earth21.txt

Katey Anthony (PI)/Natalie Tyler (FI) 
University of Alaska, Fairbanks 
20-EARTH20-0079, Understanding Methane Superseeps in Alaskan Lakes Using Remote Sensing and Field Work 
 
The proposed research combines remote sensing (RS) data with targeted field observations to better quantify greenhouse gas emissions in a warming Arctic. Specifically, I will quantify and characterize 14C-depleted methane (CH4) ebullition (bubbling) in Arctic and boreal lakes associated with deep-sourced gas migration through permafrost and near-melting glaciers using space-borne RS analysis. 
 
My objective is to develop regional maps of high flux, 14C-depleted CH4-emitting seeps, or  superseeps, using Synthetic Aperture Radar (SAR) analysis in different geologic/cryospheric environments and improve understanding of the variability among the potential superseep features detected in SAR. 
 
I will expand upon my work from a previous NASA ABoVE NNN12AA01C project, in which I explored the efficacy of mapping superseeps in three northern Alaska regions using SAR and developed landscape-scale maps of SAR-detected Potential Superseeps (SPS). My goals in the proposed work are to: (i) Characterize the variability in SPS morphologies observed during my previous work; (ii) Improve understanding of superseep flux dynamics; (iii) Evaluate SPS occurrence under different cryosphere conditions as seen in southcentral Alaska, where lakes within the footprints of retreating glaciers have confirmed superseeps emitting fossil CH4; and (iv) Quantify decadal changes in SPS over time. 
 
This project will include: (1) Collection of new and synthesis of existing field data on superseep CH4 ebullition; (2) Characterizing variable SPS morphologies observed in the previouslyanalyzed Utqiaġvik (UTQ) region based on physical presentation and radiometric parameters; (3) Applying my novel methodology to the Cordova-Katalla (CDV) region in southcentral Alaska to create a new regional SPS map; (4) Utilizing optical remote sensing as validation of SARdetected seeps; (5) Analyzing historic SAR data in UTQ to determine changes in SPS from the early 1990’s to 2006-11; (5) Presentations at UTQ schools to inspire learners of all ages; (6) Presentation of results at AGU’s Fall Meeting. Data and maps will be delivered to the Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) for sharing with the broader science community. Project goals will meet the objectives NASA’s Earth Science Division to understand changes in the Earth System and addresses NASA interests in using RS to quantify and understand the global carbon cycle. 
 
  
Dennis Baldocchi (PI)/Tianxin Wang (FI) 
University of California, Berkeley 
20-EARTH20-0225, How Advective is California's Central Valley: A Novel Approach to Quantify Sensible Heat Advection Using Bionmeteorology and Satellite Remote Sensing 
 
California, the world’s 5th largest economy, is facing unprecedented water scarcity and novel challenges to manage its increasing water demand. Agriculture in California’s Central Valley has a distinct chessboard pattern with irrigated fields surrounded by dry or fallowed fields. This special pattern can induce higher water loss through the advection of sensible heat, where the local evapotranspiration rate can often exceed the available energy. As a result, the extra amount of irrigation needs to be adjusted accordingly, but how much more?  
 
Advection can be quantified accurately via an array of eddy covariance flux towers, but it is financially impracticable to do so. Considering the operational feasibility, satellite remote sensing is an ideal tool to quantify the effect of advection. However, there is no common consensus on how to account for the effects of advection in evapotranspiration models.  
 
The overarching goal of this project is to quantify and account for the advection effects in California’s Central Valley. We will use eddy covariance measurements, Landsat 8 data (2013present), ECOSTRESS data (2018-2021), ancillary remote sensing data (GLDAS), and weather data (CIMIS) to quantify the effect of sensible heat advection locally. We will use MODIS Terra/Aqua and GLDAS to quantify the regional advection in the Central Valley. This project will improve our qualitative and quantitative understandings of features that would attract or intensify advection and the magnitude of advection. We will use the state-of-art biophysical model, CanVeg, with an Atmospheric Boundary Layer model to translate the vertical remote sensing observations into a transect view that accounts for the horizontal sensible heat advection. This project will also offer a new model, Gridded Advection Index Assimilation (GAIA), to simulate advection and provide a new gap fill method. To improve the remote sensing validation dataset, we will use the developed model to determine, if any, how advection could contribute to the energy balance closure issues for the AmeriFlux sites.  
 
This project will offer new understandings on how sensible heat advection affects evapotranspiration, as well as how to account for the advection. This proposed research aims to use NASA data and technology to improve water management strategies. It is directly related to the Terrestrial Hydrology Program, NASA Energy and Water Cycle Study, and the mission of ECOSTRESS and OpenET. By enabling the quantification of advection from space, this project brings new insights on how the water and energy cycle behaves under advection and provides a novel approach to improve the current water management. 
 
  
Adrian Borsa (PI)/Ho-man Lau (FI) 
University of California, San Diego 
20-EARTH20-0123, Inferring Hydrologic Mechanisms Through Geodetic Terrestrial Water Estimates 
 
Freshwater is vital to the development of our society, and less than 1 percent of the entire planet's freshwater reserves are in the form of surface and groundwater which can be extracted and consumed immediately. Although water is rechargeable as it circulates through the hydrologic cycle, some regions may be under water-stress due to uneven usage in time and space. With impacts from increasing human population and a changing climate, managing freshwater resources has become one of the foremost challenges on the 21st Century. Current generation of hydrologic models often lack the ability to capture the total amount of water on land (refers to terrestrial water storage), as well as year-to-year variations. Recent geodetic advances, however, provide alternate ways to estimate the terrestrial water storage. Changes in water mass on land induce gravity anomalies and vertical ground motion, which can be measured by satellite data and inferred through quantitative procedures. In this proposal, I aim to improve estimations of water mass in both time and space by combining gravity and ground movement data, and use these estimates in understanding hydrologic processes: in particular, I will address the following topics: 
 
1) Combining geodetic methods in mapping high resolution TWS estimates; and 2) Inferring seasonal and long-term runoffs patterns with geodetic TWS estimates and hydrologic data. 
 
Results of this proposed work will improve not only geodetic analysis method, but also provide insights to how water moves laterally along the Earth's surface, tackling problems relevant to society such as flood risk mitigation. These objectives will directly address NASA Earth Surface and Interior research area’s objectives of "obtaining accurate continuous observation of, and improvement in models for, global water mass balance on seasonal to decadal timescales", and "improving spatial and temporal resolution of global vertical deformation and gravity fields". Collaborations with hydrologists and climatologists will allow for wider applications of the proposed work. Results and products derived from this proposed work will be accessible and available publicly for the rest of the scientific community and beyond. 
 
  
Jodi Brandt (PI)/Nicholas Kolarik (FI) 
Boise State University 
20-EARTH20-0340, Data Fusion and Back-Casting for Describing Water Resource Dynamics and the Associated Effects of Land Management in a Semi-Arid System 
 
Spatial and temporal distributions of water resources in semi-arid systems have become increasingly uncertain due to climate change and increasing development. Earth observation (EO) data provide an efficient, cost effective way to monitor water resources over large geographies to inform improved management. However, the available products are insufficient in semi-arid environments due to coarse spatial resolution of the data used relative to the water resources that sustain these systems. In this project, I will develop methods to use freely available EO data in Google Earth Engine, an open-access platform, to measure inter- and intraannual availability of water resources in the semi-arid American West at an unprecedented spatial scale. First, I will create 10 m maps of water resources using the Sentinel-1 and -2 time series. Through a data fusion approach, I will incorporate synthetic aperture radar (SAR) data from Sentinel-1 with optical data from Sentinel-2 and quantify its value for mapping water resources in preparation for the NISAR launch. Second, using these classifications, I will develop a machine learning regression model to estimate proportions of water resources in Landsat pixels throughout the time series to create a product similar to existing products for forests (e.g. Global Forest Change) albeit limited to the study area. Finally, I will measure how land use and land management regimes have influenced water resources over time using a Bayesian modeling framework and a quasi-experimental design. 
 	This work will support NASA’s goals from both methodological and societal perspectives and will directly address objective H-2c in the Decadal Survey by quantifying how changes in land use and land cover threaten the sustainability of future water supplies. Subtle changes to water resources in the West have tremendous impacts environmentally, alter the provision of ecosystem services, and affect human livelihoods at the regional scale. Producing a high resolution time series of water resources in semi-arid systems is of tremendous importance and this project represents a way to monitor water resources that can be adapted for any study 
area and implemented by anyone with internet access. Through analyzing Sentinel C-band SAR, I am preparing for incorporation of NISAR L-band data that will be well-suited for monitoring water resources at spatial and temporal scales meaningful to semi-arid systems. I will also quantify how changes in land use and management influence the sustainability of future water supplies. The maps of water resources proportions I produce could further be used to measure changes associated with climate, wildfire, or policy. 
 
  
Roland Burgmann (PI)/Danielle Lindsay (FI) 
University of California, Berkeley 
20-EARTH20-0194, From Satellite to Serpentinite: Using Space Geodesy to Determine Mantle Dynamics at the Mendocino Triple Junction 
 
Surface uplift rates, the lithospheric structure, and the flow of mantle below are intrinsically linked. Vertical surface motions driven by geodynamic processes can be decomposed into isostatic and dynamic components sensitive to crustal thickness and mantle convection. Smallscale mantle convection has been attributed to vertical surface motions correlated at distances of up to a few hundred kilometers in Southern California and along the Western US Intermountain Belt. This project focuses on using space-geodetic observations of vertical velocities from InSAR and GPS to determine contributions from mantle dynamics at the Mendocino Triple Junction (MTJ), Northern California. We ask, 'what are the driving forces controlling active uplift around the MTJ?' 
Our approach is to take a snap-shot of the present-day surface deformation using InSAR to generate a spatially continuous, three-dimensional velocity field. Following this, we will separate contributions of earthquake-cycle deformation on the plate boundary faults from the deeper-seated geodynamic processes. In removing known fault-related deformation sources from the InSAR velocity field, we test two hypotheses; 1) Active faults are solely responsible for regional uplift, and no geodynamic contribution is required, and 2) Active faults produce some regional uplift, but a geodynamic contribution is required to account for the remainder. If, after removing known fault-related deformation sources, there remains a coherent long-wavelength spatial pattern of uplift, we can pursue further geodynamic modeling to constrain the triple junction's three-dimensional structure and dynamics. 
Our first task is to collate data and optimize InSAR processing techniques to generate a line-ofsight (LOS) times series for each data frame. The independent LOS velocity fields will be integrated with continuous GPS and decomposed into a three-dimensional velocity field. Glacial isostatic adjustments (GIA), hydrological, and viscoelastic postseismic deformation corrections will be applied. The contributions from elastic earthquake-cycle deformation on plate boundary faults will be modeled and removed from the velocity field. Assuming all short-wavelength deformation in regions characterized by tectonic activity is accounted for, we consider the residual velocity field solely due to deep-seated geodynamic driven deformation. Finally, we will determine the optimal modeling approach and implementation for geodynamic drivers at the MTJ constrained by tomographic models and our geodetic observations. 
This project applies remote sensing techniques to improve our understanding of mantle and lithospheric structure and dynamics. We will test whether InSAR derived velocity fields can image millimeter-scale vertical deformation with spatial gradients extending for 10s to 100s of kilometers driven by geodynamic processes. By better characterizing the dynamics between Earth's interior and present-day surface deformation, we can improve our capability to assess and respond to natural hazards. In approaching this problem with InSAR, we ask what presentday interseismic surface deformation tells us about the long-term driving forces responsible for building topography and stimulating seismic cycles of faults in the MTJ region. 
 
  
Seth Bushinsky (PI)/Shannon McClish (FI) 
University of Hawaii, Honolulu 
20-EARTH20-0338, Characterizing the Impact of Seasonal Sea Ice on Phytoplankton Blooms and Net Community Production with Integrated Satellite and Biogeochemical Profiling Float Observations 
 
The Southern Ocean is a strong sink of atmospheric carbon but our mechanistic understanding of what drives this carbon uptake has been limited by a lack of year-round data. This is especially true in the Southern Ocean seasonal sea ice zone (SSIZ), a region of seasonal extremes and significant projected changes. Organic carbon fixed by primary producers in the euphotic zone that is not respired in the surface layer represents a flux of carbon to the deep ocean, known as Net Community Production (NCP). This biologically driven carbon export is an important pathway that effectively sequesters atmospheric carbon dioxide. Quantifying the net annual NCP (ANCP) is key to accurately understanding air-sea CO2 fluxes and the flow of carbon and nutrients from the surface to the ocean interior. ANCP estimates have yet to be established in the SSIZ due to a historic lack of wintertime observations. Establishing circumpolar estimates of NCP and ANCP in the SSIZ and analyzing how sea ice influences regional patterns of organic carbon production by phytoplankton are critical to understanding the Southern Ocean carbon cycle and potential future change.   
 
Satellite observations of the sea surface are used to derive estimates of chlorophyll (chl-a) and particulate organic carbon (POC) that form the basis of the most spatially and temporally comprehensive net primary production (NPP) and NCP estimates. However, satellite observations are limited to ice-free conditions and the upper few meters of the ocean, potentially excluding periods of early, under-ice phytoplankton blooms and/or subsurface phytoplankton blooms resulting in underestimation of NPP, NCP, and ANCP. Newly available annual biogeochemical observations from profiling floats have greatly expanded the in-situ measurements which can be used to constrain satellite observations. I will establish the temporal correlation between float and satellite-derived chl-a and POC to identify potential seasonal biases in satellite estimates. Additionally, I will use in-situ nitrate and oxygen observations to produce circumpolar estimates of NCP and ANCP in the SSIZ. Estimates of NCP and ANCP will be compared to phytoplankton bloom characteristics to investigate the link between production estimated by satellites and subsequent carbon export. Finally, to investigate if observed regional variations in the timing and extent of sea ice are reflected in regional biological processes, I will correlate seasonal cycles in sea ice concentration observed by satellites with float and satellite-derived chl-a and POC and float-derived NCP.  
 
The proposed work supports NASA Earth Science strategic objective 1.1 and NASA's Ocean Biology and Biogeochemistry program by addressing gaps in knowledge of Southern Ocean carbon cycling by: providing novel estimates of in-situ ANCP, analysis of sea ice modulation of biological processes, and assessment of known uncertainties in satellite ocean color which can be used to improve satellite algorithms of NPP and NCP and Earth system models. Furthermore, this work will reveal drivers of ecosystem dynamics by identifying how sea ice influences phytoplankton blooms and ANCP, critical to a mechanistic understanding of how climate change will affect future Southern Ocean biogeochemistry. 
 
  
T. Caughlin (PI)/Andrii Zaiats (FI) Boise State University 
20-EARTH20-0326, Integrating UAS, Landsat, and ECOSTRESS to Fit Landscape-Level Demographic Models for Climate-Resilient Ecological Restoration 
 
Quantitative forecasts of ecosystem recovery could play a major role in spatial planning to assist restoration efforts. However, a scale mismatch between the limited spatial extent of field plots and the large spatial extent of many disturbances has impeded the development of restoration ecology as a predictive science. Sagebrush steppe of the western United States exemplifies an ecosystem where efforts to restore millions of hectares of degraded land require landscape-scale predictions. The population dynamics of big sagebrush (Artemisia tridentata) drive ecosystem function in high deserts of the western U.S. and are primarily studied using field plots. We will scale up our understanding of big sagebrush demography by developing models to quantify sagebrush population dynamics from remotely sensed data that span a wide range of spatial and temporal scales.  
 	Demographic models that integrate data from across plant life cycles represent a powerful tool for ecological forecasting. These models are most often fit using repeat measurements of plant size, including field data on individual plant growth, survival, and recruitment. Such field data are logistically challenging to collect at landscape and regional scales. We propose that emerging remote sensing data, including uncrewed aerial systems (UAS), thermal-infrared sensors (TIR), and the multi-decadal Landsat archive, can also serve as data sources for demographic models. We will leverage remote sensing to derive demographic models and focus on their application to post-fire restoration strategies.  
Uncrewed Aerial Systems (UAS) present an unprecedented opportunity to measure changes in individual plant size over time at the landscape level. We will use very high-resolution imagery from several intensively studied landscapes in the Great Basin to directly observe big sagebrush demographic rates at various stages of recovery. Our approach will apply object-oriented image analysis to detect the size and location of individual plants, including demographic changes over time (i.e., growth, survival, and recruitment). We will go beyond previous applications of UAS that have mostly focused on mapping spatial patterns to model ecological processes in dynamic landscapes.  
 	While high-resolution imagery can detect changes in plant size, novel thermal sensors enable measurements related to plant physiological stress, with relevance to demography and ecological restoration. We propose to use ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) to quantify relationships of UAS-derived growth and survival with seasonal thermal and drought trajectories. Our approach will demonstrate the capacity of ECOSTRESS to detect shifts in plant physiological function in a natural setting, where thermal stress is linked to long-term demographic changes. Regional droughts are expected to intensify in the Great Basin; our work will inform climate-resilient restoration strategies in an increasingly disturbed ecoregion. 
 	The Landsat archive provides an unmatched source of information on vegetation dynamics. The National Land Cover Database (NLCD) documents over 30 years of big sagebrush percent cover data that enable demographic analysis of plant populations at scales beyond any field monitoring efforts. We will parallel our UAS analysis informed by ECOSTRESS to develop scalable demographic models using NLCD and Landsat-derived thermal and drought time series. Spatially explicit demographic models at the regional scale will allow land managers to leverage a rich body of ecological theory and quantitative tools to understand the past and predict the future of populations. The proposed research will also advance the use of remote sensing data for ecological inference and mathematical modeling of population dynamics. Altogether, we will produce cross-scale models for ecosystem recovery to provide decision-support for land management in the western US. 
 
  
Shuyi Chen (PI)/Lucero Yakelyn Ramos Jauregui (FI) 
University of Washington, Seattle 
20-EARTH20-0039, Understanding Multiscale Air-Sea Interaction Using NASA Precipitation and Salinity Data and Coupled Modeling 
 
The Madden-Julian Oscillation (MJO) and El Nino-Southern Oscillation (ENSO) are most dominant tropical phenomena affecting global high-impact weather such as tropical cyclones, extreme precipitation and drought, heat waves, and severe storm outbreaks over a wide range of time and spatial scales. The Earth System is projected to continue experiencing warming in the coming decades. In response to this warming, the MJO and ENSO as well as high-impact weather will change, including the precipitation pattern, the distribution, frequencies and intensity of floods and droughts, all affecting the water quality and food security. Air-sea interaction plays an important role in the global hydrological water cycle. Improving the representation of the multiscale air-sea interaction processes in Earth-system models will be critical in global weather and climate predictions.  
 
The science objective of the proposed research is to better understand multiscale air-sea interaction processes (from mesoscale convection/precipitation, to the MJO and ENSO) that affect the transition from La Niña to El Niño conditions over the tropical Pacific. We will use both satellite and in-situ observations to examine the variability of precipitation, salinity, SST, and upper ocean stratification. A high-resolution, fully coupled atmosphere-ocean model simulation will be used to investigate the air-sea interaction processes, which is complimentary to the observational study. This study also explores new tools in machine learning (ML) to 
 
enhance our capability in better identify multiscale features and understand physical processes in air-sea interaction relevant to high-impact weather in a changing climate. 
  
The most critical variables to study air-sea interaction processes are provided by NASA’s satellite data products, including precipitation, sea surface temperature (SST), sea surface salinity (SSS), and surface winds. These products have enhanced our understanding of the heat and freshwater fluxes over the ocean. However, the current SSS products (Aquarius and SMAP) integrate over relatively longer time scales (~8 days). In order to fill the gap on relatively shortlived, small-meso-scale variability where the precipitation and freshwater impact air-sea interaction, a high-resolution atmosphere-ocean coupled model will be used to describe atmosphere-ocean coupled processes and their effect on SST and SSS across a range of spatial and temporal scales. 
 
  
Daniel Dawson (PI)/Qin Jiang (FI) Purdue University 
20-EARTH20-0344, Investigating How Changing Environmental Conditions Favor Tornadoes Shifting Eastward over North America 
 
Tornadoes are one of nature s most hazardous phenomena, causing extensive property damage and casualties every year. The traditional "Tornado Alley" where tornadoes and their parent storms are most frequent and intense is located in the U.S. Great Plains. However, recent studies have provided evidence that the center of tornado activity is shifting eastward from the Great Plains (GP) to the southeast U.S. (SE-US). This raises the following question: why is the center of greatest tornado activity shifting  eastward?  In particular, how do changing trends in the large-scale environments of tornadic storms associated with climate change contribute to this eastward shift? Many aspects of how the large-scale environment of tornadic storms relates to tornado occurrence and intensity are well-established, and recent studies have shown how interactions with the underlying surface through the action of surface drag can profoundly affect tornado development and intensity. However, few studies have investigated how the behavior of tornadic storms and attendant tornadoes respond to the changing largescale environments between the GP and SE-US, especially in the context of the large differences in surface roughness characteristics due to differences in land cover.  More generally, the connection between climate change and severe weather is still unclear.  This proposal seeks to fill this gap by analyzing NASA observational (MODIS) and reanalysis (MERRA-2) datasets along with performing high resolution idealized simulations of tornadic storms and attendant tornadoes. This work has the following central objectives: 
 
1.	Analyze MERRA-2 reanalysis data to understand what aspects of the temperature, moisture, and wind profiles that define the environments of tornadic storms are most associated with the eastward shift in tornado activity, 
 
2.	Use MODIS data to uncover statistical relationships between differences in the underlying surface roughness (due to differences in land cover) with historical tornado occurrence, and 
 
3.	Perform high-resolution idealized numerical simulations to investigate how these differing large-scale environments and surface roughness characteristics between the GP and SE-US together affect the structure and evolution of tornadic storms and the physical mechanisms behind tornado development within them. 
 
By generating new insights into how tornadoes and tornadic storms are responding to the changing climate of the present and recent past, this project will advance our ability to anticipate such changes in the future warming climate. This is directly aligned with the Decadal Survey Priority of NASA's Strategic goal in Earth Science to  improve the capability to predict weather and extreme weather events.  
 
  
Tim DeVries (PI)/Kana Yamamoto (FI) 
University of California, Santa Barbara 
20-EARTH20-0227, Constraining CDOM Dynamics and Reactivity in the Global Ocean 
 
Chromophoric dissolved organic matter (CDOM) is a fraction of the DOM pool that is optically active in the solar radiation band, which holds major implications for studying the ocean via satellite ocean color and light-dependent biogeochemical processes. As the dominant lightabsorbing material in the ocean, CDOM is a key parameter for ocean color algorithms used for quantifying phytoplankton biomass estimates and long-term trends. Without a mechanistic assessment of CDOM dynamics, current ocean color algorithms can only rely on assumptions and empirical relationships of optical properties which have many uncertainties. There is also rising interest in using CDOM as a biogeochemical tracer for refractory DOC dynamics in the open ocean, an important carbon reservoir, thereby taking advantage of satellite-derived CDOM measurements and expanding the utility of ocean color data in understanding biogeochemical processes governing oceanic carbon. But to do so, we require a comprehensive understanding of CDOM dynamics and its relationship to DOC throughout the global ocean. CDOM in itself is also important because it controls light attenuation in the euphotic zone, and CDOM variability can affect biogeochemical cycles of climatically important nutrients such as carbon, sulfur, and nitrogen. Understanding CDOM has become increasingly necessary as advancements in oceanography involve the use of NASA ocean color satellite sensors, but we lack a cohesive, quantitative framework for CDOM dynamics which hinder our ability to monitor changes in our marine ecosystem and the global carbon cycle. We propose to quantify the biogeochemical processes governing CDOM in the open ocean and its reactivity by creating a seasonally resolved, mechanistic model of CDOM. Our objective will be met by answering the following science questions: 
SQ1. What are the major sources and sinks of CDOM in the open ocean and the relative contributions/distribution of allochthonous and autochthonous inputs? 
SQ2. What is the turnover rate of marine CDOM and are there different pools of lability in the interior ocean? 
SQ3. How do CDOM pools relate to DOC pools of varying reactivity in the open ocean? 
 
To answer these questions, we will build a series of numerical models using satellite ocean color data, in-situ absorption measurements, machine learning models, and a seasonallyvarying, global ocean circulation model. Our work will produce the first seasonal biogeochemical model of CDOM, which will be immensely advantageous in cross-validating and aiding the separation of in-water, optical constituents from ocean color products of SeaWIFS, MODIS-Aqua, and the upcoming PACE mission and open collaborative channels with those working to develop more robust ocean color algorithms. Our study will also improve our understanding on the origin and evolution of the overall DOM pool in the interior ocean by providing insight into the dynamics of semi-labile and refractory DOM via the CDOM cycle. These aspects are directly relevant to the Carbon Cycle and Ecosystem focus area and the Climate Variability and Change program of NASA s Earth Science Division and applies to Priority 1: Exploration and Scientific Discovery and Priority 2: Innovation of NASA s Science 2020-2024: A Vision for Scientific Excellence. 
 
  
Christopher Doughty (PI)/Jenna Keany (FI) 
Northern Arizona University 
20-EARTH20-0315, Investigating African Forest Elephant Impacts on Forest Structure and Carbon Balance Using Multiscale Lidar Techniques 
 
African forest elephants (Loxodonta africana cyclotis) face severe threats from poaching and habitat loss, with estimated population losses between 62-81% in central African since the early 2000’s (Maisels et al., 2013). Forest elephants are known to transport nutrients across gradients, disperse seeds, and inflict damage to the understory through browsing (Doughty et al., 2016; Sienne et al., 2014; Blake et al., 2009). It is estimated that changes in vegetation structure from elephant disturbance significantly affect carbon stocks in the Afrotropics (Berzaghi et al., 2019), however these findings have not been validated with structural data. Due to the dense and untraversable nature of African tropical forests, direct studies on elephant distribution, biology, and ecological roles is difficult. Indirect methods such as dung counts, vegetation destruction, and biologging are typically used. As limiting as these methods are, there are few other on-the-ground methods that can be applied in tropical forests without significant field effort and resources. Therefore, the use of remotely-sensed data is essential in determining the changes in ecosystem structure made by forest elephants. This study aims to quantify the role forest elephants play as ecosystem engineers and their impact on habitat heterogeneity using terrestrial, airborne, and spaceborne lidar (light detection and ranging). Utilizing elephant location data from Wildlife Conservation Society, bio-logged tracking data from the Save the Elephants Foundation, and four types of lidar for forest composition and structure, we will address urgent questions regarding how elephants impact canopy structure. Specifically, how do forest elephants impact forest structure? At what lidar resolution can this be detected (cm, 1 m, 18 m, 25 m)? And can lidar be used to identity elephant transitory trails? As lidar has never been used for determining how elephants impact the surrounding landscape, this research will provide novel information about this endangered species. Not only will these findings address important questions related to ecological function, but they will expand our understanding of forest elephant s influence on carbon storage. Furthermore, environmental policy makers will be armed with needed information on the importance of biodiversity for ecosystem services without the need of costly and time-intensive field work. 
 
  
Ralph Dubayah (PI)/Jamis Bruening (FI) 
University of Maryland, College Park 
20-EARTH20-0235, Characterizing Forest Regrowth Biomass Accumulation in Northeastern US Forests Using Model-Data Fusion 
 
The importance of forests within the Earth system cannot be overstated, especially in mitigating anthropogenic CO2 emissions.  Scientific consensus generally acknowledges the vital yet transient role of secondary forest regrowth in global carbon sink dynamics, despite uncertainties in longevity and magnitude.  High profile global analyses that quantify forest carbon flux (both past and future) have brought much deserved attention to this area, yet estimates of forest carbon flux are potentially fraught without highly accurate and precise initial conditions.  This proposed research leverages ecosystem modeling to estimate regrowth-driven carbon sequestration of secondary forests throughout the northeast USA, and does so by constraining the predictions with high resolution LiDAR observations of forest structure from NASA’s Global Ecosystem Dynamics Investigation (GEDI).  The use of in situ data, such as GEDI, to initialize such simulations is imperative, as predictions of biomass accumulation within forests require an accurate starting point from which to calculate this change.  This work provides new insight into the magnitude and longevity of the secondary forest regrowth carbon sink within the northeast USA, where aboveground biomass development from secondary succession may increase from current levels by as much as 2x-4x in the future.  Accounting for legacies of human disturbance that still influence regional forest structure, dynamics, and secondary succession, this work highlights how past human activity that altered the natural landscape continues to play a critical role in the Earth system today, and how this part of the Earth system may change in the future. 
 
  
Laura Duncanson (PI)/Mengyu Liang (FI) 
University of Maryland, College Park 
20-EARTH20-0164, Monitoring Aboveground Biomass Recovery in Forest Restoration Areas Using GEDI and Optical Data Fusion 
 
Forests play a critical role in the global carbon cycle by sequestering carbon in the form of biomass. Tree planting and forest restoration have been lauded as solutions to combat climate change and criticized as ways for polluters to offset carbon emissions. Consistent monitoring and quantification of forest restoration can impact decisions of future restoration activities. In this study, I will develop a novel approach using remote sensing techniques to obtain objective baseline data and to monitor and to evaluate restoration project success. This research will be carried out in three phrases.  
In Phase 1, I will build empirical models to link current (2019-2021) aboveground biomass density (AGBD) estimates from the Global Ecosystem Dynamics Investigation (GEDI) with Landsat (1985-2020) and PlanetScope imagery (2020). This will allow both current wall-to-wall biomass mapping, and estimation of biomass dynamics across time. In Phase 2, I will use the annual biomass maps produced from Phase 1 models to derive AGBD recovery trajectories for literature-derived restoration sites, compare the speed of AGBD recovery for three restoration methods (Assisted Natural Regeneration (ANR), Natural Regeneration (NR), and Active Restoration (AR)) in three major biomes in East Africa. Finally, in Phase 3, I will investigate the AGBD recovery completeness of the restored sites compared to that of protected areas (PA) and old-growth forests and analyze potential drivers that influence the extent of recovery. This research is directly responsive to the SMD Earth Science division’s solicited questions of  how is the Earth system changing and what causes these changes?  and  how can Earth system science provide societal benefit?  and the four cross-cutting priorities in  Science 2020-2024: A Vision for Science Excellence  in three ways. First, the project demonstrates a research initiative utilizing remote sensing data to conduct innovative method development for understanding ecosystem and carbon dynamics in a region of strategic importance for mediating climate change. Being able to monitor changes to terrestrial carbon pools under different restoration strategies greatly aids out predictive capacities of how to maximize carbon sequestration potential in the future. Second, the development of this work has been and will continue to partner with researchers from Conservation International, SERVIR, University of Dodoma, and Wageningen University, to ensure the relevance and applicability of the results, and facilitates international scientific collaboration. Third, the FI intends to build an interactive visualization tool drawing on her web mapping expertise to communicate the research results and workflow and to engage a broader audience from the conservation, remote sensing, and local communities. 
 
  
Douglas Edmonds (PI)/Harrison Martin (FI) 
Indiana University 
20-EARTH20-0107, Testing the Hypothesis that River Discharge Variability Controls Megafan Formation in Foreland Basins 
 
This project will focus on megafans, which are large fan-shaped landscapes that can be found where rivers exit mountains-fronts. After decades of study, it remains unclear how megafans form and why only some rivers leaving mountain-fronts create megafans. Previous authors have noticed that rivers with higher discharge variability (i.e., flashier  or more-seasonal rivers) are more likely to have megafans. This inspires two complementary hypotheses. The first hypothesis is that rivers with higher discharge variability avulse more often (i.e., re-route to a new position on their floodplain) because more frequent floods provide more opportunities to  trigger  avulsions by eroding riverbanks and rerouting flow. The second hypothesis is that highly avulsive rivers are more likely to create megafans because avulsions are the main way that rivers redistribute sediment in a fan shape away from the mountain-front. I propose using a combination of satellite-based remote sensing and modeling to test these two hypotheses. The anticipated results have practical importance because avulsions on megafans can be catastrophic for those who live on them; avulsions on the Kosi (2008) and Indus (2010) rivers resulted in thousands of deaths and tens of millions displaced. The results are also important for interpreting ancient paleoenvironments, since the sediments left behind by avulsions can be buried over thousands or millions of years, preserving records of ancient river pathways. These ancient records can be used to help interpret long-scale changes in the paleoclimates and mountain-building processes in ancient mountain-front regions. There is a critical need to develop an understanding of the relation between river discharge variability, avulsions, and megafan creation. This will empower efforts to mitigate hazards, explore for subsurface natural resources, and interpret historical paleoenvironments from the stratigraphic record. 
 
The project will test the central hypothesis that fluvial discharge variability leads to increased avulsion frequency, and that increased avulsion frequency leads to megafan generation. Three actions are necessary to test this central hypothesis: 1) via remote sensing, create a map of every detectable avulsion over the last 35 years in major sedimentary basins worldwide by applying a new, automated avulsion detection method to annual surface water maps, 2) via remote sensing and literature review, create an inventory of megafans and river discharge variability for both avulsing and non-avulsing rivers in the study area, and 3) via remote sensing and modeling, measure floodplains and test whether changes in avulsion frequency alone are sufficient to generate both megafan and non-megafan landscapes over long timescales.  
 
The proposed study aligns with NASA’s Strategic Goal 1.1: to Understand The Sun, Earth, Solar System, and Universe and it fits within the Earth Surface and Interior Focus Area of the Earth Science Research Program. Megafan generation is a research question that sits at the nexus of natural disasters, climate and tectonics, plate boundaries, and earth surface water. Through this work, I aim to contribute to conversations about avulsion mechanics and prediction, landscape evolution, and climatic and tectonic interpretations of ancient paleoenvironments at plate boundaries. 
 
  
Ellyn Enderlin (PI)/Julia Liu (FI) 
Boise State University 
20-EARTH20-0336, Studying Glacier Surge Propagation with Ice Kinematics Derived from Radar and Optical Satellite Image Processing Techniques 
 
The proposed project will investigate the propagation of ice instabilities on surging glaciers using radar and optical satellite image processing techniques. We will generate records of the ice kinematics (i.e., terminus position, surface velocity, and surface elevation change) of Turner Glacier in the Wrangell-St. Elias mountain range to investigate its ongoing surge that began in early 2020. The surface velocities and elevation changes will be compared to the measurements of in situ Global Navigation Satellite System (GNSS) measurements available throughout the surge. We will use an automated glacier terminus mapping method that we previously developed to map Turner Glacier’s terminus positions throughout the surge. We will utilize feature tracking techniques on Synthetic Aperture Radar (SAR) images and optical images as well as SAR interferometry to generate surface velocity time series throughout the surge with high temporal resolution. SAR interferometry will also be used with Digital Elevation Models (DEMs) to calculate surface elevations during the surge. We will assess the accuracies of the surface velocities and elevations determined from the image processing techniques using the in situ GNSS data, which will improve the glaciological community’s understanding of the formal errors associated with these widely-used processing techniques. Furthermore, we will use the high-resolution record of ice kinematics throughout Turner Glacier’s surge to resolve its surge front propagation speeds. The proposed project will explore the link between surge front propagation speed changes and glacier geometry. Using the developed workflows, we will expand the analysis to other surge-type glaciers in the Wrangell-St. Elias mountain range, initially using optical image feature tracking to detect possible surges after 2014. For the detected surges, we supplement the surface velocity analysis with the radar image processing to increase the temporal resolution. With detailed records of surface velocities for these detected surges, we will calculate propagation speeds of the surge fronts and further evaluate the link between surge front propagation speed changes and glacier geometry. Our analysis will improve monitoring of glacier surges and contribute to the understanding of glacier surge progression and propagation of ice instabilities. These advances will improve our capability to assess and respond to the natural hazards posed by glacier surges as well as our understanding of the role and interactions of land and ice in the climate system. 
 
  
Sergio Fagherazzi (PI)/Luca Cortese (FI) Boston University 
20-EARTH20-0160, Coupling NASA UAVSAR, Air-SWOT and AVIRIS-NG Imagery with Numerical Modeling to Quantify Sediment Fluxes in Deteriorating Salt Marshes 
 
Salt marshes are important ecosystems that store carbon, host unique wildlife, and act as buffers against storm surges. In recent decades, salt marshes have been disappearing at a staggering rate in the Mississippi delta, USA, because of storms and sea level rise. In this project 
I will determine the resilience of salt marshes to sea level rise by coupling a numerical model to NASA imagery. The survival of a salt marsh is intrinsically connected to sediment supply: a healthy marsh is in fact trapping sediment and accreting vertically thus counteracting rising waters. A negative sediment budget indicates marsh degradation. The computation of sediment fluxes and an accurate sediment budget are therefore critical to determine the fate of salt marshes in a changing climate. The combination of numerical simulations and remote sensing data proposed here will enable to compute a sediment budget for several salt marsh complexes in the Mississippi delta. The hydrodynamic component of the numerical model will be informed by already available UAVSAR and Air-SWOT data, while AVIRIS-NG images will provide sediment concentration for the sediment transport component. Storms and waves cause sediment resupension and erosion, thus affecting sediment fluxes and marsh survival. We will use sediment concentration derived from Landsat-8 and Sentinel-2 images spanning several years to determine the effect of wind and waves on sediment budgets. In future years, the methodology presented here will enable the use of SWOT and NISAR imagery for assessing the resilience of coastal wetlands to climate change. 
 
  
James Garrison (PI)/Seho Kim (FI) 
Purdue University 
20-EARTH20-0274, Assessing the Impact of Root Zone Soil Moisture Observations Using MultiFrequency Signals of Opportunity on Agricultural Drought Risk Estimation 
 
Agricultural drought is probably the most severe natural disaster affecting many aspects of life and the environment. Root zone soil moisture (RZSM) (moisture profile in the top meter of soil) is a key environmental variable needed for mapping and forecasting agricultural droughts. The ESAS 2017 Decadal Survey calls for multi-frequency microwave sensors to observe RZSM with higher spatiotemporal resolutions. Direct measurement of RZSM requires the use of P-band or lower microwave frequencies to penetrate the soil more than a few cm. Spaceborne remote sensing in P-band, however, presents significant technical challenges due to the large antenna size, limited spectrum, and radio frequency interference (RFI).  
As an alternative, signals of opportunity (SoOp) has emerged in recent years with great potential to measure RZSM at high spatiotemporal scales. The most promising aspects of SoOp are the ability to make observations in frequencies outside bands allocated for science as well as the smaller size, weight, and power (SWaP) of instruments, due to the re-use of existing powerful satellite transmissions. Small instrument SWaP enables deployment of a constellation of small satellites for a reasonable cost, to achieve a higher revisit rate. The SoOp observations at multi-frequency could potentially retrieve the soil moisture profile at high temporal resolution, having different penetration depths.  
However, the utility of the RZSM observation using SoOp reflectometry has never been examined in research communities. To help advance the SoOp technique to future science mission, Observing System Simulation Experiments (OSSEs) are required to quantify the potential benefits of RZSM observations on data assimilation of hydrologic variables and actual end-use applications such as drought risk estimation. Most OSSEs for soil moisture, however, have been conducted only for surface soil moisture (SSM) observations of passive microwave systems such as Hydros, SMAP, Aquarius, and SMOS. Therefore, studies with the OSSE must be explored before the SoOp method can be applied to a spaceborne mission.  
This proposal aims to perform an OSSE to assess the potential impact of RZSM observations of a hypothetical spaceborne SoOp mission on agricultural droughts risk estimation and to determine its sensitivity to retrieval accuracy and mission parameters such as spatial resolution and temporal revisit. First, I will evaluate the synthetic RZSM retrievals from multi-frequency SoOp at I/P/L-bands by comparing them against a nature run and in-situ observations. Second, SSM and RZSM observations are separately assimilated into a land surface model to assess the added impact of RZSM observations on the model estimates. Third, I will quantify the contribution of the RZSM retrievals for improving agricultural drought estimation. Drought risk estimates of different intensities will be generated using RZSM-based percentiles and compared to each other. Lastly, the second and third phases will be repeated several times with different combinations of the resolution and the retrieval accuracy of RZSM observations to investigate the sensitivity of the RZSM estimates and end-use performance to a range of potential mission configurations.   
The significance of the proposed work is to help us understand the impact of SoOp remote sensing of RZSM observations on the end-use application in a realistic observation environment and to reduce the risk, time, and cost of advancing SoOp to a science hypothesis-driven mission for Earth remote sensing. 
 
  
Tarsilo Girona (PI)/Claire Puleio (FI) 
University of Alaska, Fairbanks 
20-EARTH20-0108, Large-Scale Thermal Unrest of Volcanoes: Causes and Implications 
 
Volcanic eruptions present serious risk to human life and infrastructure on Earth. This risk can be minimized by improving eruption forecasts, which in turn requires increasing our capabilities to detect volcanic unrest and a better understanding of the physicochemical processes governing magma-hydrothermal interactions. Volcanic unrest can be explored through remote sensing observations, including through the analysis of large-scale subtle increases in radiant heat flux along the flanks of volcanoes in the lead up to their eruptions (Girona et al. 2021; accepted for publication in Nature Geoscience). The origin of this pre-eruptive thermal unrest, however, remains unclear. Through the project outlined in this proposal, I will glean a more complete understanding of how volcanic systems operate and how subsurface processes manifest themselves at the surface. In particular, the main goal of this project is to better understand what thermal anomalies along the flanks of volcanoes can indicate regarding subsurface magma-hydrothermal interactions and impending eruptions. This will be done by developing new ways to more accurately apply satellite-based remote sensing data to eruption forecasting through the exploration of three objectives:  
 
1.	Expand the methodology already in place in order to explore, from satellite-based thermalinfrared remote sensing, how large-scale (several km2), long-term (~years), pre-eruptive thermal anomalies distribute temporally and spatially around volcanic edifices.  
2.	Design new algorithms to explore the possible emergence of large-scale thermal anomalies along the flanks of volcanoes in the short-term, i.e., from days to months prior to an eruption.  
3.	Integrate the large-scale radiant heat emissions along volcanic flanks with other observables, in particular the seismic energy released around volcanic edifices. This will provide important insights on the subsurface processes generating these thermal anomalies. 
 
These new methods will be tested in a set of case studies at target volcanoes in Alaska, USA, including Veniaminof, Cleveland, and Semisopochnoi. These volcanoes are chosen due to their frequent eruptions over the last two decades.  
 
This work will provide further insights on how volcanoes work and into the subsurface processes causing the thermal anomalies presented at the surface of volcanoes in the lead up to eruptions; will be important to complement other geophysical/geochemical data; and will be useful to assess alert levels and forecast eruptions, especially for those volcanoes where the installation of ground-based instrumentation is very challenging (e.g., Alaska) or where no classical methods are able to anticipate eruptions. Improved eruption forecasting techniques achieved by this project will aid in NASA’s Science Mission Directorate Strategic Objective for Safeguarding and Improving Life on Earth. This objective outlines how data and knowledge that NASA provides can be used to directly, positively, and currently impact life on Earth. This project is vital to complete this objective. 
 
  
Colin Gleason (PI)/Craig Brinkerhoff (FI) 
University of Massachusetts, Amherst 
20-EARTH20-0044, A First Global Analysis of Daily Riverine Gas Exchange Using the SWOT Satellite, Bayesian Remote Sensing, and Carbon Transport Modeling 
 
Decades of fieldwork and modeling have highlighted the importance of riverine gas fluxes as a significant component of the global carbon cycle. While theory is mature, our present understanding of global gas exchange from rivers is fundamentally limited by the resolution of available global hydraulics data, degrading our ability to accurately constrain greenhouse gas (GHG) evasion and production. However, soon-to-be-available data from the NASA Surface Water and Ocean Topography (SWOT) mission should contain enough hydraulic information to estimate the gas exchange rates in rivers, thereby meaningfully improving global GHG evasion and production estimates.  I propose to leverage my advisor’s positions on the SWOT Science and Cal/Val teams to map daily gas exchange rates and velocities for major rivers measured during SWOT’s fast sampling orbit using remote sensing methods I will develop and detail below. If successful, this will be the first global-scale analysis of the daily spatiotemporal dynamics of riverine gas exchange. I further seek to couple this novel gas exchange dataset with process-based transport modeling to explore the impact of remotely-sensed daily gas exchange dynamics on bulk drainage network carbon efflux. This will make use of a novel synthesis of gas exchange theory, hydraulic geometry, remote sensing, and NASA SWOT data, profoundly altering present understandings of GHG evasion and production in drainage networks and achieving the FINEEST Earth Science goals of integrating  space-based measurements [and] computational models that can be used to more fully characterize the present state and future evolution of the Earth system. 
 
  
Kaitlin Gold (PI)/Fernando Romero Galvan (FI) 
Cornell University 
20-EARTH20-0127, Detecting Plant Disease at Scale with NASA Imaging Spectroscopy 
 
Plant disease is one of the greatest threats to the environmental and financial sustainability of crop production worldwide. Even with the remarkable advances of 21st century agriculture, plant disease results in 15-30% global crop loss, equating to losses upwards of $220 billion annually. Plant disease changes how solar radiation interacts with leaves, canopy, and plant energy balance, which can be sensed with proximal and remote sensing. However, plant disease remote sensing remains underdeveloped despite its potential to revolutionize surveillance and intervention through low cost and high accuracy decision support. Forthcoming spaceborne hyperspectral instruments, such as Surface Biology and Geology (SBG), will vastly improve global coverage of important spectroscopic data products. These novel data streams coupled with emerging understanding on plant-microbe interactions on global biogeochemical cycling and CO2 fertilization responses may one day yield the ability to map high resolution plant-pathogen interaction at the global scale. SBG has the potential to one day serve as a global disease surveillance system for the broader agricultural community, but to fulfill this vision we must begin to investigate now how we can best utilize these data for early detection and warning. The goal of our proposal is to develop a scalable, remote sensing framework for detecting plant-pathogen interactions in grapevine, an economically important specialty crop, using NASA’s Airborne Visible/Infrared Imaging spectrometer Next Generation (AVIRIS-NG).   
 
Specifically, we seek to:  
1)	Compare and contrast supervised and unsupervised machine learning methods for disease detection. We hypothesize that the optimal machine learning approach will be a combination of supervised methods given their history of success in other terrestrial imaging spectroscopy applications. 
2)	Compare statistical and physiological dimensionality reduction for improving disease detection precision. We hypothesize that dimensionality reduction will improve our detection accuracy and yield a set of wavelengths with known association with plant defense chemicals and core functional traits to link to known disease biology. 
3)	Identify the minimum threshold for detection with spaceborne imaging spectroscopy. We hypothesize that 15% disease incidence, half the recommended vineyard removal threshold, is sufficient to detect disease at spaceborne imaging spectrometer resolution (30m). 
 
The use of remote sensing to advance plant disease detection represents an innovative opportunity to further the use of Earth system science research to benefit society and inform decision making while advancing applications-focused research in precision agriculture, one of the priorities outlined for Surface Biology and Geology in the 2018 NASA Decadal Survey. The ability to non-destructively sense plant disease would greatly benefit modern agriculture and food security. Early intervention is key to successful disease mitigation, and remote sensing has a rich history in early warning. Farmers can apply systemic fungicides to stop disease before it spirals out of control, but these are only effective when applied early during the infection process. Worldwide, plant disease research and early intervention efforts are often constrained by a lack of local expertise to devote to prevention, a lack of resources to devote to monitoring and/or remediation, and a lack of qualified personnel to allocate to both these tasks. Therefore, the use of remote sensing to advance plant disease research represents an opportunity to avoid these challenges and make a difference in the lives of farmers worldwide while advancing applications focused research in a NASA strategic priority area, precision agriculture, as outlined in the 2018 NASA Decadal Survey. 
 
  
Josh Gray (PI)/Xiaojie Gao (FI) 
North Carolina State University 
20-EARTH20-0156, Does Chilling Explain the Divergent Response of Spring Phenology to Urban Heat Islands? 
 
Urbanization is known to have direct impacts on plant phenology. Understanding these effects is important to biodiversity dynamics, ecosystem structure, carbon cycles, and human health. Temperature increases from the Urban Heat Island (UHI) effect are thought to be the main driver of plant phenological changes around cities. However, trends in plants  start of growing season (SOS) dates around urban areas, compared to the surrounding countryside, have diverged across the globe: some advance, and some delay. Divergent SOS trends have been observed in field measurements as well as satellite remotely sensed terrestrial vegetation seasonality land surface phenology (LSP). However, the reasons for this phenomenon remain unclear. We hypothesize that divergent SOS trends can be explained by the interaction between UHI-induced seasonal temperature changes and variable plant chilling requirements the need of plants to be exposed to sufficiently low temperatures to release dormancy in spring. The proposed project is designed to evaluate this hypothesis by accomplishing two main objectives: 
 
" Objective 1: Map long-term medium spatial resolution LSP for large cities.          o More than 50 large cities in North America that are expected to have strong UHI effects will be selected to capture variability in biome type and prevailing climate.          o LSP with pixel-wise uncertainty at 30 m spatial resolution from 1984 to present will be retrieved for each selected city using a recently developed Bayesian model. 
 
" Objective 2: Test chilling hypothesis by analyzing spring phenology models         o A suite of models describing spring emergence as a function of daily temperature exposures will be fit using the generated LSP data.          o Analysis of model fit and parameters will indicate the importance of UHI-altered chilling/warming regimes. 
 
This project responds directly to NASA’s Carbon Cycle and Ecosystems program questions: how do ecosystems, land cover and biogeochemical cycles respond to and affect global environmental change? And: What are the consequences of land cover and land-use change for human societies and the sustainability of ecosystems? Also, it is well-aligned with the broader NASA Earth Science Division goal of detecting and predicting changes in Earth s ecosystems and biogeochemical cycles, including land cover, biodiversity, and the global carbon cycle. Moreover, we expect this work will make contributions to the remote sensing, phenology, and global change science communities by providing: 1) an improved understanding of how temperature controls plant phenology in urban areas; 2) a novel approach to retrieve medium spatial resolution LSP with pixel-wise uncertainty that can be generally applied to urban or natural regions, and a long-term 30 m spatial resolution LSP dataset for large cities in North America. 
 
  
Erin Hestir (PI)/Brittany Lopez Barreto (FI) 
University of California, Merced 
20-EARTH20-0216, Wildfire Effects on Water Quality in California Water Supply Reservoirs 
 
The future of water supply is uncertain due to climate change. Droughts and wildfire are increasing in both intensity and frequency, raising water availability concerns. Areas with Mediterranean climates, such as California, are in high risk of increased soil erosion from wildfires since precipitation is followed immediately after fire season, thus endangering water supply. These potential consequences make it dire to understand the relationship wildfire has on water quality in order to prepare for the future under climate change.  
California’s unique climate and ecosystems would allow a new insight on wildfire s role on water quality in a landscape that has been facing an increasing amount of wildfire. The objective of this proposal is to determine the water quality responses across California lakes and reservoirs used for water supply to the frequency, size, duration, and severity of wildfires in their watershed.  
Using data provided from Landsat-5/7/8 and Sentinel-2 satellites, turbidity and chlorophyll-a 
(chl-a) will be mapped for California lakes. Envisat’s Medium Resolution Imaging Spectrometer’s 
(MERIS) spectral resolution allows monitoring of cyanobacteria, but unfortunately ceased in 2012. Fortunately, Sentinel-3’s Ocean Land Cover Imager (OLCI) has continued cyanobacteria collection since 2016. Field reconnaissance will be collected from governmental, state, and public agencies for identifying in depth watershed analysis for available sites in order to increase validity of chosen parameters.  
An initial statewide analysis will be conducted for all three variables. Chl-a and turbidity will have a time series conducted from 1984-2019, while cyanobacteria time range is from 20022012 and 2016-2019 due to the gap in data availability. Nature will not produce identical water quality results statewide. After the statewide analysis, we will conduct a similar analysis on a watershed basis. Watersheds will initially be separated by ecoregion in order to determine if one ecoregion is more vulnerable to wildfire. Finally, the time lag for water quality response will be determined for lakes that have shown a significant difference in water quality post-fire. For lakes found to share a similar time lag, additional variables such as land cover class, tree canopy cover, impervious surface cover, soil erosion potential, climate and vegetation type will be considered. Determining similar environmental characteristics to the duration back to baseline conditions will allow a better understanding behind the mechanisms influencing water quality response.  
This proposal’s objective addresses one of the 7 strategic goals for advancing the understanding of Earth and developing technology to improve the quality of life, enable better assessment and management of water quality and quantity to accurately predict how the global water cycle evolves in response to climate change . This research addresses the 2017 ESAS Decadal Survey question H-4,  How does the water cycle interact with other Earth System processes to change the predictability and impacts of hazardous events, such as wildfires. The survey states,  Wildfires impact water resources that are in high demand in the arid West. This proposal will allow to identify areas and specify the lakes that have shown water quality responses to wildfire, allowing to communicate to water managers and stakeholders what locations require the most monitoring and future potential mitigation under ongoing climate change. 
 
  
Jennifer Hutchings (PI)/MacKenzie Jewell (FI) 
Oregon State University 
20-EARTH20-0112, Characterizing Arctic Sea Ice Mechanics Using MODIS Imagery and Observationally-Constrained Models 
 
Sea ice is a highly dynamic material known for its amplifying role in the Arctic climate.  The interaction of sea ice with other components of the climate is controlled in part by its dynamics: sea ice drift and deformation driven by winds and ocean currents. The state and dynamics of the Arctic ice cover have changed dramatically in recent decades. Inaccurate representation of sea ice deformation could explain the failure of current Earth system models to capture the observed thinning and acceleration of the Arctic ice cover, with consequences for representation of the magnitude of air-ice-sea heat, freshwater and momentum fluxes throughout the Arctic.  
 
This proposal addresses uncertainties in the mechanical properties of sea ice and their modulation of the sea ice cover’s dynamic response to other components of the Arctic climate system. Integration of remotely sensed data and high-resolution simulations of the ice cover will be used to characterize the relationship between fine-scale mechanics and large-scale sea ice deformation. MODIS imagery will be used to identify the ice fabric (orientation and distribution of sea ice fractures) and dynamics that cause recurrent lead (large fracture) patterns to form in the Beaufort Sea ice cover from 2000-2020. Variability in the relationship between wind loading and lead geometry will be explored to identify the effects of sea ice fabric anisotropy on lead formation. Case studies will be developed using a discrete element method model of sea ice drift and interaction to simulate the roles of wind loading and prior fractures in the ice pack fabric as leads form. Each simulation will be run with both isotropic and MODIS-derived anisotropic ice fields to quantify the effects of ice fabric anisotropy on sea ice mechanical properties and resultant kinematics. The performance of isotropic vs anisotropic simulations will be evaluated to characterize the effects of fine-scale mechanical properties on large scale deformation events.  
 
Integration of remotely sensed data and high-resolutions simulations is key to understanding the fine-scale mechanics that are not resolved in current Earth system models. The results of this project will improve theoretical understanding of the dynamics of sea ice, with the goal of guiding improvements in sea ice components of Earth system models. The proposed project aligns with NASA’s goal to examine the interplay among the Earth’s atmosphere, ocean, land, and ice to determine what causes changes in the Earth system. Improving the representation of sea ice dynamics in models will reduce uncertainty in its amplifying interactions with the Earth’s oceanic and atmospheric systems. This will reduce climate uncertainty by improving projections of how the Earth system will change in the future. 
 
  
Charles Jones (PI)/Kevin Varga (FI) 
University of California, Santa Barbara 
20-EARTH20-0286, An Investigation into Wildfire Fuel Moisture Content: Dead or Alive 
 
Wildfires have been increasing in frequency, size, and intensity in many parts of the world. In particular, the state of California (CA) has experienced its deadliest, most destructive, and largest fires in the past few years. Wildfire management is difficult because of the complex interactions between humans, topography, weather, and fuels. One of the confounding factors of wildfire management is fuel moisture content (FMC), which can affect the frequency of ignition, rate of spread, and intensity. This proposal combines multiple NASA remote sensing tools (Moderate Resolution Imaging Spectroradiometer and Soil Moisture Active Passive), high resolution weather modeling (Weather and Research Forecasting model 30-year CA climatology), and in situ FMC observations (Remote Automated Weather Station and direct live FMC observations) into a hybrid model that will provide a historical 30-year FMC dataset for CA. This dataset will then be used to study how FMC has changed during extreme weather and climate events. The effects of FMC on historical fires will also be analyzed using the fire spread model, Prometheus, and historical wildfire ignition, perimeter, and intensity data. The varied topography, climate, and fuels of CA make for an ideal study site. This better understanding of FMC, especially in our changing climate, will help policy makers, land management agencies, and emergency personnel implement better wildfire management practices. 
 
  
Van Kane (PI)/Caden Chamberlain (FI) University of Washington, Seattle 
20-EARTH20-0022, When can Wildfires Improve Forest Resilience to Future Fire and Drought? A Study Using Spaceborne Data 
 
A century of forest management activities, coupled with the growing effects of climate change, have rendered millions of hectares of mixed-conifer forest in California’s Sierra Nevada susceptible to severe fire and drought disturbances, which threaten human safety and ecosystem services. Low and moderate severity wildfires are increasingly considered by land managers as a potential tool for restoring forest resilience to future disturbance events. However, a more thorough understanding of the conditions under which wildfires improve forest resilience is critical before wildfires alone are accepted as a viable alternative to traditional restoration treatments. Distinct patterns of tree clumps and openings within forest stands serve as key determinants of forest resilience to disturbance. Airborne and spaceborne lidar data offer unique advantages for identifying and quantifying these patterns at fine spatial resolution and across large spatial extents. This study will develop models using airborne lidar and NASA’s GEDI and Landsat data to predict forest structure classes representing different levels of forest resilience across contemporary reference sites and recent wildfires in California’s Sierra Nevada mixed-conifer zone. Model outputs will be used to discern the specific biophysical, weather, and forest conditions under which wildfires are more likely to improve forest resilience. This study directly supports the Carbon Cycle and Ecosystems focus area of NASA’s Earth and Science Research Program while also meeting specific objectives outlined in the 2018 Earth Science Decadal Survey by: improving our understanding of the causes of global ecosystem change; quantifying the 3D structure of terrestrial vegetation spatially and overtime; enhancing our understanding of ecosystem responses to fire; and advancing the use of NASA’s GEDI spaceborne data. 
 
  
Kristopher Karnauskas (PI)/ Mikell Warms (FI) 
University of Colorado, Boulder 
20-EARTH20-0289, Investigating Climate Change Impacts on the Galapagos Upwelling Ecosystem Using Satellite Observations and a High-Resolution Global Model 
 
The Galápagos Islands, located in the eastern Equatorial Pacific, are home to one of the largest and most significant biologically diverse hotspots on earth. Not only is the archipelago filled with a unique variety of plant and animal life endemic to the islands; it also has great historical significance. Charles Darwin’s Theory of Evolution and his famous book “The Origin of Species” are inextricably linked to the inspiration he gleaned from his visit. The largely unparalleled productivity of this region is connected to the strong upwelling currents found along the islands’ western shores, bringing cold and nutrient-rich deep ocean waters to the surface which feed phytoplankton that sustain the entire ecosystem. However, this flourishing but fragile ecosystem is at risk—the Galápagos Islands are one of many locations in the world that are already feeling the impacts of climate change. Over the past several decades, there have been a number of events of suppressed upwelling, limiting phytoplankton productivity and resulting in sharp population declines of endemic species. Despite the critical importance of upwelling and the evidence that climate change may be causing a reduction in upwelling strength or frequency in the Galápagos, the key driving forces behind upwelling remain largely uncertain. This project seeks to bring a suite of NASA remote sensing products and ultra-high resolution global model simulations to bear on this problem of high scientific and societal importance. 
 
We propose to combine analysis of available remote sensing products with a new ultra-highresolution fully-coupled global climate model that can capture the complex ocean dynamics of surface and subsurface currents coupled with a topographical obstruction (the Galápagos Islands). This project seeks to capitalize on a unique opportunity to compare remote sensing observations with a climate model at similar scales—a powerful combination that will allow us to: (1) determine how measurement and modeling scales affect our ability to understand the complexity of large- and small-scale atmosphere-ocean dynamics at play in the upwelling zone of the Galápagos; (2) provide evidence to prove or disprove hypotheses related to the main drivers of upwelling in the Galápagos archipelago and how sensitive these drivers are to climate change; and (3) validate the average conditions of the ultra-high-resolution model with remote sensing datasets in order to use the model to predict the climate change impacts to the Galápagos upwelling ecosystem for two likely emissions scenarios.  
 
This proposal is directly relevant to NASA’s Strategic Objective 1.1: Understand the Sun, Earth, Solar System, and Universe, specifically the Earth Science Research Program. In particular, this research addresses the Earth Science Division’s goal of understanding “Climate Variability and Change—to improve the ability to predict climate changes by better understanding the roles and interactions of the ocean, atmosphere, land, and ice in the climate system.” The Eastern Equatorial Pacific, and especially the Galápagos Islands, is extremely sensitive to changes in earth’s climate system. The loss of Galápagos biodiversity would also directly impact its local human communities as their livelihoods are primarily dependent nature tourism, fisheries, and agriculture—all of which are being threatened, making it an especially important area of study. 
 
  
Trevor Keenan (PI)/Xinchen Lu (FI) 
University of California, Berkeley 
20-EARTH20-0119, Using ECOSTRESS Water Use Efficiency to Quantify Vegetation Vulnerability to Water Stress 
 
Observations of the global carbon and water cycles provide vital information on ecosystem function and are essential constraints for earth system models. Water-use efficiency (WUE), which characterizes the coupling of carbon and water fluxes, is one of the key characteristics of the global carbon and water cycle. WUE is controlled by stomata, which are the gatekeepers of the tradeoff between photosynthesis and plant water use. By closing in response to water stress, they can allow plants to use water more efficiently, thus reducing the impact of stress. However, there is large diversity between species in their ability to regulate water use and even coexisting species can have very different water use efficiencies, which contributes considerable diversity in drought tolerance. By providing information on spatial and temporal changes in ecosystem water use and use-efficiency, NASA’s ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) mission represents a unique opportunity to identify vulnerable populations. ECOSTRESS provides diurnally resolved estimates of water use efficiency at high spatial resolution (70 m), which has the potential to greatly improve our understanding of ecosystem responses to stress. In this study, we propose to combine ECOSTRESS observations, flux tower observations and other land surface remote sensing products to assess the vulnerability of ecosystems to water stress using water use efficiency data. First, we will assess site-level ecosystem isohydricity using WUE plasticity from eddy covariance and leaf water potential data. Then, we will use the established relationship to map the isohydric spectrum based on ECOSTRESS data across the US. Finally, we will test the response of ecosystems to drought along the isohydric spectrum. The proposed work will provide a novel indicator of water-stress resilience, and will also directly benefit the broader scientific community by providing understanding of the mechanisms governing the vulnerability of ecosystems to drought. 
 
  
Gretchen Keppel-Aleks (PI)/Daniel Muccio (FI) 
University of Michigan, Ann Arbor 
20-EARTH20-0009, Climate and Ecosystem Impacts on High Latitude Carbon Fluxes 
 
High latitude regions of the globe are experiencing unparalleled changes in climate and the carbon cycle that can affect the net exchange of carbon between the atmosphere and the land. These changes may affect gross primary productivity, which is essentially uptake of carbon by the land through photosynthesis, and they may also affect heterotrophic respiration, which represents a source of carbon to the atmosphere from microbes decomposing dead vegetation. Gross primary productivity and heterotrophic respiration, in tandem, affect the seasonal cycle of the net ecosystem exchange of carbon, which governs the seasonal cycle of atmospheric carbon dioxide. The amplitude of the seasonal cycle of atmospheric carbon dioxide has increased, suggesting that either gross primary productivity or heterotrophic respiration has enhanced. If the flux of carbon released into the atmosphere from heterotrophic respiration was to surpass the carbon flux drawn into the land from gross primary productivity, then the high latitudes could become a net source of carbon, exacerbating climate change. There are observational limitations when quantifying the heterotrophic respiration carbon flux at large spatial scales, and satellite records of vegetation productivity are relatively short, which makes analyzing these carbon fluxes difficult. This project proposes a novel way of better understanding high latitude gross primary productivity by using a NASA data product that is merging multiple satellite records into one multi-decadal record. This data product will then be scaled and used as an input into a soil biogeochemical testbed model developed by 
collaborators at NCAR to simulate heterotrophic respiration over high latitudes of the Northern Hemisphere. Quantifying these carbon fluxes will allow us to look at relationships with variations in climate, and we will use model data to predict what will happen with these carbon fluxes in the 21st century. This proposed work will address NASA’s Carbon Cycle and Ecosystems Focus Area by  detecting and predicting changes in Earth s global carbon cycle.  Better understanding gross primary productivity and heterotrophic respiration will improve our knowledge of the seasonal cycle of atmospheric carbon dioxide and the fate of the net terrestrial carbon sink at high latitudes of the Northern Hemisphere. 
 
  
Pierre-Emmanuel Kirstetter (PI)/Noah Brauer (FI) 
University of Oklahoma, Norman 
20-EARTH20-0319, Satellite and Radar Remote Sensing of Tropical Cyclones to Quantify Microphysical and Precipitation Processes 
 
Landfalling tropical cyclones (TCs) are responsible for the most impactful excessive precipitation events worldwide. Precipitation events such as TCs that are dominated by collision-coalescence processes challenge ground- and space-based retrievals of precipitation, which often results in underestimation of the surface rainfall rate. Since 2014, NASA Global Precipitation Measurement (GPM) Mission Dual-Frequency Precipitation Radar (DPR) has sampled the surface rainfall rate and associated microphysical evolution of TCs on a global scale.  The proposed work will use a synergy between the complimentary information provided by groundbased radar and the GPM DPR to quantify microphysical properties in TCs on a global scale to construct a climatology of the aforementioned processes. The global documentation of precipitation processes addresses a targeted observable identified by the scientific community in the 2017-2027 Decadal Survey for Earth Science and Applications from Space (ESAS 2017) and anticipates the upcoming Aerosol and Cloud, Convection and Precipitation (ACCP) mission which aims to capture precipitation processes in clouds. 
 
  
Alexandra Konings (PI)/Caroline Famiglietti (FI) 
Stanford University 
20-EARTH20-0061, Quantifying and Mitigating the Role of Parametric Uncertainty in Forecasts of the Terrestrial Carbon Cycle 
 
This project seeks to inform and improve terrestrial biosphere models (TBMs) by quantifying and reducing the role of parametric uncertainty in future forecasts. Modeling the terrestrial biosphere is difficult due to the numerous processes driving variability of carbon (C) fluxes. However, because terrestrial C uptake is a first-order control on atmospheric CO2 growth, understanding the most effective ways of improving the predictability of the terrestrial biosphere is crucial. 
 
Many approaches to reduce C cycle forecast uncertainty have focused on model structure, namely by introducing additional processes and increasing overall complexity. While adding processes may improve realism, the resulting models are often encumbered by a greater number of poorly-determined or over-generalized parameters. Indeed, my prior work shows that increasing the robustness of model parameterization is a prerequisite for model structural additions to improve skill.  
 
The effectiveness of conventional model evaluation efforts to improve parameterization is limited. Parameters are often manually tuned to so that model outputs match some set of observations. However, achieving acceptable model performance in a predetermined calibration period does not imply stability thereafter. In fact, a large forecast spread between TBMs still persists. Importantly, such evaluation approaches do not guarantee a reduction in parametric uncertainty; errors induced by multiple uncertain parameters often compensate, leading the model to generate the right answers for the wrong reasons during the calibration period and to fail under novel scenarios. 
 
Here I investigate the role of parametric uncertainty in controlling the degradation in skill between calibration and forecast periods. My research will first test this hypothesis and then develop specific methods for reducing parametric uncertainty in global-scale terrestrial C cycle models. To do so, I will use a novel model data fusion (MDF) system that can retrieve spatially variable, observationally-informed estimates of internal model parameters with well-defined uncertainties. The use of MDF allows for greater structural and parametric flexibility than is currently possible with traditional TBMs and allows model parameters to be informed by a wide range of remotely sensed data. 
 
Specifically, I propose two experiments: 
1. First, I will use 16 structurally distinct C cycle models run under several hundred prescribed parametric uncertainty scenarios to explicitly test the hypothesis that parametric uncertainty controls the decline of model predictive performance between calibration and forecast periods. 2. To reduce parametric uncertainty in large-scale models, innovative new approaches are needed. My second experiment aims to derive a novel model parametrization scheme for use when MDF is not practical, using the lens of environmental filtering  (whereby parameters are predicted as a function of climate, soil, and canopy variables). I will develop the first top-down environmental filtering relationships using global, spatially variable parameter retrievals, in contrast to previous bottom-up approaches that rely on sparse in situ measurements. The resulting relationships will be tested alongside the bottom-up approach as well as traditional plant functional types to quantify their effects on parametric uncertainty and performance degradation of C cycle predictions. 
 
Knowledge gained from the proposed work would support two overarching science goals of the Earth Science Division: 1) to detect and predict changes in Earth’s ecological and biogeochemical cycles, and 2) to further the use of Earth system science research to inform decisions and provide benefits to society. Because of its strong implications for effectively reducing the spread in future terrestrial biosphere predictions, this project would be an important step forward in developing more credible predictions of key Earth system functions. 
 
  
Mehmet Kurum (PI)/Dylan Boyd (FI) 
Mississippi State University 
20-EARTH20-0003, Recycling the Radio Spectrum: Can Anthropogenic Signals in LEO Be Repurposed for Science? Advancing Soil Moisture Remote Sensing with NASA Signals of Opportunity Datasets, Forward Modeling, and Physics-Guided Machine Learning 
 
The signals of opportunity (SoOp) technique is a cost-effective microwave remote sensing technique that offers many benefits to NASA Earth sciences. Using a spaceborne receiver, the transmissions of anthropogenic signals (e.g., communication/navigation systems) can be measured after interacting with the Earth surface. SoOp is a quickly growing research area as many scientists have shown that SoOp is sensitive to ocean surface roughness, soil moisture (SM), vegetation, and snow depth. SoOp features higher spatial resolutions than passive systems and can obtain high temporal resolution from cost-efficient use of small-satellite constellations. With the launch of NASA’s SNOOPI experiment approaching, a suite of forward modeling tools, retrieval frameworks, and sensitivity studies are needed to maximize NASA’s utilization of its own SoOp datasets. 
 
This proposal outlines the development of a three-fold research plan wherein (1) advanced forward modeling tools for bistatic SoOp scattering over land, (2) a combined physics-andmachine-learning-based SM retrieval algorithm is explored, and (3) an algorithm theoretical basis document for multifrequency SoOp SM retrieval are created. The proposal seeks to address NASA SMD’s Strategic Objective 3.1 (Develop and Transfer Revolutionary Technologies to Enable Exploration Capabilities for NASA and the Nation) by creating modeling and simulation tools that can be applied to present (i.e., CYGNSS) and future (SNOOPI) NASA SoOp missions, to explore physics-guided and ML-based retrieval algorithms that fully utilize NASA datasets and create physically meaningful data products, and to provide error budgeting that will guide NASA s future SoOp-based mission for land applications. Advancing SoOp will enable NASA’s Earth science division to use active microwave frequencies that are otherwise restricted to science applications. This can allow for more efficient frequency spectrum usage and new observations of our changing planet. 
 
This research seeks to build on work done by the PI and FI over the last four years across forward modeling, SM retrieval, and sensitivity studies. Previously, the PI and FI have created the SoOp Coherent Bistatic model. This model is unique in the SoOp research community in that it is the only publicly available model for land applications that preserves both amplitude and phase information of the received signal. Over the next year, we seek to extend this model to spaceborne applications to allow for modeling of topography using the novel Facet Approach method. The upgraded model will include digital elevation maps and modeling of signal path delay and Doppler at spaceborne altitudes. This model will provide NASA with the only SoOp land modeling tool that is fully based in Maxwell’s equations. 
 
In order to fully maximize the usage of NASA’s CYGNSS dataset for SM retrieval applications, this research will explore the use of physics-guided machine learning techniques. Previously, the FI and PI have created a high spatio-temporal resolution SM dataset using CYGNSS data and machine learning datasets. That approach did not use physical modeling. This exploration of physics-guided machine learning offers a framework for future NASA retrieval mission of geophysical parameters that allows for direct learning from measurement data while ensuring the physical consistency provided by forward modeling. 
 
Finally, this research seeks to use the forward modeling tools created by this proposal in addition to NASA’s CYGNSS and SNOOPI datasets to assess SM retrieval capabilities from future, potential SoOp missions. A simulation study over the 100-2400 MHz frequency range will explore the use of other potential SoOp sources in SM retrieval. This study will assess the expected retrieval quality of SM from a multi-frequency SoOp system that will potentially be more resistant to error induced by vegetation. 
 
  
Sara Lance (PI)/Christopher Lawrence (FI) 
State University of New York, Albany 
20-EARTH20-0298, Emergence of a New Chemical Regime: Organic Carbon and Base Cations in Whiteface Mountain Cloud Water 
 
Whiteface Mountain (WFM) is an historic cloud water monitoring site in upstate New York that provides a unique opportunity for investigating in-cloud formation and chemical processing of secondary organic aerosol (SOA), which is ubiquitous in aerosols around the world. I am employing new measurements to characterize the composition of aerosols and cloud droplet residuals (i.e. the suspended particulate matter left behind after cloud droplets evaporate) at the summit of WFM. These measurements complement ongoing long-term bulk cloud water measurements and will be used to further investigate the role clouds play in the formation of SOA under a newly emerging chemical regime characterized by low sulfate and increasing abundance of base cations and organic carbon. Chemical transport modeling and box modeling are currently being conducted to better understand the sensitivities of our chemical system in recent years. For the proposed work, I will use NASA observations to better identify and characterize specific pollution events, with an emphasis on biomass burning smoke, which is frequently encountered at WFM. This research provides important insight on factors impacting aerosol chemical and physical properties, in alignment with NASA’s mission to investigate how the Earth is changing, discover what causes these changes and predict how the Earth will change in the future. 
 
  
Tristan L'Ecuyer (PI)/Julia Shates (FI) 
University of Wisconsin, Madison 
20-EARTH20-0052, Characterizing Precipitation Structure and Processes in the Satellite Radar Blind Zone 
 
Satellite observations provide near global coverage of precipitation, but ground clutter results in a satellite radar blind zone, which makes retrievals of surface precipitation unreliable and underestimates shallow precipitation. Investigations of the occurrence and macrophysical characteristics of precipitation regimes, the vertical structure of precipitation, and the precipitation phase transitions using ground-based observations provide necessary insights for satellite retrieval algorithm improvements and lend insight into future space-based instrument requirements, such as radar vertical resolution and sensitivity.  
 
This work proposes a framework for analyses to characterize precipitation vertical structure and phase transitions at two distinct sites: the North Slope of Alaska (NSA) and Marquette, Michigan (MQT). Precipitation will be examined using ground-based in-situ and remote sensing instruments, and the proposed methodology can be applied to similar instrument sites at locations across the globe. 
 
In this work, we will: 
1.	Develop methodologies to identify and characterize snowfall and snow virga regimes and  at NSA. Determine occurrence frequency of snowfall regimes and relative accumulations, connections to environmental conditions, and thermodynamic structure for distinct regimes.   
2.	Investigate radar profile relationships (e.g. vertical precipitation profile characteristics, timelagged correlations) with surface snowfall rates at NSA and MQT.  
3.	Characterize the heights of melting layers in radar profiles in relationship to surface measurements, thermodynamic profiles, and environmental conditions at NSA and MQT. 
 
Characterizing precipitation using ground-based instruments and developing an approach to continue this work at other sites contributes to a necessary step to understand the Earth System and how it will change in the future, which are topics outlined in NASA Science Mission Directorate Strategic Objective 1.1: Understanding the Sun, Earth, Solar System, and Universe.  Additionally, the proposed research complements and supports space-based precipitation observing missions. The proposed work aligns with goals and objectives of NASA global satellite observing systems including Global Precipitation Measurement and CloudSat observing systems, and future missions including the Aerosol and Cloud, Convection and Precipitation mission. 
 
  
Jan Lenaerts (PI)/Michelle Maclennan (FI) 
University of Colorado, Boulder 
20-EARTH20-0148, Constraining the Atmospheric Drivers of Thwaites Glacier Mass Balance and Implications for Future Sea Level 
 
Thwaites Glacier (TG) and the Amundsen Sea sector (AS) of West Antarctica are losing mass faster than any other region on the Antarctic Ice Sheet (AIS). Currently, mass loss is dominated by basal melt beneath floating ice shelves, which has led to the acceleration of ice discharge. The mass balance of the AIS is determined by the difference between the surface mass balance 
(SMB) and discharge across the grounding line, and so far, there has not been an increase in the SMB over TG to compensate for the ocean-driven mass loss. However, climate models strongly indicate that interior portions of the AIS, including TG, will experience more snowfall and therefore a higher SMB in a warming world, which may mitigate future mass loss. A warming atmosphere may also increase surface melt and rainfall on TG, which will act to reduce SMB in the lower parts of the glacier. This contrasting effect of future SMB increase at higher elevations and SMB decrease at lower elevations challenges a robust assessment of future changes in overall TG SMB. SMB in the AS is largely driven by extreme snowfall events and atmospheric rivers (ARs), narrow bands of warm and moist air that contribute both positively and negatively to the SMB through intense snowfall and surface melt. To constrain projections for the impact of a warming climate on the mass balance of the West Antarctic Ice Sheet (WAIS), we propose to diagnose past variability and trends in the SMB of AS, define the role of ARs in past and future mass gain, and examine the sensitivity of the WAIS mass balance to future changes in SMB. Here, we combine reanalysis products and observations to characterize the impacts of ARs on past and present-day SMB on the AS, using an Atmospheric River Detection Tool (ARDT) unique to the Antarctic region. We then apply the ARDT to CMIP6 climate models in both the historical (1980-2014) and future periods (2015-2300) to examine how ARs change in the 21st century in their frequency of landfall and associated impacts on SMB. Based on these characteristics, we will develop meaningful scenarios of SMB in 20152100, and use these datasets to force the Ice Sheet System Model (ISSM) over the AS. We will use these simulations to determine the sensitivity of the mass balance of the WAIS to atmospheric forcing in the 21st century, with a focus on the AR-driven impacts on SMB. The main deliverables of this work will be (a) the Antarctic ARDT applied to CMIP6 models for the historical (1980-2014) and future (2015-2300) time periods, (b) SMB forcing datasets for ISSM, and (c) ISSM output of monthly ice sheet thickness and floating ice masks. This proposed research leverages NASA observations, climate models, and reanalysis products to increase our understanding of the role of SMB in mitigating future mass loss from the WAIS and addresses the overarching goal of the Cryospheric Science program of  improving our understanding of ice sheet, ice shelf, and glacier processes and how those processes affect ice mass balance and ultimately sea level rise. 
 
  
Steven Margulis (PI)/Manon von Kaenel (FI) 
University of California, Los Angeles 
20-EARTH20-0252, Real-Time Diagnosis of Spatially-Continuous Snow and Snow-Driven Streamflow for Decision Support 
 
In mountainous regions where snowpack plays an important role in the water cycle, understanding the distribution of snow water resources over time and space is challenging for water managers and scientists because of high topographic variability and difficulties in getting reliable snow measurements. However, this challenge is crucial to address because of the significant contribution of snowpack to streamflow after snowmelt. Projected changes in climate also mean that water managers will have to adapt to more uncertain and variable conditions in mountainous areas. Billions of humans, and industries and ecosystems worldwide depend on snowmelt-driven water supply. However, current approaches to estimate snow water resources in real time are plagued by measurement errors, discontinuities in space and time, model uncertainty, and downscaling problems. Furthermore, snow studies typically do not link snowpack directly to streamflow in a useful way for water managers. To address these challenges, I propose leveraging a well-verified historical SWE (snow water equivalent) product to develop a framework that outputs high-resolution maps of snow water resources in realtime and uses those to improve streamflow forecasts in snow-driven river basins. The historical product provides valuable information on daily snowfall distribution over a 35-year time period that allows us to downscale remotely-sensed precipitation to a fine enough resolution where the effect of the high topographic and climatic variability in mountains is well resolved. This downscaled product will thus have an embedded uncertainty estimate, can be input to a wellverified model that also incorporates remotely-sensed observations of fractional snow cover to produce maps of spatio-temporally continuous SWE in real-time. These high-resolution maps and their uncertainty estimates can then be used to constrain a land surface model that estimates runoff and forecasts streamflow at several lead times. I propose to develop and test the entire framework in test basins that contain a variety of validation data across the Western US. The lessons learned from these test basins, which span a spectrum of climate and hydrology regimes, can be used to expand the framework to other global mountainous regions. The proposed project integrates remote sensing data, land surface modeling, and ensemble streamflow prediction into a single framework that translates readily available remote-sensing observations from NASA into valuable information about uncertainty, snow and streamflow that can facilitate decision-making for water managers in mountainous systems. The expected output of this research project has important implications for decision support applications and for answering science questions. I expect the results to provide scientific insights related to how uncertainty in snowfall distribution propagates down to streamflow estimates, the transition of snow to streamflow under a variety of land surface and climate conditions, and the relative benefits of high-resolution SWE maps for streamflow forecasting and decision making. 
 
  
J. Vanderlei Martins (PI)/Noah Sienkiewicz (FI) 
University of Maryland Baltimore County 
20-EARTH20-0111, Development of HARP-Centered Aerosol Retrievals from a CubeSat to the Global Polarimeter on PACE and Beyond 
 
HARP Cubesat is the first spaceborne polarimeter since the launch of the PARASOL mission in 2004 and is now providing the first wide field-of-view polarimetric imagery from space in over a decade. Launched in November 2019 from Wallops Virginia, HARP was deployed from the ISS in mid-March 2020, and since then, HARP has captured 35+ science ready pushbroom images across the globe, focusing on aerosol plumes and colocation events with other satellites/platforms. HARP is also the first large scale aerosol and cloud remote sensing imager to be launched in a small form factor and thereby represents not just a case study of new polarimetric systems, but also a pathway for further development of cheaper, compact instrumentation for atmospheric research from space, and the challenges they face. We propose here in-depth analysis of the HARP Cubesat imagery by the performance and analysis of optimal estimation aerosol retrievals of different events (including desert dust and smoke) captured during the HARP Cubesat data collection. To do so, we must tackle and enhance knowledge of the instrument s ongoing calibration, system degradation, preprocessing, level 1 and level 2 algorithms including cloud masking and the aerosol retrieval itself. Another important factor is the study of the error involved in instrument characterization, in the measurement itself, and how it propagates throughout the retrieval. The analysis of calibration and system degradation will directly inform the development of the PACE polarimeter (HARP2) and future systems like the ones proposed for the A&CCP mission (Mega-HARP), while the development of cloud masking and other processing algorithms will also serve those missions beyond launch. In doing this work, we expect to produce science ready level-2 aerosol products from the unique HARP Cubesat datasets and select science case studies to explore the power of multi-angle imaging polarimetry in conjunction with other available measurements from space and from the ground. We also intend to provide level 2 data products and algorithms to pace the way for other scientists to perform science analysis and develop their own level-2 algorithms. 
 
  
Lynn McMurdie (PI)/Andrew DeLaFrance (FI) 
University of Washington, Seattle 
20-EARTH20-0027, Remote Sensing of Orographically Modified Precipitation Processes 
 
 	Precipitation in mountainous regions provides vital downstream water resources that require accurate measures of global precipitation distributions. Satellite-based remote sensing platforms permit global monitoring of precipitation including over remote mountainous regions that lack ground-based instrumentation. The NASA Global Precipitation Measurement (GPM) mission Core Observatory satellite measures rain and snow from Ku- and Ka-band radar, and the future Aerosol and Cloud, Convection and Precipitation (ACCP) study will expand these capabilities through inclusion of W-band and Doppler radar measurements. Current GPM remote sensing algorithms require a priori assumed precipitation properties that do not reflect the intrinsic variability of hydrometeors, particularly in the ice phase. Additionally, ice-initiated precipitation processes can be greatly modified in orographic flow contributing to an enhancement or redistribution of precipitation. As a result, remote sensing precipitation estimates often have large biases over complex terrain. The proposed project focuses on improving our understanding, and remote sensing representation, of orographically modified ice-initiated precipitation processes over complex terrain during midlatitude winter cyclones. The project is designed to support NASA’s Earth Science Division goals of improved assessment and management of Earth’s vital water resources. Data used in this project are from three NASA-led field campaigns; the Olympic Mountains Experiment (OLYMPEX), the Investigation of Microphysics and Precipitation for Atlantic Coast-Threatening Snowstorms (IMPACTS), and the International Collaborative Experiment for PyeongChang Olympic and Paralympics (ICE-POP). Measurements from midlatitude winter cyclones over these diverse mountainous regions will be used to evaluate how ice-initiated precipitation processes are modified over varied barrier shapes and synoptic regimes. To assess the suitability of a priori assumptions of hydrometeors, remote sensing retrievals from OLYMPEX and IMPACTS Ku-, Ka-, and W-band radars will be compared with coincident in situ measurements within the ice layer. Additionally, groundbased measurements on windward slopes during OLYMPEX and ICE-POP will be compared to remote sensing retrievals of precipitation rate estimates. Statistical calculations will be used to evaluate agreement between remote sensing retrievals and in situ measurements and to determine the sensitivities to a priori precipitation assumptions over complex terrain. This project will use in situ data to investigate new remote sensing retrieval techniques that incorporate radiometer measurements with combined triple-frequency and Doppler radar measurements to more accurately represent ice-phase precipitation processes over complex terrain. This work is expected to support improvements in remote sensing precipitation rate estimates from the GPM and upcoming ACCP satellite-based platforms, advancing our understanding of the Earth’s precipitation budget and benefitting society through informed management of water resources. 
 
  
Brent Minchew (PI)/Faye Hendley Elgart (FI) 
Massachusetts Institute of Technology 
20-EARTH20-0207, Inferring Sub-Ice-Shelf Melt Rates Using ICESat-2 Altimetry and Simple Physical Models 
 
The potential future collapse of the West Antarctic Ice Sheet (WAIS) represents the single greatest source of uncertainty in projections of global sea level rise (SLR). In particular, ice shelves in the Amundsen Sea Embayment are rapidly thinning due to unsteady ocean-driven melting due to warm, salty water known as Circumpolar Deep Water (CDW) upwelling onto the continental shelf. This work will use recent revolutionary advances in both the quality and quantity of available remote sensing data and technology to constrain basal ice shelf melt rates in West Antarctica with unprecedented spatio-temporal accuracy. The proposed work brings together first-principles physical models of the grounding zone (where ice, ocean, and bedrock meet) with scalable, high resolution remote sensing data from ICESat-2 and other sources to create a novel, accurate map of ice thickness in the grounding zone. This record will then be used along with conservation of mass to calculate basal melt rate at high spatiotemporal resolution. In turn, this will improve our understanding of the causes, effects, and feedbacks of glacial melting around Antarctica, and significantly reduce uncertainty around future SLR. 
 
We have four objectives. Aim 1 is to fully quantify the effects of variable ice thickness and thickness gradient on ice shelf tidal flexure. Aim 2 is to use those results to create an inverse model of ice shelf profile in the grounding zone from ICESat-2 laser altimetry elevation data, and thus generate a record of ice shelf thickness in the grounding zone of many major ice shelves. Aim 3 is to use conservation of mass with the ice shelf thickness record along with records of ice shelf surface velocity and surface mass balance to calculate basal melt rate in the grounding zone of those ice shelves. Finally, Aim 4 is to use our observationally constrained record of basal melt rate as a test of current parameterizations of the spatial distributions of basal melt rate, and to investigate the causes and effects of glacial melt in the climate system.  
 
With this work, we endeavor to reduce critical uncertainties about the evolution of our climate system. The proposed work is highly relevant to the Earth Science Research Program focus areas of sea-level rise (SLR) and reducing climate uncertainty and informing societal response, identified by the 2017 NASA Earth Science Decadal Survey. Our work has the ultimate goal of understanding SLR (rated highest priority by the Decadal Survey), surface deformation and change (high priority for observations), and ice elevation (high priority for observations). The proposed work addresses all of the key questions raised in Strategic Objective 1.1. We seek to quantify key aspects of how the global Earth system is changing (grounding line ice thickness as a function of time and basal melt rate, Aim 2-3), and seek to extend that work to understand what causes those changes in the Earth system (Aim 4). Accurate knowledge of basal melt rate can additionally be used directly to improve projections of future global sea level rise through use as an input to global and regional climate models to quantify how the Earth system will change in the future. Furthermore, reducing the uncertainty in future sea level rise, which currently varies from centimeters to up to a meter on decadal time scales, will provide direct societal benefit, as the ways in which coastal societies prepare for a centimeter versus a meter of sea level rise will be very different. 
 
  
Peter Neff (PI)/Julia Andreasen (FI) 
University of Minnesota 
20-EARTH20-0173, Constructing Radar Snow Accumulation Time Series to Improve Understanding of Surface Mass Balance Processes and Recent Climate in Coastal West Antarctica 
 
The West Antarctic Ice Sheet (WAIS) is a dynamic and critical region of Antarctica, where ice, ocean, and atmosphere converge to drive change--persistent ice loss along the Pacific coast of WAIS is of global concern for sea level rise. Reliable weather observations, climate reanalysis products, and satellite remote sensing datasets for this region are largely unavailable prior to the mid-20th century, limiting understanding of ice-ocean-atmosphere interactions along the WAIS coast. NASA’s Operation IceBridge (OIB) can provide new observational constraints on coastal snow accumulation in this region. Utilizing OIB snow radar data, multi-decadal records of interannual snow accumulation variability will be produced and used to validate climate reanalysis products, providing better understanding of regional ocean-atmosphere influences on snowfall particularly at coastal ice rises. Spatiotemporal correlations between multi-decadal radar snow accumulation time series, ice cores (where available), and reanalysis variables (e.g., geopotential height, temperature, etc.) will provide new constraints on the climate processes driving surface mass balance patterns and trends along the WAIS coast. 
 
  
Kim Novick (PI)/Qing Chang (FI) 
Indiana University 
20-EARTH20-0143, Incorporating Land-Atmosphere Feedbacks into Agricultural Drought Monitoring and Forecasting 
 
Many widely used drought indices have been developed for detecting and forecasting meteorological drought; however, the vast majority of those drought indices often fail at detecting drought impacts on plants, since they treat plants as static participants in the hydrologic cycle. In fact, plants respond quickly and dynamically to evolving water stress in ways that  control land-atmospheric water and carbon exchanges, and thus strongly affecting the evolution of drought. The overall goal of this proposal is to understand the role of plants in affecting drought evolution and processes and to incorporate land-atmosphere feedbacks into plant drought monitoring and forecasting frameworks using both in-situ and satellite datasets. This project represents a novel approach for making substantial gains in  our ability to characterize plant water limitation and detect early drought signals. This project will answer several questions relevant to NASA’s Earth Science research goals aiming to  Understand of ecosystem interactions with the atmosphere and hydrosphere leading to comprehensive modeling of the exchange of gases, water, and energy among the components of the earth system, to  Improve the ability to predict climate changes by better understanding the roles and interaction of the atmosphere and land in the climate system , and to  Detect and predict changes in Earth’s ecosystems and biogeochemical cycles, including land cover, biodiversity, and the global carbon cycle. 
 
  
Gregory Okin (PI)/Francisco Ochoa (FI) 
University of California, Los Angeles 
20-EARTH20-0214, Imaging Spectroscopy of Surface Soil Mineralogy and Vegetation Cover: Sensitivity and Validation 
 
Drylands, which cover 40% of the Earth s land surface and house over 40% of the Earth s population, 2 billion of which are dependent upon the ecosystem for sustenance, are especially ripe for the application of spaceborne imaging spectroscopy. Low and discontinuous cover in many of the world s drylands means that the spectral signature of vegetation and the soil background is nearly always mixed in Landsat-scale pixels. The composition of the vegetation and soil in drylands is critical to the role these lands play in the Earth system. For vegetation, both the functional type (with more than one often co-dominant, as in savannas) and status (i.e., green vegetation, GV vs. nonphotosynthetic vegetation, NPV) determine the form and function of dryland ecosystems. For soils, the mineral composition of the soil is reflected in the dust produced in world s drylands and influences both the magnitude and sign of radiative forcing caused by mineral aerosols. Quantification of the mineral content of the soil as well as vegetation cover requires continuous spectra from the entire reflected, with many important diagnostic spectral features occurring in the short-wave infrared (SWIR). 
The proposed research will use spectral unmixing, field measurements, and modeling to contribute to the assessment of imaging spectroscopy s ability to accurately and simultaneously quantify vegetation fractional cover and soil mineral composition. This analysis will include propagation of uncertainty. Specifically, three activities are planned: 
Activity 1: Derive and validate unmixing-derived fractional cover of GV, NPV, and soil, 
Activity 2: Estimate and validate mineral abundance from laboratory spectra, and 
Activity 3: Evaluate the error in mineral abundance estimates caused by vegetation cover and the error in GV/NPV fractional cover caused by different mineral soil substrates. The proposed research is directly applicable to the Surface Biology and Geology (SBG) 
Designated Observable identified in the National Academies  Decadal Survey. The Algorithms 
Working Group of the SBG research and applications study has identified Proportion Cover (including GV and NPV cover) and Substrate Composition (including mineral areal fractional abundance) as Key Products. These products contribute to the Decadal Survey s Objective E-1a 
(Composition of vegetation), Objective E-1c (Dynamics of primary producers), Objective S-7a 
(Surface mineralogic composition), Objective E-2c (Ecosystem subsidies from solid Earth), Objective C-2h (Aerosol radiative forcing), Question C-5 (Aerosols  impact on radiation budget), Objective H-2b (Anthropogenic processes that cause changes in radiative forcing). The research will also benefit from the Earth surface Mineral dust source InvesTigation (EMIT) project through the FI s mentoring by members of the Science Team, the use of EMIT algorithms, and access to EMIT data. The research will contribute to the EMIT project by providing insight into the interaction between vegetation cover and mineral soil signatures in the retrieval of both.  Activity 1 involves collection of in situ data proportion cover validation data concurrent with EMIT or Airborne Visible Infrared Imaging Spectrometer (AVIRIS) image acquisition in drylands of the United States Southwest. These data will be used to validate vegetation cover retrievals from imaging spectroscopy.  Activity 2 involves comparison of soil mineral  spectral abundance  with known concentrations of minerals to validate and/or create usable calibrations. Activity 3 will use spectral simulations to estimate the impact that soil substrate has on vegetation cover retrievals and the impact of vegetation cover on mineral abundance retrievals from spaceborne imaging spectrometer data. Activities 1 - 3 will be conducted throughout a period of three years (9/01/2021   9/1/2024).  Findings from activities will be presented in publications, conferences, and in activities for K-12 students. 
 
  
Michael Oskin (PI)/Alba Mar Rodriguez Padilla (FI) 
University of California, Davis 
20-EARTH20-0313, Testing Constitutive Laws for the Evolution of Off-Fault Deformation over the Earthquake Cycle 
 
A growing body of geologic and geodetic observations, including folding around faults, steep slip gradients at fault step-overs, and deficits in shallow earthquake slip, requires a distributed off-fault failure mechanism in the upper crust. Neglecting stress dissipation by this off-fault deformation restricts the ability to simulate earthquake cycles and hampers the comparison of current plate motion with long-term geologic deformation. I will add constraints to a constitutive law for off-fault deformation by modeling observations of fault-tip folding collected at coseismic, postseismic, and geologic timescales from locations spanning a variety of strain rates and lithologies. I will use inSAR (interferometric synthetic aperture radar) time series to measure the postseismic evolution of folding. To calibrate the behavior of off-fault folding over multiple earthquake cycles, I will compare folding measured from high-resolution altimetry data, collected by NASA and the National Center for Airborne Laser Mapping, to model predictions under different candidate constitutive laws using open-access finite element code Pylith. I will test elastic, elastoplastic, linear viscoelastic, and non-linear viscoelastic rheologies to describe off-fault deformation. Following NASA’s Earth and Planetary SMD s objectives to improve the response to earthquake hazard and quantify the nature of deformation associated with plate boundaries, this project will provide a constitutive law that accounts for stress dissipation by off-fault deformation for incorporation into the next generation of geodynamic and earthquake-cycle models, link coseismic strains to long-term crustal deformation, and guide future radar observations of seismic and aseismic deformation over the earthquake cycle. 
 
  
Volker Radeloff (PI)/Natalia Rogova (FI) 
University of Wisconsin, Madison 
20-EARTH20-0139, Agricultural Abandonment Across the Eurasian Steppe: Effect on Fires, Vegetation Succession and Habitat Quality for Rare Waterfowl Species 
 
The Eurasian steppe were converted to an agricultural landscape in the middle of the 20th century and then abandoned after the collapse of the Soviet Union in 1991. These rapid changes have numerous consequences such as rewilding of former fields, burned area increase, vegetation succession and changes in habitat structure affecting steppe inhabitants and migratory wildlife. Remote sensing data provide a unique opportunity to study spatiotemporal land cover changes while bird GPS tracking and in-the-field data collection with drones allow to evaluate ecological response to land cover changes. 
 
The overall goal of my proposed research is to study the effects of agricultural abandonment in the Eurasian steppe on fires, vegetation successions, and habitat use by rare migratory geese. Specifically, I will:  
1)	evaluate fire dynamics since the late Soviet period in unplowed steppes and abandoned fields and identify relationships between fires and land use changes;  
2)	describe vegetation succession due to fires in unplowed steppe and on abandoned fields; and  
3)	analyze habitat use by rare waterfowl species, the spatial-temporal variability of their stopover sites, and the role of land cover changes in this variability in order to model bird distributions in the Eurasian steppe during migrations.  
 
I will use my already developed land cover maps to extract the necessary information (unplowed steppe, abandonment, active agriculture and water bodies) for my proposed research. For the first objective I will create annual fire maps using all available Landsat images from 1986 to 2020 complemented by MODIS if necessary. For each year, I will evaluate the amount of fires, their mean areas, and the proportion of burned and unburned areas in steppe and abandonment. Then I will evaluate differences in fire regime between steppe and abandonment, across various ecoregions, and if there is a specific fuel accumulation period on abandoned fields for fires to emerge. For the second objective, I will conduct massive fieldwork in Kazakhstan and collect botanical data in unburned and regularly burned areas. I will assess fire effect on vegetation structure and plant biodiversity and test if there is a positive feedback loop in that fires foster successional pathways in favor of more flammable communities, resulting in even more fires. For the third objective, I selected three model species (Taiga been goose; Red-breasted goose; and Bewick’s Swan) that are species of conservation concern. I will map their migration stopover sites (using birds GPS data), compare them with the land cover maps and estimate what environmental factors are more important for waterfowl habitat selection. Then I will utilize these data to model distribution of these species in the Eurasian steppe during their migrations. 
 
Results of my research will provide new insights into grassland ecology, relationships of different processes (fires, vegetation successions) in open landscapes and waterfowl biology.  Practically, the results will be helpful for fire and animal population management, improving protection of rare species and high conservation value landscapes, hunting regulations and creating protected areas. My research will be relevant to the NASA Earth Science Research program goal «Detect and predict changes in Earth’s ecosystems and biogeochemical cycles, including land cover, biodiversity, and the global carbon cycle. Results will reveal humantriggered land cover changes and their environmental consequences, so could serve as a basis for to predicts the effects of of climate change or human landscape transformations elsewhere. 
 
  
Kristen Rasmussen (PI)/Marquette Rocque (FI) 
Colorado State University 
20-EARTH20-0195, Orographic Impacts on Electrical Properties of Severe Storms in Subtropical South America in a Current and Future Climate 
 
Satellite-based studies have shown that some of the most intense storms on Earth occur downstream of the Andes mountains in subtropical South America (SSA). These storms are characterized by high lightning flash rates and deep convective cores often reaching at least 15 km in height. However, the lack of ground-based observations in this region has made it particularly challenging to thoroughly investigate convective characteristics of these intense storms. Collecting invaluable observations of the convective lifecycle in SSA was a primary goal of the Remote sensing of Electrification, Lightning, And Mesoscale/microscale Processes with Adaptive Ground Observations (RELAMPAGO) and the Clouds, Aerosols, and Complex Terrain Interactions (CACTI) field campaigns in 2018-2019. The datasets obtained from these field campaigns will be used to address several objectives. First, we will document the electrical characteristics of severe storms in SSA and determine how well the Geostationary Lightning Mapper (GLM) onboard the Geostationary Operational Environmental Satellite (GOES-16) detects lightning in this region. Additionally, several studies that analyzed convection in the U.S. 
have shown that lightning can be used as a predictor for severe weather. Thus, we will also investigate the predictability of severe weather in SSA using data from GLM.  
 
This geographic area is also characterized by the complex terrain of the Andes, which most likely plays a significant role in the convective lifecycle. The second objective this research aims to address is determining how important terrain features such as the Sierras de Córdoba (SDC) or the foothills of the Andes are in convective initiation and electrification. Past studies of SSA have shown lightning flash rates are highest close to the SDC, but given the lack of observations in this region, it is still unknown why this is the case. This objective will be explored by conducting terrain-modification experiments using the Weather Research and Forecasting (WRF) model. Lightning parameterizations will be tuned to this region based on ground-based observations from RELAMPAGO, and the spatiotemporal evolution of lightning throughout the convective lifecycle will be investigated. 
 
The third objective of the proposed research is to compare the kinematics, microphysics, and electrification of current severe storms in SSA to those in a future warmer climate. Regional climate simulations run over South America and forced by a climate perturbation derived from the RCP8.5 scenario will be analyzed and compared with results from RELAMPAGO. Potential changes in convective characteristics would likely play a major role in modulating the water and energy cycles in this region and would have large societal impacts. 
 
The proposed research will expand the scientific community s understanding of the relationship between storm electrification, kinematics, and microphysics and will provide insight into how satellite-based lightning observations from GLM can improve forecasts and the predictability of severe weather. Furthermore, this research will explore the impacts a warming climate will have on storm location, duration, and intensity in SSA, and will allow for the preparation of vulnerable populations to climate change impacts. 
 
  
Eric Rignot (PI)/Jae Hun Kim (FI) 
University of California, Irvine 
20-EARTH20-0213, Iceberg Calving Dynamics and Ice-Ocean Interaction in West Greenland Glaciers Combining Space- and Ground-Based Observations 
 
The ice sheets in Greenland and Antarctica are major contributors to sea level at present, but projections of their contributions to future sea level change are affected by large uncertainties. Many of these uncertainties are bound to our incomplete knowledge of glacier dynamics and how fast these glaciers could discharge mass into the ocean in the coming decades. The main processes of mass loss from these glaciers are increased production of icebergs, enhanced melting of ice in contact with ocean waters, and surface melt. The first two are the least well known components of the mass balance, especially iceberg calving. Calving has been associated to longitudinal stretching, subaerial melting, submarine calving, brittle fracture with marine ice sheet instability and marine ice cliff instability, and full-thickness block rotation. To detect and quantify iceberg calving dynamics, research has been conducted using remote sensing data such as synthetic aperture radar interferometry (InSAR) and GPS data. Calving occurs on timescales spanning hours to decades. Satellite data cover time scales from weeks to decade while terrestrial radar interferometers (TRI) cover time scales from minutes to weeks.       In this study, we will use TRI data and satellite InSAR and optical data to improve our understanding of iceberg calving dynamics in West Greenland. We will use TRI data from 4 glaciers already surveyed, 2 more to be surveyed with independent funding in 2022-2023, and 2 more available surveyed by colleagues who will likely share the data at some point during this project. The study area stretches from Jakobshavn Isbrae  to the northern end of Uumammaq fjord, West Greenland. TRI data will constrain ice speed, ice front positions, grounding line positions, calving processes, tensile stresses, which, when combined with precision bathymetry, will determine the fraction of calving events that remove grounded ice vs floating ice. We will complement these data with satellite SAR (Sentinel-1a/b, Cosmo Skymed, and starting in 2022 NISAR), optical (Landsat-8, Sentinel-2), and time series of digital elevation models 
(WorldView/GeoEye, ICESat, ICESat-2) data on glacier velocity,  surface topography, ice front positions for the past 25 years. We will calculate glacier undercutting by the ocean using historical temperature data, bathymetry and regional atmospheric climate models combined with a parametrization from the MITgcm ocean model. We will test how we can improve  our representation of calving processes and undercutting by modeling the 4-8 TRI glaciers with the  UCI/JPL Ice Sheet Sea-level system Model (ISSM) and comparing the results with short-term and long-term observations. The modeling work will then be extended to a set of 15 glaciers in the region and eventually extended to run projections for this Century. If successful, the results will inform ice sheet models on how to better represent calving and ocean-induced melt processes at glacier margins, and in turn improve projections of sea level rise. This work supports NASA's Earth Science overarching goal to study the melting of the Earth's land ice masses and their impact on sea level rise. 
 
  
Dar Roberts (PI)/Christopher Kibler (FI) University of California, Santa Barbara 
20-EARTH20-0327, Drought Sensitivity of Evapotranspiration and Carbon Uptake in Riparian Woodlands 
 
Riparian woodlands are among the most productive plant communities in dryland ecosystems because they grow in river corridors where alluvial water tables provide a constant supply of water. As a result, the evapotranspiration (ET) and carbon uptake of riparian tree species are relatively insensitive to normal climate variability. During drought conditions, however, groundwater recharge decreases and alluvial water tables decline. Riparian tree species are extremely vulnerable to xylem cavitation, so when their root systems lose access to groundwater, they close their stomata at moderate levels of atmospheric water demand. These changes substantially reduce both ET and carbon uptake by riparian trees. It is likely that drought-induced decreases in the productivity of riparian woodlands have substantial impacts on carbon cycling at regional scales, but the drought sensitivity of ET in riparian woodlands is poorly constrained. As a result, their impact on dryland carbon cycling is not well understood. In this analysis, I will use thermal remote sensing imagery from MASTER and ECOSTRESS to quantify the drought sensitivity of ET in riparian woodlands at diurnal, seasonal, and interannual time scales. I will focus on the period during and after the 2012-2019 California drought, which was unprecedented in the paleoclimate record. I will compare the ET measurements to in situ measurements of groundwater depth, soil moisture, air temperature, and vapor pressure deficit. These measurements will enable me to identify the environmental triggers that cause riparian tree species to downregulate ET during drought conditions. I will also use species maps derived from AVIRIS and AVIRIS-NG imagery to compare drought responses across different species and plant types. This research will transform our understanding of dryland carbon cycling under drought conditions, and it will improve our ability to model water and carbon fluxes in dryland watersheds. These findings are especially important as drought conditions become more frequent and more severe under anthropogenic climate change. 
 
  
David Siegel (PI)/Nathalie Eegholm (FI) 
University of California, Santa Barbara 
20-EARTH20-0223, Multi-Scale Demographic Modeling of an Emergent Macrophyte in the Santa Barbara Channel Informed by Remote Sensing 
 
This project aims to use our emerging understanding of giant kelp life cycle and physiology to develop novel techniques for predicting the temporal and spatial dynamics of giant kelp population dynamics from satellite data. We propose to use high spatial and spectral resolution sUAS imagery to assist in the development of an age-structured demographic model of giant kelp in the Santa Barbara Channel, which will consider variables including nutrient and light availability, recruitment, physical forcing and disturbances. This proposed project supports the objectives outlined in the Surface Biology and Geology Designated Observable to quantify the physiological dynamics of terrestrial and aquatic primary producers  and to  quantify the distribution of the functional traits, functional types, and composition of vegetation and marine biomass, spatially and over time.  
 
  
Deepti Singh (PI)/Dmitri Kalashnikov (FI) 
Washington State University, Pullman 
20-EARTH20-0120, Dry Thunderstorms in the Western United States: Meteorological Conditions, Teleconnections, Subseasonal-To-Seasonal (S2S) Predictability, and Future Projections 
 
This proposal aims to investigate the physical drivers of dry thunderstorms in the western United States (WUS), their teleconnections, predictability on subseasonal-to-seasonal (S2S) timescales, and response to projected warming in the WUS. Dry thunderstorms occur during the summer across this region and are characterized by cloud-to-ground lightning without meaningful rainfall. Since vegetation is seasonally dry across many parts of this region during the summer, dry thunderstorms present a major wildfire ignition risk as was witnessed in central California during August 2020.  
 
My proposed work has three main research objectives. First, I will use a combination of ground sensor-based lightning data and NASA’s IMERG satellite-based precipitation dataset in order to identify dry thunderstorms on a 0.1° grid resolution across the WUS, and use reanalysis products to identify the combination of meteorological conditions associated with dry thunderstorm days. Second, I will investigate teleconnections of dry thunderstorm-producing meteorological conditions with natural modes of climate variability to assess S2S prediction potential of dry thunderstorm risk in the historical record. Finally, I will evaluate the response of dry thunderstorm-producing meteorological conditions and consequently, dry thunderstorm risk in the WUS to warming using CMIP6 climate model simulations.  
 
Dry thunderstorms will be characterized using statistical and machine learning approaches in order to develop a robust classifier of days that are likely to produce a dry thunderstorm, at each grid cell, based on concurrent meteorological conditions. This approach will account for potential differences in meteorological factors across the region and will allow for a quantification of trends associated with dry-thunderstorm producing meteorological conditions over the reanalysis era. The association of climate variability modes such as the Madden-Julian Oscillation (MJO) and El Niño-Southern Oscillation (ENSO) will be explored for their ability to predict dry thunderstorm-producing meteorological conditions on S2S timescales using statistical and machine learning-based predictive models. The classification algorithm for dry thunderstorms developed based on observations will be applied to historical climate model simulations to assess model biases associated with the occurrence of these meteorological conditions. Model evaluation metrics will be quantified and then applied to future projections from CMIP6 models in order to build bias-corrected future climatologies of dry thunderstorm conditions and assess projected changes in their characteristics (frequency, persistence, spatial extent).  
 
Outcomes of the proposed work will provide useful information to fire weather forecasters, agencies responsible for fire suppression, and other planners and decision makers by providing 
1) a comprehensive understanding of dry thunderstorm climatology particularly in understudied locations across the varied geography of the WUS, 2) an identification of meteorological drivers for improving predictions, 3) quantification of the long-term variability and trends associated with dry thunderstorm-producing meteorological conditions across this region, and 4) projections of dry thunderstorm risks in a changing climate. 
 
This work is aligned with the NASA Earth Science Division’s mission to advance our knowledge of the Earth System, and addresses the goals of the Modeling, Analysis, and Prediction (MAP) program, falling under stated MAP research themes “Extremes in the Earth System “ and  “Predictability in the Earth System.”  Additionally within the Earth Science Research Program, this work is relevant to two focus areas: 2.2 Climate Variability and Change, question “How can predictions of climate variability and change be improved?”, and 2.5 Weather and Atmospheric Dynamics, question “How can weather forecast duration and reliability be improved?”  
 
  
Sergii Skakun (PI)/Mohammad Khan (FI) 
University of Maryland, College Park 
20-EARTH20-0167, Climate Induced Agriculture Change Hotspots and Its Implication to Global Food Security in the Former Soviet Union (Russia and Ukraine) 
 
In the past two decades, Russia and Ukraine (hereby referred as Rus/Ukr) has emerged as the major food producers of the world. One of the reasons is attributed to the influence of climate change in providing more favorable conditions. The warmer climate expanded the growing season length and allowed the use of more productive crop varieties. Higher temperatures also shifted bio-climatically suitable areas for intensive agriculture poleward, with significant increase in agricultural output especially for summer crops. There are also other studies which have shown that climate change will negatively impact the crops and the general perception of beneficiary effect of warmer climate is unlikely to hold, primarily due to increasing risk of droughts in the main agricultural areas of Rus/Ukr. However, most of the previous studies are conducted at coarse spatial resolution at the country or regional level and lack spatial inhomogeneity to assess the climate change impact on selected crop yields at local or field level. 
Also, the changing climate had altered the cropping patterns and had given rise to more extreme weather events with increased frequency. The drought of 2010 had a significant impact on the crop production and further altering the cropping patterns, with a major shift from winter to summer crops in Rus/Ukr. As the field of climate change proceeds to finer spatial and temporal scales there are more opportunities to examine the interaction of climate modelling and crop studies now. However, due to limited field data availability and accessibility in Russia, satellite data remains the only source of objective and synoptic information. Hence, this research is undertaken with the objective to understand the climate induced agriculture change hotspots and the adaptability of crops for extreme weather events within these hotspots. The above-mentioned research objectives can be achieved by addressing the following research questions:: 
i. 	Where are the hotspots of climate change induced agriculture changes located in Russia and Ukraine? ii. 	How accurate radar-based  transfer learning  will perform for crop type mapping in Russia? 
iii. 	How winter and summer crops response and adapt to extreme drought events in Russia and Ukraine? 
We will focus on NASA’s satellites/sensors of MODIS, Landsat, ECOSTRESS and will supplement it with ESA’s Sentinel 1 & 2 and high resolution Planet for calibration and validation. New methods of using machine learning based  transfer learning  approach for crop type mapping in data sparse regions of Russia will also be explored. ECOSTRESS satellite data onboard ISS will also be used for understanding the crop stress and adaptability for different winter and summer crops during extreme weather (drought) events for mid and higher latitudes in Rus/Ukr.  
The project directly addresses the NASA’s strategic objective  Advance knowledge of Earth as a system to meet the challenges of environmental change, and to improve life on our planet . This work supports the Science Mission Directorate (SMD) through the strategic objectives within the Earth Sciences Research Program, related to both research and applied sciences.  It will advance our scientific understanding of the issues and opportunities of climate change and environmental sensitivity by answering the key question of  how the global earth system is changing  with focus on the mid and higher latitudes regions of Northern Hemisphere. It will also answer  how the earth system science provides societal benefit  with focus on food security issues in the Rus/Ukr region. 
 
  
Brian Soden (PI)/Sisam Shrestha (FI) 
University of Miami, Key Biscayne 
20-EARTH20-0136, Investigating Large-Scale Atmospheric Circulation Changes and Their Implications for Climate Sensitivity Using Satellite Observations 
 
The tropics make up 40 % of the global surface area and is home to large-scale circulations like the Hadley circulation and the Walker Cell. Climate models suggest both an overall weakening of the tropical circulation, as well as a tightening of the ascending branch of the tropical circulation with rising temperatures. The weakening arises from energy and water balance considerations (Held and Soden 2006) and is manifest primarily as a weakening of the zonallyasymmetric circulation (i.e., Walker cell). The tightening is associated with a strengthening of the circulation, reduction in cloud coverage and an increase in extreme precipitation in the strongest convective regions. These model projected changes in the large-scale ascent have huge implications for both the hydrological cycle and climate sensitivity. Recent studies have suggested a long-term drying in formerly wet regions along the boundaries of the ascending branch due to a contraction of the tropical ascent.  Unfortunately, since mid-tropospheric vertical velocity (a measure of circulation strength) is virtually impossible to observe directly, most of the past studies have relied on reanalysis datasets for which trends are often spurious due their reliance on heterogeneous data sets.  
 
In this proposal, we seek to leverage long-term satellite data sets of upper tropospheric relative humidity (UTH) as a proxy to infer multi-decadal changes in the large-scale tropical circulation. 
An intercalibrated climate record of UTH from the High Resolution Infrared Radiation Sounder 
(HIRS) goes back to 1979, providing a 40+ year record of variations. This data will be supplemented by utilizing an intercalibrated archive of satellite microwave measurements at 183 GHz from sensors on operational satellites: AMSU-B (1998-present), MHS (2003-present), ATMS sensor on Suomi NPP (2012-present) that provide a 20+ year homogenized, data set of upper tropospheric water vapor radiances from all-sky conditions.  
 
By performing Observing System Simulation Experiments (OSSEs) using multiple reanalyses, we will determine the extent to which satellite observations upper-tropospheric relative humidity (UTH) can provide a proxy for diagnosing changes in the large-scale circulation over the tropics. These results will then be compared to climate model simulations from CMIP6. Using a hierarchy of coupled and atmospheric-only model simulations, we will investigate the influence of internal and external forcing on the changes in tropical circulation and the implications of the observed changes for climate sensitivity. 
 
  
Paul Staten (PI)/Samuel Smith (FI) 
Indiana University 
20-EARTH20-0094, Determining the Dynamical Drivers of Present and Future Changes in the Atmospheric Water Cycle 
 
As Earth warms, the atmospheric water cycle will intensify. Rising surface temperatures will boost the atmosphere’s moisture storage capacity, causing much of this water cycle intensification. However, the water cycle will not intensify at the same rate everywhere on Earth due to changes in the atmospheric circulations which produce or inhibit precipitation. Presently, it is unclear in many regions how strong this intensification will be because it is difficult to predict how the atmospheric circulation will change at higher temperatures. Furthermore, it is unclear how modified circulations will impact the water cycle.  
 
To improve our understanding of how the atmospheric circulation connects to the water cycle in present and future climates, we will analyze satellite observations, reanalysis, and model simulations using a novel diagnostic framework. This framework enables us to determine the role of atmospheric waves in producing or inhibiting precipitation. Since waves are intimately connected to the circulation and frequently contribute to extremes in the water cycle, the proposed work will aim to evaluate the contributions of the large-scale circulation to water cycle extremes and the fidelity of these connections in current Earth system models. We will analyze both interannual variability and the forced response in the circulation and the water cycle to examine the root causes for uncertainty in water cycle predictions. This proposal addresses the NASA Earth Science Research Program goal of “…accurately predict[ing] how the global water cycle [will] evolve in response to climate change.”  
 
  
David Sutherland (PI)/Nicole Abib (FI) 
University of Oregon 
20-EARTH20-0076, Characterizing Environmental Controls on Ice Mélange Distributions and Associated Ice-Ocean Feedbacks in Greenland 
 
Recent observations have documented the rapid breakup of Greenland’s ice tongues and the associated dynamic thinning and retreat of its marine-terminating outlet glaciers. The rapid retreat of these glaciers can be attributed to ocean and ice forcings occurring at the ice-ocean boundary. A key component of ice forcing is thought to be a reduction in the persistence of rigid ice mélange, a (semi)permanent conglomeration of icebergs, brash ice, and sea ice, at the glacier termini. Recent studies have shown that sustained cool ocean temperatures have led to a persistent rigid ice mélange at Greenland’s largest outlet glacier, Jakobshavn Isbræ. This development has slowed the glacier’s retreat, providing evidence that the glacier-mélange system is actually a glacier-mélange-ocean system. Beyond ice mélange’s influence on glacier dynamics, changes in ice mélange distribution may significantly affect circulation within Greenland’s glacial fjords. Changes in the freshwater flux exiting these glacial fjords can enhance exchange with warm ocean shelf waters by influencing fjord stratification and the underlying buoyancy-driven circulation. This enhanced exchange can in turn influence the behavior of Greenland’s outlet glaciers by increasing submarine melting of glaciers and ice mélange. Understanding the relationship between ice mélange persistence and buoyancydriven circulation driven by subglacial discharge, therefore, has important ramifications for long term glacier stability.  
 
To date, studies investigating the relationship between ice mélange presence and iceberg calving have focused on a few large glaciers in Greenland, such as Jakobshavn Isbræ, Helheim, and Store Glacier. In addition, the influence of meltwater from ice mélange on fjord circulation has only been examined in Sermilik Fjord. Although these studies have been able to resolve individual systems at high spatial resolution, these records provide little evidence on the regional spatial and temporal variability of ice mélange extent, persistence, and physical characteristics. To overcome this lack of observations, this study aims to extend detailed observations of ice mélange and its relationship to buoyancy-driven circulation beyond the three previously studied glaciers to a system of eleven glaciers in the Uummannaq Bay region of Central West Greenland. 
 
This study will integrate optical, radar, and LiDAR remote sensing to obtain a broader set of data for ice mélange characterization. In addition, the proposed study hopes to utilize this improved quantification of ice mélange properties to advance understanding of the glaciermélange-ocean system by: 1) creating a statistical model to describe how environmental forcing and geographic variables control ice mélange presence; and 2) pair ice mélange and environmental data with novel high-resolution numerical simulations of fjord circulation to investigate the relationship between ice mélange distribution and buoyancy-driven flow. 
 
By constraining environmental controls on ice mélange distribution and predicting the oceanic response to changes in ice mélange properties, this project directly addresses the NASA Earth Science Division’s goal to  improve the ability to predict climate changes by better understanding the roles and interactions of the oceans, atmosphere, land, and ice in the climate system. The remote sensing techniques used in this study will serve as a test bed for developing methods to conduct ice mélange characterization around the Greenland Ice Sheet. The results of our ocean circulation model will improve our understanding of processes transforming heat and freshwater from the Greenland Ice Sheet to the global ocean, providing more accurate spatiotemporal predictions of how glacier melt contributes to changes in global ocean circulation. In addition, these findings will provide insight into how enhanced exchange of fjord and shelf water could contribute to glacier destabilization. 
 
  
Jeffrey Thayer (PI)/Kevin Sacca (FI) 
University of Colorado, Boulder 
20-EARTH20-0095, Developing Uncertainty Quantification Methods for Space-Borne Lidar Surface Topography and Vegetation Applications 
 
In the Earth sciences field, lidar emerged as a breakout technology for the collection of highresolution topographic maps used in landscape change detection, emergency response, biomass inventory, wildfire fuel models, shallow water bathymetry, arctic ice monitoring, etc. from air- and space-borne platforms. The 2017-2027 Earth Science decadal survey calls for highresolution topography on a global scale, with many applications requiring high resolution at a high revisit cadence. In response, NASA organized an incubation group for Surface Topography and Vegetation (STV) to assess the state of the field and identify knowledge gaps and strategic technologies from subject matter experts and users of NASA data products. Earth observation satellite lidar systems such as ICESat-2/ATLAS and GEDI are uniquely qualified to fulfill a role in collecting topographic data with repeated global coverage for STV science applications. Spaceborne lidar instrumentation is challenged by a host of design, operational, and data processing obstacles that contribute to measurement uncertainty and must be addressed to be viable for high resolution topographic applications. Large errors in topographic lidar data products leads to inaccurate derived terrain maps, biomass estimates, and coastlines, which could have a negative impact on scientific research and disaster response efforts such as wildfire or hurricane management that rely on such data.  
 
Uncertainty quantification (UQ) of lidar measurements is a challenging problem, and the current methods commonly used to quantify lidar point position uncertainties are limited and often ignore coupling of input variables. The STV roadmap lists UQ as a future strategic technology development effort, however UQ methods have advanced rapidly and can be improved upon for specific applications such as lidar point cloud mapping. Modern UQ techniques seek to improve error estimation accuracy while reducing the total computational cost. Reducing the cost can be achieved with smarter sampling. In contrast to random sampling by traditional Monte Carlo techniques, stochastic collocation sampling techniques optimally sample high dimensional spaces with numerical quadrature relationships, resulting in orders of magnitude fewer samples to represent a random variable input. Increasing the accuracy of UQ techniques is accomplished using fewer assumptions and using uncertainty propagation models that include cross coupling of input parameters. Traditional methods such as Total Propagated Uncertainty (TPU) typically assume Gaussian random inputs and ignore cross correlation terms, while lidar signals are typically Poissonian and coupled inputs could have a significant effect on error estimates. Newer methods such as generalized Polynomial Chaos Expansion (gPCE) can accommodate any distribution, can incorporate cross coupled inputs, and produce high dimensional response surfaces that identify complex relationships between inputs and quantitatively assess the highest contributors of error from a system level. 
 
The proposed research seeks to advance the standards for lidar error analysis by applying modern UQ methods that simultaneously increase the fidelity of error estimates and reduce the cost of dynamically computing error for individual measurements. Like machine learning techniques, UQ can be applied to a diverse range of applications to improve error estimates of complex data products and in the case of lidar mapping, the error could provide an additional layer of contextual information to improve generative machine learning clustering and classification strategies. The ultimate goal is to advance the UQ standards used in the field of lidar to improve resulting data products with a long-lasting impact on the capabilities of spacebased lidar systems such as ICESat-2, GEDI, and future systems. 
 
  
Andrew Thompson (PI)/Yue Bai (FI) 
California Institute of Technology 
20-EARTH20-0314, Altimetry-Informed Variability in Southern Ocean Ventilation: Applications for SWOT and Surface Current Missions 
 
The Southern Ocean has an outsized impact on Earth s climate as it is the principal site where deep water is ventilated at the ocean surface and intermediate water masses, formed from airsea-ice interactions, are subducted into the interior.  Theories underpinning the dynamics of Southern Ocean ventilation largely assume that the flow properties of the Antarctic 
Circumpolar Current (ACC) are zonally-symmetric.  Yet, this is contrary to decades of evidence from satellite altimetry showing elevated regions of eddy kinetic energy localized within standing meanders arising from the interactions of the ACC with major topographic features.  Recent numerical studies have suggested that these standing meanders may also be preferred pathways for surface-interior exchange.  A systematic analysis of how the dynamical properties of these standing meanders have varied over time scales from seasons to decades has yet to be been carried out.  Therefore, building on previous work, we will apply a range of dynamical properties, diagnosed from the surface velocity field, to infer the statistical properties of smaller-scale (submesoscale) fronts and vertical velocities along energetic meanders and attempt to link these to Southern Ocean ventilation.  We believe that our project has the potential to unveil the sensitivity and importance of Southern Ocean ventilation to the global climate and Earth system evolution. 
 
Thus, we aim to develop and evaluate techniques that apply multiple remotely-sensed data products to identify coherent features associated with hot spots of ventilation in the Southern Ocean.  These techniques will evaluate spatial and temporal variability in ventilation rates as well as their impact on the global overturning circulation and oceanic heat and carbon dioxide uptake.  This proposal tests the hypothesis that ventilation is localized to  hot spots  of subduction and upwelling associated with an enhanced eddy energy field in the lee of major standing meanders of the ACC.  This work builds on new, promising evidence that dynamical features of the surface flow fields are closely related to interior properties   an idea tested via numerical simulations and a key focus of the NASA-funded S-MODE project.  Our work would extend the efforts of S-MODE to a climatically-relevant region of the global ocean.  
 
This project has three objectives.  First, we will determine a relationship between remotelysensed dynamical surface properties and vertical exchange between the surface and interior ocean.  This will involve the development and validation of mechanically-informed methods for identifying Southern Ocean ventilation hot spots.  Second, we will assess seasonal- to decadalscale variations in Southern Ocean surface dynamical properties and their impact on ventilation with a focus on identifying links to climate modes.  This objective will also explore the sensitivity of ventilation to surface forcings and modes of climate variability.  Finally, this project will study the link between surface ventilation and air-sea fluxes using a combination of numerical models and in situ data, including SOCCOM floats and S-MODE observations. 
 
We anticipate that the analysis techniques described here will support multiple NASA efforts.  Our analysis techniques will be directly applicable to future SWOT observations, leading to an improved mechanistic understanding of Southern Ocean ventilation.  Through involvement and collaboration with the EV-S S-MODE project, we will explore the applicability of our ventilation estimates to regions outside of the Southern Ocean.  Finally, we anticipate that our study will highlight how a future air-sea flux mission could improve long-term numerical climate predictions by observationally constraining surface heat and momentum fluxes. 
 
  
Mirela Tulbure (PI)/Vinicius Perin (FI) 
North Carolina State University 
20-EARTH20-0133, Quantifying On-Farm Reservoirs' Impacts on Surface Hydrology Using a Multi-Sensor Approach 
 
Fresh water stored by on-farm reservoirs (OFRs) is a fundamental component of surface hydrology and is critical for meeting global irrigation needs. Farmers use OFRs to store water during the wet season for crop irrigation during the dry season. There are more than 2.6 million OFRs in the US alone, and many of these OFRs were constructed during the last 40 years. Despite their importance for irrigating crops, OFRs can contribute to downstream water stress by decreasing stream discharge and peak flow in the watersheds where they are built, thereby exacerbating water stress intensified by climate change and population growth. However, modeling the impact of OFRs on surface hydrology remains a challenge because they are so abundant and have frequent fluctuations in surface area and water volume. Prior to the recent availability of satellite data, widespread monitoring of OFRs  surface area and water volume across space and time was impossible due to temporal latency of satellite observations. The goal of this project, therefore, is to harness a multi-sensor satellite imagery approach to reduce observation latency and improve surface hydrology modeling, with the aim of supporting more efficient management of OFRs and mitigation of their downstream impacts. Our objectives are: 1) Develop a multi-sensor imagery approach to reduce latency and obtain sub-weekly OFRs surface area and volume change; and 2) Input sub-weekly OFRs volume change into the Soil Water and Assessment Tool (SWAT) model to estimate OFRs  impact on surface hydrology. Specifically for Objective 1, a novel method based on the Kalman filter will be used to harmonize data from multiple optical sensors and to provide sub-weekly OFRs surface area change, which will be converted to volume change using area-elevation equations. Then for Objective 2, we will carry out hydrological simulations in SWAT to quantify OFRs  impact on simulated daily and monthly stream discharge, simulating stream discharge with and without the OFRs. We will perform yearly simulations, based on satellite imagery availability, to measure OFRs  impact during low and peak flows in each watershed of our study region, which will account for both intra- as well as inter-annual variability in flows. This project will monitor 
OFRs  surface area and volume change to enable better assessment and management of water quantity, and further the use of Earth system science to inform decisions and provide benefits to society regarding preservation of surface water resources, both of which are overarching science goals that guide NASA’s Earth Science Division program. 
 
  
Jun Wang (PI)/Meng Zhou (FI) 
University of Iowa, Iowa City 
20-EARTH20-0265, Nighttime Aerosol Optical Depth Retrieval from VIIRS DNB and Its Application for Surface PM2.5 Estimation 
 
The observation of nighttime aerosol conditions is important because of aerosol effects on radiative forcing, visibility, air quality, and public health, and is of high interest for improving the prediction of aerosol transport by complimenting the daytime aerosol observation. However, compared to the relatively abundant daytime measurements from space, there have been only very limited aerosol measurements at nighttime, and hence, the knowledge of nighttime aerosols is lacking and needs to be further investigated. By measuring visible light at night from space, the Visible Infrared Imaging Radiometer Suite Day/Night Band (VIIRS DNB) sensor onboard the Suomi National Polar-orbiting Partnership (Suomi-NPP) and National Oceanic and Atmospheric Administration (NOAA-20) satellites provide the research and operational communities the capability to explore the opportunity to study nighttime atmospheric optical and aerosol properties.  
 
Here, to quantitatively use the data from DNB for nighttime remote sensing of aerosols and provide air quality information for rural regions, we propose to (1) develop and add components to simulate Nighttime shortwave radiative transfer into the in-house Unified Linearized Vector Radiative Transfer Model (http://unl-vrtm.org), hereafter NRTM; (2) design the algorithm to retrieve AOD from VIIRS DNB radiances measured at night with a look-up table approach in which NRTM is used with consideration of DNB spectral response function and ancillary information about aerosol optical properties and surface reflectance;  (3) develop a neural-network-based algorithm to improve estimates of nighttime PM2.5 concentration by the combined use of the retrieved nighttime AOD and other ancillary data (such as GEOS-FP). The project will also emphasize the uncertainty and error analysis through sensitivity experiments and (either independent or cross) validation. The development of the NRTM is already completed and the model evaluations of the NRTM are promising. Preliminary results of nighttime AOD retrieval for the rural regions show the feasibility of using nighttime observation to derive AOD. The results of PM2.5 estimation using a backpropagation neural network justifies the promising potential of the proposed objectives.  
 
The study region of the proposed project is the rural areas of the continental U.S. where moonlight, in terms of spatial coverage, is the major nighttime illumination source, and the surface PM2.5 observation sites are sparse, both of which make a compelling case for the proposed objectives. 
 
  
Yuxuan Wang (PI)/Tabitha Lee (FI) 
University of Houston 
20-EARTH20-0185, Identifying Unreported NO2 Hotspots from Satellite Data and Quantifying Their Effects 
 
Nitrogen dioxides (NO2) are influential in chemical reactions that lead to the formation of tropospheric ozone (O3), a criteria pollutant harmful to human health, the nitrate radical (NO3), a part of fine particulate matter, and nitric acid (HNO3), an acid rain contributor. Yet, because of their abundant sources and short lifetime, not all emission events are indicated by bottomup emission inventories or seen by sparse in situ measurements. Satellite instruments (e.g., TROPOMI 3.6x5.6 km2 at nadir) have the potential of capturing unreported emission events through the observation of high NO2 signals.  
 
To understand the impacts of unreported NO2 hotspots, the proposed project aims to improve our ability to identify unreported signals in satellite observations using TROPOMI and provide an assessment of their influence on regional air quality. With the quantity of satellite-based NO2 data reaching an exceptional amount from spatial resolution improvements on polarorbiting instruments and upcoming geostationary instruments, it is imperative to evaluate and improve current hotspot identification techniques to increase their efficiency and accuracy when applied to large quantities of satellite observations. The proposal will focus on densitybased clustering algorithms for reproducibly detecting unreported NO2 hotspots in TROPOMI. The TROPOMI data will further be applied to estimate NOx emissions from unreported NO2 hotspots, and a chemical transport model (GEOS-Chem) will be used to simulate the impacts of these unreported emissions on regional air quality. We will first focus on the state of Texas as it has the potential for many unreported signals due to its diverse landscape where industrial, agricultural, and other processes occur, then we will move to other areas of the United States. The specific objectives are: 
1.	To evaluate current density-based clustering algorithms and implement improvements in the clustering detection strategies.  
2.	To apply the clustering method to the TROPOMI NO2 dataset to capture 'unreported' signals and determine signals' source origin.  
3.	To assess the influence of unreported hotspots on the tropospheric composition and air quality through top-down emission estimates and inclusion in a chemical transport model.  
 
With the improved understanding of unreported hotspots' influence on atmospheric composition, the proposed project will increase our understanding of what processes control the spatiotemporal structure of key atmospheric constituents and how these processes change the Earth system in overlooked locations where air quality standards may not be addressed. Therefore, the proposal is a direct response to the questions posed by the Earth Science 
Division "What cause these changes in the Earth system?", and question W-5 in the NASEM 2017 Decadal Strategy for Earth Observation from Space "What processes determine the spatiotemporal structure of important air pollutants and their concomitant adverse impact on human health, agriculture, and ecosystems?". 
 
  
Mark Zondlo (PI)/Daniel Moore (FI) Princeton University 
20-EARTH20-0290, Characterization of Reactive Nitrogen Emissions from Agriculture 
 
Ammonia (NH3), the primary base in the atmosphere emitted primarily through volatilization during agricultural activity, reacts to form particles that can be detrimental to air quality. Nitrogen dioxide (NO2) is a product of the nitrification/denitrification process in agricultural soils, and can lead to ozone formation, a key atmospheric pollutant, in the presence of volatile organic compounds (VOCs) and sunlight. Moreover, the deposition of reactive Nitrogen (Nr), mainly in the form of NH3 and nitrogen oxides (NOx), can negatively impact ecosystems downwind of source regions. Current knowledge of the spatiotemporal variability of the agricultural emissions of these important trace gases is lacking. Where surface observations are temporally and spatially sparse, recent strides in satellite observations allow for more frequent measurements and greater coverage of concentrations. Additionally, with atmospheric lifetimes on the order of hours to a day, daily satellite observations of these trace gas species can be used for general source determination at the daily scale during large events. 
 
I propose to employ remotely sensed data from the CrIS and IASI satellite-based instruments for NH3, and OMI, TROPOMI and future TEMPO data for NO2. I will analyze daily column abundances to characterize regional spatial and temporal scales of springtime agricultural emissions across the United States, including regional differences between agricultural activity types (i.e. livestock husbandry, heavily fertilized cropland). Through this work, I will develop a climatology (i.e. timing, frequency and spatial extent) of these intermittent, high-magnitude events, or 'blooms', and analyze them in the context of observed environmental control parameters, such as soil moisture, temperature and relative humidity from reanalysis data and merged satellite products to investigate the processes that govern these emissions. Specifically, volatilization of nitrogen fertilizer inputs to early-growth crops is sensitive to these environmental variables, and previous work has established important relationships between rainfall and NO2 enhancements from agriculture. I will also examine temporal relationships between the timing of emission events, which are a key signature of fertilizer application, and satellite-based crop health parameters, such as leaf area index (LAI) and the fraction of absorbed photosynthetically active radiation (FAPAR), on a yearly basis. Finally, I will examine individual events to determine the spatial coherence of NH3 and NO2 emissions and develop emission factors at the event-scale. Overall, this project aims to improve scientific understanding around emissions of these important trace gases, which are critical to regional nitrogen budgets, as well as the processes that govern them. Doing so is fundamental to establishing agriculture practices with a focus on food security and sustainability.