Laura Louise Bourgeau-Chavez,  Michigan Technological University,  lchavez@mtu.edu (Presenter)
Jacquelyn K. Shuman,  National Center for Atmospheric Research,  jkshuman@ucar.edu
Paul Robert Siqueira,  University of Massachusetts,  siqueira@umass.edu
Michael Battaglia,  Michigan Technologcal University,  mjbattag@mtu.edu
Michael Billmire,  Michigan Tech Research Institute (MTRI),  mgbillmi@mtu.edu
Nancy HF French,  Michigan Tech Research Institute (MTRI),  nhfrench@mtu.edu
Bruce Chapman,  JPL,  bruce.d.chapman@nasa.gov
Mike Flannigan,  University of Alberta,  mike.flannigan@ualberta.ca
Jennifer Baltzer,  Wilfrid Laurier University,  jbaltzer@wlu.ca

Wildfire is the most prevalent disturbance factor within boreal regions of North America followed by permafrost thaw. There is evidence that climate change is exacerbating both of these variables and the related feedback mechanisms. The interaction of wildfire and hydrology on successional trajectories in lowlands and uplands, particularly in the peat-rich ecosystems of the boreal-taiga, is poorly understood. The distribution of soil moisture across the landscape has links to both the potential for fire, wildfire effects, as well as post-fire recovery. Developing and refining microwave soil moisture retrieval algorithms for the arctic-boreal zone (ABZ) will fill an important data gap. To better understand, diagnose and predict ecosystem dynamics, especially the interaction of hydrology with wildfire and post-fire recovery, integration of field and remote sensing data observations with coupled landscape scale ecosystem and hydrological models is needed. As a fundamental component of biological, hydrological, and biogeochemical processes, soil moisture is a seasonally dynamic and spatially variable driver of ecosystem respiration and C fluxes. L-band SAR is well suited for retrieving soil moisture from low to densely vegetated sites with its 24 cm wavelength, but retrieval algorithms need to be developed specifically for the deep organic soils and unique vegetation structure of boreal ecosystems. The overall goal of the research proposed is to improve our understanding of the controls and impacts of a changing climate on the vulnerability and resiliency of boreal-taiga ecosystems to wildfire through process-based ecological and hydrological modeling of the ABZ with a focus on the intensive study of 2014-2016 wildfires of southern Northwest Territories (NWT) and northern Alberta (a peat-rich subregion of the ABoVE study domain). The proposed work builds from NASA ABoVE Phase 1 grant #NNX15AT83A, under which the team focused on analysis of field data characterizing fire-affected ecosystems and developing algorithms from coincident remote sensing and in-situ data for mapping a variety of variables to lay the empirical foundations for modeling C emissions and ecosystem processes post-fire for the peat-rich study area. Our proposed objectives are to: 1) develop and refine algorithms for soil moisture retrieval from L-band UAVSAR in preparation for NASA’s NISAR mission; and 2) use field and remote sensing data collected in ABoVE Phase 1 project and Obj. 1 as the basis to parameterize ecosystem, hydrology and fire effects models for lowlands and uplands to improve our understanding of the interactions of wildfire and hydrology. The proposed project will allow for a more thorough understanding of the vulnerability of a variety of ecosystems to wildfire, the hydrological effects (including permafrost changes) and the trajectories of succession that will ensue. This is particularly important for landscape ecosystems that historically do not burn, such as peatlands, but also for determining thresholds that may change successional trajectories. The proposed work addresses two of the ABoVE Phase 2 research focus areas by analyzing remote sensing data collected during the 2017 ABoVE Airborne Campaign to develop the data products required to improve understanding of ecosystem dynamics and integrating research results from ABoVE into a coherent modeling framework to diagnose and predict ecosystem dynamics. Outputs from this project will include: 1) ABZ specific L-band algorithms for retrieving soil moisture; 2) SAR maps of trends in soil moisture; 3) time series of ABZ tuned SMAP maps across the ABoVE domain; 4) quantification of C-consumption and emissions from empirical model CanFIRE for 2014-16 wildfires of NWT/Alberta; 5) improved mechanistic understanding of the interactions of wildfire and hydrology on the boreal lowland and upland systems; and 6) model predictions of the effects of future climate scenarios on successional trajectories.

Associated Project(s): 

• Bourgeau-Chavez (TE 2018)

Presentation: ASTM5_Poster_BourgeauChavez_1_16_21.pdf 

Natalie T Boelman,  Lamont-Doherty Earth Observatory, Columbia Univ.,  nboelman@ldeo.columbia.edu (Presenter)
Todd Brinkman,  University of Alaska, Fairbanks,  tjbrinkman@alaska.edu
Glen Liston,  Colorado State University,  glen.liston@colostate.edu
Laura Prugh,  University Of Washington,  lprugh@uw.edu
Adele Reinking,  Colorado State University,  adele.reinking@colostate.edu
Peter Mahoney,  University of Washington,  pmahoney29@gmail.com

Snow and ice are key features of Arctic and boreal regions (ABRs) for approximately nine months each year, and thus play a central role in governing its social-ecological systems in myriad ways. In recent decades, ABR warming has caused changes in the spatial extent and distribution, seasonal timing and duration, and physical characteristics and properties, of snow and ice covers. These multi-dimensional changes in snow- and ice-scapes are occurring concurrently and with a high degree of spatial heterogeneity, creating a complex mosaic of conditions over which people, and the big game species and transportation networks on which they rely, traverse. Furthermore, snow- and ice-scape conditions have a complex and comprehensive influence on biological processes in ABRs, with significant ramifications for carbon cycling that feedback and regulate climate. This project’s overarching science goal is to: Quantify the socio-ecological consequences of changing snow- and ice-scapes across the ABoVE study domain. The poster presentation will describe our four main objectives designed to achieve this goal.

Associated Project(s): 

• Boelman (TE 2018)

Nancy HF French,  Michigan Tech Research Institute (MTRI),  nhfrench@mtu.edu
Liza K Jenkins,  Michigan Tech Research Institute (MTRI),  lliverse@mtu.edu (Presenter)
Michael Battaglia,  Michigan Technological University,  mjbattag@mtu.edu
Laura Louise Bourgeau-Chavez,  Michigan Technological University,  lchavez@mtu.edu
Kevin Smith,  Ducks Unlimited,  k.smith@ducks.ca
Bruce Chapman,  JPL,  bruce.d.chapman@nasa.gov
Michael Allan Merchant,  Ducks Unlimited Canada,  m_merchant@ducks.ca

Wetlands provide unique ecosystem services and thereby have great significance for human subsistence in the vast regions of northern North America. This vast 1.4 billion acre area is home to 23 species of waterfowl, of which at least 10 have more than half of their continental breeding season population within the region. The project goal is to characterize changes in waterfowl habitat using remote sensing approaches developed using field, airborne, and satellite data collected during the ABoVE research campaigns. The goal will be addressed through two objectives that will utilize field and remote sensing data and research outcomes from previous ABoVE activities as well as data
and expertise of partners with the Jet Propulsion Lab (JPL), Ducks Unlimited Canada (DUC), and other Canadian Collaborators. The research will be developed for the core ABoVE study region, and demonstrated at specific waterfowl study sites where habitat and duck populations are well documented. The research will address two topics under the solicitation requests under section 4.1: Airborne Science Using Data Collected During 2017 AAC as well as topics under 4.3: Societal Effects of Environmental Change. We propose
two specific objectives. The first is to develop wetland mapping algorithms across diverse wetland types in the boreal region using multiple remote sensing approaches, but focusing on L-band SAR (NISAR w/UAVSAR and ALOS-2, SAOCOM, if available) as well as mapping of wetland inundation. This objective will include efforts to verify performance of emerging NISAR inundation mapping capability (developed for ALOS-PALSAR) for wetlands in the ABoVE region. We will look at integrating different wavelength SAR (e.g. Radarsat, Sentinel1, PALSAR) for exploring inundation algorithm development and validation. The value of Radarsat 2 and Sentinel 1 C-band SAR is to provide a long time series of imagery. Our second objective will be to use the new maps of wetland type, change, and inundation for waterfowl habitat suitability assessment. The expectation is that improved wetland characterization
will be useful for forecasting habitat suitability and, hence, waterfowl distribution and trends under climate change. We will test this approach in the Mackenzie River Delta region Peace-Athabasca Delta and Slave River Delta regions of Canada, which are high waterfowl use areas with long temporal records of waterfowl breeding population data, remote sensing information (satellite and aerial, including 2017/18 ABoVE Airborne Campaigns), and current partnerships with indigenous communities. Wetland type maps will be developed using methods currently underway within the ABoVE program (Bourgeau-Chavez, PI). The map accuracy and value will be assessed with regard to needs for waterbird habitat, as defined with our Ducks Unlimited research partners. In addition to the focus of wildlife habitat assessment, the research will be beneficial broadly by developing methodologies for using NISAR for
wetland characterization, and specifically for developing techniques of validating SAR-derived estimates of inundation extent by using Sentinel-1 and Radarsat-2 as surrogates for the future NISAR mission. The outcomes of this project will contribute to the development of comprehensive, consistent wetland maps and hydrological condition products, which are valuable to spatial data modeling activities for wildlife habitat and populations, biodiversity, carbon, and ecological processes. These approaches will be important in later phases of ABoVE where ecological modeling and decision support tools will be developed.

Associated Project(s): 

• French (TE 2018)

Presentation: ASTM5_Poster_French_1_19_83.pdf 

Min Chen,  Pacific Northwest National Laboratory,  min.chen@pnnl.gov (Presenter)
Ben Bond-Lamberty,  Pacific Northwest National Laboratory,  bondlamberty@pnnl.gov
Corinne Hartin,  Pacific Northwest National Laboratory,  corinne.hartin@pnnl.gov

The overarching scientific questions of the study are: how will the key ecosystem services in the Arctic respond to the climate change in the past decade and in the future? How big are the socioeconomic impacts of these changes in the 21st century? And how valuable are current observed ecological datasets, in terms of reducing the uncertainties of the estimated responses and changes? This poster will demonstrate the structure of an integrated human-Earth system modeling and data assimilation framework for quantifying socioeconomic impacts of changing Arctic ecosystems. The framework consists of an advanced land surface model (the Community Land Model, CLM), a sophisticated integrated assessment model (the Global Change Assessment Model, GCAM), and the Data Assimilation Research Testbed (DART), which will be used to quantify how ABoVE data can constrain the terrestrial component of Human-Earth system models, characterize uncertainties in their projections, and assess the socioeconomic impacts of such improved projections at regional and global scales.

Associated Project(s): 

• Chen (TE 2018)

Presentation: ASTM5_Poster_Chen_1_20_35.pdf 

Joshua B. Fisher,  NASA JPL,  jbfisher@jpl.nasa.gov (Presenter)

The Arctic-Boreal Region (ABR) is the source of among the largest uncertainties to global climate projections [Chapin et al., 2000; McGuire et al., 2006; Koven et al., 2011; IPCC, 2014]. The lack of observational data has limited model improvements, testing, and evaluation for the ABR [Fisher et al., 2014; Fisher et al., 2018]. NASA’s Arctic-Boreal Vulnerability Experiment (ABoVE) was motivated in part because of these uncertainties, and has now since produced airborne and satellite observations with in situ measurements necessary to improve Earth System Model (ESM) uncertainties for the ABR [Goetz et al., 2011; Griffith et al., 2012; Kasischke et al., 2013]. Here, we outline the structural framework for infusing ABoVE data into models to reduce these uncertainties.

Associated Project(s): 

• Fisher (TE 2014)
• Fisher (TE 2014) Extension

Presentation: ASTM5_Poster_Fisher_1_21_60.pdf 

Andrew M Fox,  University of Arizona,  andrewfox@email.arizona.edu (Presenter)
David JP Moore,  University of Arizona,  davidjpmoore@email.arizona.edu
William Kolby Smith,  University Of Arizona,  wksmith@email.arizona.edu

In our new Phase 2 project we address the question of “How are the magnitudes, fates, and land-atmosphere exchanges of carbon pools responding to environmental change, and what are the biogeochemical mechanisms?”. We will improve the mechanistic representation of biochemical processes within the Community Earth System Model (CESM) to evaluate whether warming in the arctic has and is likely to cause amplifying feedbacks to climate, having first used data assimilation to ensure that the model agrees with historical LAI in the region. Where the highly tuned model disagrees with appropriately scaled observations we will infer a biogeochemical mechanism is misspecified, mis-parameterised or missing. By revising the representation of mechanisms in the model subroutines we can test a series of hypotheses to develop our understanding of feedbacks between global scale climate forcing and regional and local scale impacts. Our data assimilation system brings together CESM, one of the most highly developed and extensively used ESMs with a leading-edge, community ensemble data assimilation tool, the Data Assimilation Research Testbed (DART). Together, they represents one of the most advanced DA systems yet developed for such a complete, coupled ESM.
Here, in addition to outlining our project we present some preliminary results: (i) showing the impact of assimilating 10 years of of MODIS LAI observations were assimilated into CLM globally. Comparing the free run (no assimilation) with the assimilation run over the ABoVE domain, it is clear that whilst simulation of LAI is improved, mean error is reduced only by about 0.5 m2 m-2; and (ii) for a site at the Barrow Environmental Observatory, we have conducted a sensitivity analysis of 6 aboveground CLM5.0 model parameters by varying them systematically around their default values as a first step towards targeted model improvements.

Associated Project(s): 

• Moore (TE 2018)

Presentation: ASTM5_Poster_Fox_1_22_56.pdf 

David Lutz,  Dartmouth College,  david.a.lutz@dartmouth.edu (Presenter)
Manuel Lerdau,  University of virginia,  mlerdau@virginia.edu
Michael Palace,  University of New Hampshire,  michael.palace@unh.edu
Xi Yang,  University of Virginia,  xiyang@virginia.edu
Herman Henry Shugart,  University of Virginia,  hhs@virginia.edu
Adrianna Foster,  Northern Arizona University,  af2358@nau.edu

Over the past several decades there have been notable changes to patterns of climate, hydrology, and disturbance across North American boreal forests. Though these changes have distinct influences on boreal forests individually, complex interactions among these drivers result in magnified responses and a dramatic impact on forest species composition, structure, and functioning. While concerns over these interactions have attracted the interest of researchers, less attention has been paid to the corresponding impacts on ecosystem services and subsequent repercussions for long-term management of forests in the region. Our proposed cross-institutional project brings together interdisciplinary expertise to construct a modeling and evaluation framework through the lens of ecology, remote sensing, and ecological economics. Our overarching goals are to transform our understanding of how rapid change in North American boreal forests may alter the delivery of ecosystem services and to utilize predictive computational modeling to construct effective adaptive management strategies in conjunction with local stakeholders. Specifically, we seek to:

1. Simulate the dynamics of forest patterns and processes under future climate and disturbance scenarios at two distinct case study sites using an individual-based forest gap model that has been calibrated and validated with remotely sensed products, and
2. Assess tradeoffs in the delivery of ecosystem services under a set of stakeholder-informed management strategies at each study site using coupled forest-gap and ecological economics model output.

To add a new dimension to the extensive work underway on changes in boreal forest ecosystems within the ABoVE project, we will focus specifically on two case study areas encompassing public lands with documented management plans, thereby allowing for direct assessments of how changes in climate, ecosystem response, and management strategy may alter the flow of ecosystem services to stakeholders.

Associated Project(s): 

• Lutz (TE 2018)

Presentation: ASTM5_Poster_Lutz_1_23_77.pdf 

Troy Sehlin Magney,  California Institute of Technology,  tmagney@caltech.edu (Presenter)
Christian Frankenberg,  Caltech,  christian.frankenberg@jpl.nasa.gov
Jochen Stutz,  University of California Los Angeles,  jochen@atmos.ucla.edu
David Bowling,  University of Utah,  david.bowling@utah.edu
Nicholas Parazoo,  JPL,  nicholas.c.parazoo@jpl.nasa.gov
Barry Logan,  Bowdoin College,  blogan@bowdoin.edu
Zoe Pierrat,  University of California, Los Angeles,  zpierrat@atmos.ucla.edu

Understanding the feedbacks associated with climate change in Northern Boreal forests is critical for predicting how these ecosystems will change into the future. A rigorous understanding of the phenology of carbon cycling in evergreen Boreal forests has been challenging with traditional vegetation indices such as NDVI and EVI, because they retain foliage and remain green year-round, but the light use efficiency (LUE) of photosynthesis changes rapidly and dramatically. Seasonal changes in pigment composition associated with changing LUE can be detected via reflectance (PRI and CCI) and provide information on plant photoprotective mechanisms, but their utility from space-borne platforms has been limited. Solar induced fluorescence (SIF) reflects dynamical changes across the growing season driven by internal and external factors, and thus may provide a critical tool to evaluate ecophysiological mechanisms driving changing GPP. While spaceborne detection of SIF, PRI, and CCI has yielded important insights, the spatial and temporal resolution is currently too coarse to resolve detailed seasonal changes. We propose to link observations of vegetation reflectance and SIF collected at high temporal resolution (continuous ground-based hyperspectral reflectance and SIF with custom PhotoSpec spectrometers) and high spatial resolution (airborne data – CFIS, AVIRIS-NG) with carbon fluxes across a range of hydrologic and climate regimes in the Northern Boreal forest. This will require advancing the mechanistic understanding of how physical (canopy architecture, viewing and illumination geometry) and ecophysiological (seasonal acclimation of photosynthesis and needle pigment composition) drivers affect top-of-canopy SIF and reflectance. To accomplish this, we will leverage the strength of both SIF and reflectance synergistically to interpret their relations to GPP, and dependence on observation geometry. Towards this goal, we will install our tower-based spectrometer system at a new site (Delta Junction, Alaska), which will be augmented by our existing network, representative of different climatological and hydrological regimes (sites Niwot Ridge, Colorado; Southern Old Black Spruce, Saskatchewan). These three evergreen-dominated sites have substantially different climatic, radiative, and hydrological regimes, which enables us to characterize much of the wide range of evergreen ecosystems in the ABoVE domain. Tower-canopy and airborne spectroscopy data will be coupled with in-situ¬ measurements and 3-D radiative transfer modeling to link processes at a range of scales to remote sensing measurements. This multi-scale approach will inform satellite observations from MODIS, OCO-2, and TROPOMI, to scale GPP estimates across the Boreal region. Our primary objectives are as follows:
1) Establish a quantitative framework to describe the ecophysiological and physical mechanisms linking surface measurements of SIF and vegetation reflectance to GPP at established tower sites and coincident airborne campaigns.
2) Reduce uncertainties in satellite-based estimates of GPP across the Boreal region by applying a process-based understanding of both the temporal and spatial dynamics of SIF and vegetation reflectance at fine scales.

Associated Project(s): 

• Frankenberg (TE 2018)

Presentation: ASTM5_Poster_Magney_1_24_59.pdf 

Amanda Armstrong,  NASA GSFC / USRA GESTAR,  aharmstrong12@gmail.com (Presenter)
Batuhan Osmanoglu,  NASA GSFC / USRA,  batuhan.osmanoglu@nasa.gov
Paul Montesano,  NASA/SSAI,  paul.m.montesano@nasa.gov (Presenter)
Kenneth Jon Ranson,  NASA GSFC,  kenneth.j.ranson@nasa.gov
Howard Epstein,  University of Virginia,  hee2b@virginia.edu
Herman Henry Shugart,  University of Virginia,  hhs@virginia.edu

Shifts in the TTE are occurring and anticipated as Arctic warming persists through the next century (Elmendorf et al. 2012, Evju et al. 2012, Kaarlejarvi et al. 2012, Diffenbaugh et al. 2013). Some studies predict that up to ~50% of the tundra could be colonized by trees by 2100 (IPCC, 2014). However, the spatial distribution of TTE change is nonlinear and heterogeneous, exhibiting different rates, directions, and drivers (e.g. permafrost, elevation, micro-climate, soils, etc.). These and future changes, driven by multi-scale influences, need to be examined in the context of site-scale patterns of current structure and productivity in both tundra and forests.

The goal of this ABoVE Phase II research is to examine and quantify the likelihood of predicted changes in TTE forest structure patterns occurring within the ABoVE extended domain, using airborne imagery and lidar observations, site-scale (i.e., high resolution spatially-explicit individual-based) forest and tundra vegetation modeling and a Landsat-derived map of the extent and pattern of the TTE. Using the mapped extent of the TTE to spatially guide our modelling, our specific objectives
are to:
1. Parameterize the spatially-explicit individual-based gap model SIBBORK for the TTE forest study sites using leveraged field data and forest structure derived from LVIS and
GLiHT datasets.
2. Improve permafrost, allometry routines, introduce litterfall and integrate the ArcVeg (tundra) model parameters into the SIBBORK (forest) model to extend the simulation domain across the full tundra-taiga boundary at the individual scale.
3. Predict future forest and tundra productivity and ecotone location using CMIP6 projections.
4. Quantify the ability of the tree cover abruptness variable to predict patterns of TTE change. Calibrated with an ABOVE airborne product (LVIS) and initialized with a Landsat-derived map, our modeling framework will deliver the first high resolution, spatially-explicit forest and tundra model that has the capability to model ecotone shift across the ABoVE study domain.
Our research will provide a deeper understanding of the current productivity dynamics along the TTE, and allow for informed prognostication about the shift in the extent of tree cover, and the spatial variability in the direction, rate, and magnitude of these shifts.

Associated Project(s): 

• Armstrong (TE 2018)

Presentation: ASTM5_Poster_Armstrong_1_25_10.pptx 

Adrianna Foster,  Northern Arizona University,  af2358@nau.edu (Presenter)
Scott J. Goetz,  Northern Arizona University,  scott.goetz@nau.edu
Brendan Morris Rogers,  Woods Hole Research Center,  brogers@whrc.org
Michelle Cailin Mack,  Northern Arizona University,  michelle.mack@nau.edu
Matthew Macander,  Alaska Biological Research, Inc.--Environmental Research & Services,  mmacander@abrinc.com
Peter Nelson,  University of Maine,  peter.nelson@maine.edu
Jacquelyn K. Shuman,  National Center for Atmospheric Research,  jkshuman@ucar.edu
Herman Henry Shugart,  University of Virginia,  hhs@virginia.edu

The University of Virginia Forest Model Enhanced (UVAFME) is an individual tree-based forest model which simulates the establishment, growth, and mortality of individuals trees in response to external factors such as environmental and climate conditions, competition for resources, and disturbances. Because UVAFME operates at the individual-tree scale, it is able to capture the species- and tree-level interactions between vegetation, soil, and disturbances, and the resulting competitive dynamics at the inherent scale at which they operate. UVAFME has been extensively updated for use within the North American boreal region. Results from validation tests with the model against forest inventory and other field data show that output compares well with measurements of forest biomass, species composition, and stand structure, as well as solar radiation and evaporation, soil moisture, active layer depth, organic layer depth, and other important abiotic drivers. Notably, UVAFME output for current conditions only compared well with field data when all updates to vegetation-soil-climate-wildfire interaction were incorporated, highlighting the importance of these fine-scale interactions within the boreal zone. Following these updates, we simulated forest response to climate change (RCP 4.5 and 8.5) across the Tanana Valley River Basin. Results suggest that climate change and the concomitant impacts on wildfire and permafrost dynamics will result in overall decreases in biomass (particularly for spruce) within the region, and a resulting shift towards higher deciduous fraction. Simulation results also predict increases in biomass at cold, wet locations and at high elevations, and decreases in biomass in dry locations. These simulations demonstrate that a highly detailed, species interactive model can be used across a large region within Alaska to investigate interactions between vegetation, climate, wildfire, and permafrost. The vegetation changes predicted here have the capacity to feed back to broader scale climate-forest interactions in the North American boreal forest, a region which contributes significantly to the global carbon and energy budgets.

Associated Project(s): 

• Goetz (TE 2014)
• Goetz (TE 2018)

Scott J. Goetz,  Northern Arizona University,  scott.goetz@nau.edu (Presenter)
Adrianna Foster,  Northern Arizona University,  af2358@nau.edu
Matthew Macander,  Alaska Biological Research, Inc.--Environmental Research & Services,  mmacander@abrinc.com
Michelle Cailin Mack,  Northern Arizona University,  michelle.mack@nau.edu
Peter Nelson,  University of Maine,  peter.nelson@maine.edu
Brendan Morris Rogers,  Woods Hole Research Center,  brogers@whrc.org

Our Phase-1 and Phase-2 applications focus on investigating the role of vegetation dynamics and vegetation change across the ABoVE study domain. Building on our extensive work documenting changes in arctic and boreal vegetation productivity over the past three decades, we analyze recent and higher resolution satellite data products and lidar, hyperspectral and SAR data from the ABoVE airborne campaigns. These analyses are coupled with extensive field measurements and species-specific modeling to capture areas of documented changes in boreal tree productivity and mortality, and tundra vegetation changes across climate and environmental gradients.
Our investigations are in the context of exploring multiple lines of evidence for the progression of an arctic-boreal biome shift, where tree productivity decreases and mortality increases in the southern boreal while suitability for range expansion and densification of woody vegetation (e.g. shrubs) increases in the northern boreal and arctic tundra. We leverage existing resources, focusing on aspects of boreal forest dynamics that make them particularly vulnerable to productivity declines and tree mortality, mapping those with multi-scale remote sensing imagery and modeling these processes using forest demography models with a long history of well-developed capabilities. Through these analyses we have produced satellite-based maps of fractional cover of deciduous versus evergreen cover through time, developed a remote-sensing based framework for detecting early warning signals of tree mortality, investigated interactions between insects and climate change and the subsequent response of forests, and modeled biomass and species composition change across the ABoVE domain. We also conduct more detailed regional analyses focused on faunal habitat changes associated with vegetation composition and structure changes in both the arctic and boreal biomes, with a focus on areas of interior Alaska and western Canada (boreal biome), and the Seward Peninsula and North Slope of Alaska and northwestern Canada (tundra biome). We have produced maps of lichen cover for 2000 and 2015, and shrub canopy height and Light Macrolichen extent within the Arctic. These regions include the winter range of the declining western Arctic and eastern arctic Porcupine caribou herds, and the increasing Fortymile caribou herd of the eastern interior. Thus, our activities focus on some of the largest issues of concern to management agencies in the region.

Associated Project(s): 

• Goetz (TE 2014)
• Goetz (TE 2018)

Laura Duncanson,  University of Maryland,  lduncans@umd.edu
Amy Neuenschwander,  University of Texas,  amyn@arlut.utexas.edu (Presenter)
Paul Montesano,  NASA/SSAI,  paul.m.montesano@nasa.gov
John Armston,  University of Maryland,  armston@umd.edu
Joanne C White,  Canadian Forest Service, Natural Resources Canada,  joanne.white@canada.ca
Michael Wulder,  Canadian Forest Service, Natural Resources Canada,  mike.wulder@canada.ca
Steven Hancock,  University of Edinburg,  steven.hancock@ed.ac.uk

Significance:
Aboveground biomass is a critical element of the global carbon cycle, both in terms of the large magnitudes of carbon stored in aboveground woody material, and the ecological carbon-climate feedbacks related to disturbances from pests and fire which are particularly important for carbon cycling in the boreal system (Kurz et al. 2013). Baseline aboveground biomass maps provide modeling groups with important constraints on climate and carbon cycle models, and enable the long term monitoring of changes to aboveground biomass in the future. To date, high resolution biomass maps have been largely unavailable across the boreal system due to a dearth of active remote sensing data sensitive to forest structure at the spatial scales of boreal forest processes (~30m – 1 ha). Common approaches have generated biomass maps using various combinations of passive optical data (e.g. Landsat), regional zones, sparse spaceborne lidar, and airborne lidar transects (Margolis et al., 2015; Neigh et al., 2013; Matasci et al., 2018), but these maps often do not resolve relevant spatial variation in forest biomass and those using ICESat GLAS data require updating to reflect changes in forested systems over the past decade.

Several upcoming missions will collect spaceborne data directly relevant to forest biomass mapping (e.g. NASA’s GEDI, ICESat-2, NASA/ISRO’s NISAR, ESA’s BIOMASS), however only ICESat-2 and NISAR will collect data across the full boreal domain. ICESat-2, launching in September of 2018, with be temporally coincident with GEDI (launching in November 2018), but GEDI will only capture data below ~51.6 degrees North and South. As such, ICESat-2 data will provide a new opportunity to resolve the spatial variation of boreal biomass that would otherwise be omitted from the next generation accounting of forest biomass using spaceborne lidars. The resulting map for the ABoVE domain will demonstrate the application of these data for circumpolar mapping of biomass density for high-latitude forests.

This research will produce an ABoVE-wide biomass map with ICESat-2 data, and provide the algorithms necessary to extend this product across the entire boreal system. This product will also allow a comparison to biomass maps representing 2010 conditions (e.g. Neigh et al. 2013, Matasci et al. 2018) and from which to compare future products such as those generated using NISAR data, thus moving toward the development of a boreal-wide forest biomass monitoring system.

Objectives:
1) Develop ICESat-2 biomass models representative of the range of forest structures found in the ABoVE domain using existing field plot and discrete return airborne lidar data (lidar-plots);
2) Produce a boreal forest biomass density map for the ABoVE domain at 1 ha spatial resolution representing 2019-2021 using ICESat-2 data and Landsat imagery;
3)Utilize data from the 2017 LVIS campaign data for product validation, and error characterization

Methods:
We will curate field and airborne lidar data across the ABoVE domain and other representative boreal sites, focusing on data from across Canada and Alaska. These data will be used to generate ICESat-2-specific biomass models through the simulation of ICESat-2 from the existing discrete return lidar datasets. This follows the approach adopted for biomass product development by the GEDI Science Definition Team, and indeed uses the GEDI data system for estimating field biomass. These models will be applied to real ICESat-2 data over the full ABoVE domain for the period of 2019-2021. A gridded product will be generated at a 1 ha resolution, using Landsat land cover maps for stratification and imputation. For validation, we will produce an independent airborne lidar biomass map from the LVIS 2017 acquisition. The production of these local lidar biomass maps, as well as the ICESat-2 product validation, will follow recommendations from the forthcoming CEOS Land Product Validation protocol for biomass.

Associated Project(s): 

• Duncanson (TE 2018)

Scott Hensley,  JPL,  scott.hensley@jpl.nasa.gov (Presenter)
Bruce Chapman,  JPL,  bruce.d.chapman@nasa.gov
Brian Hawkins,  JPL,  brian.p.hawkins@jpl.nasa.gov
Naiara Pinto,  JPL,  naiara.pinto@jpl.nasa.gov
Razi Ahmed,  JPL,  razi.u.ahmed@jpl.nasa.gov
Konstantinos Papathanassiou,  DLR,  kostas.papathanassiou@dlr.de
Matteo Pardini,  DLR,  matteo.pardini@dlr.de

SAR tomographic methods have proven extremely adept at measuring vegetation vertical structure at a variety of wavelengths including L and P-bands. Measuring the three dimensional structure of vegetation and its changes resulting from either natural or anthropogenic causes are key parameters in monitoring ecosystems and in mapping the spatial variability of the 3-D structure of boreal forests provides information about their biomass and habitat suitability. The NASA/JPL UAVSAR system deployed to Alaska in August/September 2017 and conducted tomographic SAR observations at L-band and P-band. The aim of this proposal is the generate tomographic products using the UAVSAR L-band and P-band tomographic data and combine this data with hyperspectral G-LiHT data that provides information on species variation and canopy characteristics and their spatial gradient. Tomographic profiles will be compared with either G-LiHT or LVIS lidar measurements whenever they are available.

The ABoVE tomographic observations in 2017 near Delta Junction, Alaska consists mostly of relatively short vegetation with mean heights less than 20 m and maximal height less than 25 m. Present within the site are small ponds and stands of lower stature trees and bushes, as well as manmade features such as paved and dirt roads. The underlying terrain topography is mostly gently undulating, with a small mount of mostly grassy ground foliage. The second tomography data acquisition at the BERMS site near Saskatoon, Saskatchewan was planned jointly in cooperation DLR who flew the F-SAR radar and acquired data at L-band and S-band. Forest at this site consisted mostly of Jack pine and Aspen. Many of the stands have substantial undergrowth and ground litter left over from previous logging operations. Both the NASA/JPL UAVSAR and F-SAR flight lines were designed to have overlap with existing LVIS data at the site and to overlap each other.

We plan to use a variety of tomographic profile techniques including Fourier, Capon and compressive sensing algorithms to generate and calibrate vertical profiles in the vegetation for all polarization and compare these with lidar structure profiles. We also intend to extract biophysical parameters like biomass from a combination of structural parameters such as height and canopy thickness derived from the tomographic data and species information from the G-LiHT hyperpsectral data. Additionally, we will compare these estimates with the NISAR biomass estimation algorithm in low biomass (< 100 Mg/ha) regions. Metrics for 3-D structure spatial gradients will be derived and compared across species.

Associated Project(s): 

• Hensley (TE 2018)

Presentation: ASTM5_Poster_Hensley_1_30_57.pptx 

Karl Fred Huemmrich,  NASA GSFC/UMBC,  karl.f.huemmrich@nasa.gov (Presenter)
Petya Krasteva Entcheva Campbell,  JCET/UMBC,  petya.campbell@nasa.gov
Joanna Joiner,  NASA GSFC,  joanna.joiner@nasa.gov
Yasuko Yoshida,  SSAI,  yasuko.yoshida-1@nasa.gov
Craig E. Tweedie,  University of Texas at El Paso,  ctweedie@utep.edu
Elizabeth M. Middleton,  NASA GSFC,  elizabeth.m.middleton@nasa.gov

Describe the goals an methodology of this project to determine in situ relationships between solar induced fluorescence (SIF) and vegetation photosynthetic capacity under different environmental conditions at tundra and boreal forest sites, and to scale these observations from leaf-level to the plot/canopy level and finally to apply these relationships to the landscape level across the ABoVE domain using satellite data. The satellite imagery will be used to describe SIF spatial variability along with diurnal, seasonal and multi-year changes; and we will apply appropriate radiative transfer and physiologically-based models to derive gross primary productivity (GPP) and describe its variability across the ABoVE domain, validating the satellite-derived GPP estimates against flux tower data.

Associated Project(s): 

• Huemmrich (TE 2018)

Presentation: ASTM5_Poster_Huemmrich_1_32_63.pdf 

Philip A Townsend,  University of Wisconsin,  ptownsend@wisc.edu (Presenter)
Eric L Kruger,  University of Wisconsin,  elkruger@wisc.edu
Ryan P Pavlick,  NASA JPL,  ryan.p.pavlick@jpl.nasa.gov
Charles E Miller,  NASA JPL,  charles.e.miller@jpl.nasa.gov
David Schimel,  NASA JPL,  david.schimel@jpl.nasa.gov
Zhihui Wang,  University of Wisconsin,  zwang896@wisc.edu
Adam Chlus,  University of Wisconsin,  chlus@wisc.edu
Ting Zheng,  University of Wisconsin,  tzheng39@wisc.edu
Fabian Schneider,  NASA JPL,  fabian.schneider@jpl.nasa.gov

Foliar functional traits are widely used to describe the chemical, morphological and physiological characteristics that drive plant photosynthesis, growth and responses to stress. A large body of research has shown that imaging spectroscopy data (hyperspectral imagery) can be used to measure/map many important foliar traits, which then can be used to estimate functional diversity, parameterize ecosystem models or track vegetation response to a changing environment. We plan to leverage the ongoing AVIRIS-NG acquisitions in the ABoVE domain to quantify trait variation between and within ecosystem types, and across environmental gradients in the region. Focusing on LMA, nitrogen, lignin, non-structural carbohydrates, chlorophyll and possibly phosphorus, this work will extend and validate existing cross-biome imaging spectroscopy trait models, and will build on concurrent work being developed using the NEON Airborne Observation Platform (AOP) and AVIRIS-Classic data throughout Alaska and the US. The resulting maps (and their uncertainties) will help fill gaps in existing field sampling networks and expand our grasp of the full functional dimensionality of Arctic and boreal vegetation. As part of this effort, we will implement an open-source image processing workflow developed collaboratively between UW-Madison and JPL on the record of ABoVE AVIRIS-NG imagery to produce for the community a complete set of topographically and BRDF corrected AVIRIS-NG imagery. Our work will 1) yield a more complete understanding of the variation in plant traits that drive ecosystem functioning, suitable as a foundation to track responses to environmental variability and change of Arctic and boreal ecosystems, and 2) extend existing imaging spectroscopy algorithms already applied and validated for temperate and sub-boreal forests, agricultural systems and Mediterranean ecosystems to Arctic tundra and boreal forests, in preparation for recommended future missions such as the Surface Biology and Geology (SBG) concept in the recent Decadal Survey. This second result represents a significant step in the development of broadly applicable algorithms needed to implement spectroscopy-based trait mapping (e.g., by SBG) at the global scale. Comprehensive mapping of functional traits in Arctic and boreal systems will also serve to inform models projecting future changes in these critical biomes.

Associated Project(s): 

• Townsend (TE 2018)

Clayton Drew Elder,  Jet Propulsion Laboratory,  clayton.d.elder@jpl.nasa.gov (Presenter)
Katrin Kohnert,  Leibniz Institute for Freshwater Ecology and Inland Fisheries,  kohnert@igb-berlin.de
Andrew Kenji Thorpe,  JPL,  andrew.k.thorpe@jpl.nasa.gov
David Ray Thompson,  Jet Propulsion Laboratory / Caltech,  david.r.thompson@jpl.nasa.gov
Torsten Sachs,  GFZ German Research Centre for Geosciences,  torsten.sachs@gfz-potsdam.de
Charles Miller,  NASA JPL,  charles.e.miller@jpl.nasa.gov

Arctic CH4 emissions could more than double by mid-century as the region adapts to rapid sea-ice decline and amplified warming1,2. Yet despite decades of research, the high spatial and temporal heterogeneity of CH4 emissions and the inaccessibility of Arctic regions has complicated efforts to reconcile top-down and extrapolated bottom-up CH4 emission budgets. As a result, these approaches rarely converge, scaling of site-level observations across landscapes or regions suffers large uncertainties, and forecasts of CH4 emissions from high latitudes span two orders of magnitude through year 2300 (ref. 2). Here, we leverage two novel approaches: the GFZ/AWI AirMeth airborne eddy covariance observation and NASA’s Airborne Visible/Infrared Imaging Spectrometer- Next Generation (AVIRIS-NG) in the Mackenzie Delta, NWT, Canada to directly observe Arctic CH4 emission patterns at unprecedented spatial scales. The AirMeth survey, conducted in summer of 2012 and 2013, derives a regional scale (10,000 km2) CH4 flux map at 100 x 100 m spatial resolution using airborne eddy covariance and multiple flight transects3. Alternatively, the AVIRIS-NG observes excess CH4 absorption in the atmospheric column between the aircraft and the surface (CH4 enhancement), and is sensitive enough to detect CH4 hotspots at 5 x 5 m spatial resolution4. In summer of 2017, the AVIRIS-NG imaged over 2.0 x 108 pixels in a 5,000 km2 mosaic designed to overlap with the prior AirMeth survey in the Mackenzie Delta.

We exploit spatial and temporal differences in each survey, and the fact that AVIRIS-NG is only sensitive to CH4 emission hotspots on land, to differentiate temporal variability in CH4 emissions as well as dispersed vs. concentrated hotspot emission patterns across the Delta landscape. AVIRIS-NG pixels were aggregated to the AirMeth pixel grid and the median CH4 enhancement was compared to the AirMeth-derived fluxes. Preliminary results indicate that there is no correlation between the two observations. This suggests that either dispersed, uniformly elevated emissions (below the AVIRIS-NG detection limit), or emissions over water (undetectable by AVIRIS-NG), are the primary CH4 emission modes in the Delta -as opposed to concentrated hotspots on land. In instances where the independent observations were consistent (< 10% of study area), emissions are thus temporally persistent between the observation periods and are predominantly CH4 hotspots. This suggests that CH4 emissions in these regions are likely dominated by geologic seeps, which are relatively insensitive to the interannual environmental variability that regulates ecological-type CH4 emissions. Ongoing work further investigates the relationship between the two observation techniques and aims to coordinate AirMeth and AVIRIS-NG surveys in time and space in the summer of 2019. This research has the ability to bridge scale gaps in existing CH4 observation systems and reshape our understanding of CH4 emission patters across regional scales.

References:

1. McGuire, a. D. et al. Biogeosciences 9, 3185–3204 (2012).
2. Schneider Von Deimling, T. et al. Biogeosciences 12, 3469–3488 (2015).
3. Kohnert, K., Serafimovich, A., Metzger, S., Hartmann, J. & Sachs, T. Sci. Rep. 7, 1–6 (2017).
4. Thompson, D. R. et al. Atmos. Meas. Tech. 8, 4383–4397 (2015).

1. McGuire, a. D. et al. Biogeosciences 9, 3185–3204 (2012).
2. Schneider Von Deimling, T. et al. Biogeosciences 12, 3469–3488 (2015).
3. Kohnert, K., Serafimovich, A., Metzger, S., Hartmann, J. & Sachs, T. Sci. Rep. 7, 1–6 (2017).
4. Thompson, D. R. et al. Atmos. Meas. Tech. 8, 4383–4397 (2015).

Associated Project(s): 

• Miller (TE 2016)
• Miller (TE 2018)

Presentation: ASTM5_Poster_Elder_1_34_19.pdf 

Lei Hu,  NOAA/CIRES,  lei.hu@noaa.gov (Presenter)
Colm Sweeney,  NOAA Earth System Research Laboratory,  colm.sweeney@noaa.gov
Arlyn Elizabeth Andrews,  NOAA Earth System Research Laboratory,  arlyn.andrews@noaa.gov
Kathryn McKain,  University of Colorado,  kathryn.mckain@noaa.gov
Stephen A. Montzka,  NOAA,  stephen.a.montzka@noaa.gov
John Miller,  NOAA Earth System Research Laboratory,  john.b.miller@noaa.gov
John Henderson,  AER,  jhenders@aer.com
Aleya Kaushik,  NOAA/CIRES,  aleya.kaushik@noaa.gov
Andrew R Jacobson,  Cooperative Institute for Research in Environmental Sciences (CIRES),  andy.jacobson@noaa.gov
Ed J. Dlugokencky,  National Oceanographic and Atmospheric Administration,  ed.dlugokencky@noaa.gov
Pieter Tans,  NOAA,  pieter.tans@noaa.gov

As the atmospheric CO2 burden grows, substantial increases have been observed in the amplitude of the CO2 seasonal cycle, especially in the Arctic. Earlier studies have suggested multiple possible causes for the increase, including an earlier onset and lengthening of the plant growing season, enhanced photosynthetic uptake, increased respiration from labile carbon, and transport from lower latitudes. However, it is still under huge debate on how much each individual possible cause has contributed to the enhanced seasonal amplitude, and which cause dominates this change in the Arctic. In our ABoVE Phase II project, we will leverage the rich in situ and remote sensing datasets collected in recent years (from ABoVE, CARVE, OCO2, SMAP, TROPOMI etc.) and high resolution inverse modeling to provide more quantitative information on the magnitudes and variations of gross primary production (GPP) and respiration in the Arctic. We will also quantify influences associated with transport of mid-latitude fluxes using the global CarbonTracker model. In this presentation, we will discuss our detailed plan on high-resolution inverse modeling analyses using CO2 and COS over the well-sampled North American Arctic, how we will extrapolate the regional inversion results to the pan-Arctic and quantify the impact of transport from lower latitudes. Finally, we will present some preliminary results from analyzing long-term atmospheric CO2, COS, and δ13CO2 observations collected from various locations in the Arctic and some of the exciting new features and implications.

Associated Project(s): 

• Hu (TE 2018)

David Ellison Butman,  University of Washington,  dbutman@uw.edu (Presenter)
Tamlin Pavelsky,  University of North Carolina Chapel Hill,  pavelsky@unc.edu
Laurence C. Smith,  UCLA,  lsmith@geog.ucla.edu
Robert G.M. Spencer,  Florida State Univ.,  rgspencer@fsu.edu
Rob Striegl,  USGS,  robstriegl@gmail.com
Kimberly P. Wickland,  United States Geological Survey,  kpwick@usgs.gov

Inland waters represent greater than 3% of the total continental surface of the pan-arctic, with densities of surface waters exceeding 10% in shield bedrock and low slope, deltaic environments dominated by lakes. Carbon emissions from high latitude lakes can exceed 340Tg-C yr-1, and release upwards of 16.5 Tg-C-CH4 yr-1. This represents one of the largest natural sources of atmospheric methane from the Arctic-boreal region. Surface water significantly impacts landscape-scale estimates of carbon emissions. Currently the magnitude and extent of Arctic-Boreal seasonally inundated land remains unknown, and we hypothesize that the region of regularly inundated soils are hotspots for the cycling of carbon and represent a component of the landscape highly vulnerable to change. Airborne Campaigns (AAC) acquired airborne remote sensing data crossing broad environmental gradients. UAVSAR and AirMOSS, are both capable of measuring inundation extent under vegetation. AVIRIS-NG, can quantify vegetation extent and type. AirSWOT measures water surface elevation (with radar interferometry) and aquatic vegetation boundaries (with an infrared camera) simultaneously. Finally, LVIS measures water and land elevation and evidence of emergent aquatic plants. These previously acquired datasets offer a unique opportunity to study seasonal Arctic-Boreal inundation patterns through vegetation and link these patterns to the cycling of carbon. Here we propose to quantify inundation extent across four important classes of hydrologic connectivity representative of the ABoVE domain: open water, permanent inundation, transient inundation, and dry uplands. Using ABoVE airborne remote sensing data we will map these classes across a hydrological gradient of landscapes: fluvial-connected lowland (the Peace-Athabasca Delta, Canada); fluvial-disconnected lowland (Yukon Flats, Alaska), and bedrock controlled (Yellowknife-Daring Lake, Canada). With additional support from the U.S. Geological Survey, we will conduct a series of comprehensive field campaigns to quantify the variability in carbon flux among these different classes and landscapes, with the goal of identifying the importance of inundated lands in the net carbon balance of Arctic-Boreal aquatic environments.

Associated Project(s): 

• Butman (TE 2018)
• Miller (TE 2018)
• Striegl (TE 2014)

Presentation: ASTM5_Poster_Butman_1_37_69.pdf 

Yonghong Yi,  JPL,  yonghong.yi@jpl.nasa.gov (Presenter)
Richard Chen,  University of Southern California,  chenrh@usc.edu
Mahta Moghaddam,  University of Southern California,  mahta@usc.edu
John S Kimball,  University of Montana,  john.kimball@mso.umt.edu
Charles Miller,  NASA JPL,  charles.e.miller@jpl.nasa.gov

The contribution of cold-season soil respiration to the Arctic-boreal carbon cycle and its potential feedback to global climate remain poorly quantified, partly due to a poor understanding of soil moisture variation and its effects on the seasonal freezing and thawing (F/T) of the active layer in the permafrost landscapes. Longwave radar can potentially provide useful information on subsurface soil properties regulating active layer F/T process. Extensive airborne P-band (430 MHz) polarimetric SAR (PolSAR) backscatter data from the NASA AirMOSS airborne sensor were collected during the recent ABoVE campaigns. In this study, we examine the sensitivity of the airborne P-band radar to active layer conditions in northern Alaska tundra using both airborne radar measurements and a modelling framework. The modelling framework mainly includes a permafrost hydrology model and a radar scattering model, which are linked through a soil dielectric module. A key challenge in capturing active layer dynamics is in accurately representing thermal, hydraulic and dielectric profiles in highly organic tundra soils; therefore, a new organic soil parameterization scheme was developed to harmonize key parameters used in both models. Preliminary results show that simulated soil dielectric profile and radar backscatter were closely linked to soil organic carbon (SOC) decomposition state; areas with less decomposed SOC in the surface generally show drier surface and wetter deeper active layer, and overall higher radar backscatter at P-band. This characteristically different soil dielectric profile also affects the sensitivity of P-band radar backscatter to the seasonal active layer freezing process, modulated by early snow cover conditions. More accurate representation of soil organic carbon and soil dielectric profiles helps reduce uncertainties in both the radar retrievals and soil model simulations of active layer thickness and saturation, enabling improved monitoring and assessment of permafrost active layer dynamics over large extent.

Associated Project(s): 

• Kimball (TE 2018)