Michael Keller, USDA Forest Service, mkeller.co2@gmail.com
Paul Duffy, Neptune and Company, Inc., paul.duffy@neptuneinc.org
Douglas Morton, NASA GSFC, douglas.morton@nasa.gov
David Schimel, Jet Propulsion Laboratory, david.schimel@jpl.nasa.gov
Maiza dos Santos, EMBRAPA-CNPM, maizanara@gmail.com
Ekena Pinagé, EMBRAPA-CNPM, ekenapinage@hotmail.com

Carbon accounting for REDD+ requires knowledge of deforestation, degradation, and associated changes in forest carbon stocks. Degradation is more difficult to detect than deforestation so SilvaCarbon, an US inter-agency effort, has set a priority to better characterize forest degradation effects on carbon loss. We have initiated efforts to quantify carbon stocks and changes and associated uncertainties in Paragominas, a jurisdiction in the eastern Brazilian Amazon with a high proportion of logged and burned degraded forests where political change has opened the way for REDD+. We build on a long history of research including our extensive studies of logging damage. Our study currently uses recent forest inventories and airborne lidar and will be extended in the future to include Landsat remote sensing that will incorporate a novel use of time series data to estimate the structural properties of degraded forests using bidirectional reflectance information. We identify two objectives for forest carbon accounting at the jurisdictional level: * Quantify spatially explicit above-ground carbon stocks and the changes in carbon stocks; * Quantify spatially explicit uncertainties in above-ground carbon stocks and changes in carbon stocks We are working to meet these objectives by employing innovative data assimilation methods. Our approach employs a hierarchical Bayesian modeling (HBM) framework where the assimilation of information from multiple sources is accomplished using a change of support (COS) technique. The COS problem formulation allows data from several spatial resolutions to be assimilated into an intermediate resolution. This approach provides a mechanism to assimilate information from multiple sources to produce spatially-explicit maps of carbon stocks and changes with corresponding spatially explicit maps of uncertainty. Importantly, this approach also provides a mechanism that can be used to assess the value of information from specific data products. Hence future data collection can be optimized in the context of the reduction of uncertainty. The spatially explicit quantification of uncertainties naturally provides insights into effective sampling designs.

Associated Project: 2013 - Keller (CMS 2013) - Michael Keller

Warren Cohen, USDA Forest Service, warren.cohen@oregonstate.edu
Hans Andersen, USDA Forest Service, handersen@fs.fed.us
Sean Healey, USDA Forest Service, seanhealey@fs.fed.us
Gretchen Moisen, US Forest Service, gmoisen@fs.fed.us
Christopher Woodall, USDA Forest Service, cwoodall@fs.fed.us
Grant Domke, USDA Forest Service, gmdomke@fs.fed.us
Zhiqiang Yang, Oregon State University, zhiqiang.yang@oregonstate.edu
Stephen Stehman, SUNY College of Environ Sci & Forestry, svstehma@syr.edu
Robert Kennedy, Oregon State University, rkennedy@coas.oregonstate.edu
Curtis Woodcock, Boston University, curtis@bu.edu
Zhe Zhu, Boston University, zhuzhe@bu.edu
Todd Schroeder, US Forest Service, tschroeder@fs.fed.us
James Vogelmann, USGS Sioux Falls, vogel@usgs.gov
Daniel Steinwand, U.S. Geological Survey/EROS, steinwand@usgs.gov
Chengquan Huang, University of Maryland, cqhuang@umd.edu

We are developing a REDD MRV system that tests different biomass estimation frameworks and components. Design-based inference from a costly field plot network is compared to sampling with lidar strips and a smaller set of plots in combination with Landsat for disturbance monitoring. Biomass estimation uncertainties associated with these different datasets in a design-based inference framework is examined. We are also testing estimators that rely primarily on Landsat within a model-based inference framework. Contributions from Landsat are current (e.g., 2014) spectral response and metrics describing disturbance history derived from a time series leading up to the current date. An advantage of the model-based framework is its extension back in time (e.g., to 1990) using a consistent approach based on disturbance history as an indicator of biomass density. This requires use of the older, MSS archive to be fully effective in estimating biomass for the 1990 baseline. The US, while not a REDD country, is party to the UNFCCC and has a need for similar NGHGI baseline information. The various components of our MRV system will be tested in the US, where the best data are available for parsing the uncertainty contributions of the several system components we are testing.

Associated Project: 2013 - Cohen (CMS 2013) - Warren Cohen

Steven Lohrenz, University of Massachusetts Dartmouth, slohrenz@umassd.edu
Hanqin Tian, Auburn University, tianhan@auburn.edu
Wei-Jun Cai, University of Delaware, wcai@udel.edu
Ruoying He, North Carolina State University, rhe@ncsu.edu
Shufen Pan, Auburn University, panshuf@auburn.edu

Our prior NASA-funded research employs a combination of models and remotely-sensed and in situ observations to develop georeferenced products and associated uncertainties for land-ocean exchange of carbon, air-sea exchanges of carbon dioxide, and coastal to open ocean exchanges of carbon. The primary region of study has been the Mississippi River watershed and northern Gulf of Mexico. Our approach involves using coupled models of terrestrial and ocean ecosystem dynamics and associated carbon processes to assess how societal and human-related land use, land use change and forestry (LULUCF) and climate-related change affect terrestrial carbon storage and fluxes, as well as export of materials through watersheds to the coastal margins. Here, we propose to extend the domain of our observational and integrated terrestrial-ocean ecosystem model system to include the southeastern U.S. and South Atlantic Bight. In addition to land-ocean and sea-atmosphere exchanges, we will utilize satellite observations together with the capabilities of the terrestrial ecosystem model to characterize and quantify terrestrial carbon storage and fluxes, including land-atmosphere fluxes of both carbon dioxide and methane. Our approach will include assembling model products along with associated uncertainties and errors in a geospatial framework that will facilitate decision support for carbon and land use management. Objectives of the proposed research include the following: 1) Expand the spatial domain of our observational and integrated modeling approach to include the Mississippi River basin and southeastern U.S., and examine terrestrial carbon storage and fluxes including characterization and quantification of biomass and carbon stocks in and land-atmosphere, land-ocean, and sea-atmosphere fluxes of carbon dioxide and methane; 2) Examine different LULUCF scenarios within the terrestrial domain and different climate scenarios to assess effectiveness of carbon management strategies; 3) Engage with other CMS projects and stakeholders (e.g., USDA, National Climate Assessment, etc.) to identify user needs related to carbon management and MRV activities, modify and expand the scope of information based on user feedback, and explore possible transition of prototype products to fully operational status.

Associated Project: 2014 - Lohrenz (CMS 2014) - Steven Lohrenz

Julie Wolf, JGCRI, Julie.Wolf@pnnl.gov
Yannick le Page, JGCRI, yannick.lepage@pnnl.gov
Tristram West, White House Council on Environmental Quality, tristram_o_west@ceq.eop.gov

Quantification of biogenic carbon fluxes from agricultural lands is needed to generate comprehensive bottom-up estimates of net carbon exchange for global and regional carbon monitoring. We quantify global agricultural carbon fluxes associated with annual crop net primary production (NPP), harvested biomass, consumption of biomass by humans and livestock, and net carbon exchange (NCE), and distribute them to 0.05 degree resolution. In 2011, global crop NPP was 5.25 ± 0.46 Pg carbon, of which 2.05 ± 0.05 Pg carbon was harvested. In addition to primary crop harvests, 0.54 Pg carbon was collected from crop residues for livestock fodder. In 2011, total livestock feed intake was 2.42 ± 0.21 Pg carbon, of which 2.31 ± 0.21 Pg carbon was emitted as CO2 and 0.07 ± 0.01 Pg carbon was emitted as CH4. Livestock grazed an estimated 1.27 Pg carbon of forage in 2011, representing 52.4% of total livestock feed intake. Global human food intake was 0.57 ± 0.03 Pg carbon in 2011, with an additional 0.26 Pg carbon lost or wasted from the food supply chain. A completed global agricultural carbon budget for 2009 accounts for the ultimate use of 79.6% of the 2009 harvest, and indicates that 5.01 Pg carbon was taken up by crop plants, 2.56 Pg was decomposed on-field, and 2.45 Pg was harvested and emitted elsewhere. The spatial distribution of these fluxes may be used for carbon monitoring purposes, to understand uncertainty around agricultural flux estimates, and for use in Earth system models.

Associated Project: 2012 - West (CMS 2011) - Tristram West

Steven Lohrenz, University of Massachusetts Dartmouth, slohrenz@umassd.edu
Wei-Jun Cai, University of Delaware, wcai@udel.edu
Hanqin Tian, Auburn University, tianhan@auburn.edu
Ruoying He, North Carolina State University, rhe@ncsu.edu
Stephan Howden, University of Southern Mississippi, stephan.howden@usm.edu

This research has employed a combination of models and remotely-sensed and in situ observations to develop georeferenced products and associated uncertainties for land-ocean exchange of carbon, air-sea exchanges of carbon dioxide, and coastal to open ocean exchanges of carbon. Such information is critically needed to better constrain the contribution of coastal margins to carbon sources and sinks and improve capabilities to attribute sources and sinks to different regions as well as reducing uncertainties in estimates. A major aspect of this project has involved establishing and populating geospatial portals for sharing and analysis of carbon data sets and products. The primary region of study is the Mississippi River watershed and northern Gulf of Mexico. However, the model domain also includes the continental margins of Florida and the South Atlantic Bight. The work is closely aligned with objectives of the NASA Carbon Monitoring System scoping effort and of the North American Carbon Program and will support National Climate Assessment activities. The effort also contributes to NASA Coastal Carbon Synthesis effort and international efforts to develop a North American carbon budget (CarboNA). The unique nature of our approach, coupling models of terrestrial and ocean ecosystem dynamics and associated carbon processes, will allow for assessment of how societal and human-related land cover and land use changes, as well as climate change, affect terrestrial carbon sources and sinks, export of materials to coastal margins, and associated carbon processes in the continental margins. Results will also benefit efforts to describe and predict how land cover and land use changes impact coastal water quality, including possible effects of coastal eutrophication, hypoxia, and ocean acidification.

Associated Project: 2012 - Lohrenz (CMS 2011) - Steven Lohrenz

Wayne Walker, WHRC, wwalker@whrc.org
Alessandro Baccini, Woods Hole Research Center, abaccini@whrc.org
Fabio Goncalves, Woods Hole Research Center, fgoncalves@whrc.org
Curtis Woodcock, Boston University, curtis@bu.edu
Luis Carvalho, Boston University, lecarval@math.bu.edu
Alicia Peduzzi, US Forest Service, apeduzzi@fs.fed.edu
Javier Corral Rivas, Universidad Juárez del Estado de Durango, jcorral@ujed.mx
Carlos Lopez Sanchez, Universidad Juárez del Estado de Durango, calopez@ujed.mx

In response to the high uncertainties associated with traditional approaches to forest carbon accounting, this project seeks to investigate the potential for annual changes in the aboveground carbon density (ACD) of forests to be estimated directly, consistently, and with measurable accuracy across large areas using an array of existing commercial off-the-shelf and NASA remote sensing assets. The geographic focus is the country of Mexico where members of the project team have been working closely with the Mexican government as part of the USAID-supported M-REDD+ project to assist in advancing Mexico’s forest monitoring capacity. The specific objectives focus on quantifying the certainty with which extensive field, off-the-shelf airborne LiDAR (G-LiHT and M-REDD+), and NASA satellite data sources (MODIS, VIIRS, and Landsat) can be used synergistically to estimate wall-to-wall changes in ACD at varying resolutions (i.e., 500 - 30 m) across five Mexican states (Chihuahua, Oaxaca, Campeche, Yucatan, and Quintana Roo) over a ~15-year period (2001-2015). An independent accuracy assessment of the ACD change products will be conducted leveraging permanent plot data from the Mexico National Inventory of Forest and Soil (INFyS), intensive field and micrometeorological measurements from the Mexico network of eddy covariance flux towers (MexFlux), and deforestation data from Hansen et al. (2013). The anticipated results of this project represent a fundamentally new way of quantifying carbon fluxes that will significantly reduce uncertainty while leading to a more complete understanding of terrestrial carbon cycling. Unlike conventional approaches, which focus on deforested areas leaving degradation unaccounted for, the proposed approach provides for a unique estimate of gross emissions at the pixel level, integrating losses attributable to deforestation, degradation, and other forms of disturbance with gains attributable to growth.

Associated Project: 2014 - Walker (CMS 2014) - Wayne Walker

Mark Cochrane, South Dakota State University/GIScCE, mark.cochrane@sdstate.edu
Bambang Saharjo, Bogor Agricultural University, bhherosaharjo@gmail.com
Robert Yokelson, University of Montana, Bob.Yokelson@mso.umt.edu
Grahame Applegate, Consultant, grahame.applegate@gmail.com
Andrew Vayda, Consultant, apvayda@gmail.com
Sulistyo Siran, Consultant, sulistyo.siran@yahoo.com
TImothy Jessup, Consultant, tcjessup@gmail.com
Kevin Ryan, Consultant, kryan.wildland.fire@gmail.com

In Indonesia, drained peat-swamp forests, with deep organic soils that store vast amounts of carbon, are now being lost to decomposition and combustion. Episodic uncontrolled fires within these areas have contributed to the country ranking as the 3rd largest CO2 emitter in recent decades. We are developing a prototype peat-fire emissions module for the Indonesian Carbon Accounting System (INCAS) to incorporate into the nation’s Measuring, Reporting and Verification (MRV) efforts. The project leverages off of historical field data collection from the former Kalimantan Forest and Climate Partnership (KFCP) between the Australian and Indonesian governments, continuing and augmenting extensive collection efforts for hydrology, fuels, land uses and fire occurrence in the 120,000ha study area. We utilize Landsat, MODIS and TRMM data and products to quantify land cover changes, burned area, estimate the timing of fire activity, and relate precipitation history to observed water table changes that impact peat-fire activity. Ongoing Lidar data collection efforts (2014) will be integrated with existing aerial KFCP Lidar (2007 & 2010) to provide quantified temporal topographic change maps to validate modeled amounts of fire-related peat consumption. Modeling this phenomenon requires defining 1) the annual surface area burned, 2) the available fuel fraction (burnable) at each location through time, and 3) the amount of fuel consumed per unit area. We will provide Monte Carlo estimates of type, depth, and area of burning, with emissions quantitatively weighted by appropriate emission factors (EFs) derived for surface, shallow and deep peat smoke amounts, validated using the 2014 Lidar data collection. In 2015, we will conduct detailed emissions field sampling of in-situ smoldering surface, shallow (<20 cm) and deep (>20 cm) peat fires, with on-site gas chromatography for quantifying reactive species, whole air sampling for precise lab measurements of non-reactive gases, and simultaneous filter sampling of particulates.

Associated Project: 2013 - Cochrane (CMS 2013) - Mark Cochrane

Steven Pawson, NASA GMAO, steven.pawson@nasa.gov
Lesley Ott, NASA GSFC, lesley.e.ott@nasa.gov
David Baker, CIRA Colorado State University, dfbaker66@gmail.com
George Collatz, NASA GSFC, jim.collatz@nasa.gov
Janusz Eluszkiewicz, Atmospheric and Environmental Research, jel@aer.com
Watson Gregg, NASA GSFC, watson.w.gregg@nasa.gov
Stephan Kawa, NASA GSFC, stephan.r.kawa@nasa.gov
Thomas Nehrkorn, AER, Inc, tnehrkor@aer.com
Tomohiro Oda, USRA at NASA/GSFC, tomohiro.oda@nasa.gov
Christopher O'Dell, Colorado State University, odell@atmos.colostate.edu
Cecile Rousseaux, GMAO/USRA, cecile.s.rousseaux@nasa.gov
Andrew Schuh, Colorado State University, aschuh@kiwi.atmos.colostate.edu
Brad Weir, NASA GSFC & USRA, brad.weir@nasa.gov

This poster gives an overview of the work performed during the CMS Phase 2 GEOS-Carb project. GEOS-Carb continued NASA GSFC's contributions to CMS which began as part of the CMS Flux Pilot Project and included advances toward an integrated carbon modeling system. We include an overview of the modeling system as well as highlights of results of several components.

Associated Project: 2012 - Pawson (CMS 2011) - Steven Pawson

Katherine Haynes, Colorado State University, kdhaynes@atmos.colostate.edu
Ian Baker, Colorado State University, baker@atmos.colostate.edu
Scott Denning, Colorado State University, denning@atmos.colostate.edu

We simulate fully self-consistent global carbon flux and biomass pools using the latest version of the Simple Biosphere Model (SiB4). SiB4 is an enzyme-kinetic/process-based model that simulates biophysical mechanisms important for surface-atmosphere exchange. Environmental cues and simulated photosynthesis determine phenology, and carbon uptake is allocated into multiple above- and belowground live and dead pools; transfer coefficients determine exchange between pools, and mechanistic respiration ‘closes the loop’ for a fully self-consistent carbon cycle. SiB4 diagnostics are evaluated against eddy covariance flux and site-based observations across multiple Plant Functional Types (PFTs). We also confront the model with spectrally-based observations of Leaf Area Index (LAI), fraction of Photosynthetically Active Radiation absorbed (fPAR) and Solar Induced Fluorescence (SIF) for evaluation across broad spatiotemporal scales. We can also isolate the photosynthesis component of the carbon cycle using evaluations of surface uptake of carbonyl sulfide (OCS). SiB4 simulates reasonable estimates of global Gross Primary Productivity (GPP) and global aboveground biomass. Seasonality in mid- and high-latitudes is reproduced. Biomass in tropical forests is reproduced, but simulated seasonality in carbon flux conflicts with some observations. Some regions of tropical savanna are overproductive, but overall SiB4 shows excellent fidelity when evaluated across multiple observational platforms.

Associated Project: 2012 - Bowman (CMS 2011) - Kevin Bowman

Daniel Jacob, Harvard, djacob@fas.harvard.edu

We use satellite observations to improve knowledge of methane fluxes, with particular focus on North American anthropogenic emissions and on wetlands. Satellite observations of atmospheric methane (SCIAMACHY, GOSAT, TROPOMI) are a unique resource for constraining methane fluxes in combination with suborbital observations and state-of-science bottom-up inventories. We use advanced inverse methods based on the GEOS-Chem chemical transport model and its adjoint to exploit the power of satellite observations for constraining methane fluxes with unprecedented spatial resolution and full error characterization. We also produce a new gridded version of the US EPA national inventory, and a new gridded global inventory of wetland emissions based on satellite inundation data and the MsTMIP terrestrial ecosystem model ensemble. These gridded bottom-up inventories including uncertainties serve as a priori in our inversions, and correction factors from the inversions can then be used to guide inventory improvements. We conduct inverse analyses using the full GOSAT record from 2009 through present. Our work with GOSAT will put us in position to exploit the TROPOMI data as soon as they are available.

Associated Project: 2012 - Jacob (CMS 2011) - Daniel Jacob

George Collatz, NASA GSFC, jim.collatz@nasa.gov
Stephan Kawa, NASA GSFC, stephan.r.kawa@nasa.gov
Lesley Ott, NASA GSFC, lesley.e.ott@nasa.gov
Alvaro Ivanoff, NASA GSFC/ADNET, alvaro.ivanoff-1@nasa.gov
Fanwei Zeng, NASA GSFC/SSAI, fanwei.zeng@nasa.gov
Yuping Liu, NASA GSFC/SSAI, yuping.liu-1@nasa.gov

The focus of this NASA CMS project is to provide land surface biological and fire fluxes as boundary conditions for atmospheric transport models. We use the CASA-GFED3 model driven by MERRA meteorology and GIMMS3g NDVI. Physiological (NPP, RH) and fire fluxes are simulated at monthly, 0.5o resolutions. Physiological fluxes are further decomposed in time to 3 hourly by scaling with 3 hourly solar radiation and temperature. Fire fluxes are scaled to quasi-daily using a MODIS active fire product. Our flux products span 2003-2011 with recent updates for 2012 and 2013. The monthly (0.5o) and 3 hourly (1ox1.25o) fluxes are available at the CMS and NACP websites (http://nacp-files.nacarbon.org/nacp-kawa-01/?C=M;O=D). Model output also includes biomass and detrital carbon pools. The impacts of uncertainties in model parameters on fluxes and carbon pools are estimated by selecting values from probability distribution functions for key parameter and propagating these in Monte Carlo ensemble simulations. Our CMS products are used by a number of other CMS projects (e.g. Bowman, Ott, French). Model products are evaluated against atmospheric CO2 observations, atmospheric inversion fluxes, eddy covariance fluxes, and independent estimates of biomass. Comparisons with atmospheric CO2 measurements (flask, continuous, TCCON, GOSAT) show the model is good at matching the phase and amplitude of the measured CO2 seasonal cycle but is less successful at capturing interannual variability. The baseline model overestimates the atmospheric CO2 growth rate because it lacks representation of long term land sink mechanisms (e.g. CO2 fertilization). We have also developed model setups that are forced to reproduce the magnitude of the observed global land sink. Baseline modeled above ground biomass is within the uncertainties of independent estimates though biased low in northern latitudes. These and future evaluations provide bases for improving the model.

Associated Project: 2013 - Collatz (CMS 2013) - George Collatz

Robert Kennedy, Oregon State University, rkennedy@coas.oregonstate.edu
Neeti Neeti, Woods Hole Research Center, neeti@whrc.org
Matthew Gregory, Oregon State University, matt.gregory@oregonstate.edu
Heather Roberts, Oregon State University, heather.roberts@oregonstate.edu

For the NASA Carbon Monitoring System (NASA CMS) to function as an integrated framework, we must learn from agreement and disagreement among products. Here, we describe spatial and temporal strategies to compare and integrate biomass products at national and regional scales. At the national scale, we document geographic methods to compare among the four currently-available maps of aboveground biomass, and show how spatial patterns of agreement and disagreement can lead to insight into strengths and weaknesses of the different approaches. At the regional scale, we are integrating Landsat time series and FIA plot data to produce yearly estimates of biomass with associated uncertainties for every year for the forested areas of Washington, Oregon, and California. These yearly estimates of biomass can act as a common currency to compare expected differences between the national maps, which were produce for different epochs, and further distinguish forest types where maps agree or disagree.

Associated Project: 2012 - Kennedy (CMS 2011) - Robert Kennedy

Manvendra Dubey, Los Alamos National Laboratory, dubey@lanl.gov
Zachary Butterfield, Los Alamos National Laboratory, zbutterfield24@gmail.com
Rodica Lindenmaier, PNNL, rodica.lindenmaier@pnnl.gov
Jacob Hedelius, Caltech, jhedeliu@caltech.edu
Debra Wunch, Caltech, dwunch@caltech.edu
Paul Wennberg, Caltech, wennberg@gps.caltech.edu
Harrison Parker, Los Alamos National Laboratory, hparker@lanl.gov
Frank Hase, Karlsruhe Institute of Technology, frank.hase@kit.edu

Recent studies demonstrate that satellite and ground based observations of column CH4 and CO2 can be used to verify their fluxes to correct bottom up inventories. Ground based Total Column Carbon Observing Network (TCCON) measurements and satellite observations (SCIAMACHY, GOSAT and OCO-2) are providing data to achieve verification. Unfortunately due to the large size, high cost and complexity of TCCON that uses a high resolution solar Fourier Transform Spectrometer (Bruker 125HR) its coverage is confined to the developed world. Our goal is to fill gaps in the TCCON in the developing regions in Asia, South America and Africa. To achieve this we acquired the first commercial low resolution solar FTS with an inbuilt camtracker (EM27/SUN). The EM27 is compact, inexpensive, mobile and easy to operate. We compared the performance of the EM27 in side-by-side deployments at 3 TCCON (125HR) sites their standard retrievals. They included a power plant (Four Corners, NM), urban (Caltech), remote mountain (Los Alamos, NM) and coastal (Armstrong, CA). Comparisons of changes in diurnal CH4 and CO2 the EM27 with the 125HR TCCON are good and promising in a diverse set of environments. However, there is a 20 ppb high bias in EM27. Our system was calibrated in the factory in Germany and has experienced international shipment. It has been driven extensively from Los Alamos to Four Corners and Los Angeles for 4 campaigns over a period of 1 year. Given this the EM27 appears to have good stability. However, more careful analysis of the Instrument Line Shape and its long-term stability are needed and calibration and operational protocols are being developed at KIT as part of the proposed Collaborative Column Carbon Observing Network (COCCON). We plan to perform campaigns at other TCCON sites in Lamont OK, Park Falls WI and possibly Eureka, Canada in 2015.

Associated Project: 2013 - Dubey (CMS 2013) - Manvendra Dubey

Thomas Nehrkorn, AER, Inc, tnehrkor@aer.com
Steven Wofsy, Harvard University, wofsy@fas.harvard.edu
Lucy Hutyra, Boston University, lrhutyra@bu.edu
Bill Callahan, Earth Networks, Inc., BCallahan@earthnetworks.com
Philip DeCola, Sigma Space Corp., pdecola@sigmaspace.com
George Collatz, NASA GSFC, jim.collatz@nasa.gov
Charles Miller, NASA JPL, charles.e.miller@jpl.nasa.gov
Crystal Schaaf, University of Massachusetts Boston, Crystal.Schaaf@umb.edu
Marikate Mountain, AER, Inc, mmountai@aer.com
Kathryn McKain, Harvard University, kmckain@fas.harvard.edu
Maryann Sargent, Harvard University, mracine@fas.harvard.edu
Yanina Barrera, Harvard University, 1topcheme@gmail.com
Brady Hardiman, Boston University, brady.hardiman@gmail.com
Conor Gately, Boston University, cgately@gmail.com
Amanda Long, Earth Networks, Inc., along@earthnetworks.com
Christopher Sloop, Earth Networks, Inc., cdsloop@aws.com
Steve Prinzivalli, Earth Networks, Inc., sprinzivalli@earthnetworks.com
Taylor Jones, Harvard University, taylorjones@g.harvard.edu

The world's population growth is increasingly concentrated in urban areas and this trend is expected to continue in the future. Urbanization has a profound impact on carbon dynamics, leading to increases in anthropogenic carbon dioxide (CO2) emissions and decreases in biogenic fluxes from these areas. We present Year 1 results from a measurement network and an accompanying atmospheric modeling framework for downscaling the current NASA CMS flux products to regional and local scales. These spatially and temporally resolved estimates of biogenic and anthropogenic fluxes are central to meeting greenhouse gas emissions reductions goals, and they complement Monitoring, Reporting, and Verification (MRV). Our research focuses on the Northeast corridor (Boston MA - Washington DC megalopolis), where about 17% of the U.S. population lives on less than 2% of the nation's land area, making it a key source of US anthropogenic CO2 emissions. The research has progressed along three main lines: 1) High-resolution transport modeling (WRF-STILT) customized and verified for the region, 2) High-resolution CO2 flux model incorporating anthropogenic emissions estimates and the CASA model (including its 0.5-deg resolution variant that provides the foundational biosphere model for the current CMS Flux Product and nested higher resolution runs to represent the scale sensitivity within heterogeneous urban areas), and 3) Inverse CO2 flux estimates corresponding to in-situ and remote CO2 observations in and around Boston, New York City, and Washington DC. During the first year we have completed baseline WRF simulations, analyzed sensitivity experiments for varying WRF configurations, and deployed two mini Micro Pulse LiDARs for remotely sensing planetary boundary layer characteristics; quality-controlled and cross-calibrated CO2 measurements from the Boston and Earth Networks measurement sites; generated high-resolution a priori biospheric and anthropogenic flux estimates; and tested aspects of the inversion framework using data from a methane study for Boston.

Associated Project: 2013 - Nehrkorn (CMS 2013) - Thomas Nehrkorn

Marc Fischer, Lawrence Berkley National Laboratory, mlfischer@lbl.gov
Heather Graven, Imperial College London, h.graven@imperial.ac.uk
Ralph Keeling, UCSD Scripps Institution of Oceanography, rkeeling@ucsd.edu
Christian Frankenberg, Jet Propulsion Laboratory / Caltech, christian.frankenberg@jpl.nasa.gov
Tom Guilderson, Center for AMS, tguilderson@llnl.gov
Seongeun Jeong, Lawrence Berkeley National Lab, sjeong@lbl.gov
Nick Parazoo, UCLA, nicholas.c.parazoo@jpl.nasa.gov
Bill Callahan, Earth Networks, BCallahan@earthnetworks.com
Ying Hsu, California Air Resources Board, yhsu@arb.ca.gov
John Miller, NOAA Earth System Research Laboratory, John.B.Miller@noaa.gov
Tim Lueker, UCSD Scripps Institution of Oceanography, tlueker@ucsd.edu
Sally Newman, Caltech, sally@gps.caltech.edu
Brian LaFranchi, Sandia National Laboratory, bwlafra@sandia.gov
Hope Michelson, Sandia National Laboratory, hamiche@sandia.gov
Ray Bambha, Sandia National Laboratory, rpbambh@sandia.gov

We report progress on the development of a prototype data collection and analysis system to leverage remote sensed data from the OCO-2 satellite and provide regional fossil and biosphere CO2 fluxes and state-annual evaluation of fossil CO2 emissions inventories in California. In the first year of the project, we have developed and deployed a flask sampling network encompassing 10 collaborative tower measurement sites across California. We performed a field campaign to collect air samples every three days for the month of May, 2014 and started a second campaign covering October 15 to November 14, 2014. We also began a pseudo-data modeling experiment that evaluates the potential to estimate fossil CO2 emissions using only radiocarbon-derived fossil fuel CO2 concentration measured in flask samples, and later the potential to estimate both fossil and biosphere CO2 fluxes using the OCO-2 column CO2 retrievals supplemented by flask sampling and in situ total CO2 measurements. Initial results from the flask-only pseudo-data experiment suggest that sub-weekly sampling at the towers included in this experiment might be expected to yield a posterior uncertainty in monthly mean state-total fossil fuel CO2 emissions comparable to or better than those provided by bottom-up inventory approaches.

Associated Project: 2013 - Graven (CMS 2013) - Heather Graven

Sangram Ganguly, NASA ARC BAERI, sangramganguly@gmail.com
Saikat Basu, Louisiana State University, sbasu8@tigers.lsu.edu
Ramakrishna Nemani, NASA ARC, rama.nemani@nasa.gov
Cristina Milesi, NASA ARC, Cristina.Milesi-1@nasa.gov
Supratik Mukhopadhyay, Louisiana State University, supratik@csc.lsu.edu
Andrew Michaelis, NASA ARC UCMB, amac@hyperplane.org
Petr Votava, NASA ARC UCMB, petr.votava-1@nasa.gov
Sassan Saatchi, NASA JPL, sasan.s.saatchi@jpl.nasa.gov

Several studies to date have provided an extensive knowledge base for estimating forest aboveground biomass (AGB) and recent advances in space-based modeling of the 3-D canopy structure, combined with canopy reflectance measured by passive optical sensors and radar backscatter, are providing improved satellite-derived AGB density mapping for large scale carbon monitoring applications. A key limitation in forest AGB estimation from remote sensing, however, is the large uncertainty in forest cover estimates from the coarse-to-medium resolution satellite-derived land cover maps (present resolution is limited to 30-m of the USGS NLCD Program). As part of our CMS Phase II activities, we have demonstrated the use of Landsat-based estimates of Leaf Area Index and ICESat Geoscience Laser Altimeter System (GLAS) derived canopy heights for estimating AGB at a 30-m spatial resolution, which compare relatively well with inventory based plot level estimates. However, uncertainties in forest cover estimates at the Landsat scale result in high uncertainties for AGB estimation, predominantly in heterogeneous forest and urban landscapes. We have successfully tested an approach using a machine learning algorithm and High-Performance-Computing with NAIP air-borne imagery data for mapping tree cover at 1-m over California. In a comparison with high resolution LiDAR data available over selected regions in the two states, we found our results to be promising both in terms of accuracy as well as our ability to scale nationally. In this project, we propose to estimate forest cover for the continental US at spatial resolution of 1-m in support of reducing uncertainties in the AGB estimation. The generated 1-m forest cover map will be aggregated to the Landsat spatial grid to demonstrate differences in AGB estimates (pixel-level AGB density, total AGB at aggregated scales like ecoregions and counties) when using a native 30-m forest cover map versus a 30-m map derived from a higher resolution dataset. The process will also be complemented with a LiDAR derived AGB estimate at the 30-m scale to aid in true validation. The proposed work will substantially contribute to filling the gaps in ongoing NASA CMS research and help quantifying the errors and uncertainties in NASA CMS products.

Associated Project: 2012 - Saatchi (CMS 2011) - Sassan Saatchi

Arlyn Andrews, NOAA Earth System Research Laboratory, Arlyn.Andrews@noaa.gov
John Miller, NOAA Earth System Research Laboratory, John.B.Miller@noaa.gov
Kirk Thoning, NOAA Earth System Research Laboratory, Kirk.W.Thoning@noaa.gov
Marikate Mountain, AER, Inc., mmountai@aer.com
Thomas Nehrkorn, AER, Inc., tnehrkor@aer.com
Anna Michalak, Carnegie Institution for Science and Stanford University, michalak@carnegiescience.edu
Vineet Yadav, Stanford University, vineety@stanford.edu
Christopher O'Dell, Colorado State University, odell@atmos.colostate.edu
Chris Sloop, Earth Networks, Inc., csloop@earthnetworks.com

Two CMS-2012 projects, “North American Regional-Scale Flux Estimation and Observing System Design for the NASA Carbon Monitoring System” (A. Andrews, PI) and “In situ CO2-based evaluation of the Carbon Monitoring System flux product” (J. Miller, PI), have been combined under CMS-2014. Both CMS-2012 projects leveraged available in situ measurements of CO2 and used high-resolution regional inverse modeling tools to quantify CO2 fluxes on regional scales and to investigate consistency among in situ and remote sensing datasets and flux products. We have incorporated remote sensing measurements of CO2 into CarbonTracker-Lagrange (CT-L), a NOAA-led effort to implement a regional inverse modeling framework that uses footprints from a suite of Lagrangian transport models and a flexible inversion scheme with geostatistical and Bayesian capability. Under CMS-2014, CT-L inversions will be further developed for North America and Amazonia. The CT-L inversions complement the CMS Flux Pilot estimates, because they are obtained for a regional domain and at higher resolution (1o), using different transport models (i.e. Lagrangian vs. Eulerian), augmented CO2 data sets (in situ and remote sensing), and using explicit matrix inversions rather than a data assimilation approach. The North American inversions leverage dense datasets and models that were developed for the North American Carbon Program, while the Amazon component emphasizes use of vertical profiles from aircraft above the Brazilian Amazon, a critically important yet under-sampled region where extensive cloud and aerosol contamination limit the usefulness of satellite data. Efficient inversion algorithms enable ensemble calculations to test the sensitivity of inferred fluxes to uncertainties caused by possible satellite retrieval errors and model inadequacies, such as errors in simulated atmospheric transport and assumed prior flux error. This work will further develop strategies for incorporating diverse CO2 observations in CMS data assimilation efforts and for quantifying fluxes at scales relevant for Monitoring, Reporting and Verification (MRV) and quantifying uncertainties of CMS products.

Associated Project(s):
2012 - Andrews (CMS 2011) - Arlyn Andrews
2012 - Miller (CMS 2011) - John Miller
2014 - Andrews (CMS 2014) - Arlyn Andrews

Josef Kellndorfer, Woods Hole Research Center, josefk@whrc.org
Curtis Woodcock, Boston University, curtis@bu.edu
Richard Houghton, The Woods Hole Research Center, rhoughton@whrc.org
Pontus Olofsson, Boston University, olofsson@bu.edu
Oliver Cartus, Woods Hole Research Center, ocartus@whrc.org
Neeti Neeti, Woods Hole Research Center, neeti@whrc.org
Chris Holden, Boston University, ceholden@bu.edu

A core need for the development of REDD+ Monitoring, Reporting, and Verification (MRV) systems is the annual assessment of land area change from forest lost or gained, and forest degradation. The goal of this project is 1) to develop an approach for accurate assessment of land area change (i.e., activity data) making optimal use of spaceborne optical and radar time series (Landsat, ALOS PALSAR), and 2) to assess the uncertainty of carbon emission estimates derived from Landsat/PALSAR-derived Activity Data and various sources for Emission Factors for test sites in Mexico, Colombia and Peru. The work carried out in year 1 of the project was focused on: 1) Identification of test sites in Mexico, Peru, Colombia. 2) Preprocessing of Landsat and PALSAR data for the selected test sites. 3) Analysis of Landsat time series based on the change detection algorithm that is under active development at Boston University (Zhu & Woodcock, 2014) to understand the multitemporal signal over different types of land cover/use change in the tropics, and to evaluate where the fusion of optical and radar data can help to obtain timely and accurate activity data (e.g., in areas with persistent cloud cover). 4) An algorithm was implemented that identifies persistent backscatter changes across a time series of PALSAR observations. Initial tests of the algorithm demonstrated 1) the potential of PALSAR data to complement Landsat time series for detecting change, and 2) PALSAR time series are often not sufficiently dense to effectively distinguish between changes in backscatter that are a consequence of actual land cover/use changes and those related to small-scale moisture dynamics, agriculture, etc. A crosscheck of change signals in the Landsat and PALSAR time series improves the accuracy/timeliness of change detection.

Associated Project: 2013 - Kellndorfer (CMS 2013) - Josef Kellndorfer