic Wildfire Management and Science

Alison York, Alaska Fire Science Consortium, ayork@alaska.edu
Randi Jandt, Alaska Fire Science Consortium, rjandt@alaska.edu
Robert Ziel, Alaska Fire Science Consortium, rhziel@alaska.edu

With support from the NASA Applied Sciences Program, the Alaska Fire Science Consortium (AFSC; part of the International Arctic Research Center at UAF) organized an international workshop in April 2017 to advance the application of remote sensing tools and data by Alaska fire managers and scientists. The 85 workshop participants included regional fire and resource managers, with representation from state, federal, and Canadian agencies, as well as scientists with expertise in remote sensing and related disciplines, including several ABoVE-affiliated investigators. Topics of discussion at the workshop included: 1. Potential fire risk: Can remotely sensed data (e.g., daily snow extent, others) estimate spring soil moisture and surface and subsurface fuel moisture and fuel conditions, and thus provide critical inputs for fuel moisture indices used to predict fire danger and risk? 2. Near real-time fire behavior: Which remotely sensed data are best and most timely for fire detection, plume tracking of fire emissions, fire behavior modeling, mapping of flaming fronts, fire intensity, active fire perimeters, and response for ongoing fires? 3. Post-fire effects: Can we improve analytical methods for remotely sensed data to assess fire severity, consumption/CO2 balance, active-layer changes, and successional trajectories of high latitude vegetation communities? The outcomes of this workshop include: • increased understanding of current and potential uses and limitations of remote sensing data in the fire and resource management context in Alaska and Canada; • improved communication and coordination among agency personnel and the research and tech transfer communities, including several projects underway to explore new management and scientific uses of remote sensing data in high latitudes; and • input from Alaska and Canadian managers into ongoing user requirements and tech transfer efforts such as Geospatial Technology and Applications Center (USFS) and the Tactical Fire Remote Sensing Advisory Committee (joint NASA and USFS). Specific outcomes from the workshop can assist ABoVE investigators seeking to apply their data and products to improve the scientific foundations and operational efficiencies of Alaska fire and resource management and advance the use of remotely sensed data for fire management and fire science in high northern latitudes.

55:

Tundra Fire Accelerates De-frosting of America’s Icebox

Randi Jandt, University of Alaska, rjandt@alaska.edu
Eric Miller, Bureau of Land Mgmt, Alaska Fire Service, eamiller@blm.gov
Carson Baughman, USGS, cbaughman@usgs.gov
Ben Jones, USGS, bjones@usgs.gov
Go Iwahana, University of Alaska Fairbanks, giwahana@alaska.edu

Can fire accelerate the changes in the arctic that climate is already inducing and could a single fire event trigger a threshold change in arctic vegetation communities, with far-reaching implications? Ten years following a large and severe wildfire in the arctic foothills north of the Brooks Range, Alaska, tundra is experiencing rapid biophysical changes. The Anaktuvuk River Fire burned about 104,000 ha in 2007, spreading over broad ranges in parent soils, topography, hydrography, and permafrost features. Plant communities are responding to primary disturbance by fire but also to permafrost degradation, terrain subsidence, and apparent increase in soil drainage or evapotranspiration. Re-burns were documented within the fire area (in an ecotone where fire return intervals are estimated close to 1,000 years) and large increases in biomass (fuels) may be contributory. In order to track the diverse landscape scale changes occurring on large disturbances, collaboration with remote sensing scientists is needed to scale up field data collected by agencies both spatially and temporally. Application of remote sensing technology as part of the future monitoring is also desirable due to its cost effectiveness in a time of shrinking agency budgets. Key questions include: what are changes in fuelbed height and moisture content? How significant are changes in surface roughness and topology (due to subsidence) and do they correlate with burn severity? How does surface and subsurface layer moisture content change? Is snow depth altered in burn scars due to surface and/or vegetation change (impacting wildlife, subsistence and permafrost)? Several (listed) ABoVE investigations may be collaborative.

[poster]

96:

Drivers of post-fire albedo across Alaska and Canada: implications for climate feedbacks

Stefano Potter, Woods Hole Research Center, spotter@whrc.org
Kylen Solvik, Woods Hole Research Center, ksolvik@whrc.org
Angela Erb, University of Massachusetts Boston, Angela.Erb001@umb.edu
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu
Jill Johnstone, University of Saskatchewan, jill.johnstone@usask.ca
Michelle Mack, Northern Arizona University, michelle.mack@nau.edu
James Randerson, University Of California, Irvine, jranders@uci.edu
Miguel Román, NASA Goddard Space Flight Center, Miguel.O.Roman@nasa.gov
Crystal Schaaf, University of Massachusetts Boston, Crystal.Schaaf@umb.edu
Merritt Turetsky, University of Guelph, mrt@uoguelph.ca
Sander Veraverbeke, Vrije Universiteit, Amsterdam, The Netherlands, sveraver@uci.edu
Zhuosen Wang, NASA GSFC, zhuosen.wang@nasa.gov
Brendan Rogers, Woods Hole Research Center, brogers@whrc.org

Understanding how altered fire regimes impact vegetation composition and energy budgets is critical to forecasting regional and global climate change. High-severity fires cause winter and spring albedo to increase due to increased snow exposure and replacement of evergreen conifers by deciduous broadleaf trees. Although summer albedo decreases initially due to the deposition of black carbon and charred surfaces, it typically increases for several decades thereafter when younger and brighter deciduous trees dominate. The net effect of these albedo changes is expected to result in substantive radiative cooling, but there has been little research to examine how albedo trajectories differ spatially and temporally as a result of differences in topography, climate and soil properties, and what the associated implications for future energy balances are. Here we investigate drivers of post-fire monthly albedo trajectories across Canada and Alaska using a new Collection V006 500 m MODIS daily blue-sky albedo product and historical fires from the Canadian and Alaskan National Fire Databases. The impacts of landscape position, soils, climate, time since fire, and ecoregion on monthly albedo trajectories are explored using a Random Forest model. The resulting monthly models are then used to predict long term albedo under the Representative Concentration Pathway (RCP) 4.5 and 8.5 scenarios. Models show that temperature and ecoregion are the most influential drivers of post fire albedo in the spring and autumn. In the summer ecoregion is again important, but year since fire and soil texture are more important than climate variables. In the future, the model predicts that as climate change intensifies, spring albedo will decrease due mainly to declining snow cover and warmer temperatures. In the summer this trend is reversed, although variation between future scenarios is small. These results indicate that there will be diminished post fire radiative cooling in the future. Future steps are to quantify how the radiative forcing will change under future climate, and to integrate albedo projections into a comprehensive fire-forcing framework that also considers biogeochemical feedbacks.

Associated Project: Rogers-01

113:

Mapping fractional coverage of major fuel types for wildland fire research in Alaskan tussock tundra

Jiaying He, University of Maryland, hejiaying0608@gmail.com
Tatiana Loboda, University of Maryland, loboda@umd.edu
Liza Jenkins, Michigan Tech Research Institute (MTRI), lliverse@mtu.edu
Dong Chen, University of Maryland, itscd@umd.edu

Wildland fires can cause dramatic changes in vegetation, soil and water properties of ecosystems in a short period of time, leading to strong impacts on ecosystem functions and services. Though comparatively less frequent, wildfires in Alaskan tundra are a major disturbance agent with significant impacts on climate through carbon cycle and surface albedo. Tundra fires could not only release vast underground carbon from permafrost thawing, but also introduce shrub expansion, which affects the regional surface energy balance and carbon storage capability. Additionally, recent studies indicate that climate warming has the potential to increase future wildfire frequency and burned area in the tundra. Hence global warming could further enhance tundra fires' impacts on the local ecosystems and climate change. However, better understanding of factors and mechanisms governing wildland fires in the Alaskan tundra is needed to improve our predictive capabilities. As a major components of the fire environmental triangle, fuel conditions have been commonly considered as significant factor affecting fire ignitions and spread. Paleorecords suggest that change of vegetation type could play an important role in tundra fires. To enhance the understanding of tundra fire activities, this study aims to explore the spatial distribution of major fuel types in Alaskan tundra. We develop continuous fractional coverage maps of shrub, sedge and grass in multiple years with Level-2 Landsat 8 Operational Land Imager (OLI) multispectral surface reflectance data, field observations collected in the Noatak River Basin and the Seward Peninsula in our ABoVE project and other remote sensing products. Using spectral bands and metrics calculated with Landsat 8 data, we apply Random Forest algorithm to develop models for the fractional coverage of major fuel types with 70~80% accuracies. We further map the spatial distribution of these fuel types in Alaskan tussock tundra with Landsat 8 data. This result will function as important input data for future research of wildfire behaviors in the Alaskan tundra.

Associated Project: Loboda-03

116:

Legacy carbon combustion in boreal black spruce forests of the Northwest Territories, Canada

Xanthe Walker, Northern Arizona University, xanthe.walker@nau.edu

Associated Project: Mack-01

Vegetation Dynamics and Distribution

29:

Airborne Solar-Induced Chlorophyll Fluorescence to Characterize Arctic Boreal Zone Productivity

Darren Drewry, JPL, ddrewry@jpl.nasa.gov
Charles Miller, NASA JPL, charles.e.miller@jpl.nasa.gov
David Schimel, JPL, david.schimel@jpl.nasa.gov
Oliver Sonnentag, Université de Montréal, oliver.sonnentag@gmail.com
Adrian Rocha, Univ of Notre Dame, arocha1@nd.edu
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Syndonia Bret-Harte, University of Alaska, Fairbanks, msbretharte@alaska.edu
Ryan Pavlick, NASA JPL, rpavlick@jpl.nasa.gov

Recent demonstrations of the retrieval of vegetation solar-induced fluorescence (SIF) emission from satellite platforms have opened up the possibility of remotely monitoring photosynthetic function, in addition to the structural and biochemical parameters that characterize the current capabilities of vegetation observing systems. The Chlorophyll Fluorescence Imaging Spectrometer (CFIS) was recently developed for OCO-2 validation purposes and provides an airborne capability to help fill the spatial gap between leaf- or canopy-level observations of SIF flux and extensive satellite footprints. CFIS is a high-resolution (<0.1nm) spectrometer covering the 740-770 nm wavelength range, optimized for SIF quantification. The flexibility of an airborne instrument allows for studies of the temporal variability of SIF emission over consecutive days, or with meteorological variability throughout a day. Here we present an overview of the instrument design and capabilities, along with the retrieval methodology. We discuss the CFIS airborne campaign conducted across Northwestern Canada and Alaska as part of the 2017 Arctic-Boreal Vulnerability Experiment (ABoVE). This campaign provided an opportunity to acquire data over a variety of northern latitude ecosystems. We provide an overview of near- and long-term plans for analysis in the context of field data and multi-sensor synthesis.

Associated Project: Drewry-01

30:

Identifying forest patterns across the circumpolar taiga-tundra ecotone: linking multi-scale estimates of structure to explore biome boundary dynamics

Paul Montesano, NASA/SSAI, paul.m.montesano@nasa.gov
Christopher Neigh, NASA GSFC, christopher.s.neigh@nasa.gov
Min Feng, GLCF, fengm@umd.edu
Joseph Sextion, GLCF, jsexton@umd.edu
Saurabh Channan, GLCF, schannan@umd.edu
William Wagner, NASA GSFC - SSAI, william.c.wagner@nasa.gov
Margaret Wooten, NASA GSFC - SSAI, margaret.wooten@nasa.gov
Benjamin Poulter, NASA GSFC, benjamin.poulter@nasa.gov

In the northernmost portion of the circumpolar boreal, the taiga-tundra ecotone (TTE), the variety of forest patterns are the result of interactions between broad-scale climate and local-scale site factors. Patterns of forest extent, height, and cover help describe forest structure spatial transitions that influence future and reflect past dynamics. Ground and airborne observations of structure provide important calibration for tuning spaceborne estimates of forest height and cover. These spaceborne estimates reveal patterns at a range of scales, including patches with similar forest structure properties and the spatial rates of change of these properties. They provide both site-level samples of height and continuous cover estimates that span the circumpolar domain. Continuous domain-wide structure estimates are important for consistent modelling of circumpolar biome boundary dynamics. We highlight (1) our integration of ground and airborne observations to calibrate forest structure estimates; (2) use of sub-meter spaceborne stereogrammetry for sampling structure along the taiga-tundra ecotone; (3) Landsat-derived patterns of tree cover and its spatial rate of change across the circumpolar ecotone; and (4) how we’re linking remote sensing structure estimates to the modelling of circumpolar dynamics such as the variability in boreal forest age structure & carbon fluxes, and the snow-albedo feedback to climate.

Associated Project(s):
Neigh-01
Ranson-02

31:

Estimating vegetation height from WorldView-02 and ArcticDEM data for broad ecological applications

Arjan Meddens, University of Idaho, ameddens@uidaho.edu
Lee Vierling, University of Idaho, leev@uidaho.edu
Jan Eitel, University of Idaho, jeitel@uidaho.edu
Jyoti Jennewein, University of Idaho, jjennewein@uidaho.edu
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Joanne White, Canadian Forest Service, joanne.white@canada.ca
Michael Wulder, Canadian Forest Service, mike.wulder@canada.ca

Boreal and arctic regions are warming at an unprecedented rate, and at a rate higher than in other regions across the globe. Ecological processes are highly responsive to temperature and therefore substantial changes in these northern ecosystems are expected. Recently, NASA initiated the Arctic-Boreal Vulnerability Experiment (ABoVE), which is a large-scale field campaign that aims to gain a better understanding of how the arctic responds to environmental change. High-resolution data products that quantify vegetation structure and function will improve efforts to assess these environmental change impacts. Our objective was to develop and test an approach that allows for mapping vegetation height at a 5m grid cell resolution across the ABoVE domain. To accomplish this, we selected three study areas across a north-south gradient in Alaska, representing an area of approximately 130 km2. We developed a RandomForest modeling approach for predicting vegetation height using the ArcticDEM (a digital surface model produced across the Arctic by the Polar Geospatial Center) and high-resolution multispectral satellite data (WorldView-2) in conjunction with aerial lidar data for calibration and validation. Vegetation height was successfully predicted across the three study areas and evaluated using an independent dataset, with R2 ranging from 0.58 to 0.76 and RMSEs ranging from 2.2 to 2.5 m. This predicted vegetation height dataset also led to the development of a digital terrain model using the ArcticDEM digital surface model by removing canopy heights from the surface heights. Our results show potential to establish a high resolution pan-arctic vegetation height map, which will provide useful information to a broad range of ongoing and future ecological research in high northern latitudes.

Associated Project(s):
Boelman-01
Eitel-01
Vierling-01

33:

Detecting early warning signals of tree mortality in the ABoVE domain using multi-scale satellite data

Brendan Rogers, Woods Hole Research Center, brogers@whrc.org
Kylen Solvik, Woods Hole Research Center, ksolvik@whrc.org
Ted Hogg, Canadian Forest Service, ted.hogg@canada.ca
Junchang Ju, NASA Goddard Space Flight Center, junchang.ju@nasa.gov
Jeffrey Masek, NASA GSFC, Jeffrey.G.Masek@nasa.gov
Michael Michaelian, Canadian Forest Service, michael.michaelian@canada.ca
Logan Berner, Northern Arizona University, logan.berner@nau.edu
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

Increasing tree mortality from global change drivers such as drought and biotic infestations is a widespread phenomenon, including in the ABoVE domain where climate changes and feedbacks to the Earth system are relatively large. Despite the importance for science and management communities, our ability to forecast tree mortality at landscape to continental scales is limited. However, two independent information streams have the potential to inform and improve mortality forecasts: repeat forest inventories and satellite remote sensing. Time series of tree-level growth patterns indicate that productivity declines and related temporal dynamics often precede mortality years to decades before death. Plot-level productivity, in turn, has been related to satellite-based indices such as the Normalized Difference Vegetation Index (NDVI). Here we link these two data sources to show that early warning signals of mortality are evident in several NDVI-based metrics up to 24 years before death. We focus on two repeat forest inventories and three NDVI products across the ABoVE domain where productivity and mortality dynamics are influenced by periodic drought. These data sources capture a range of forest conditions and spatial resolution to highlight the sensitivity and limitations of our approach. Overall, results indicate potential to use satellite NDVI for early warning signals of mortality. Relationships are broadly consistent across inventories, species, and spatial resolutions, although coarse-scale imagery in the heterogeneous aspen parkland was of limited utility. Longer-term NDVI data and annually re-measured sites with high mortality levels generate the strongest signals, although we still found robust relationships at sites re-measured at a typical five-year frequency. The approach and relationships developed here can be used as a basis for improving forest mortality models and monitoring systems.

Associated Project: Goetz-03

35:

Describing Tundra Vegetation Cover Using Spectral Data

Karl Huemmrich, NASA GSFC/UMBC, karl.f.huemmrich@nasa.gov
Sergio Vargas, University of Texas at El Paso, savargas@utep.edu
Petya Campbell, JCET/UMBC, petya.campbell@nasa.gov
Craig Tweedie, University of Texas at El Paso, ctweedie@utep.edu
John Gamon, University of Nebraska, jgamon@gmail.com

Non-vascular plants (lichens and mosses) are significant components of tundra landscapes and may respond to climate change differently from vascular plants affecting ecosystem function. Remote sensing provides critical tools for monitoring plant cover types, as optical signals provide a way to scale from plot measurements to regional estimates of biophysical properties, for which spatial-temporal patterns may be analyzed. Partial Least Squared Regression is applied to ground measured spectral reflectance to develop algorithms to estimate vegetation functional type coverage, along with other biophysical characteristics, such as chlorophyll concentration. These algorithms are then used with AVIRIS NG imagery to map tundra variation across regions of the North Slope of Alaska.

Associated Project(s):
Gamon-01
Huemmrich-01

36:

Fingerprinting Three Decades of Changes in Interior Alaska (1982-2014) Using Field Measurements, Stereo Air Photos, and G-LiHT Data

Bruce Cook, NASA GSFC, bruce.cook@nasa.gov
Hans Andersen, USDA Forest Service, handersen@fs.fed.us
Douglas Morton, NASA GSFC, douglas.morton@nasa.gov
Michael Alonzo, NASA GSFC, michael.g.alonzo@nasa.gov
Anika Halota, NASA GSFC, anika.halota@nasa.gov

Associated Project: Cook-B-02

37:

Shifting Patterns of Boreal Forest Succession and Browning Over the Last 30 Years

Michael Goulden, University Of California, Irvine, mgoulden@uci.edu
Claudia Czimczik, University Of California, Irvine, czimczik@uci.edu
James Randerson, University Of California, Irvine, jranders@uci.edu

Climate and fire largely control the productivity (“greenness”) and biodiversity of boreal forests in North America. Our research focuses on better understanding: 1) the patterns of, controls on, and recent changes in North American Boreal Forest “Browning" and the declining Normalized Difference Vegetation Index (NDVI) observed in satellite records, and 2) the patterns of, controls on, and recent changes in North American Boreal Forest fire recovery and succession. Much of our effort has used the Landsat archive to analyze the patterns of wildfire and forest recovery along a transect cutting across central Canada; this study areas covers 3 Landsat rows x 25 paths with 2500 summer images. Key findings include: 1) Most (80-90%) of the recent NDVI trends in our study area are attributable to wildfire (areas that burned after ~1995 and also before ~1975 show browning; areas that burned in ~1975-1995 show greening). 2) There are a significant number of non-fire related patches that show either browning or greening; some of these patches are related to fires or human disturbances that aren't in our disturbance database, but others occur in wetter areas, where there is a general tendency toward browning with many specific cases of greening. 3) Various remote sensing metrics yield complementary information providing a clearer sense of the biophysical trends during succession. 4) We see evidence of accelerating succession from 1985-1995 to 2005-2015. This acceleration isn't dramatic, just 1-3 years during early recovery and more during later succession, but it is a consistent feature of the analysis. We are not seeing a systematic decline in old-stand LAI. While NDVI declines in old stands with the loss of deciduous trees, we are not seeing a systematic decrease in old stand LAI or wide spread mortality.

Associated Project: Goulden-02

38:

Datasets of Disturbance, Phenology, Peak Summer Greenness and Land Cover for Landsat for ABoVE

Curtis Woodcock, Boston University, curtis@bu.edu
Mark Friedl, Boston University, friedl@bu.edu
Damien Sulla-Menashe, Boston University, sullamenashe@gmail.com
Eli Melaas, Boston University, emelaas@bu.edu
Jonathan Wang, Boston University, jonwang@bu.edu
Letitia Lee, Boston University, leticial@bu.edu

The purpose of our poster is to update the ABoVE Science Team on our efforts to provide datasets on disturbance (date and magnitude), phenology (onset and senescence), peak summer greenness, and land cover. All of these datasets are derived primarily from the entire Landsat archive of TM and ETM+ data for the region at 30m spatial resolution. The first major step was to “tile” the archive so that each available observation for each pixel could be used in processing. A second major step has been to run the Continuous Change Detection and Classification (CCDC) algorithm for the entire region. To date, the change detection component, which is the most computationally intensive, has been completed for most of the ABoVE study area. Production of the disturbance datasets will be based on the results from CCDC and their final formulation remains under development. Development of the peak summer greenness, phenology and land cover products also depend on CCDC results, and are being developed for implementation upon completion of the disturbance datasets. We will show maps of progress to date, sample products and descriptions of the algorithms being used.

Associated Project: Woodcock-02

39:

Arctic tundra greening and browning by continent and latitudinal subzone

Howard Epstein, University of Virginia, hee2b@virginia.edu
Leah Reichle, University of Virginia, lmr8ws@virginia.edu
Coleman Dickerson, University of Virginia, csd7yw@virginia.edu
Uma Bhatt, University of Alaska, Fairbanks, usbhatt@alaska.edu
Donald Walker, University of Alaska, Fairbanks, dawalker@alaska.edu
Martha Raynolds, University of Alaska, Fairbanks, mkraynolds@alaska.edu

Until recently the scientific literature has seen an abundance of papers describing the “greening” of the Arctic; from a remote sensing perspective this has meant an increase in the Normalized Difference Vegetation Index (NDVI), or a similar satellite-based index. More recently, there have been more widespread observations of tundra “browning.” Here, we use a circumpolar remote sensing dataset to evaluate the spatio-temporal patterns of arctic tundra vegetation dynamics (greening and browning), and its control by summer warmth, at the circumpolar, continental (North America, Eurasia), and tundra subzonal (i.e. latitude) scales over the past 35 years. Significant warming trends, significant greening trends, and significant inter-annual relationships between NDVI and summer warmth were not spatio-temporally consistent. Significant warming trends tended to occur further north, whereas significant greening trends tended to occur more in the southern tundra. Some significant browning trends were observed in the northern tundra subzones. Significant relationships between NDVI and SWI were more likely found in the middle tundra latitudes. Over the satellite record, the number of years of greening was similar to the number of years of browning, with the exception of the most southern tundra subzone (Subzone E). The spatio-temporal dynamics of tundra vegetation and the controls on greening and browning appear to be highly complex and in need of continued study.

Associated Project: Epstein-01

43:

A New Version of CANAPI for Mapping Changes in Tall Shrub Canopies in Arctic Tundra

Mark Chopping, Montclair State University, choppingm@mail.montclair.edu
Rocio Duchesne-Onoro, University of Wisconsin - Whitewater, duchesnr@uww.edu
Angela Erb, University of Massachusetts Boston, Angela.Erb001@umb.edu
Zhuosen Wang, ESSIC, University of Maryland, zhuosen.wang@nasa.gov
Crystal Schaaf, University of Massachusetts Boston, Crystal.Schaaf@umb.edu
Christopher Chopping, West Orange High School, tchopp00@gmail.com

A new version of the Canopy ANalysis with Panchromatic Imagery (CANAPI) code was developed following tall shrub mapping tests with WorldView-2 imagery over Alaskan Arctic tundra. CANAPI results have some dependence on user-determined settings, so the project team performed iterative tests on the impact of analyst subjectivity by having four team members plus one naïve user perform multiple runs with different settings (each time attempting to find an optimal set) and subsequently labeling the set that produced the subjectively "best" result. Suitable test imagery was located using SQL queries in ArcGIS under the ABoVE Science Cloud Windows VM. QuickBird (QB02) panchromatic and multispectral imagery from June 20, 2003 and WorldView-2 (WV02) panchromatic and multispectral imagery from July 14, 2015, and a smaller QuickBird image subset previously used were used in a series of CANAPI runs. Only tall shrub canopy measurement results from the "best" CANAPI runs from each user were considered. The relative uncertainty in the estimates of mean crown radius was lowest at 4.3%, 3.2%, and 4.5% for the test, QB02, and WV02 image subsets, respectively, while the corresponding values for %tall shrub cover are 25.6%, 58.0%, and 35.2%; and for mean shrub height 25.2%, 24.7%, and 30.5%. The estimated absolute and % changes over 2003–2015 (again considering only the "best" CANAPI runs) showed increases in the number of crowns detected, though there was a wide disparity from user to user: from 122% through 630% (though the latter was from the naïve user). Changes in mean crown radius were far less variable but showed changes in both directions, from -5% through +8%. Changes in estimated tall shrub cover were highly divergent, ranging from 7% to 105% (ignoring the 349% result from the naïve user), while changes in mean shrub height varied from no change through 129% (again, ignoring the result from the naïve user). Visual inspection of the results for the QB02 and WV02 imagery from the most experienced user (Duchesne) indicated unambiguous increases in shrub number, size, cover, and height over the 2003-2015 period, consistent with the quantitative results (increase of 196%, 5%, 59%, and 15%, respectively). These values are outside the uncertainty range, with the exception of the change estimate for mean height (15% vs 25% and 31% uncertainty for QB02 and WV02, respectively). In order to reduce the impact of analyst subjectivity and improve the results, a new version of the code was developed to exploit the multispectral bands as well as the panchromatic imagery (Canopy ANalysis with Multispectral and Panchromatic Imagery, CANAMPI). To this end, orthorectification was performed for the WV02 July 14, 2015 scene on the NCCS ADAPT linux VM using the Polar Geospatial Center ortho processing and the Alaska NED mosaic digital elevation model (approximate example syntax: python pgc_ortho.py --epsg 32606 --resolution 0.5 --dem /DEM/alaskaned_mosaic_wgs84.tif --format GTiff --stretch ns --outtype UInt16 ~/nrIvishak_River/ /att/nobackup/ascuser/orthoout/). This produced accurately-geolocated panchromatic and multispectral image files on a 0.5 m grid; a Normalized Difference Vegetation Index image was also produced using the NIR and Red bands. Tests were performed on CANAPI-like code that first identifies candidate tall shrubs in the usual way (i.e., by locating crescent-shaped areas of bright pixels arising from shrub crown illumination), then evaluates the mean NDVI of the pixels of each candidate crown, and finally flags objects unlikely to be shrubs using a fixed threshold value. Although this does improve accuracy by removing false positives, it was found that some areas with tall shrubs have unexpectedly low NDVI values, potentially limiting the utility of this approach. An alternative strategy that uses image spectral information will be investigated: spectral vectors may be extracted automatically from crowns initially identified using the panchromatic imagery and a distance metric (e.g., Mahalonobis distance) subsequently used as the criterion.

[poster]

Associated Project: Chopping-03

45:

Shrub sensitivity to recent warming across Arctic Alaska from dendrochronological and remote sensing records.

Laia Andreu-Hayles, LDEO, lah@ldeo.columbia.edu
Ben Gaglioti, Lamont Doherty Earth Observatory, rdd@ldeo.columbia.edu
Rosanne D'Arrigo, Lamont Doherty Earth Observatory, rdd@ldeo.columbia.edu
Kevin Anchukaitis, University of Arizona, kanchukaitis@email.arizona.edu
Scott Goetz, NAU, scott.goetz@nau.edu

Shrub expansion into Arctic and alpine tundra ecosystems has been documented during the last several decades based on repeat aerial photography, remote sensing, and ground-truthed estimates of vegetation cover. Today, summer temperatures limit the northern limit of Arctic shrubs, and warmer summers have been shown to have higher NDVI in shrub tundra zones. Although global warming has been considered the main driver of shrub expansion, soil types, shrub species and non-linear responses can moderate how sensitive shrub growth is to climate warming. Here, we assess the sensitivity of shrub growth to inter-annual climate variability using a newly generated network of 18 shrub ring-width chronologies in the tundra regions of the North Slope of Alaska. We then test whether the dendroclimatic patterns we observe at individual sites are representative of the broader region using remotely sensed productivity data (NDVI). The common period of both satellite and shrub ring data from all sites was 1982 to 2010. Instrumental daily data from Toolik Lake and interpolated products was compared to detrended growth rates of Salix spp. (willow) and Alnus sp. (alder), located on and to the west of the Dalton Highway (68- 70ºN 148ºW). Whereas summer temperatures were found to enhance shrub growth, warm temperatures outside the core of the growing season have the inverse effect in some chronologies. All tundra shrub chronologies shared a common strong positive response to summer temperatures despite growing in heterogeneous site conditions and belonging to different species. In this work we will discuss shrub climate sensitive across Alaska and how NDVI data compared to the shrub ring-width network.

Associated Project: D'Arrigo-01

46:

Characterizing Arctic plant traits with near-surface and unmanned aerial system (UAS) remote sensing

Shawn Serbin, Brookhaven National Laboratory, sserbin@bnl.gov
Ran Meng, Brookhaven National Laboratory, ranmeng@bnl.gov
Andrew McMahon, Brookhaven National Laboratory, amcmahon@bnl.gov
Kim Ely, Brookhaven National Laboratory, kely@bnl.gov
Alistair Rogers, Brookhaven National Laboratory, arogers@bnl.gov
Stan Wullschleger, Oak Ridge National Laboratory, wullschlegsd@ornl.gov

The inadequate representation of plant traits and trait variation across space and time in terrestrial biosphere models (TBMs), including many that underlie the land-surface component of Earth System Models (ESMs), is a key driver of uncertainty in model hindcasts and forecasts of terrestrial carbon, water, and energy cycling and storage. This is particularly relevant for biomes with only sparse observational data availability such as the Arctic and tropics. In the Arctic, uncertainty in the modeling of carbon uptake and associated processes and fluxes has been tied to the lack of key data on plant properties that regulate these processes. What is needed is an approach to bridge the scales between detailed ongoing in-situ observations of Arctic vegetation in remote locations and the larger, landscape context needed to inform models on parameter variation in relation to climate, soils, topography, perturbations and other edaphic conditions. Remote sensing approaches, particularly spectroscopy, imaging spectroscopy, high resolution imaging, and thermal infrared (TIR) thermography, represent powerful observational datasets capable of scaling plant properties and capturing broad-scale spatial and temporal dynamics in many important vegetation properties related to terrestrial ecosystem functioning, offering a an important and direct data constraint on model projections or as critical benchmarks against prognostic model predictions. In temperate ecosystems we have shown how leaf and imaging spectroscopy (IS) can be used to map a broad range of plant traits across large areas of the continental U.S. and through time. Here we extend these approaches to the high Arctic to evaluate the capacity to scale and map vegetation properties, including biochemical, morphological and physiological leaf traits from the leaf to landscape scales. We focus on the development of linkages between a range of plant species and remote sensing within our two core study areas within the Barrow Environmental Observatory (BEO), Barrow, and Nome Alaska regions. We coupled measurements of leaf chemistry and physiology, including leaf-level gas exchange, with measurements of full range (i.e. 0.3 to 2.5 microns) leaf optical properties (reflectance and transmittance), TIR, and optical imagery from near-surface (leaf, tram) to unmanned aerial system (UAS) platforms. We show how leaf-level spectra-trait models for Arctic vegetation, developed using data collected in the BEO during the 2014-2016 growing seasons, are comparable with existing models from other biomes. In addition, tram and UAS platforms show a strong capacity to scale leaf-level traits to the larger landscape and capture patterns through time. Importantly, despite strong variation in leaf morphology and physiology, we are finding a good potential for spectral models to capture trait variation and highlights the possibility to map traits in the high Arctic. Our next steps include the use of the existing and future NASA ABoVE airborne campaign data to scale from tram/UAS to the larger regions to develop algorithms for mapping key traits across broad regions in the Arctic. Shawn P. Serbin; sserbin@bnl.gov

[poster]

Associated Project: Wullschleger-01

49:

Toward Understanding Dynamics in Shifting Biomes: An Individual Based Modeling Approach to Characterizing Drought and Mortality in Central Western Canada

Amanda Armstrong, University of Virginia, amanda.h.armstrong@nasa.gov
Adrianna Foster, NASA GSFC / USRA, adrianna.c.foster@nasa.gov
Brendan Rogers, Woods Hole Research Center, brogers@whrc.org
Ted Hogg, Canadian Forest Service, ted.hogg@canada.ca
Michael Michaelian, Canadian Forest Service, michael.michaelian@canada.ca
Jacquelyn Shuman, National Center for Atmospheric Research, jkshuman@ucar.edu
Herman Shugart, University of Virginia, hhs@virginia.edu
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

The Arctic–Boreal zone is known be warming at an accelerated rate relative to other biomes. Persistent warming has already affected the high northern latitudes, altering vegetation productivity, carbon sequestration, and many other ecosystem processes and services. The central-western Canadian boreal forests and aspen parkland are experiencing a decade long drought, and rainfall has been identified as a key factor controlling the location of the boundary between forest and prairie in this region. Shifting biome with related greening and browning trends are readily measureable with remote sensing, but the dynamics that create and result from them are not well understood. In this study, we use the University of Virginia Forest Model Enhanced (UVAFME), an individual-based forest model, to simulate the changes that are occurring across the southern boreal and parkland forests of west-central Canada. We present a parameterization of UVAFME for western central Canadian forests, validated with CIPHA data (Climate Change Impacts on the Productivity and Health of Aspen), and improved mortality. In order to gain a finescale understanding of how climate change and specifically drought will continue to affect the forests of this region, we simulated forest conditions following CMIP5 climate scenarios. UVAFME predictions were compared with statistical models and satellite observations of productivity across the landscape. Changes in forest cover, forest type, aboveground biomass, and mortality and recruitment dynamics are presented, highlighting the high vulnerability of this region to vegetation transitions associated with future droughts.

Associated Project: Goetz-03

56:

The Canadian High Arctic Research Station - A 12-Month-a-Year Partner for Science Cooperation

DONALD McLennan, Polar Knowledge Canada, coenosis@hotmail.ca
Adam Houben, Polar Knowledge Canada, Adam.Houben@polar.gc.ca
Ioan Wagner, Polar Knowledge Canada, johann.wagner@polar.gc.ca

The Canadian High Arctic Research Station (CHARS) is within a new government agency called Polar Knowledge Canada (POLAR), and is located in Cambridge Bay, Nunavut, Canada (https://en.wikipedia.org/wiki/Cambridge_Bay). CHARS is the largest research facility in the Canadian Arctic, featuring the Main Research Building and the Field and Maintenance Building to be open by summer 2018, and free, comfortable accommodation in two adjacent triplexes for up to 48 researchers that is open now. The CHARS Experimental and Reference Area (CHARS ERA) is a multi-scalar area around the Station where baseline studies to support research in terrestrial, freshwater and coastal-marine ecosystems are underway. Experimental sites within the ERA range from a paired-watershed Intensive Monitoring Area (IMA) adjacent to the station where instrumented, long-term hypothesis-based monitoring experiments are being initiated, to linked terrestrial-freshwater-marine-community studies in the 1,500 km2 Greiner watershed, and in the Regional CHARS ERA that includes most of the Kitikmeot Region of Nunavut. CHARS has an evolving in-house science capacity to support professional and technical aspects of science partnerships. Additionally, POLAR is presently funding a broad range of ecosystem studies in the CHARS ERA and is actively seeking science partnerships that support science objectives closely aligned with the NASA ABoVE Program. Please contact Donald McLennan (donald.mclennan@polar.gc.ca) or Adam Houben (adam.houben@polar.gc.ca) for further information, or come talk to us at our poster during the Poster Sessions during the Seattle Science Meeting.

Associated Project: McLennan-01

60:

Calibration and Validation of Fractional Lichen Cover Mapping

Matthew Macander, Alaska Biological Research, Inc.--Environmental Research & Services, mmacander@abrinc.com
Eric Palm, University of Montana, e2palm@gmail.com
Peter Nelson, University of Maine - Fort Kent, peter.nelson@maine.edu
Mark Hebblewhite, University of Montana, mark.hebblewhite@umontana.edu
Jim Herriges, Bureau of Land Management, jherrige@blm.gov
Gerald Frost, Alaska Biological Research, Inc.--Environmental Research & Services, jfrost@abrinc.com
Christopher Swingley, Alaska Biological Research, Inc.--Environmental Research & Services, cswingley@abrinc.com
Emily Holt, University of Northern Colorado, Emily.Holt@unco.edu

Cover maps are being developed for selected tundra plant functional types (PFTs) across >500,000 sq. km of arctic and boreal Alaska and adjacent Canada at 30 m resolution. Training and validation data include a field-based training dataset based on both point-intercept and ocular estimation sampling methods at thousands of plots spanning bioclimatic and geomorphic gradients. In 2017, we also compiled over 20 blocks of 1-5 cm resolution RGB image mosaics in Alaska (White Mountains) and the Yukon Territory to provide supplementary training and validation data for mapping lichen cover in the range of the Fortymile Caribou Herd. The mosaics and associated surface and canopy height models were developed using a consumer drone and structure from motion processing. We summarized both the in situ measurements and drone imagery to determine cover of Light Macrolichens. We applied these data to train 2 m (limited extent) and 30 m (wall to wall) maps of fractional cover for lichen for c. 2015. Predictors for 2 m models were commercial satellite imagery such as WorldView-2 and Worldview-3, analyzed on the ABoVE Science Cloud. Predictors for 30 m models were percentile reflectance composites and spectral metrics, developed from Landsat imagery using Google Earth Engine. Next steps include extending the mapping to Arctic Alaska and Canada; expanding to include mapping of shrub PFTs; and applying models to historical Landsat data to estimate c. 2000 shrub and lichen cover.

[poster]

Associated Project(s):
Boelman-01
Goetz-03

61:

Plant and shrub aboveground biomass mapped across the North Slope of Alaska using Landsat

Logan Berner, Northern Arizona University, logan.berner@nau.edu
Patrick Jantz, Northern Arizona University, patrick.jantz@nau.edu
Kenneth Tape, University of Alaska, Fairbanks, kdtape@alaska.edu
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

Rising temperatures are increasing plant biomass and shrub dominance in parts of the Arctic tundra, including northern Alaska. These changes are in turn affecting regional wildlife, climate feedbacks, and northern communities, yet the amount and spatial distribution of plant biomass remains uncertain in tundra ecosystems. In this study, we mapped plant and shrub aboveground biomass (AGB; kg m^-2) across the North Slope of Alaska by combining biomass harvests (n = 24 sites) and a large collection of 30 m resolution Landsat observations (n > 2,000 scenes). We first built non-linear regression models to predict plant and shrub AGB based on Landsat peak summer NDVI (r² ≈ 0.80, P < 0.01). We then predicted regional plant AGB, shrub AGB, and shrub dominance (shrub AGB / plant AGB) by applying these models to a Landsat composite mosaic that covered the North Slope. We quantified pixel-wise uncertainty using a Monte Carlo approach that incorporated sampling and sensor calibration errors. The new maps track landscape variation in AGB visible in high-resolution satellite and aerial imagery, such as the contrast between shrubby water tracks and barren ridgelines. Modeled shrub AGB was closely correlated with a regional map of shrub cover (r_s = 076) and with field measurements of shrub height (r_s = 0.88, n = 25 sites). The spatial distribution of plant and shrub AGB was shaped by local topography, summer temperatures, and prior disturbances, including fires. Future warming will likely increase plant AGB and shrub dominance, yet ecological response to warming will likely be mediated by topography and disturbance. These biomass maps can aid in assessing ecosystem-climate feedbacks associated with ongoing environmental change, and may also inform management of North Slope ecosystems.

Associated Project: Goetz-03

63:

Forest composition, structure and productivity of browning and greening forests in Interior Alaska

Claudia Czimczik, University Of California, Irvine, czimczik@uci.edu
Nicole Fiore, UC Irvine, nmfiore@uci.edu
Shawn Pedron, UC Irvine, spedron@uci.edu
Clayton Elder, Jet Propulsion Laboratory, cdelder@uci.edu
Michael Goulden, University Of California, Irvine, mgoulden@uci.edu

Previous remote sensing analyses of boreal forest structure have yielded several key findings, including a widespread trend toward declining Normalized Difference Vegetation Indices (NDVIs), which suggests decreasing forest productivity, and large changes in surface reflectance and brightness temperature with and following wildfire, which implies large shifts in biophysical properties during fire recovery. These observations provide a powerful tool for assessing boreal forest structure and function at large spatial scales, but also underscore the challenge of linking remotely-sensed observations to the actual conditions on the ground. Direct field observations are needed to link remote sensing observations to biophysical properties, and to further explore recent possible declines in boreal productivity, as well as the biophysical changes during succession. We surveyed the composition and structure of 20 forest stands in interior Alaska (63.8004 to 65.5720 N and 149.3965 to 149.9545 W) that over the past 30 years showed increasing, decreasing or no change in Landsat TIR, NDII, NDVI, Brightness, Greenness and, or Wetness. Some of these sites had recently burned, while others were considered old growth. We surveyed the forest at each site along 3-5 100-m transects, and assessed the forest structure, including tree height, diameter per canopy, vegetative cover (stratified by canopies and ground), fraction and type of standing and downed dead trees, and leaf area index (LAI). Data were recorded by tree species and ground cover type (moss, lichen, forb, grass). Fire history and productivity of each stand were assessed with increment cores or disks for 2-5 trees per species. Key findings include: 1) Various remote sensing proxies from Landsat are tied to distinct biophysical properties. The local TIR relative to the scene average was correlated with canopy height. Wetness and NDII were correlated with canopy cover and LAI. The ratio of greenness or NDVI to wetness was correlated to the relative abundance of deciduous vs. evergreen foliage. These correlations proved useful for quantifying the biophysical changes with fire recovery. 2) Fire recovery can be distinguished into three stages: i.) A rapid increase in NDVI and greenness with an expansion of deciduous foliage, ii.) A gradual increase in wetness and decrease in brightness and TIR with an expansion of total LAI and a co-dominant evergreen and conifer canopy, and iii.) A gradual decrease in NDVI and greenness despite a high wetness and LAI as deciduous species senesce and evergreen dominate. 3) Some of the recent NDVI decline in the area appears to reflect this mid to late succession transition from high NDVI, deciduous stands to low NDVI, evergreen stands. Several of the "browning" stands we visited had a healthy evergreen overstory that had recently overtopped a senescent deciduous stratum. However, “browning” was also observed in mature forests transitioning to sphagnum moss-dominated peat plateaus or recently burnt forests, where permafrost thaw and collapse promoted shrubs, and “greening” in a forest stand that had been recently drained by regional flood-control projects. Together, our data show that fire history and changes in hydrology strongly modify the reflective properties of boreal forests. Thus, quantifying effects of climate change on these forests requires knowledge of typical succession trajectories in relation to aspect, and drainage.

[poster]

Associated Project: Goulden-02

68:

Terrestrial Lidar and Chlorophyll Fluorescence Reveal Structure-to-Function Relationships of Spruce Saplings at the Forest-Tundra Ecotone

Andrew Maguire, University of Idaho, magu7563@vandals.uidaho.edu
Jan Eitel, University of Idaho, jeitel@uidaho.edu
Lee Vierling, University of Idaho, leev@uidaho.edu
Daniel Johnson, University of Idaho, danjohnson@uidaho.edu
Kevin Griffin, Columbia University, griff@ldeo.columbia.edu
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Johanna Jensen, Columbia University, jej2141@columbia.edu
Arjan Meddens, University of Idaho, ameddens@uidaho.edu

Purpose and Methods The effect of climate – and microclimate – on treeline position at the latitudinal forest-tundra ecotone (FTE) is poorly understood. While the Arctic FTE is expansive (~13000 km), understanding relationships among climate, tree location, and tree function may depend on very fine scale processes, therefore tools are needed to appropriately characterize the leading (northernmost) edge of the FTE. We hypothesized that microstructural metrics obtainable from lidar remote sensing can explain variation in the physical growth environment that governs sapling photosynthetic processes. To test our hypothesis, we used terrestrial laser scanning (TLS) to collect highly spatially resolved (<1 cm) 3-D structural information of white spruce (Picea glauca) saplings and their aboveground growth environment at the leading edge of a FTE in northern Alaska. Coordinates of sapling locations were extracted from the 3-D TLS data along with sapling heights. Dark-adapted chlorophyll fluorescence was measured using a handheld fluorometer on three needle bundles per sapling. Digital terrain models (DTMs) and digital surface models (DSMs) were interpolated based on ground- and canopy-classified laser returns of the TLS data, respectively. Five terrain attributes were modeled from interpolated DTMs: average aspect, average slope, average curvature, wind-shelter index, and solar insolation. Two canopy attributes were modeled from interpolated DSMs: wind-shelter index and solar insolation. Average canopy depth was calculated as the height difference between DTMs and DSMs. Ground roughness and canopy roughness were calculated as the standard deviation of the heights of ground- and canopy-classified laser return. Each microstructural metric was extracted within variable search radii (0.1 – 2.0 m) from the sapling coordinates. Results and Discussion Random-Forest statistical modeling was conducted to determine which microstructural metrics were most strongly related to chlorophyll fluorescence (Fv/Fm, the maximum quantum yield of primary photochemistry). Wind-shelter index from the dominant wind direction (south-southeast) and solar insolation from the seven day period prior to sampling within 2.0 m radius from the saplings explained the most variance in Fv/Fm. Both terrain attributes were found to be moderate predictors of Fv/Fm using simple linear regression (r2 = 0.40; RMSE = 0.024). The observed Fv/Fm values were low (0.67 – 0.79), potentially indicating chronic stress commonly observed at other ecotones. The solar radiation environment and the wind environment, as characterized based on TLS data, both constrain sapling photosynthesis. These findings corroborate broader research on the importance of the Arctic light environment (360 solar azimuth and 24 h daylight during growing season) and treeline wind environment on plant function. Our results indicate that high resolution structural information from TLS remote sensing may provide valuable insights on sapling function at the FTE. In particular, the observed negative relationship with wind-shelter and Fv/Fm is surprising given ample evidence of the importance of positive sheltering effects for immature trees at treeline. The coupling of wind and light environments suggests that fine-scale exposure plays a critical role in the condition of spruce trees during vulnerable development stages. These results reveal a sensitivity of the photosynthetic performance of spruce saplings to fine-scale exposure, with respect to both wind and light, suggesting that temperature regulation may be a critical process for photosynthesis at the FTE. Characterizing the relationships between the remotely sensible structural growth environment and ecological responses is of great interest to broader research endeavors of ecological vulnerability at the FTE.

Associated Project: Eitel-01

69:

Towards lidar-based mapping of tree age at the Forest Tundra Ecotone

Johanna Jensen, Columbia University, jej2141@columbia.edu
Andrew Maguire, University of Idaho, magu7563@vandals.uidaho.edu
Rose Oelkers, Lamont Doherty Earth Observatory, roelkers@ldeo.columbia.edu
Laia Andreu-Hayles, Lamont Doherty Earth Observatory, lah@ldeo.columbia.edu
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Rosanne D'Arrigo, Lamont Doherty Earth Observatory, rdd@ldeo.columbia.edu
Kevin Griffin, Columbia University, griff@ldeo.columbia.edu
Carlos Silva, University of Idaho, csilva@uidaho.edu
Jyoti Jennewein, University of Idaho, jjennewein@uidaho.edu
Arjan Meddens, University of Idaho, ameddens@uidaho.edu
Micah Russell, University of Idaho, russ5140@vandals.uidaho.edu
Lee Vierling, University of Idaho, leev@uidaho.edu
Jan Eitel, University of Idaho, jeitel@uidaho.edu

Climate change may cause spatial shifts in the forest-tundra ecotone (FTE). To improve our ability to study these spatial shifts, information on tree demography along the FTE is needed. The objective of this study was to assess the suitability of lidar-derived tree heights as a surrogate for tree age. We calculated individual tree age from 62 tree cores collected at basal height from white spruce (Picea glauca) within the FTE in northern Alaska. Tree height was obtained from terrestrial lidar scans (< 1 cm spatial resolution). The relationship between age and height was examined using a linear regression model. We found a very strong predictive relationship between tree height and age (R² = 0.6951, RMSE = 29.32 years) for trees that ranged between 14 to 230 years old and 0.29 to 15.2 m tall. Regression models were also developed for small (height < 3 m) and large trees (height >= 3 m), however, these models captured less variance in the data (R² = 0.4292, RMSE = 14.05 years and R² = 0.2383, RMSE = 41.15 years, respectively). Although a strong, predictive relationship between age and height is uncommon in light-limited forest environments, we hypothesize that the sparseness of trees within the FTE may explain the strong tree height-age relationships found herein. Using this relationship and tree height extracted from an individual tree detection algorithm (Silva et al., in review), we predicted age for six 2,500 m² plots within white spruce (Picea glauca) stands previously determined to be along the FTE. From these predictions, we generated age distributions to assess population dynamics. Age distributions show substantially fewer trees which are < 40 years old with increasing latitude. Further analysis of 36 additional tree cores recently collected within the FTE near Inuvik, Canada will be performed. Our analysis suggests that lidar-derived tree height could be a reliable proxy for tree age at the FTE, thereby establishing a new technique for scaling tree structure and demographics across larger portions of this sensitive ecotone.

Associated Project(s):
D'Arrigo-01
Eitel-01

78:

Forest change at the southern limits of the boreal zone

Elizabeth Campbell, Natural Resources Canada, elizabeth.campbell@canada.ca
Joseph Antos, University of British Columbia, jantos@uvic.ca
Tongli Wang, Universitiy of British Columbia, tongli.wang@ubc.ca
Lara VanAkker, Natural Resources Canada, lara.vanakker@canada.ca

Altered disturbance regimes due to climate change will be important drivers of projected ecosystem shifts at the southern limits of the boreal zone. We are using field data and models to characterize forest responses to intensifying or novel bark beetle disturbances, and other effects of continued climate change. To assess boreal forest resilience to a severe spruce beetle (Dendroctonus rufipennis) outbreak in the southern Yukon (1994-2005), we sampled 21 stands dominated by white spruce (Picea glauca) in 2000, 2010 and 2016. In each stand, we collected data on tree species composition, tree size structure, radial growth, and site/soil conditions. We found that the impact of the spruce beetle outbreak varied considerably among stands, reducing tree density (stems/ha) and basal area (m2/ha) between 5 to 85% from 2000 to 2010. Deciduous tree species abundance increased significantly in only 3 stands. Radial growth increment (mm/yr) of nearly half of the surviving canopy (43%) and subcanopy (45%) spruce trees increased by at least 50% since the last re-measurement in 2016. Advanced regeneration of spruce (30-130 cm tall) was abundant but only a small percentage (19%) had increased growth after beetle-caused canopy removal. Despite abundant regeneration and increases in growth among some trees, canopy tree basal area and density reached pre-outbreak values in only 4 of the 21 stands 10 years after the outbreak. More time is needed for these forests to recover. Our models of climate suitability for spruce and spruce-dominated ecosystems, suggest warming temperatures could ─ in the absence of other disturbances ─ favour accelerated forest recovery until about 2055 but after that, the climatic habitats for spruce and boreal forest become very unsuitable. We continue our exploration of climate change effects in these forests using modular landscape simulation modelling approaches.

81:

Multi-scale modeling of boreal forest vegetation growth under the influence of permafrost and wildfire interactions

Adrianna Foster, NASA GSFC / USRA, adrianna.c.foster@nasa.gov
Amanda Armstrong, University of Virginia, amanda.h.armstrong@nasa.gov
Jacquelyn Shuman, National Center for Atmospheric Research, jkshuman@ucar.edu
Kenneth Ranson, NASA GSFC, kenneth.j.ranson@nasa.gov
Herman Shugart, University of Virginia, hhs@virginia.edu
Brendan Rogers, Woods Hole Research Center, brogers@whrc.org
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

Global temperatures have increased about 0.2°C per decade since 1979, and the high latitudes are warming faster than the rest of the globe. Climate change within Alaska is likely to bring about increased drought and longer fire seasons, as well as increases in the severity and frequency of fires. These changes in disturbance regimes and their associated effects on ecosystem C stocks, including permafrost, may lead to a positive feedback to further climate warming. As of now, it is uncertain how vegetation will respond to ongoing climate change, and the addition of disturbance effects leads to even more complicated and varied scenarios. Through ecological modeling, we have the capacity to examine forest processes at multiple temporal and spatial scales, allowing for the testing of complex interactions between vegetation, climate, and disturbances. The University of Virginia Forest Model Enhanced (UVAFME) is an individual tree-based forest model that has been updated for use in interior boreal Alaska, with a new permafrost model and updated fire simulation. These updated submodels allow for feedback between soils, vegetation, and fire severity through fuels tracking and impact of depth of burn on permafrost dynamics. We present these updated submodels as well as calibration and validation of UVAFME to the Yukon River Basin in Alaska, with comparisons to inventory data. We also present initial findings from simulations of potential future forest biomass, structure, and species composition across the Yukon River Basin under expected changes in precipitation, temperature, and disturbances. We predict changing climate and the associated impacts on wildfire and permafrost dynamics will result in shifts in biomass and species composition across the region, with potential for further feedback to the climate-vegetation-disturbance system. These simulations advance our understanding of the possible futures for the Alaskan boreal forest, which is a valuable part of the global carbon budget.

Associated Project(s):
Goetz-03
Ranson-02

85:

Testing the correspondence among satellite-observed tundra greenness and on-the-ground vegetation monitoring using drones

Isla Myers-Smith, University of Edinburgh, isla.myers-smith@ed.ac.uk
Jeff Kerby, Dartmouth College, jeffrey.t.kerby@dartmouth.edu
Jakob Assmann, University of Edinburgh, j.assmann@ed.ac.uk
Andrew Cunliffe, University of Edinburgh, Andrew.Cunliffe@ed.ac.uk

Satellite-based observations of the Arctic indicate that tundra vegetation productivity is increasing, which is often referred to as the ‘greening’ of the Arctic. This increased productivity has been associated with the rapid warming experienced by terrestrial Arctic ecosystems over recent decades. However, coarse observational scales and strong regional variation in the satellite datasets have resulted in repeated calls for validation of the observed trends. High-resolution multispectral imagery obtained with Unmanned Aerial Vehicles (drones) holds the potential to bridge this gap. Here, we present time-series of multispectral and RGB imagery acquired with drones across the growing season 2016 and 2017 on Qikiqtaruk – Herschel Island, YT Canada, a region that is undergoing rapid permafrost thaw and vegetation change. We have developed a standardised workflow over three field seasons to capture plant phenology, shrub tundra biovolume, retrogressive thaw slumps and coastal erosion across the growing season using drones. Our preliminary analyses indicate correspondence in tundra vegetation greenness across scales. However, that data from ~5-15 cm grain-sizes better capture certain biological and physical processes that govern tundra greening patterns. Our results provide valuable insights into how drones can be used to monitor vegetation in rapidly changing tundra environments. For more information see: https://teamshrub.wordpress.com/ https://droneecology.wordpress.com/

Associated Project: Myers-Smith-01

86:

A multitemporal 30m land cover map for the ABoVE domain using dense Landsat time series and high resolution imagery.

Jonathan Wang, Boston University, jonwang@bu.edu
Damien Sulla-Menashe, Boston University, dsm@bu.edu
Curtis Woodcock, Boston University, curtis@bu.edu
Oliver Sonnentag, University of Montreal, oliver.sonnentag@umontreal.ca
Mark Friedl, Boston University, friedl@bu.edu

The rapid climate change in arctic and boreal ecosystems is resulting in drastic changes to land cover composition, including woody expansion in the arctic tundra, successional shifts following boreal fires, and permafrost thaw-induced wetland expansion. The impacts on physical climate and the carbon cycle of these land cover transformations are well-documented in field and modeling studies, but there have been few attempts to estimate overall rates of land cover on decadal and continental scales. Previous studies were either too coarse in spatial resolution or too limited in temporal range to analyze relevant rates of change. As part of NASA’s Arctic Boreal Vulnerability Experiment (ABoVE), we employ dense time series of Landsat remote sensing data to map disturbances and classify land cover types across the ABoVE domain, spanning western Canada and Alaska, over the last three decades (1982-2014) at 30m resolution. We utilize regionally-complete and repeated acquisition high-resolution (<2 m) remote sensing imagery to build a database of training data that samples from geographically diverse regions and follows a nested, hierarchical classification scheme encompassing vegetation form, wetland status, vegetation cover density, and land use. We use the Continuous Change Detection and Classification (CCDC) algorithm to estimate change dates and stable temporal-spectral features in the Landsat data for training and prediction using random forest classifiers. We predict land cover types in order to estimate the distribution of land cover and land cover change rates, focusing primarily on woody expansion in the tundra, post-fire succession, and wetland expansion in the boreal forest. The hierarchical classification scheme provides flexibility in assessing different relevant plant functional types in different regions of the arctic and boreal ecosystems. We resample from the high resolution imagery to assess the accuracy of the land cover change maps.

[poster]

Associated Project: Woodcock-02

93:

Fusing Digital Elevation Datasets to Create a Best Currently Available Composite DEM for ABoVE

Patrick Burns, Northern Arizona University, pb463@nau.edu
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu
Richard Massey, Northern Arizona University, rm885@nau.edu

Digital elevation models (DEMs) are necessary for a variety of Earth Science applications, including image orthorectification and topographic correction, as well as hydrologic and vegetation modeling. Working at very large scales, such as the entire NASA ABoVE domain, it is often beneficial to perform analyses using contiguous (gap-free) DEMs that have an associated vertical accuracy metric. Currently no single, publicly-available elevation dataset covers the entire ABoVE study domain. For this reason we created a best currently available composite using the following publicly-available elevation datasets, ranked in order of overall quality: airborne ifsar (Alaska only), ArcticDEM, ALOS World 3D, Shuttle Radar Topography Mission DEM, Canada DEM, and ASTER Global Digital Elevation Model. The input datasets have variable spatial resolution and temporal acquisition dates, as well as horizontal and vertical accuracy. Furthermore, the input datasets are referenced from different horizontal and vertical datums (not to mention reference frame epochs), and also span the USA-Canada border. The input datasets are first harmonized to a common reference frame (same horizontal and vertical datums: WGS84, EGM2008) and then composited as one ESRI master mosaic dataset on the ABoVE Science Cloud. This master mosaic dataset allows users flexibility when exporting their sub-project-specific composites. For example, a user could filter the master mosaic dataset to only include DEMs acquired between 2010 and 2015 with a vertical accuracy less than 10m (LE90). The initial version of this ABoVE best currently available elevation composite (v1) will be tiled based on the ABoVE grid system at 10m spatial resolution and include all datasets made publicly available through the end of 2017. Vertical accuracy metrics are tied to individual input DEMs (where available) and overall vertical accuracy of the composite is assessed using airborne lidar and airport runway elevations.

Associated Project: Goetz-03

95:

Arctic vegetation mapping using UAV RGB imagery scaled up to AVIRIS imagery using ground spectroradiometry and ocular vegetation estimates

David Paradis, University of Maine - Fort Kent, david.p.paradis@maine.edu
Peter Nelson, University of Maine - Fort Kent, peter.nelson@maine.edu

We are using AVIRIS ng imaging spectrometer data as a means of scaling up cover estimates of lichens and other plant functional types to a fine taxonomic resolution using UAV-based RGB mosaics and ground measurements. We flew our commercial UAV within 2-3 weeks and under AVIRIS ng flight lines over seven 100x100 m areas (plots) in Alaska wherein we systematically sampled vegetation on 100m transects using target plant functional groups. We scanned each ground sample unit (1m2 quadrat where vegetation cover was ocularly estimated) with a spectroradiometer (PSR-3500 spectral evolution) in four 30cm diam FOVs, which were averaged to the quadrat. We used a contact probe with an active light source to make high quality scans of individual species from which we built a spectral library of arctic plants and lichens. We made lichen fractional cover maps produced from ML classification of UAV orthomosaics, located across Alaska, as training data for scaling up to AVIRIS flight lines. We also use the spectral library as the spectral endmember parameters which are passed to filtering algorithms which ultimately produce binary lichen/no lichen classification using AVIRIS data. Results from AVIRIS vegetation mapping are tested against fractional lichen cover maps derived from the independent UAV imagery. Our end product will be a large number of AVIRIS flight lines with high precision vegetation cover estimates that will be scaled to Landsat and larger extents in Alaska and NW Canada.

Associated Project: Goetz-03

98:

Mapping deciduous fraction in interior Alaska using seasonal composites from Landsat imagery

Richard Massey, Northern Arizona University, rm885@nau.edu
Logan Berner, Northern Arizona University, logan.berner@nau.edu
Brendan Rogers, Woods Hole Research Center, brogers@whrc.org
Scott Goetz, Northern Arizona University, Scott.Goetz@nau.edu

One of the most fundamental properties for understanding vegetation in the ABoVE domain is leaf habit, i.e. deciduous vs. evergreen. Deciduous trees such as aspen and birch have substantially different impacts on ecosystem function and land-atmosphere interactions compared to conifers, including carbon cycling, albedo, and energy partitioning. Climate warming and increased disturbances, primarily wildfire, are thought to have been increasing the fraction of deciduous vegetation, but this remains largely unquantified. Here we derive spatial estimates of deciduous fraction at 30 m for Alaska using phenology-based metrics. We initially focused on 2010 due to the availability of reference data sets and dense 30 m Landsat 5 TM and Landsat 7 ETM+ data stacks. We trained a Random Forest classifier using maximum NDVI over three periods: 16th May – 30th June, 1st July – 15th August, and 16th August – 30th September. Training and validation samples were derived from multiple forest inventory sources, including the Cooperative Alaska Forest Inventory (CAFI) and the Bonanza Creek long term ecological research transects. Initial results indicate relatively high accuracy and are consistent with the NLCD 2011 categorical land cover map. Future steps include mapping and validating in different time periods, assessing the major drivers of change, extending our analysis into Canada, and comparing against coarse-scale AVHRR estimates of deciduous fraction.

Associated Project(s):
Goetz-03
Rogers-02

101:

Generation of large-scale forest height and disturbance maps through the fusion of NISAR and GEDI along with TanDEM-X/L

Yang Lei, Universities Space Research Associations; Jet Propulsion Laboratory, yang.lei@jpl.nasa.gov
Robert Treuhaft, JPL, California Institute of Technology, robert.treuhaft@jpl.nasa.gov
Paul Siqueira, University of Massachusetts, siqueira@umass.edu
Nathan Torbick, Applied Geosolutions, ntorbick@appliedgeosolutions.com
Richard Lucas, University of New South Wales, richard.lucas@unsw.edu.au
Michael Keller, USDA Forest Service @ JPL, michael.m.keller@jpl.nasa.gov
Michael Schmidt, Remote Sensing Centre, Brisbane, Australia, michael.schmidt@dsiti.qld.gov.au
Mark Ducey, University of New Hampshire, mark.ducey@unh.edu
William Salas, Applied GeoSolutions, wsalas@agsemail.com

Large-scale products of forest height and disturbance are essential for understanding the global carbon distribution as well as its changes in response to natural events and human activities. Regarding this scientific need, both NASA’s GEDI and NASA-ISRO’s NISAR are going to be launched in the 2018-2021 timeframe in parallel with DLR’s current TanDEM-X and/or the proposed TanDEM-L, which provides a lot of potential for global ecosystem mapping. A new simple and efficient method of forest height mapping has been developed for combining spaceborne repeat-pass InSAR and lidar missions (e.g. NISAR and GEDI) which estimates temporal decorrelation parameters of repeat-pass InSAR and uses the lidar data as training samples. An open-access Python-based software has been developed for automated processing. As a result, a mosaic of forest height was generated for US states of Maine and New Hampshire (11.6 million ha) using JAXA’s ALOS-1 and ALOS-2 HV-pol InSAR data and a small piece of lidar training samples (44,000 ha) with the height estimates validated against airborne lidar and field inventory data over both flat and mountainous areas. In addition, through estimating and correcting for the temporal decorrelation effects in the spaceborne repeat-pass InSAR coherence data and also utilizing the spaceborne single-pass InSAR phase data, forest disturbance such as selective logging is not only detected but also quantified in subtropical forests of Australia using ALOS-1 HH-pol InSAR data (validated against NASA’s Landsat), as well as in tropics of Brazil using TanDEM-X and ALOS-2 HH-pol InSAR data (validated against field inventory data). The operational simplicity and efficiency make these methods a potential observing/processing prototype for the fusion of NISAR, GEDI and TanDEM-X/L.

103:

Seasonal and Inter-annual Phenological Variability is Greatest in Low-Arctic and Wet Sites Across the North Slope of Alaska as Observed from Multiple Remote Sensing Platforms

Sergio Vargas Zesati, University of Texas at El Paso, savargas@utep.edu
Gesuri Ramirez, University of Texas at El Paso, gramirez12@utep.edu
Geovany Ramirez, New Mexico State University, geoabi@gmail.com
Christian Andresen, Los Alamos National Laboratory, candresen@lanl.gov
Nathan Healey, NASA Jet Propulsion Laboratory, natehealey@hotmail.com
Jeremy May, Florida International University, jmay010@fiu.edu
Steve Oberbauer, Florida International University, oberbaue@fiu.edu
Bob Hollister, Grand Valley State University, hollistr@gvsu.edu
Craig Tweedie, University of Texas at El Paso, ctweedie@utep.edu

Rates of climate change have been documented to be highest in the Arctic when compared to other ecosystems across the globe, however the magnitude and variability of change is not well known. Plant phenological trends can shift in response to climate change and has the potential to elucidate seasonal and inter-annual shifts in ecosystem properties and processes. Traditionally, ecosystem phenology has been quantified using satellite-based systems and ground-based observations but each approach has limitations especially in high latitude regions. Mid-scale sensing platforms that measure plot to landscape scale optical properties (e.g. robotic tram systems, unmanned aerial vehicles (UAVs), pheno-cams) have shown to provide alternative, and in most cases, low-cost solutions with comparable results to those acquired traditionally. This study contributes to the US Arctic Observing Network (AON) and assesses the effectiveness of using plot-level images (PLI), imagery acquired from pheno-cams, and kite aerial photography (KAP) for deriving measures of phenological variability (e.g. start of season (SOS), greening and end of season (EOS)) for dominant vegetation communities near Utqiaġvik (formerly Barrow) and Atqasuk, Alaska. Using five growing seasons of digital imagery acquired from these platforms, the Green Chromatic Coordinate (GCC)) was derived from RGB digital numbers (DN) and compared to the normalized difference vegetation index (NDVI) calculated from ground-based reflectance measurements. NDVI has been shown to be an effective proxy of primary productivity across multiple ecosystems including the Arctic. The three low-cost sensor platforms showed trends that tracked the traditionally preferred NDVI but showed an improved capacity to document fine-scale species-level phenological changes at high temporal frequencies. Seasonal and inter-annual variability in GCC and NDVI were greatest in low arctic and wet sites while high arctic and dry sites showed less variability. Preliminary results suggest that the strong seasonal and inter-annual variability in arctic landscapes, similar to those sampled for this study, is driven by moist to wet land cover types. Future studies will extend cross-scale analysis to a variety of satellite platforms (i.e. WorldView, Landsat, MODIS) to understand how such patterns transcend sampling at different spatial scales and sensor platforms.

Associated Project: Huemmrich-01

104:

The ABoVE Spectral Library (ASTRAL) – A Web-mapping Application to Enhance the Discovery, Visualization, and Sharing of Spectral Reflectance Data

Sergio Vargas Zesati, University of Texas at El Paso, savargas@utep.edu
Ari Kassin, University of Texas at El Paso, akassin@utep.edu
Mauricio Barba, University of Texas at El Paso, mbarba3@utep.edu
Ryan Cody, University of Texas at El Paso, rpcody@utep.edu
Karl Huemmrich, NASA's Goddard Space Flight Center, karl.f.huemmrich@nasa.gov
Petya Campbell, NASA's Goddard Space Flight Center, petya.k.campbell@nasa.gov
John Gamon, Universtiy of Nebraska at Lincoln, jgamon@unl.edu
Stephen Escarzaga, University of Texas at El Paso, smescarzaga@utep.edu
Craig Tweedie, University of Texas at El Paso, ctweedie@utep.edu

The Arctic is experiencing among the most dramatic impacts from climate change on the planet. Observed large-scale responses include but are not limited to loss of sea ice and snow cover, sea level rise, enhanced coastal erosion, increases in near-surface air temperature and satellite-derived green biomass, permafrost thaw and degradation, subsidence and geographical shifts in vegetation distribution. NASA’s Terrestrial Ecology Program launched the Arctic-Boreal Vulnerability Experiment (ABoVE) in an effort to better understand the vulnerability and resilience of Arctic and Boreal ecosystems and societies to environmental variability and change. Remote sensing of natural targets, in particular spectral reflectance, from ground, airborne and space borne platforms is a widely used method for monitoring changing arctic landscapes, but remains poorly documented for much of the Arctic. Additionally, there exists a need for consistent web based tools that aid the discovery, sharing and visualization of spectral reflectance data for these regions specifically. During the summer of 2017 our project collected over 1000 field spectra at the leaf and plot level for validation of AVIRIS spectral reflectance and we have sourced more than 300,000 spectra from our archives and those of others that would likely be useful for ABoVE efforts. Managing and visualizing large amounts of spectral data originating from a variety of sensors and data collection methods remains a challenge for the majority of the ABoVE projects as discussed at previous meetings. In an effort to alleviate these issues and help bring together our understanding, we have prototyped a spectral reflectance library that allows for the discovery, visualization, and sharing of spectral data. These include web-based spectral libraries and informational tools such as pre-existing BAID- the Barrow Area Information Database, AOV- the Arctic Observing Viewer, EcoSIS, ASU Spectral Library, JPL HyspIRI Spectral Library, the USGS Digital Spectral Library, and the ASTER Spectral Library. Here, we present a beta application and welcome constructive suggestions on how to better design the application to be most useful for ABoVE Science Team members and other stakeholders.

Associated Project: Huemmrich-01

106:

Multi-sensor Remote Sensing for Large-area Forest Inventory in the Northwest Territories

Guillermo Castilla, Canadian Forest Service, guillermo.castilla@canada.ca

The Canadian Forest Service (CFS) and the Government of Northwest Territories (GNWT) are creating a broad forest inventory across 44 million hectares encompassing 80% of NWT forests. Specifically, this collaborative project is creating full coverage estimates (in the form of 30 m rasters) of stand height, crown closure, stand and total volume, aboveground biomass, and stand age derived from a multi-scale sampling design that includes field data, limited airborne and satellite LiDAR, Landsat and RADAR imagery, and other ancillary data. Some of these raster maps will serve as input to CFS spatially explicit Carbon Budget Model (gCBM) that will estimate C pools and fluxes in each forest pixel by combining information on biomass, growth and fire disturbance. In addition, the project will also deliver an integrated polygon map that resembles a traditional forest inventory map and that seamlessly integrates the existing GNWT forest vegetation inventory (FVI, covering < 10% of NWT forests) to obtain a wall to wall product (the Multisource Vegetation Inventory, or MVI) that retains the higher quality information where available. Phase 1 of the project (southern half) was completed last year, and Phase 2 will be delivered in 2018.

115:

Airborne LVIS LiDAR measurements of surface topography and structure for ABoVE

Michelle Hofton, University of Maryland, mhofton@umd.edu
Bryan Blair, NASA GSFC, james.b.blair@nasa.gov
Dave Rabine, GSFC, david.l.rabine@nasa.gov
Helen Cornejo, NASA GSFC / SGT, helen.g.cornejo@nasa.gov
Sarah Story, NASA GSFC / SGT, sstory@sgt-inc.com
Colleen Brooks, NASA GSFC / SSAI, colleen.brooks@nasa.gov

In June-July 2017, NASA's Land, Vegetation and Ice Sensor (LVIS) Facility was deployed to sites in northern Canada and Alaska in support of NASA's ABoVE 2017 airborne campaign flying a total of 11 science flights in the region. LVIS-F is NASA's high-altitude airborne lidar sensor, collecting a nominal ~2km wide swath of data from 10km altitude above the ground. Footprints are contiguous both along and across track and for ABoVE operations, were ~6m in diameter. LVIS is a full waveform sensor. These data are collected for every footprint and georeferenced to provide a true 3 dimensional view of overflown terrain from whcih elevation, height and structure measurements and metrics are derived. After accounting for cloud cover, approximately 20,000km2 of surface data were collected by the mission. Level1B (geolocated waveform) and Level2 (elevation and height) data products are available to investigators via a NASA DAAC and the LVIS website at lvis.gsfc.nasa.gov/ABoVE2017map.html.

Poster Session C

Thursday 4:30 PM

Permafrost and Hydrology

10:

Differentiating surface water change from seasonal and inter-annual variability in North Western Canada using multi-decadal time series of data

Mark Carroll, NASA GSFC / SSAI, mark.carroll@nasa.gov
Tatiana Loboda, University of Maryland, loboda@umd.edu
Margaret Wooten, NASA GSFC / SSAI, margaret.wooten@nasa.gov

Over the last several decades, warming in the Arctic has outpaced the already impressive increases in global mean temperatures. The impact of these increases in temperature has been observed in a multitude of ecological changes in North American tundra including changes in vegetative cover, depth of active layer, and surface water extent. The low topographic relief and continuous permafrost create an ideal environment for the formation of small water bodies—a definitive feature of tundra surface. In this study, water bodies in Nunavut territory in northern Canada were mapped using a long-term record of remotely sensed observations at 30 m spatial resolution from the Landsat suite of instruments. The temporal trajectories of water extent between 1985 and 2015 were assessed. Over 675,000 water bodies have been identified over the 31-year study period with over 168,000 showing a significant (p < 0.05) trend in surface area. Approximately 55% of water bodies with a significant trend were increasing in size while the remaining 45% were decreasing in size. The overall net trend for water bodies with a significant trend is 0.009 ha per year per water body.

Associated Project: Carroll-01

14:

Impacts of thawing permafrost at the Scotty Creek Research Station, NWT, Canada

Ryan Connon, Wilfrid Laurier University, rfconnon@gmail.com
Élise Devoie, University of Waterloo, egdevoie@uwaterloo.ca
Ashley Rudy, Wilfrid Laurier University, arudy@wlu.ca
Caren Ackley, Wilfrid Laurier University, ackl2230@mylaurier.ca
Olivia Carpino, Wilfrid Laurier University, ocarpino@wlu.ca
Michael Braverman, Wilfrid Laurier University, michael.braverman@ghd.com
Bill Quinton, Wilfrid Laurier University, wquinton@wlu.ca

Rapid climate warming in northwestern Canada has led to unprecedented rates of permafrost thaw. Near its southern edge, the occurrence of permafrost is often restricted to peatlands, where the large thermal offset between the ground surface and permafrost table preserves permafrost when mean annual temperatures approach, and sometimes exceed, 0°C. The Scotty Creek Research Station (SCRS; 61.3°N, 121.3°W) was established in 1999 to investigate the controls that permafrost exerts on hydrology and other ecosystem functions in peatland terrains underlain by permafrost. The SCRS is located in the headwaters of the Scotty Creek watershed (approximately 50 km south of Fort Simpson, NT). Here, permafrost exists solely beneath forested peat plateaus, which are interspersed with permafrost-free wetlands (channel fens and flat bogs), where saturated conditions typically limit the development of trees. Permafrost at the SCRS has been thawing rapidly over the past 60 years, and has changed the cycling and storage of water and solutes, transforming biophysical land cover units from forested, permafrost-cored peat plateaus to wetlands. A similar trend is observed using a space-for-time approach along a North-South transect from Northern BC to the SCRB. The processes that give rise to permafrost thaw are initiated by either climate warming or natural or anthropogenic disturbances. Within plateau systems, climate warming-induced thaw has led to the development of significant talik networks (perennially thawed regions in a permafrost environment) that allow for the perennial storage and transport of mass (water) and energy. The impact of these talik networks on the stability of underlying permafrost and the transport of water and solutes is not yet well understood. In the Northwest Territories, the most common natural disturbances are forest fires, a phenomenon that is becoming more frequent as the region warms. A relatively small fire occurred at the SCRS in 2014, and has been instrumented and monitored to assess the impacts on permafrost stability and changes to the physical structure of the near-surface peat. Anthropogenic disturbances in the region often take the form of linear cut lines used for either winter roads or seismic exploration. The density of these disturbances is approximately 7 times higher than the natural drainage density in the Scotty Creek basin, and involve the complete removal of the tree canopy, a process that leads to the degradation and loss of the underlying permafrost. Permafrost degradation along linear disturbances transforms these features into preferential flowpaths and may fundamentally change the hydrology of the basin. Data collected by the Arctic Boreal Vulnerability Experiment (ABoVE) will play an integral role in enhancing our understanding and improving our capacity to predict the impacts of thawing permafrost in peatlands in subarctic ecosystems. In addition to ongoing monitoring efforts, repeated thaw depth and soil moisture measurements were taken in a defined forest plot in the summer of 2017 to provide ground-truthing data for ABoVE missions.

Associated Project: Quinton-01

15:

Climate change and permafrost thaw-induced boreal forest loss in Canada’s fringe permafrost zone

Olivia Carpino, University of Guelph, ocarpino@uoguelph.ca
Aaron Berg, University of Guelph, aberg@uoguelph.ca
Bill Quinton, Wilfrid Laurier University, wquinton@wlu.ca
Justin Adams, Wilfrid Laurier University, jadams@wlu.ca

Permafrost distribution throughout the Canadian subarctic is not particularly well understood due to a combination of the remoteness and size of the region, spatial and temporal heterogeneity, limited data availability, and incomplete monitoring networks. These factors not only highlight the challenges associated with establishing a comprehensive understanding of the changing distribution of permafrost under the impacts of climate change, but also further emphasize the need to improve techniques of remotely capturing and analyzing permafrost distribution. Landcover, which is highly visible and easily identified through remote sensing data, has been proposed as an emerging method; where forest cover is often indicative of permafrost plateaus, while wetlands are underlain by permafrost-free ground. Recent warming throughout the subarctic boreal peatlands has led to rapid and widespread permafrost degradation and has also corresponded with a significant decrease in forest cover and wetland expansion. This study quantifies landcover change and net forest loss at 10 subarctic boreal peatland sites in the southern Northwest Territories and northeastern British Columbia between 1970 and 2010. Historical air photos and optical remote sensing images were assessed using a change detection approach over 10 square kilometre areas of interest. Variable patterns of net forest loss at each site ranged from 6.9% to 11.6% over the 40-year study period. These differential rates of landcover change can be explained in part through climatic and environmental factors that vary latitudinally across the selected sites. Change statistics – net change, forest gain and forest loss were significantly correlated with an assortment of factors that varied across the 10-site transect.

Associated Project: Quinton-01

16:

Influence of wildfire on the hydrology and runoff chemistry of a peat plateau, Scotty Creek, NWT

Caren Ackley, Wilfrid Laurier University, ackl2230@mylaurier.ca
Bill Quinton, Wilfrid Laurier University, wquinton@wlu.ca
Suzanne Tank, University of Alberta, suzanne.tank@ualberta.ca
Fereidoun Rezanezhad, University of Waterloo, frezanezhad@uwaterloo.ca
Colin McCarter, University of Waterloo, cmccarter@uwaterloo.ca

Wildfire trends throughout the arctic and subarctic show increasing fire frequency, size and severity. Occurrence of wildfire in the zone of discontinuous permafrost can impact the health and stability of boreal peatland ecosystems. Disturbance of the vegetative groundcover and tree canopy influence critical hydrological processes in this region that can lead to ecological changes. As snowmelt provides a significant source of freshwater in northern regions, changing fire regimes coupled with climate warming urgently call for an improved understanding of the impacts of wildfires on water resources. Here, we examine a tree-covered permafrost plateau at Scotty Creek in the Northwest Territories, Canada, where roughly half of the plateau was affected by a localized burn in July 2014. This study provides key insights into both the hydrological and geochemical impacts of wildfires. Water samples and field measurements were collected between April and August 2016 within study plots located in adjacent burned and unburned areas. These measurements were complimented by laboratory analysis of peat cores to examine the impacts of fire on physical, hydraulic and solute transport properties of soils within the plots. Increased SWE and melt rate in the burned area lead to a greater volume of runoff and earlier time to snow-free. Greater depth of seasonal thaw under the burn indicates permafrost degradation and flow path alteration. The concomitant assessment of changes in peat physical and hydraulic properties offer new insight into the mechanisms governing the partitioning of water between runoff and storage, and water quality changes that result from wildfire.

Associated Project(s):
Quinton-01
Tank-01

17:

Influence of Topography and Disturbance on Water and Energy Flow in the Active Layer

Taylor Sullivan, University of Wyoming, tsulli12@uwyo.edu
Andrew Parsekian, University Of Wyoming, aparseki@uwyo.edu
Kevin Schaefer, National Snow and Ice Data Center, kevin.schaefer@nsidc.org
Roger Michaelides, Stanford University, rmich@stanford.edu

We examine how relative elevation influences Active Layer Thickness and soil moisture on North Ridge—a slope near Toolik Lake, AK—by investigating the relationship of ALT and elevation above the bottom of a ravine. We measured ALT and soil moisture using ALT probes, Ground Penetrating Radar (GPR), Electromagnetic Induction (EMI), and hand-held soil moisture probes to learn about ALT and soil moisture trends on hill slopes. These data, in conjunction with ABoVE airborne measurements and digital elevation data, will help inform and validate Remotely Sensed Active Layer Thickness techniques. This study addresses the science question: can a relationship between ALT and slope gradient improve ReSALT calculations? This study hypothesizes that a correlation between ALT and hill slope characteristics can be predicted in permafrost regimes, and this implemented correlation can improve ReSALT models. In the case of North Ridge, we observe a 35% decrease in soil moisture as elevation increases 93 meters above a wet ravine. We will use these and similar relationships to create topographically-based correction factors to remotely sensed retrievals of ALT.

[poster]

Associated Project: Schaefer-06

18:

The Permafrost Dynamics Observatory (PDO)

Kevin Schaefer, National Snow and Ice Data Center, kevin.schaefer@nsidc.org
Albert Chen, JPL, Albert.C.Chen@jpl.nasa.gov
Jingyi Chen, University of Texas, jingyi.ann.chen@utexas.edu
Richard Chen, University of Southern California, chenrh@usc.edu
Kazem Dogaheh, University of Southern California, bakiando@usc.edu
Elchin Jafarov, Los Alamos National Laboratory, elchin@lanl.gov
Lin Liu, University of Hong Kong, liulin@cuhk.edu.hk
Roger Michaelides, Stanford University, rmich@stanford.edu
Mahta Moghaddam, University of Southern California, mahta@usc.edu
Andrew Parsekian, University Of Wyoming, aparseki@uwyo.edu
Alireza Tabatabaeenejad, University of Southern California, alirezat@usc.edu
Jeffery Thompson, University of Colorado, jeffery.a.thompson@colorado.edu
Howard Zebker, Stanford University, zebker@stanford.edu

The Permafrost Dynamics Observatory (PDO) combines backscatter measurements of soil moisture with surface deformation measurements using Interferometric Synthetic Aperture Radar (InSAR). The Remotely Sensed Active Layer Thickness (ReSALT) product uses the L-band InSAR to measure seasonal ground subsidence and active layer thickness. Airborne Microwave Observatory of Subcanopy and Subsurface (AirMOSS) uses p-band backscatter to measure soil moisture. Together, they form a powerful tool to study permafrost dynamics. We show how to use radar to study processes that dominate the permafrost landscape: active layer development, thermokarst, hydrology, and fire. We present examples of products in Alaska to highlight the untapped potential of the combined InSAR and backscatter technique to understand permafrost dynamics, with a strong emphasis on the underlying processes that drive change.

[poster]

Associated Project(s):
Schaefer-03
Schaefer-04
Schaefer-05
Schaefer-06
Tabatabaeenejad-01

19:

Estimation of Belowground Biomass and Permafrost Active Layer Properties Using Radar and Lidar Measurements: Progress and Challenges

Richard Chen, University of Southern California, chenrh@usc.edu
Alireza Tabatabaeenejad, University of Southern California, alirezat@usc.edu
Kazem Bakian Dogaheh, University of Southern California, bakiando@usc.edu
Mahta Moghaddam, University of Southern California, mahta@usc.edu

This project advances methods of retrieving belowground biomass and permafrost active layer properties by developing accurate radar scattering and inverse scattering models of layered ground that include a combination of organic and mineral soils, overlain by vegetation, and containing vegetation roots. We use radar forward scattering models to develop corresponding methods of simultaneously retrieving root biomass, permafrost active layer thickness (ALT), and soil moisture profiles for both organic and mineral layers using the ABoVE Foundational Airborne Measurements, particularly P-band (70 cm) and L-band (24 cm) measurements from the AirMOSS and UAVSAR instruments, respectively. Our team had a leading role in planning the flight lines for the L-band UAVSAR and the P-band AirMOSS radar instruments in coordination with the ABoVE Science Team, in particular the radar Working Group and JPL. The Alaska lines used the legacy flights lines from Mahta Moghaddam’s IDS project as a baseline, adding several more lines to accommodate the requirements for the science proposed by other selected radar teams. The P- and L-band campaigns were flown 2 weeks apart in both the spring and summer time frames (due to unavailability of one of the data recorders). Nearly all of the planned lines were flown and radar data were successfully acquired. We will present an overview of the flown lines and the corresponding sensors. We also carried out a 7-day fieldwork at 4 different sites (Coldfoot forest, Imnavait Creek Watershed near Toolik Lake, Happy Valley near Sagwon, and Prudhoe Bay) along Alaska’s Dalton Highway in August 2017 and measured active and organic layer thicknesses, soil moisture, vegetation parameters, and soil roughness. Preliminary analysis of the collected field data will be presented. Moreover, we will present the latest advances to our radar scattering model, which include two new approaches: (1) considering three layers to represent the organic-to-mineral soil transitions more accurately, and (2) characterizing soil texture and moisture vertical profile heterogeneity with continuous profile functions. We investigate the optimal profile function by minimizing the error between predicted and measured radar backscatter signals as well as between in-situ and fitted profiles. Finally, based on our findings so far, there is no one reliable organic soil dielectric model that can be used at both P and L bands across the ABoVE sites. Therefore, while at this point, we are inclined to use Mironov model as a starting point to proceed with the planned retrievals in 2018, we have come to conclusion that we need to consider development of our own model based on the soil samples collected during the August 2017 fieldwork and possibly a future fieldwork. We will present an overview of this investigation as well.

Associated Project(s):
Moghaddam-03
Tabatabaeenejad-01

28:

SoilSCAPE Sensor Network in Alaska: Studying Permafrost Active Layer Dynamics in the Arctic

Richard Chen, University of Southern California, chenrh@usc.edu
Agnelo Silva, Decagon Devices, Inc., agnelors@gmail.com
Kazem Bakian Dogaheh, University of Southern California, bakiando@usc.edu
Alireza Tabatabaeenejad, University of Southern California, alirezat@usc.edu
Mahta Moghaddam, University of Southern California, mahta@usc.edu

The Soil moisture Sensing Controller And oPtimcal Estimator (SoilSCAPE) is an ultra-low-power wireless sensor network technology, developed under NASA/ESTO Advanced Information Science Technology (AIST) support, for measurement of soil properties including soil moisture and temperature profiles. The SoilSCAPE network technology, initially developed to support calibration and validation activities of NASA missions such as SMAP and AirMOSS, has been enhanced in several ways to allow its operation in the arctic environment in support of ABoVE. The redesign included various power management and environmental robustness considerations for operating the network in the harsh cold and dark arctic environment, as well as redesigning the hardware and software communications interfaces to seamlessly integrate with satellite (Iridium) links. Data compression schemes and dynamic scheduling were also implemented to optimize cost vs. science information content of the data delivered to our gateway at USC. Two separate networks were installed in the Alaska North Slope with a total of 13 network nodes. These new SoilSCAPE-arctic networks in Happy Valley and Prudhoe Meadow have been delivering near-real-time high-quality in-situ field data, including dielectric constant, electrical conductivity, and temperature throughout the permafrost active layer soil profile. With the novel power management and robust environmental design strategies that we have now tested under a number of dynamic environmental conditions, we expect to achieve sustained unattended network operation for 1- 2 years. What we have learned from the data has been unique: For example, our data have directly shown that the permafrost active layer freeze-thaw transition involves much more than a temperature transition, and that instead it is marked very strongly by a transition in the dielectric constant and conductivity of the soils. During the extended zero-curtain period of time in the Fall, even though the temperatures hover around zero-C, the soil still keeps a relatively high dielectric constant and conductivity. We have shown that the final transition to frozen state is indicated by a sharp switch in dielectric constant and conductivity. The SoilSCAPE-arctic data are also being used directly for validation of radar-derived retrievals of active layer thickness and soil moisture. The data are shedding light on the soil profile spatial and temporal dynamics that are essential not only for accurate formulation of radar retrieval models, but also for parameterizing recently developed process models for characterizing distributions and dynamics of the active layer.

Associated Project(s):
Moghaddam-03
Tabatabaeenejad-01

42:

Retrieving Soil Moisture from Satellite Microwave Sensors for Fire Danger Assessment in Boreal and Arctic Regions

Laura Bourgeau-Chavez, Michigan Tech Research Institute, lchavez@mtu.edu
Michael Battaglia, Michigan Technological University, mjbattag@mtu.edu
William Buller, Michigan Tech Research Institute, wtbuller@mtu.edu
Michael Billmire, Michigan Tech Research Institute (MTRI), mgbillmi@mtu.edu
Kyle McDonald, The City College of New York, kmcdonald2@ccny.cuny.edu
Chele Hanes, Canadian Forest Service, chelene.hanes@canada.ca
Liza Jenkins, Michigan Tech Research Institute (MTRI), lliverse@mtu.edu
John Kimball, University of Montana, john.kimball@mso.umt.edu
Randi Jandt, University of Alaska, rjandt@alaska.edu
Joseph Buckley, Royal Military College of Canada, buckley-j@rmc.ca

In Alaska and Canada, the Canadian Forest Fire Danger Rating System (CFFDRS) is used to estimate moisture in the organic soil layers of the ground from the surface (nominally 1.2 cm) down to deeper, more compact organic layers of the duff horizons (10-20 cm depth). The moisture status of the organic soil is a key driver of the potential for wildfire and C- and L-band Polarimetric SAR as well as NASA’s Soil Moisture Active Passive (SMAP) satellite sensor have potential to provide complementary data to CFFDRS. As a weather-based point source system, CFFDRS has inherent limitations that could be greatly improved with synoptic moisture information from a satellite sensor at high repeat frequency but low spatial resolution, such as SMAP, as well as complementary high spatial resolution but low repeat frequency SAR data. Research is underway to assess the utility of the L-band 9 & 36 km resolution SMAP moisture products for organic layer fuel moisture monitoring in boreal and Arctic ecosystems for fire danger prediction including: a) comparison to the broadscale network of weather-based CFFDRS fuel moisture estimates; b) soil moisture databases; and d) the actual occurrence of wildfire. At the same time, algorithm development is underway to use PolSAR data to predict organic layer fuel moisture in a variety of boreal and arctic ecosystems. Using polarimetric data has been shown to improve moisture retrieval by more than 35% over using only the backscatter information from multiple bands. In addition, a comparison of the passive SMAP data with high-resolution SAR imagery is being evaluated to: a) address the impact of scene heterogeneity and surface water on SMAP results; and b) investigate methods for downscaling SMAP to a finer resolution (0.2 to 3 km) soil moisture product through development of hydrological modeling and integration of high resolution SAR data from Sentinel-1 and/or PalSAR-2. The overall investigation will yield a more complete understanding of the relationship between field measurements, the CFFDRS Fire Weather Index, SAR and SMAP soil moisture. A key goal of this project is the development of a refined assessment of fire danger for boreal and Arctic regions. The outcomes of this project will be valuable for fire management and prediction across the vast region where weather-based information is scarce. While the research is focused on improving modeling of fire danger to understand spatial and temporal patterns of organic soil moisture in high northern latitude ecosystems, something that has not been able to be monitored synoptically before now, the impact of this research extends beyond fire decision-making into needs for ecosystem modeling and monitoring climate change vulnerability.

Associated Project: Bourgeau-Chavez-04

44:

The Changing Cold Regions Network: Observation, Diagnosis, and Prediction of Environmental Change in the Saskatchewan and Mackenzie River Basins, Canada

Chris DeBeer, Changing Cold Regions Network, chris.debeer@usask.ca
Merritt Turetsky, University of Guelph, mrt@uoguelph.ca
Howard Wheater, University of Saskatchewan, howard.wheater@usask.ca

The cold interior of Northwestern Canada has one of the world's most extreme and varied climates and, as with other regions across the Arctic, is experiencing rapid environmental change. The Changing Cold Regions Network (CCRN; www.ccrnetwork.ca) is a Canadian research network devoted to addressing key challenges and globally-important issues facing the Arctic by improving the understanding of past and ongoing changes in climate, land, vegetation, and water, and predicting their future integrated responses, with a geographic focus on the Saskatchewan and Mackenzie River Basins. The network is funded for 5 years (2013-18) by the Natural Science and Engineering Research Council of Canada, and combines the unique expertise of over 40 Canadian scientists representing 8 universities and 4 Federal government agencies, as well as 15 international researchers from the United States, China, Australia, the UK, France, and Germany. CCRN has been integrating existing and new experimental data with modelling and remote sensing products to understand, diagnose and predict changing land, water and climate, and their interactions and feedbacks, for Northwestern Canada’s cold interior. It uses a network of world class Water, Ecosystem, Cryosphere and Climate (WECC) Observatories to study the detailed connections among changing climate, ecosystems and water in the permafrost regions of the Sub-arctic, the Boreal Forest, the Western Cordillera, and the Prairies. More detailed information can be found at http://ccrnetwork.ca/science/index.php. The network is strongly linked to several key international organizations and projects. Under the auspices of the Global Energy and Water Exchanges (GEWEX) project of the World Climate Research Programme, CCRN constitutes a regional hydroclimate project (RHP)—currently the only active project in North America. CCRN is linked with NASA and the Arctic–Boreal Vulnerability Experiment as an Affiliated Project of the Science Team. In particular, surface observations from CCRN’s WECC Observatories contribute an important source of information to the ABoVE airborne campaign. CCRN also has strong collaborative links with the US National Center for Atmospheric Research regarding the development and application of high resolution weather and climate modelling over western North America. It is expected that the knowledge and tools developed through this research will benefit not only Canada, but also many other countries in cold regions that face similar challenges in the face of such uncertainty. In particular CCRN, and its larger follow-on project, the Global Water Futures Programme (GWF; https://gwf.usask.ca/), welcome the opportunity for broader collaboration with the international arctic research community.

Associated Project: Wheater-01

53:

Understanding the interplay between permafrost conditions and groundwater flow in boreal headwater catchments, interior Alaska (USA)

Michelle Walvoord, U.S. Geological Survey, walvoord@usgs.gov
Clifford Voss, U.S. Geological Survey, cvoss@usgs.gov
Brian Ebel, U.S. Geological Survey, bebel@usgs.gov
Burke Minsley, U.S. Geological Survey, bminsley@usgs.gov

Permafrost distribution exerts a major control on subsurface flowpaths and fluxes of water and dissolved constituents. This study examines linkages between permafrost conditions and groundwater flow in boreal headwater catchments, where subsurface flow is primarily limited to the active layer. We use an integrated approach to assimilate findings from headwater catchment sites located along a north to south gradient in the boreal region of interior Alaska (USA) with coupled heat transfer and fluid flow modeling using the USGS SUTRA-Ice code. The study sites represent a range of shallow subsurface geology, fire history, permafrost, and air temperature conditions. Field characterization of soil hydraulic and thermal properties constrain sensitivity analyses designed to assess hillslope conditions conducive to active layer thickening and talik development and to quantify rates of change. Geophysical data, including borehole nuclear magnetic resonance (NMR) and electrical resistivity tomography (ERT), provides information relevant to permafrost conditions at the study sites and identifies areas of degrading permafrost and talik development. Model simulations reproduce temporal patterns of liquid water content as shown in the NMR data and spatial patterns of permafrost as inferred from the ERT data. Groundwater flux results highlight the impact of thaw-induced changes on baseflow and terrestrial-aquatic transport of dissolved species.

Associated Project: Striegl-01

64:

Integrated evaluation of the vulnerability to thermokarst disturbance and its implications for the regional carbon balance in boreal Alaska

Helene Genet, Institute of Arctic Biology, hgenet@alaska.edu
Heather Greaves, Institute of Arctic Biology, hegreaves@alaska.edu
Mark Lara, University of Alaska, Fairbanks, mjlara@alaska.edu
A. McGuire, USGS, admcguire@alaska.edu
Bob Bolton, University of Alaska, Fairbanks, bbolton@iarc.uaf.edu
Eugenie Euskirchen, University of Alaska, Fairbanks, seeuskirchen@alaska.edu
Vladimir Romanovsky, University of Alaska, Fairbanks, veromanovsky@alaska.edu

Our capacity to project future ecosystem trajectories in northern permafrost regions depends on our ability to characterize complex interactions between climatic and ecological processes at play in the soil, the vegetation, and the atmosphere. We present a study that uses remote sensing analyses, field observations, and data synthesis to inform models for the prediction of ecosystem responses to climate change in the boreal zone of Alaska. Recent warming, altered precipitation and fire regimes are driving permafrost degradation, threatening to mobilize vast reservoirs of ancient carbon previously protected from decomposition. Although large scale, progressive, top-down permafrost thaw have been well studied and represented in high-latitude ecosystem models, the consequences of abrupt and local thermokarst disturbances (TK) are less well understood. To fill this gap, we conducted a detection analysis characterizing 60 years of land cover change in the Tanana Flats, a wetland complex subjected to TK disturbance in Interior Alaska, using aerial and satellite images. We observed a nonlinear loss of permafrost plateau forest associated with TK and driven by precipitation and forest fragmentation. The results of this analysis were integrated into the Alaska Thermokarst Model (ATM), a state-and-transition model that simulates land cover change associated with TK disturbance. Thermokarst-related land cover change was simulated from 2000 to 2100 across the Tanana Flats. By 2100, the model predicts a mean decrease of 26.5% (sd 7.2%) in permafrost plateau forests associated with an increase in TK fens and bogs. Transitions from permafrost plateau forests to TK wetlands are accompanied with changes in physical and biogeochemical processes affecting ecosystem carbon balance. We evaluated the consequences of TK disturbances on the regional carbon balance by coupling outputs from the ATM and from a process-based biogeochemical model. We used long-term field observations of vegetation and soil physical and biogeochemical attributes to develop new parameterizations for TK wetlands and permafrost plateau forest land cover types. Preliminary simulations from 2000 to 2100 estimate that the conversion of permafrost plateau forest to young TK wetlands would result in a 28.0% (sd 3.5%) increase in Net Ecosystem Exchange.

Associated Project: Genet-01

65:

Quantification of Thermokarst and Carbon Release: Field Surveys

Go Iwahana, University of Alaska Fairbanks, giwahana@alaska.edu
Reginald Muskett, University of Alaska Fairbanks, reginald.muskett@gmail.com
Robert Busey, International Arctic Research Center, rcbusey@alaska.edu
Kazuyuki Saito, JAMSTEC, ksaito@jamstec.go.jp
Seungbum Kim, JPL, Seungbum.Kim@jpl.nasa.gov
Stan Wullschleger, Oak Ridge National Laboratory, wullschlegsd@ornl.gov

Permafrost degradation includes the mobilization of water, melting of ground ice and release of greenhouse gases and organic carbon that had once been stored on regional scales. It is of great scientific interest and social concern to know where and to what extent permafrost degradation may occur, especially in ice-rich permafrost land where subsidence by thaw (thermokarst) will cause large changes in surface ecology, landscape evolution and hydrological processes and will also affect human life and subsistence. Lack of knowledge about the fate of permafrost and the amount of carbon release represents a major source of uncertainty for future climate projection. In order to improve the qualification of thermokarst development and to establish a new approach to monitoring carbon release upon permafrost thaw we proposed multiple remote sensing techniques and field observations together with detailed permafrost sample analysis. During the thawing season in 2017 we conducted field surveys in four areas in Alaska including the Anaktuvuk River Fire scar, North Dalton Highway, coastal land near Utqiagvik (Barrow) and Kougarok in the Seward Peninsula. These areas cover various ages and spatial extent of permafrost with a wide range of ice contents. Our fieldwork was designed to obtain surface frozen soil cores and acquire datasets of thaw depth, ground micro-topography and surface moisture along newly established survey transects at each site to validate surface subsidence due to thermokarst by airborne and spaceborne RADAR remote sensing. In this presentation I will give an overview about our 2017 fieldwork and future plans including our preliminary results regarding the spatial variations in seasonal surface displacement (thaw settlement).

Associated Project(s):
Iwahana-01
Wullschleger-01

67:

Identifying landscape changes and associated impacts on lakes and rivers in Old Crow Flats, Yukon, Canada

Kevin Turner, Brock University, kturner2@brocku.ca
William Thorne, Brock University, bt10yi@brocku.ca
Daniel Hughes, Brock University, dan.hughes@brocku.ca
Mohammad Ahmed, University of Calgary, mohammad.ahmed2@ucalgary.ca
Ian McDonald, Parks Canada, ian.mcdonald@pc.gc.ca

Old Crow Flats (OCF), Yukon, Canada is a lake-rich permafrost landscape that is regarded for its rich ecology and cultural heritage. Local observations of landscape changes include drastically fluctuating lake water levels, unpredictable weather, increasing shrub growth, fire and shoreline slumping. The influence of these changes on lake and river hydroecological conditions and carbon mobility remains uncertain. Our research aims to refine our knowledge of the influence of changing catchment characteristics on lake and downstream conditions so that we can improve predictions of how the hydrology and biogeochemistry (e.g., carbon mobility) will respond in the future. We are integrating findings from a decade of lake and river hydrological monitoring with additional field measurements of lake and landscape conditions, and several spaceborne and airborne remote-sensing products. Recent landscape change mapping initiatives include 1) developing a shrub proliferation map for the entire OCF 14,500-km2 watershed, 2) surveying active layer and vegetation properties at six plots of varying land cover types, 3) utilizing UAV images and photogrammetry to quantify sediment and nutrient export from retrogressive thaw slumps to the Old Crow River, and 4) identifying post-drainage lake hydrological and limnological responses. The hydrology and biogeochemistry of 22 creeks and 14 lakes are being monitored using water isotope tracers (δ13C DIC and DOC, δ18O, and δ2H) and a suite of water chemical parameters including DIC and DOC concentrations. Mapping of coupled results are elucidating the relative importance of different thermokarst geomorphic processes on exported water quality and carbon concentrations, including the influence of lake drainage and retrogressive thaw slumping along the Old Crow River network. Field measurements will be used to evaluate the landscape and hydrology modeling utility of several remote sensing products from the 2017 NASA-ABoVE Airborne Campaign. Comprehensive knowledge generated here will be valuable for informing northern stakeholders and land management strategies as climate continues to change.

Associated Project: Turner-K-01

70:

Augmentation of the USArray sites with temperature profilers

Dmitry Nicolsky, University of Alaska Fairbanks, djnicolsky@alaska.edu
Vladimir Romanovsky, University of Alaska Fairbanks, veromanovsky@alaska.edu

The ground temperature variability across the Arctic landscape depends on the air temperature, snow cover, moisture content, vegetation, terrain, soil properties, and related environmental variables. A juxtaposition of all these factors results in a highly heterogeneous distribution of the ground temperature, active layer thickness and permafrost conditions. As a result, prediction of subsurface temperature dynamics remains challenging, and mean temperatures for a study region may not account for "hot spots" of change, which alone could significantly contribute to thaw and associated carbon emissions. A solution is to record temperature regimes within different ecotypes, temperature and precipitation conditions, and build a portfolio of subsurface thermal regimes across various ground conditions. The ground temperature profilers installed across Alaska at the USArray sites supplement the existing data loggers and provide means to sample the ground temperature regime in currently underrepresented ecotypes and increase our knowledge of permafrost variability across Alaska and Northern Canada.

[poster]

Associated Project: Nicolsky-01

72:

Multi-site observations of near-surface soil temperature in northwestern Canada: implications for high-resolution permafrost mapping

Yu Zhang, Canada Centre for Mapping and Earth Observation, yu.zhang@canada.ca
Ridha Touzi, Canada Centre for Mapping and Earth Observation, ridha.touzi@canada.ca
Wanpeng Feng, Canada Centre for Mapping and Earth Observation, wanpeng.feng@canada.ca
Gang Gong, Canada Centre for Mapping and Earth Observation, gang.hong@canada.ca
Trevor Lantz, University of Victoria, tlantz@uvic.ca
Steve Kokelj, GNWT Geological Survey, steve_kokelj@gov.nt.ca
Wenjun Chen, Canada Centre for Mapping and Earth Observation, wenjun.chen@canada.ca

Permafrost temperature observations are sparse, which limits our understanding and mapping permafrost at landscape scale. From August 2016 to August 2017, we measured near-surface (5 cm depth) soil temperatures (Tnss) at 107 sites around Inuvik and Tuktoyaktuk in northwestern Canada. They are in northern boreal area and low-arctic tundra, respectively. The data show strong site variations and allow us to quantify the statistics about their relations with ecotopes. The site variation of Tnss mainly occurs in snow-cover period, indicating the importance of snow. Ecotopes are effective for stratifying Tnss in snow-cover period and annual mean Tnss, but not for Tnss in thawing months. Ecotopes are useful to stratify organic layer thickness and active-layer thickness (ALT) in low-arctic tundra, but not in northern boreal area. The site variation of ALT within each area does not correlate with thawing season Tnss but is determined by the edaphic factor. There is a significant and positive correlation between ALT and Tnss in snow-cover period in both areas. The soil near the highway is warmer due to snow accumulation near the embankment and changes of land cover during highway construction. These results are useful for understanding and mapping permafrost at landscape scale and for assessing the impacts of highway on permafrost.

Associated Project(s):
Chen-W-01
Touzi-01

76:

Thermokarst in pingos and adjacent collapse scar bogs in interior Alaska

Thomas Douglas, US Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, thomas.a.douglas@usace.army.mil
Merritt Turetsky, University of Guelph, mrt@uoguelph.ca

A region of discontinuous permafrost 50 kilometers southeast of Fairbanks, Alaska exhibits rapid thermokarst and landscape change. The area contains a dozen pingos (hydrolaccoliths), mounds of ice covered by earth material typically 100 meters across and 20 meters above the surrounding ground surface. The pingos have sunken craters in their centers formed through melting and subsequent collapse of an inner ice lens core. Adjacent to the pingos are collapse scar bogs in various states of formation and ice wedge terrain undergoing active thaw subsidence to polygons and thermokarst mounds (baydzherakhs). With a mean annual temperature of -1 degree C the area contains warm ecosystem-protected permafrost highly vulnerable to thaw triggered by changes in temperature, precipitation, or vegetation. We analyzed historical imagery to the 1970s to track water features in a subset of pingos. The craters appear to have expanded over the past few decades suggesting melting and collapse of the ice cored center and potential permafrost degradation along pingo margins. Collapse scar bogs in adjacent low-elevation terrain are roughly the same size as the pingos but have little vertical elevation gradient compared to the surrounding terrain and are barely discernable in airborne LiDAR imagery. Electrical resistivity tomography (ERT) measurements, high resolution GPS surveys, SIPRE coring, and thaw depth probing were focused along nine 100-400 meter transects across three of the pingos to identify relationships between geophysical properties, permafrost composition, seasonal thaw, and ecological state. A large (~40 meters across and 20 meters thick) lens shaped region of thawed permafrost is evident in the ERT results about 10 meters below the ground surface in the center of one pingo we surveyed in detail. This is believed to be the original ice cored region of the pingo that has melted. A thin (1-5 meters thick) layer of permafrost is present above this thawed region while the rampart margins surrounding the pingo are underlain by thick (10-30 m) permafrost. The pingo and thermokarst features reside in a location where rapid permafrost thaw in response to warming or changing hydrology could provide a hot spot for landscape change, particularly given a projected climate warming of 5 degrees C over the next 80 years in the area. Their future thermal, geomorphological, and ecological states may be a harbinger for how discontinuous permafrost in the region responds to projected climate warming.

Associated Project: Douglas-01

77:

CRREL Permafrost Field Sites in Interior Alaska

Thomas Douglas, US Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, thomas.a.douglas@usace.army.mil

For the past five years, through a variety of projects and collaborations, we have been developing research infrastructure at a variety of sites around Fairbanks, Alaska. We will invest in these sites further over the next ten years. Sites are road accessible and represent a variety of ecotype and terrain types common in the boreal biome of interior Alaska. The goal is to accumulate a long term record of permafrost geomorphology, seasonal thaw, ground thermal state, vegetation and ecological processes, hydrology, snowpack composition, geophysical properties, and airborne and remote sensing characteristics. CRREL owns and operates facilities at two of the sites (Permafrost Tunnel and Farmer’s Loop) and these are suitable for manipulation experiments, installing permanent equipment, and establishing long-term measurements. We have acquired airborne LiDAR over all sites and at some we have 2 or 3 years of measurements. We have acquired airborne hyperspectral measurements at all sites. Each fall we measure seasonal thaw along 500 meter transects across each site. Thermistors measuring ground temperatures have been continuously measuring at each site for the past 3-5 years. For most sites we have deep (15 meter) boreholes and have measured electrical resistivity tomography and ground penetrating radar across site transects. End of season snow pack measurements are made at the sites.

Associated Project: Douglas-01

80:

InSAR Experiments in Arctic Alaska

Reginald Muskett, University of Alaska Fairbanks, reginald.muskett@gmail.com

Geodetic methods to measure centimeter to millimeter-scale changes using aircraft- and spacecraft deployed Synthetic Aperture RADAR cannot ignore volume scattering. Backscatter and coherence at L-frequency and others possess both surface and volumetric scattering. On lowland tundra underlain by permafrost volume scattering is dominant. Measurement of the L-frequency penetration depth for evaluation of mass change (loss and transport) through permafrost thaw-degradation with erosion is necessary. Data from the NASA Ice, Cloud, and land Elevation Satellite Geoscience Laser Altimeter System (ICESat GLAS), JAXA Advanced Land Observing Satellite Phased Array type L-frequency Synthetic Aperture RADAR (ALOS PALSAR), aircraft-deployed NASA L-frequency UAVSAR and in-situ observations are employed. Collocation of ICESat GLAS exact-repeat profiles for elevation change (surface scattering) with PALSAR InSAR Line-Of-Sight changes (volume scattering) and UAVSAR Polarimetry Cross-Pole HHVV (volume scattering) confirms the dominance of volume scattering on lowland tundra (RADAR soft targets) and surface scattering on river channel deposits and rock outcrops (RADAR hard targets). NASA NNX17AC57A References: Muskett, R.R. (2015), ICESat GLAS Elevation Changes and ALOS PALSAR InSAR Line-Of-Sight Changes on the Continuous Permafrost Zone of the North Slope, Alaska. International Journal of Geosciences, 6 (10), 1101-1115. doi:10.4236/ijg.2015.610086 Muskett, R.R. (2017), L-Band InSAR Penetration Depth Experiment, North Slope Alaska. Geoscience and Environment Protection, vol. 5, no.3, p. 14-30. doi: 10.4236/gep.2017.53002.

Associated Project: Iwahana-01

83:

Tracking seasonal evolution of surface water extent in the Yukon Flats, Alaska and the Canadian Shield

Sarah Cooley, Brown University, sarah_cooley@brown.edu
Laurence Smith, UCLA, lsmith@geog.ucla.edu
Lincoln Pitcher, UCLA, lincolnpitcher@g.ucla.edu
Tamlin Pavelsky, University of North Carolina Chapel Hill, pavelsky@unc.edu
Simon Topp, University of North Carolina, sntopp@live.unc.edu

Quantifying spatial and temporal variability in surface water storage at high latitudes is critical for assessing environmental sensitivity to climate change. Traditionally the tradeoff between high spatial and high temporal resolution space-borne optical imagery has limited the ability to track fine-scale changes in surface water extent. However, the recent launch of hundreds of earth-imaging CubeSats by commercial satellite companies such as Planet opens up new possibilities for monitoring surface water from space. In this study we present a comparison of seasonal evolution of surface water extent in two study areas with differing geologic, hydrologic and permafrost regimes, namely, the Yukon Flats in Central Alaska and the Canadian Shield north of Yellowknife, NWT. Using near-daily 3m Planet CubeSat imagery, we track individual lake surface area from break-up to freeze-up during summer 2017 and quantify the spatial and temporal variability in inundation extent. We find that the magnitude and timing of surface water recession varies significantly by watershed. In the Yukon Flats, lakes are more stable over the summer season in continuous permafrost and less stable in discontinuous permafrost. The results of this analysis demonstrate the value of CubeSat imagery for dynamic surface water research particularly at high latitudes and illuminate fine-scale drivers of cold regions surface water extent.

Associated Project: Smith-L-01

84:

Measuring the Impact of Wildfire on Active Layer Thickness in a Discontinuous Permafrost region using Interferometric Synthetic Aperture Radar (InSAR)

Roger Michaelides, Stanford University, rmich@stanford.edu
Kevin Schaefer, National Snow and Ice Data Center, kevin.schaefer@nsidc.org
Howard Zebker, Stanford University, zebker@stanford.edu
Lin Liu, The Chinese University of Hong Kong, liulin@cuhk.edu.hk
Jingyi Chen, University of Texas at Austin, jingyi.ann.chen@utexas.edu
Andrew Parsekian, University Of Wyoming, aparseki@uwyo.edu

In permafrost regions, the active layer is defined as the uppermost portion of the permafrost table that is subject to annual freeze/thaw cycles. The active layer plays a crucial role in surface processes, surface hydrology, and vegetation succession; furthermore, trapped methane, carbon dioxide, and other greenhouse gases in permafrost are released into the atmosphere as permafrost thaws. A detailed understanding of active layer dynamics is therefore critical towards understanding the interactions between permafrost surface processes, freeze/thaw cycles, and climate-especially in regions across the Arctic subject to long-term permafrost degradation. The Yukon-Kuskokwim (YK) delta in southwestern Alaska is a region of discontinuous permafrost characterized by surface lakes, wetlands, and thermokarst depressions. Furthermore, extensive wildfires have burned across the YK delta in 2006, 2007, and 2015, impacting vegetation cover, surface soil moisture, and the active layer. Using data from the ALOS PALSAR, ALOS-2 PALSAR-2, and Sentinel-1A/B space borne synthetic aperture radar (SAR) systems, we generate a series of interferograms over a study site in the YK delta spanning 2007-2011, and 2014-present. Using the ReSALT (Remotely-Sensed Active Layer Thickness) technique, we demonstrate that active layer can be characterized over most of the site from the relative interferometric phase difference due to ground subsidence and rebound associated with the seasonal active layer freeze/thaw cycle. Additionally, we show that this technique successfully discriminates between burned and unburned regions, and can resolve increases in active layer thickness in burned regions on the order of 10’s of cms. We use the time series of interferograms to discuss permafrost recovery following wildfire burn, and compare our InSAR observations with GPR and active layer probing data from a 2016 summer field campaign to the study site.

[poster]

Associated Project(s):
Schaefer-05
Schaefer-06

87:

AirSWOT Measurements of Water Surface Elevations and Hydraulic Gradients over the Yukon Flats, Alaska

Lincoln Pitcher, UCLA, lincolnpitcher@g.ucla.edu
Tamlin Pavelsky, University of North Carolina Chapel Hill, pavelsky@unc.edu
Laurence Smith, UCLA, lsmith@geog.ucla.edu
Delwyn Moller, Remote Sensing Solutions, dkmoller@remotesensingsolutions.com
Elizabeth Altenau, University of North Carolina Chapel Hill, ealtenau@unc.edu
Christine Lion, University of North Carolina Chapel Hill, clion@unc.edu
George Allen, University of North Carolina Chapel Hill, geoallen@unc.edu
Mark Bertram, US Fish and Wildlife Service, mark_bertram@fws.gov
Sarah Cooley, Brown University, sarah_cooley@brown.edu

AirSWOT is an airborne, Ka-band synthetic aperture radar interferometer (InSAR) intended to quantify surface water fluxes by mapping water surface elevations (WSE). In July and August 2017, as part of the NASA ABoVE airborne campaign, AirSWOT successfully mapped WSEs in lakes and rivers spanning ~20° of latitude. Prior to this, AirSWOT acquired data over the Yukon Flats, Alaska, USA on 15-June-2015. Coincident with this collection, we conducted in situ GPS surveys of WSE. We use this field data to assess the accuracy of AirSWOT WSE measurements in lakes and rivers. Next, we demonstrate that AirSWOT can be used to estimate large-scale hydraulic gradients across wetland complexes. Third, we use AirSWOT to assess how lake levels across the Yukon Flats spatially vary with permafrost presence or absence. Finally, in anticipation of forthcoming processed 2017 AirSWOT data, we present an overview of the custom GPS system we developed for surveying WSEs and that we deployed >100 times in support of 2017 NASA ABoVE airborne campaigns.

Associated Project: Smith-L-01

88:

Alaska Transportable Array: Capabilities and Interdisciplinary Instrumentation

Robert Busby, IRIS--Incorporated Research Institutions for Seismology, busby@iris.edu
Bob Woodward, IRIS--Incorporated Research Institutions for Seismology, woodward@iris.edu
Kasey Aderhold, IRIS--Incorporated Research Institutions for Seismology, kasey@iris.edu

The Alaska Transportable Array (ATA) is a network of 280 autonomous earthquake monitoring stations spaced every ~85 km across Alaska and western Canada. Stations operate continuously, powered by photovoltaic panels and a combination of lithium ion and lead acid batteries. Data is telemetered to the Array Network Facility at UC San Diego in real-time (latency less than 10 seconds in summer, within the hour in winter) and made freely available through the IRIS Data Management Center. Along with high quality seismic data, EarthScope has collaborated with the NASA ABoVE field campaign and other partners to add Vaisala meteorological sensors to 132 of the most remote stations. Sampling rates are 1 sample/second and timing is accurate to within a few milliseconds for all data channels including seismic, pressure, infrasound, and meteorological. Self contained soil temperature probes provided by NASA ABoVE and Yukon Geological Survey were also collocated at 78 of the ATA sites. Although that data is not telemetered, it is collected and archived by the UAF Permafrost group led by Vladimir Romanovsky. The ATA serves as a multi-disciplinary observation platform across a vast region of the Arctic. Stations will continue to be operated through 2019 and then a process of removal will commence. Data collected from the ATA will be used to image the geological structure under Alaska, detect and locate local and regional earthquakes, map out active faults, track rupture patterns of large earthquakes globally, detect volcanic eruptions, record landslides, and monitor weather events. We expect to see methods developed for measuring sea ice extent, detecting ice jams, and other new research applications. Additional instrumentation related to space weather and cameras are considered for future integration if support for such partnerships emerges.

91:

Identifying relations among active layer properties and land cover types in Old Crow Flats, Yukon, Canada

W. Thorne, Brock University, bthorne2@brocku.ca
Kevin Turner, Brock University, kturner2@brocku.ca

Ground-based measurements indicate widespread increasing ground temperatures across northern regions during recent decades. There have also been associated changes in land cover and hydrological conditions across lake-rich landscapes, however, additional studies are required to investigate relations among these integrated landscape components. For instance, it is unclear how warming ground conditions or shrub vegetation proliferation are affecting important northern lake-rich environments. Our research investigates these relationships in Old Crow Flats (OCF), Yukon, which is a lake-rich Arctic landscape widely regarded for its cultural and ecological integrity. Here, we use a combination of remote sensing, in-situ soil moisture and active layer thickness (ALT) probe measurements, and vegetation sampling to investigate ground-land cover characteristics among six plots spanning varying land cover types. Plots were initially grouped according to dominant land cover types including tundra/bog, shrub/spruce, and burned. Preliminary results from 2017 show high late thaw-season variability in ALT among tundra/bog (mean = 25.95 cm), shrub/spruce (mean = 35.79 cm), and burned (mean = 49.57 cm) sites. Soil moisture in-situ volumetric water content (VWC) measurements, with 20 cm probe depth, show similar late-thaw season variability among tundra/bog (mean = 58.77%), shrub/spruce (mean = 43.44%), and burned (mean = 47.55%). Ongoing analysis will incorporate use of unmanned aviation vehicle (UAV) acquired high-resolution aerial photography and additional remote sensing products (acquired as part of NASA’s Arctic Boreal Vulnerability Experiment airborne campaign) to refine maps of lake and river catchments that have been monitored in collaboration with Parks Canada since 2007. Integrated approaches being developed here will enhance our knowledge of the complex relations affecting lake-rich permafrost landscapes as climate continues to change.

[poster]

Associated Project: Turner-K-01

94:

Active layer and water geochemistry data throughout the Yukon River Basin

Ryan Toohey, Alaska Climate Science Center, rtoohey@usgs.gov
Edda Mutter, Yukon River Inter-Tribal Watershed Council, emutter@yritwc.org
Nicole Herman-Mercer, US Geological Survey, nhmercer@usgs.gov
Paul Schuster, US Geological Survey, pschuste@usgs.gov

The hydrology of the Yukon River Basin has changed over the last several decades as evidenced by a variety of discharge, gravimetric, and geochemical analyses. The Indigenous Observation Network (ION), a community-based project, was initiated by the Yukon River Inter-Tribal Watershed Council and the US Geological Survey. Capitalizing on existing USGS monitoring and research infrastructure while supplementing USGS collected data, ION investigates changes in surface water geochemistry and active layer dynamics throughout the Yukon River Basin. Over 1600 samples of surface water geochemistry (i.e., major ions, dissolved organic carbon, and 18O and 2H) have been collected at 35 sites throughout the Yukon River and its major tributaries over the past decade. Active layer dynamics (maximum thaw depth, soil temperature and moisture) have been collected at 20 sites throughout the Yukon River Basin for the past eight years. Important regional differences in geochemistry and active layer parameters linked to permafrost continuity and tributaries will be highlighted. Additionally, annual trends and seasonal dynamics describing the spatial and temporal heterogeneity of the watershed will be presented in the context of observed hydrological changes. These data assist the global effort to characterize arctic river fluxes and their relationship to the carbon cycle, weathering and permafrost degradation.

99:

Arctic lake classification from NASA ABoVE digital color infrared airborne imagery

Ethan Kyzivat, Brown University, ethan_kyzivat@brown.edu
Laurence Smith, UCLA, lsmith@geog.ucla.edu
Sarah Cooley, Brown University, cooleysarahw@gmail.com
Lincoln Pitcher, UCLA, lincolnpitcher@g.ucla.edu
John Arvesen, Cirrus Digital Systems, arvesen@cirrus-designs.com
Tamlin Pavelsky, University of North Carolina Chapel Hill, pavelsky@unc.edu

One of the novel sensors deployed in ABoVE field compaings is AirSWOT, which aims to map surface water elevations in lake, wetland and river environments. The AirSWOT instrument suite includes a Ka-band experimental interferometric synthetic aperture radar and a NASA color infrared (CIR) Digital Camera System (DCS). In July and August 2017, NASA AirSWOT flew from southern Canada to Arctic Alaska, mapping >3,000 km² area including hundreds of lakes underlain by spatially varying geologic and permafrost conditions. This research aims to statistically assess the spatial variability of lake size and elevation changes to determine their sensitivity to broad-scale environmental gradients. To this end, we mapped lakes from 1 m spatial resolution, 16-bit, multiband, CIR orthomosaics. We classified water boundaries using a probabilistic adaptive thresholding technique applied to a normalized difference water index (NDWI). Next, we validated the derived lake water mask using in-situ shoreline maps collected with GPS. Results will identify spatial variability in lake size, distribution and quality across ~20 deg. of latitude. The resulting water mask will also assist in validation of the coincident radar interferometry data. Our findings should enhance the understanding of changing Arctic hydrologic systems and associated landscape controls on surface water variability.

Associated Project: Smith-L-01

102:

Combining AirSWOT with Direct and Isotope Hydrological Measurements for Landscape Scale Assessment of Mackenzie Region Thermokarst Lakes

Evan Wilcox, Wilfrid Laurier University, wilc0150@mylaurier.ca
Philip Marsh, Wilfrid Laurier University, pmarsh@wlu.ca
Branden Walker, Wilfrid Laurier University, bwalker@wlu.ca
Gabriel Hould - Gosselin, Université de Montréal, ghgosselin@gmail.com
Brent Wolfe, Wilfrid Laurier University, bwolfe@wlu.ca

Thousands of small thermokarst lakes cover up to 50% of the landscape across the forest – tundra transition zone between Inuvik and Tuktoyaktuk in the western arctic of Canada. The yearly water balance of these lakes has not been quantified before, nor has the variability in water balance across the region been assessed. Water flux to the atmosphere, permafrost degradation, and limnological conditions are all affected by the lake hydrology, and climate change induced temperature increases and changes to precipitation patterns could affect lake water balances in this region. This necessitates an evaluation of current lake water balance conditions, and an assessment of landscape variability of water balance, in order to predict how the landscape will change in the future. To characterize the water balance of these lakes, direct measurements of lake level, lake discharge, evaporation, basin snow storage, and precipitation were made at two adjacent lakes near the Trail Valley Creek Research Station, 45km N of Inuvik. Each lakes had similar sized catchments, but one lake has a surface area twelve times larger than the other. This larger lake fell below its sill level shortly after snow melt and was only fully recharged by large rainfall events, while the smaller lake was maintained to sill level by runoff the entire year. This lead to the hypothesis that the water balance of a lake can be approximated by dividing the area of its catchment by the area of the lake. To test whether this variable is the dominant control on lake water balance, water levels changes between the two AirSWOT flights in the region are compared to the lake size and its basin size. Stable water isotopes will be sampled from a variety of lakes in the region next year pre-snowmelt, post-snowmelt, and at the end of the summer to describe freeze-up, snowmelt recharge, and rainfall recharge, respectively. This will help describe landscape scale variability while relating results back to the dominant hydrological processes observed from the direct measurements of water balance.

Associated Project: Marsh-01

105:

Investigating Lake Area Dynamics as a Function of Permafrost Degradation Using Airborne Electromagnetic Surveys in Yukon Flats, AK

David Rey, Hydrologic Science and Engineering Program, Colorado School of Mines, U.S. Geological Survey, drey@mymail.mines.edu
Michelle Walvoord, U.S. Geological Survey, walvoord@usgs.gov
Burke Minsley, U.S. Geological Survey, Crustal Geophysics and Geochemistry Science Center, bminsley@usgs.gov
Jennifer Rover, Earth Resources Observation and Science (EROS) Center, U.S. Geological Survey, jrover@usgs.gov
Kamini Singha, Hydrologic Science and Engineering Program, Colorado School of Mines, ksingha@mines.edu

High arctic warming rates are changing the routing and storage of water above and below the ground surface, potentially impacting surface-water dominated regions such as boreal lowlands, where evolving permafrost dynamics are affecting the regional hydrology and biogeochemical cycling of carbon. The Yukon Flats, Alaska, is an expansive boreal lowland primarily underlain by discontinuous permafrost, making the region particularly sensitive to a warming climate. Consequently, the shallow, continuous, fluvial gravel layer existing within the Yukon Flats may be transitioning from relatively impermeable to highly permeable if thawed. This makes the Yukon Flats and other boreal lowlands with high permeability material susceptible to increasing shallow subsurface connectivity between lakes and stream networks as permafrost degrades. Increased subsurface connectivity as a function of permafrost degradation could potentially alter lake dynamics within these regions, perturbing terrestrial surface conditions and changing energy and carbon partitioning in affected boreal lowlands. This study presents a synthesis of new data products and methods with previous work, identifying linkages between permafrost distribution, hydrologic connectivity (surface and subsurface), and decadal-scale trends in surface water dynamics. New Airborne Electromagnetic (AEM) data acquisitions coupled with expanded, co-located Landsat data provided an ice-rich analog to an existing Yukon Flats AEM survey, allowing for comparisons between lakes in more continuous and discontinuous permafrost settings not previously possible. Landsat derived lake surface area time-series behaviors were clustered to correlate individual lake surface area trends, measure lakes proximity to adjacent rivers and streams, and evaluate average lake-surface area extent. Landsat data was coupled with AEM data to identify permafrost distribution underlying the identified clusters, elucidating five physical processes governing observed lake surface-area heterogeneity. Processes driving lake surface area dynamics differed between regions of continuous and discontinuous permafrost, providing important context for projecting future spatial and temporal patterns of lake change in boreal lowlands.

Associated Project: Striegl-01

114:

Quantification of Surface Subsidence Using Real Time Kinematic GPS in Experimentally Warmed Permafrost

Heidi Rodenhizer, Northern Arizona University, hgr7@nau.edu
Marguerite Mauritz, Northern Arizona University, marguerite.mauritz@nau.edu
Susan Natali, Woods Hole Research Center, snatali@whrc.org
Emily Romano, Northern Arizona University, elr88@nau.edu
Elaine Pegoraro, Northern Arizona University, efp23@nau.edu
Meghan Taylor, Northern Arizona University, meghan.taylor@nau.edu
Temuulen Sankey, Northern Arizona University, temuulen.sankey@nau.edu
Edward Schuur, Northern Arizona University, ted.schuur@nau.edu

Permafrost soils contain twice as much Carbon (C) as is currently in the atmosphere and can be composed of a high proportion ice. As temperatures rise, ice loss in permafrost causes subsidence, while warmer soil temperatures cause the release of stored C to the atmosphere. In addition, subsidence has the potential to speed permafrost thaw when water accumulates in subsided areas and speeds heat absorption. Using real-time kinematic GPS, we tracked subsidence at a permafrost warming experiment from 2009-2016 near Healy, Alaska, at a site underlain by continuous permafrost within the discontinuous permafrost zone. Soil warming is achieved by the installation of snow fences that cause snow to drift on the leeward side in winter, providing insulation from cold Arctic air temperatures. Environmental variables such as water table depth (WTD), active layer thickness (ALT), and CO2 fluxes are measured on the control and warming sides of the fence. Additionally, soil moisture and thaw depth were collected along a transect twice in the summer of 2017 as ground-truthing for ABoVE flights. Over the course of the experiment in the warming plots, WTD decreased by 3 times, ALT doubled and CO2 fluxes increased compared to the control plots. At the beginning of the experiment in 2009, slope corrected warming and control plot elevation surfaces showed differences of < 0.7 m between lowest and highest values, with no discernable spatial pattern. By 2016, warming plot elevation surfaces were clearly lower than control plot elevation surfaces, with differences of nearly 1 m between lowest and highest values. Concurrent changes in WTD, ALT, and CO2 fluxes at the site indicate interactions between several soil environmental variables could impact CO2 fluxes. Future research will investigate the ability to model ALT using remotely sensed digital terrain models, NDVI, and soil moisture available through ABoVE and NEON products.

Wildlife and Ecosystem Services

Precipitation as a potential climate driver of NDVI trends in Southwest Alaska

Amy Hendricks, University of Alaska Fairbanks, ashendricks@alaska.edu
Uma Bhatt, University of Alaska, Fairbanks, usbhatt@alaska.edu
Gerald Frost, Alaska Biological Research, Inc.--Environmental Research & Services, jfrost@abrinc.com
Matthew Macander, Alaska Biological Research, Inc.--Environmental Research & Services, mmacander@abrinc.com
Peter Bieniek, University of Alaska, Fairbanks, pbieniek@alaska.edu
Mark Jorgenson, Alaska Ecoscience, ecoscience@alaska.net

Changes to climatic properties such as precipitation are likely to modulate the effects of climate warming on Arctic vegetation. Satellite imagery has indicated a recent decline in vegetation productivity in southwest Alaska, in contrast to the rest of circumpolar Arctic where vegetation has mainly shown increasing productivity. The AVHRR-derived Normalized Difference Vegetation Index (NDVI), a measure of vegetation productivity, has indicated a recent significant negative trend in southwest Alaska’s Yukon-Kuskokwim Delta (YKD), since 1999. Across much of the Arctic, higher NDVI values have been linked to sea ice decline as diminishing sea-ice permits increased land surface warming near coastal regions. While the southwest Alaska (East Bering Sea) region has experienced a comparatively large increase in the summer warmth index, >10°C month, the region’s tundra vegetation has become less productive. Here evaluate whether changes in precipitation and total rainfall have been an important covariate with temperature increase in driving changes to Arctic vegetation productivity. Six standard readily available, gridded, monthly datasets were chosen to examine trends in precipitation: CPC Merged Analysis of Precipitation, Global Precipitation Climatology Centre, Global Precipitation Climatology Project, National Centers for Environmental Prediction/National Center for Atmospheric Research Reanalysis, NOAA’s Precipitation Reconstruction over Land, and University of Delaware Precipitation. Downscaled ERA-Interim data were used as a 30-year comparison climatology for the growing season, defined here as June–August. There is significant variability among the data sets, both in the spatial patterns of total precipitation and trends. The large variability between the data suggests high uncertainty in precipitation climatology across the state of Alaska. We are investigating a 34-year period (1982-2015) of summer precipitation and NDVI to understand if the two trends are correlated. ERA-Interim shows a negative trend of approximately 30 mm over the 34-year period, while the ensemble average shows a very small positive trend of approximately 1 mm over the same period. NDVI trends have shown stronger negative values in the YKD for the later half of the period, 1999-2015, compared to the early half, 1982-1998. Preliminary results show stronger negative precipitation trends in the later period for the majority of datasets compared to the earlier period, however, ERA-Interim shows a slightly positive precipitation trend in the later period. Currently we are dissecting the precipitation trend variability and discrepancies before correlating the spatial and temporal correlation calculations to NDVI.

[poster]

Associated Project: Frost-01

Ecosystem Dynamics and Succession after Tundra Fire, Yukon-Kuskokwim Delta

Gerald Frost, Alaska Biological Research, Inc.--Environmental Research & Services, jfrost@abrinc.com
Lisa Saperstein, U.S. Fish and Wildlife Service, Lisa_Saperstein@fws.gov
Rachel Loehman, U.S. Geological Survey, rloehman@usgs.gov
Kristine Sowl, USFWS Yukon Delta National Wildlife Refuge, kristine_sowl@fws.gov
Matthew Macander, Alaska Biological Research, Inc.--Environmental Research & Services, mmacander@abrinc.com
Peter Nelson, University of Maine, peter.nelson@maine.edu
David Paradis, University of Maine, david.p.paradis@maine.edu

The Yukon-Kuskokwim Delta (YKD) encompasses the southernmost, warmest parts of the arctic tundra biome. Ice-rich permafrost currently is widespread and strongly influences terrestrial and aquatic environments. In 2015, the YKD experienced large wildfires across >1,200 km2 of permafrost-affected upland tundra. Although the 2015 fire season was exceptional, tundra fire is common in this region with episodes of historical fire circa 2005, 1985, and 1971, offering a natural laboratory for understanding the ecosystem impacts of tundra fire in a discontinuous permafrost region during a period of warming air and ground temperatures. In 2017, we collected field data on vegetation, soils, and burn severity within recent and historical burns and unburned tundra. Using these data we analyzed patterns of correspondence between vegetation species-composition and structure, soil properties, and fire history. We also tested for differences in biophysical properties among the tundra fire epochs and unburned tundra. Vegetation in unburned tundra was dominated by lichens, whereas burned areas support enhanced cover of shrubs and mosses; however, post-fire shrub cover was composed of the same low-statured species common to unburned tundra and we seldom observed sites colonized by taller, canopy-forming species. Geomorphology and soils were similar in burned and unburned tundra, likely because thick peat layers protected ice-rich permafrost and conferred ecosystem resilience after fire. While this historical perspective suggests that peaty soils will moderate the impact of the 2015 fires, we did observe secondary impacts related to permafrost degradation in circa 2005 fires that were not evident in older burns, such as thaw-settlement, increased surface wetness, complex microtopography, and progressive mortality of shrubs. These contrasts represent persistent, rather than successional shifts and suggest that upland ecosystems of the YKD may be less resilient to wildfire disturbance than they were in the past.

[poster]

Associated Project: Frost-01

20:

New tools for environmental annotation of animal movement tracks at the ABoVE domain and beyond

Gil Bohrer, Ohio State University, bohrer.17@osu.edu
Sarah Davidson, Max Planck Institute for Ornithology, sdavidson@orn.mpg.de
Scott LaPoint, Lamont-Doherty Earth Observatory, Columbia Univ., sdlapoint@gmail.com
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Martin Wikelski, Max Planck Institute for Ornithology, wikelski@orn.mpg.de

Climate and the Earth's land-surface and changing in the Arctic and globally. The impact of climate and land-surface changes on wildlife is difficult to assess, as they typically occur over decades, while wildlife monitoring programs have been in place for relatively short periods. Further challenges stem from the lack of uniformity in animal movement data collection approaches and methods. To analyze the impact of long-term changes in weather and habitat conditions on wildlife movements in the ABoVE domain, we built an archive of avian, predator and ungulate movements throughout the region. The archive is compiled and hosted in Movebank (www.movebank.org), a free, web-based service for managing animal movement data. Using Movebank allows us to manage data from multiple sources within a single database while supporting project-specific data security, terms of use, and access. By importing the data to the Movebank database, they are converted to a uniform metadata, reviewed for quality and completeness, and made easily accessible for analysis through the R package 'move'. An integral part of Movebank, the Env-DATA System allows easy annotation of these and related time-location records with hundreds of environmental variables provided by global remote sensing and weather data products, including MODIS Land, Snow and Ice products, the ECMWF and NARR weather reanalyses, and others. Recent developments in Env-DATA allow users to receive annotated raster data, and to compare these with observed movement tracks that were annotated using the same remote sensing dataset, and using the same QA and interpolation method. Users can also annotate modelled tracks to test hypotheses about the movement's interactions with the environment. The package R-MoveWindSpeed was recently developed to allow interpretation of atmospheric conditions from high-frequency GPS observations in flying-soaring birds. Environmental data interpreted from the track, combined with data from remote sensing and weather reanalysis can help elucidate the impacts of climate change on animal movement, in the Arctic and beyond.

Associated Project: Boelman-01

21:

Assessing the Utility of a Temporally Dynamic Wind Index in Alaskan Moose Resource Selection

Jyoti Jennewein, University of Idaho, jjennewein@uidaho.edu
Peter Mahoney, University of Washington, pmahoney29@gmail.com
Arjan Meddens, University of Idaho, ameddens@uidaho.edu
Sophie Gilbert, University of Idaho, sophiegilbert@uidaho.edu
Mark Hebblewhite, University of Montana, mark.hebblewhite@umontana.edu
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Scott Brainerd, Alaska Department of Fish and Game, scott.brainerd@alaska.gov
Kyle Joly, National Park Service, kyle_joly@nps.gov
Kim King Jones, Alaska Department of Fish and Game, kim.jones@alaska.gov
Kalin Seaton, Alaska Department of Fish and Game, kalin.seaton@alaska.gov
Lee Vierling, University of Idaho, leev@uidaho.edu
Jan Eitel, University of Idaho, jeitel@uidaho.edu

Accelerated warming in high northern latitudes (HNL) has led to warmer, wetter, and more variable environments. Wildlife living in these systems must adapt to these changes. For instance, moose (Alces alces) are cold-adapted ungulates who can experience heat stress year-round. In winter, moose begin experiencing heat stress at -5C, and temperatures exceeding 0C require excess energy to be expended to maintain normal body conditions. To combat heat stress, moose adopt thermoregulatory behaviors that include selection of taller and denser forest canopies, conductive cooling from lying on snow, and convective cooling from exposure to wind. Despite its notable role in thermoregulation, wind parameters are rarely included in resource selection studies because of difficulties associated with spatial and temporal interpolation of weather station wind data. One possible approach to estimate the effect of wind in resource selection modeling is the use of dynamic terrain attributes, such as a wind shelter index, which model long-term, highly-dynamic meteorological data across a digital elevation model. To investigate the utility of dynamic terrain attributes in winter moose resource selection, we modelled a wind shelter index as a thermoregulation mechanism in three game management units in Alaska. Next, we compared a null winter model that focuses on structural habitat features related to thermoregulation to a windshelter model using conditional-logistic regression to account for the dynamics of changing resource availability throughout the winter. Three moose datasets (n=128 total; 93 cows, 35 bulls), were analyzed independently to assess location-specific resource selection. The ArcticDEM (5m pixels) served as the terrain input, which enables dynamic, fine-scale topographic features to be resolved. Wind shelter was programmed with meteorological data from the North American Regional Analysis (NARR) database. A resource selection function in a used-available sampling design estimated the proportional probability of resource use. A step-selection function generated ten available locations per used location. Matched case-control logistic regression contrasted the used and available points. Results from this analysis suggest that the dynamic wind shelter index is a statistically significant predictor (p<0.01) in Alaskan winter moose resource selection. Additionally, in four out of five models wind shelter enhances model fit (delta BIC from -24 to -195). Taken together, these results suggest that the wind shelter index provide insight into winter moose resource selection and thermoregulation that could not be gained when using static landscape characteristics. Future applications of such dynamic terrain indices that incorporate time-varying meteorological data may be increasingly important in modelling habitat selection under continued climate change scenarios.

Associated Project: Boelman-01

22:

Arctic warming and Golden eagle migrations: potential for desynchrony between spring and eagle arrival dates

Scott LaPoint, Lamont-Doherty Earth Observatory, Columbia Univ., sdlapoint@gmail.com
Eliezer Gurarie, University of Maryland, egurarie@umd.edu
Gil Bohrer, Ohio State University, bohrer.17@osu.edu
Sarah Davidson, Max Planck Institute for Ornithology, sdavidson@orn.mpg.de
Peter Mahoney, University of Washington, pmahoney29@gmail.com
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu

Understanding how animals adapt to climate change is a conservation priority, particularly in arctic landscapes where these changes are accelerated. Doing so however, remains challenging because animal behavior datasets are typically conducted at site- or population-specific scales and are often short term (e.g., 2-3 years). We have overcome this challenge by compiling a long-term (~25 years), large-scale (northwestern North America) dataset of > 0.5 million locations collected via 80 adult-aged golden eagles (Aquila chrysaetos) fitted with satellite and GPS data loggers. We used mechanistic range shift analyses to identify the locations and dates when each eagle performed a behavioral switch from a stationary phase (e.g., over-wintering or breeding) to migration and vice-versa. We annotated these spatio-temporal data with a suite of environmental data, including: % snow cover, snow depth, air temperature, and barometric pressure. We find age-based differences in Golden eagle spring migration phenologies. Adults begin and end migration earlier, with less intra- and inter-annual variability than subadults. The date adults arrive at their breeding grounds is consistent year after year, despite slight variations in departure date from their over-wintering grounds. We will use a survival analysis (e.g., Cox proportional-hazard regression model) to quantify the influence of the environmental variables on these dates. The age-based differences are interesting. Adults initiate migration when triggered by a fixed cue (i.e., photoperiod), preventing them from capitalizing on optimal migration conditions and also from arriving at their breeding grounds in spring earlier to synchronize with earlier spring arrival dates in the Arctic. Subadults are known to be more responsive to local environmental conditions, yet are unable to reproduce and contribute to their populations. Golden eagles exhibit some resiliency to changes in the arctic climate, but the impact of a potential desynchrony between adult arrival to and the conditions (e.g., local prey population pulses) at their Arctic breeding grounds warrants further investigation.

Associated Project: Boelman-01

23:

Mapping lichen coverage across caribou herds using high-resolution imagery from consumer unmanned aerial vehicles

Eric Palm, University of Montana, e2palm@gmail.com
Matthew Macander, Alaska Biological Research, Inc.--Environmental Research & Services, mmacander@abrinc.com
Peter Nelson, University of Maine - Fort Kent, peter.nelson@maine.edu
David Paradis, University of Maine, david.p.paradis@maine.edu
Mark Hebblewhite, University of Montana, mark.hebblewhite@umontana.edu

For caribou populations in northern boreal forests of Alaska and the Yukon, changing fire regimes have the potential to alter resource selection patterns and limit food availability, leading to declines in body condition, recruitment and population sizes. To understand the mechanisms by which fires may affect caribou habitat selection and demography, we require better data on the spatial distribution and abundance of terrestrial lichens, a major caribou food source. We used consumer unmanned aerial vehicles (UAVs) and in situ ground cover measurements to sample lichen cover at 22 sites across three caribou ranges in interior Alaska and western Yukon. We classified <1 cm pixels from the UAV image mosaics as either light-colored terrestrial lichen or not using ground-based plot photos as references. Our classified UAV imagery is aggregated to produce fractional lichen cover estimates for 2 m and 30 m pixels, which will function as training data for models estimating lichen cover for larger areas using commercial satellite imagery (e.g. WorldView-2) and entire caribou ranges using Landsat composites. Data from UAVs bridge the gap between in situ ground measurements and satellite remote sensing data by providing cost-effective, high-resolution imagery that allows for interpretation of lichen and vegetation features and also functions to calibrate and validate satellite-based cover models.

Associated Project(s):
Boelman-01
Goetz-03

24:

Navigating snowscapes: scale-dependent responses of mountain sheep to snowpack properties

Peter Mahoney, University of Washington, pmahoney29@gmail.com
Glen Liston, Colorado State University, glen.liston@colostate.edu
Scott LaPoint, Lamont-Doherty Earth Observatory, Columbia Univ., sdlapoint@gmail.com
Eliezer Gurarie, University of Maryland, egurarie@umd.edu
Buck Mangipane, National Park Service, buck_mangipane@nps.gov
Adam Wells, University Of Idaho, Moscow, adamwells26@hotmail.com
Todd Brinkman, University of Alaska, Fairbanks, tjbrinkman@alaska.edu
Jan Eitel, University of Idaho, jeitel@uidaho.edu
Mark Hebblewhite, University of Montana, mark.hebblewhite@umontana.edu
Anne Nolin, Oregon State University, nolina@science.oregonstate.edu
Natalie Boelman, Lamont-Doherty Earth Observatory, Columbia Univ., nboelman@ldeo.columbia.edu
Laura Prugh, University Of Washington, lprugh@uw.edu

Snow enshrouds up to one third of global land mass and exerts a major influence on the demography and movements of animals that occupy these snowscapes. The snow-covered season is often limiting for many terrestrial mammals as cold temperatures, restricted resources, and constrained mobility can increase physiological stress relative to summer conditions. The accelerated pace of warming in arctic and boreal systems is altering the phenology and composition of snowscapes, with profound implications for wildlife. Yet, linking animal movement and resource selection to snowscape properties is challenged by a paucity of snow products that reflect spatiotemporal heterogeneity at scales relevant to biological data. A lack of knowledge regarding how animals respond to snowscapes represents a critical gap in understanding vulnerability and resilience of natural systems to climate change. Thus, we used a step-selection framework to evaluate the influence of snowscape properties, as characterized by a remotely sensed product (MODIS Snow Cover) and a snowpack evolution model (SnowModel), on scale-dependent movement and resource selection of Dall sheep (Ovis dalli) in Lake Clark National Park, Alaska. We monitored movement and resource selection in 30 sheep (12 rams and 18 ewes) over the course of three winters (2006 – 2008), defined as January through May of each year. We defined eight spatiotemporal scales of selection characterized by steps taken over seven-hour (our shortest fix interval) through 38-day intervals. We utilized MODIS daily fractional snow cover at a 500-m resolution (MOD10A1) and generated snow depth and density outputs from SnowModel at resolutions of 25-m through 10-km. Snow depth and density at fine-to-moderate resolutions (25-m – 500-m) consistently performed best at almost all scales of selection except the longest step intervals, indicating that MODIS fractional snow cover may be insufficient for describing animal movements at all but the largest spatiotemporal scales. Although SnowModel was developed primarily for hydrological applications, we demonstrate its utility in describing relevant variation in sheep resource selection at finer scales during winter. However, our findings highlight a critical need for further development of fine-scaled, process-driven snow metrics for use in animal movement modeling, particularly in light of increased climatic variability at higher latitudes and subsequent impacts on animal fitness and demography.

Associated Project(s):
Boelman-01
Prugh-01

25:

Seasonal snowpack quality and hazardous event detection in Dall Sheep habitat; modelling and remote sensing approaches to examine long-term patterns of variability.

Christopher Cosgrove, Oregon State University, cosgrovc@oregonstate.edu
Anne Nolin, Oregon State University, nolina@science.oregonstate.edu

Dall sheep are an emblematic species of alpine northwestern N. America and a large ungulate, reliant on year-round access to forage in open-alpine terrain. Their population decline, 20% range-wide since 1990, may be a description of broad-scale alpine ecosystem change in Alaska and northwestern Canada, with changing seasonal snow cover thought to be a principal cause. For a study area in the north Wrangell St Elias National Park, a physically based, spatially explicit snow evolution model (SnowModel) is forced by surface level meteorological data from the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) product. Producing a time series of high resolution (30 m, daily) snow and metrological variables (e.g. snow depth, total precipitation) from 1980 to 2017, metrics potentially important to Dall Sheep movement and forage (e.g. percentage area under a threshold snow depth) are examined for long term trends and patterns of temporal and spatial variability. September to November snowfall is shown to have greater volume and year-on-year variability than later winter months where snowfall is limited. This suggests that the volume of snowfall in these months is critical for determining the available area accessible for Dall Sheep forage during the entirety of the snow season. Total days per month with rain-on-snow (RoS) is shown to be slightly increasing in the month of September from the model data, these events potentially cause forage-inhibiting, season-lasting ice layers within the snowpack. Following this, a methodology using the National Aeronautics and Space Administration’s (NASA) Calibrated Enhanced Resolution Passive Microwave Brightness Temperature (CETB) product is presented for detecting RoS events at spatial resolutions up to 3.125 km2. Spanning a record from 1979 to 2016, this approach has potential for novel use in areas of variable terrain, such as mountainous Dall Sheep habitat.

Associated Project: Prugh-01

26:

Climatic drivers of Dall sheep survival rates

Madelon van de Kerk, University of Washington, madelon@uw.edu
Laura Prugh, University Of Washington, lprugh@uw.edu

Dall sheep population declines have been attributed primarily to harsh spring weather and snow conditions. However, the mechanisms behind the declines are not known, and no quantitative analyses have examined why some populations appear to be resilient while others appear to be vulnerable. We investigated Dall sheep survival rates among several populations and how they were affected by a variety of environmental factors. We used GPS and VHF telemetry data from Dall sheep collected from 9 different sites for a total of 229,815 sheep days. We estimated annual survival from May to May for sheep of different age classes (lamb: 0-1 yrs, adult: 1-12 yrs, old: >12 yrs) using Cox proportional hazard regression. We then used a model selection approach to examine how survival was affected by mean temperature, mean precipitation, mean and maximum NDVI, and total number of freeze/thaw events in summer and winter. We combined covariates that were uncorrelated into additive and interactive models and ranked the models based on the AIC. We compared the model fit of the base model with and without a random effect of site to see if there were any significant differences in survival rates among sites. We then compared the fit of the top model with and without a random effect of site to investigate if the included covariates adequately accounted for these differences. We estimated the overall survival rates by age class to be 29% (SE: 14) for lambs, 80% (SE: 3) for adults, and 50% (SE: 19) for old sheep. Including a random effect of site in the base model (with no environmental covariates) improved the fit, indicating substantial differences in survival rates among sites. Based on a total model set of 453 models, the top model included an interactive effect of the mean precipitation in summer and the total number of winter thaw events. Including a random effect of site in the top model did not improve the fit, indicating that the variation among sites was adequately captured by the included environmental factors. Adult survival was generally high and stable, except in years with dry summers and multiple thaws in winter. Lambs survival, and to a lesser extend the survival of old individuals, was good in years with wet summers and multiple thaws in winter as well as years with dry summers and few thaws in winter. Their survival was relatively poor under the opposite circumstances; years with wet summers and few thaws in winter as well as years with dry summers and multiple thaws in winter.

Associated Project: Prugh-01

62:

Changing river ice seasonality and impacts on interior Alaskan communities

Dana Brown, University of Alaska, drbrown11@alaska.edu
Todd Brinkman, University of Alaska, Fairbanks, tjbrinkman@alaska.edu
Dave Verbyla, University of Alaska Fairbanks, dlverbyla@alaska.edu
Helen Cold, University of Alaska, Fairbanks, hscold@alaska.edu
Caroline Brown, University of Alaska, Fairbanks, caroline.brown@alaska.gov
Teresa Hollingsworth, USDA Forest Service, tnhollingsworth@alaska.edu

Subsistence harvesters in the North rely on frozen rivers for winter access to local resources. Interior Alaskan residents have observed changes in river ice regimes that are significant hindrances to travel and subsistence practices. In this study, remote sensing and local observations were used to examine changes in seasonality of break-up and freeze-up seasons on major rivers in interior Alaska and assess the implications for subsistence harvesters. Spring and autumn air temperatures, respectively, were found to impact timing of break-up (-2.0 days/˚C) and freeze-up (+2.0 days/˚C). Spring air temperatures have increased by 0.02-0.06 ˚C/year over the last ~53-93 years. Accordingly, the break-up season has advanced by 6 days over the last century. The beginning of safe travel on river ice has been delayed by 10 days over the last century. Freeze-up timing was positively related to both air temperature and river discharge. Fall river discharge increased by 920 m3/s over the last 40 years, and was unrelated to precipitation change. Changing hydrology resulting from climate-related permafrost thaw and glacial melt may have contributed to the delay in freeze-up. The duration of river inaccessibility during freeze-up was double to triple that of the break-up, thus, understanding controls over freeze-up remains an important goal. The duration of safe travel on river ice has declined over the last century, and is expected to decline further as the climate continues to warm, thereby presenting new challenges to accessing subsistence resources at traditional times of the year. The mismatch in timing will necessitate community adaptation.

Associated Project: Brinkman-01

71:

Climate Sensitivity of High Arctic Permafrost Demonstrated by Widespread Ice Wedge Thermokarst on Banks Island, Canadian Arctic Archipelago

Robert Fraser, Canada Centre for Mapping and Earth Observation, robert.fraser@canada.ca
Steve Kokelj, GNWT Geological Survey, steve_kokelj@gov.nt.ca
Trevor Lantz, University of Victoria, tlantz@uvic.ca
Morgan McFarlane-Winchester, Canada Centre for Mapping and Earth Observation, morgan.mcfarlane-winchester@canada.ca
Ian Olthof, Canada Centre for Mapping and Earth Observation, ian.olthof@canada.ca
Wenjun Chen, Canada Centre for Mapping and Earth Observation, wenjun.chen@canada.ca

Ice-wedge networks underlie polygonal terrain and comprise the most widespread massive ground ice type in continuous permafrost. Wedge ice is typically encountered at the top of permafrost, so near-surface thawing can result in subsidence of the terrain surface, ponding, and the development of high centred polygons. Studies have documented that recent permafrost warming is associated with degradation of ice-wedges at local scales (10-100 km2) in several arctic environments leading to trough subsidence and ponding of water. In this analysis we use a combination of the 1985-2017 Landsat satellite image archive, high resolution optical satellite imagery, air photos, detailed digital elevation models, and field surveys to show that ice-wedge thermokarst has been extensive over Banks Island’s 70,000 km2 landmass during a period of recent warming. End-members of this thermokarst continuum range from high-centred polygon field development on well-drained hilltops, to potential thermokarst lake initiation on flatter uplands where ice-wedge ponds have rapidly coalesced. We demonstrate that cold Arctic permafrost landscapes can be among the most thaw sensitive permafrost environments because well-developed epigenetic and anti-syngenetic ice-wedges truncated by the active layer lack the strong ecosystem protection that buffers summer warming and drives feedbacks that stabilizes thermokarst in the low Arctic.

Associated Project: Chen-W-01

73:

How might satellite remote sensing help the monitoring of the cumulative impacts on barren-ground caribou? Case studies for the Bathurst herd

Wenjun Chen, Canada Centre for Mapping and Earth Observation, wenjun.chen@canada.ca
Sylvain Leblanc, Canada Centre for Mapping and Earth Observation, sylvain.leblanc@canada.ca
H. White, Canada Centre for Mapping and Earth Observation, hpeter.white@canada.ca
Carla Schmitt, Canada Centre for Mapping and Earth Observation, carlavanessa.schmitt@canada.ca

The barren ground caribou has played an important role in the culture, economy, and way of life of aboriginal peoples in Arctic North America for thousands of years. The widespread decline of caribou populations around the Arctic raised concerns among northern communities and local governments. For example, the Bathurst caribou herd declines 93% since the mid-1980s. Why has the caribou population not recovered? What are the underlining causes that resulted in the decline? How has industrial developments, such as mining, been affecting caribou? These are some of the questions raised by community members on a regular basis. Many factors might affect caribou, such as habitat, harvest, predators, diseases and parasites, extreme weather, climate change, industrial development, forest fires, and pollution, in a complex manner spatially and temporally. In order to find out the underlining causes of the population decline so that effective recovery plan can be designed and implemented, the first and the most important task is to develop verifiable and comprehensive long-term data sets for these factors. With long-term historical records and large spatial coverages, satellite remote sensing is uniquely equipped to contribute to this goal. As a part of NWT CIMP (Northwest Territories Cumulative Impact Monitoring Program) and the NASA ABoVE (Arctic-Boreal Vulnerability Experiment), we have been working towards this goal. In this poster, we will show four case studies of developing satellite datasets for monitoring and assessing the cumulative impacts on the Bathurst caribou. The first case study was the long-term winter range indicators developed from remote sensing and climate data. The second study developed satellite-derived summer range indicators and their linkage with caribou net productivity (i.e., the late-winter calf: cow ratio). The third study detected plant phenology changes on the summer range and calving ground of the Bathurst caribou herd using long-term satellite data and community-based field observations, and their impacts on caribou calving dates. Finally, the fourth study demonstrated the role of remote sensing assist for quantifying the zones of mining disturbances (e.g., dust, noise, PM2.5, sight, and changes in vegetation and soil).

Associated Project: Chen-W-01

Contact Meeting Support

Meeting Planning Committee

Download Acrobat Reader