Remote Sensing and Modeling Carbon Fluxes

Overview

Quantifying the movement, or flux, of carbon dioxide (CO2) into and out of the atmosphere is important for understanding carbon sinks and sources. Mapping and monitoring carbon fluxes in different terrestrial environments is essential for understanding the contribution of these environments to the global carbon cycle. It is also a prerequisite for making decisions regarding various carbon-related climate change mitigation strategies, such as carbon sequestration activities and justification for carbon credit payments for removal of carbon from the atmosphere as the result of land management changes that utilize the ability of plants to reduce atmospheric CO2.

Much of this work has been accomplished as part of the USAID GL-CRSP project "Livestock Development and Rangeland Conservation Tools for Central Asia (LDRCT)" Grant No. PCE-G-00-98-00036-00 to University of California, Davis. This project focused on the Central Asian countries of Kazakhstan, Uzbekistan, and Turkmenistan. The project focus was on broad livestock development issues such as animal production, socio-economic conditions, GIS technologies, resource availability, range forage, and carbon flux. The range forage and carbon flux component of this project included the scaling-up of flux tower measurements to the regional scale in a collaborative effort by the US Geological Survey, the USDA Agricultural Research Service (http://www.usu.edu/forage/frrl.htm), University of California, Davis, host country institutions, the Office of Agriculture and Food Security, Global Bureau, US Agency for International Development (under Grant No. PCE-G-98-00036-00) The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID, UC Davis, USDA ARS, or USGS.

USGS/EROS has created and validated models that employ near-real-time remotely sensed data, spatial data regarding soil attributes and land management, and the biophysical mechanisms that account for carbon fluxes between the land surface and the atmosphere. Using these models, quantitative maps are created for the seasonal and annual dynamics of net ecosystem exchange (NEE) of carbon. The models can be manipulated to explore carbon fluxes at regional and national scales.

Much of this research has focused on carbon flux in grasslands. Grassland ecosystems (rangelands) store most of their carbon below ground. Because rangelands constitute 40% of the Earth's terrestrial surface, they could be a significant factor in the global carbon budget. To assess the role of rangelands in the global carbon cycle, the exchange of CO₂ between rangelands and the atmosphere must be quantified. A method for quantifying CO₂ fluxes between rangeland soils and plants and the atmosphere involves the use of flux towers. These instruments measure the vertical changes in atmospheric CO₂ concentrations, along with wind speed and direction above the vegetation over a distance upwind (fetch) that is about one hundred times the height of CO₂ sensors. In order to quantify CO₂ fluxes over large expanses of rangeland, it is necessary to extrapolate data gathered from one or a few flux towers to entire regions or ecosystems. In doing so, it is also necessary to consider variability in climate, soils, and management conditions that often occur in rangelands. Remote sensing provides a feasible and economical means to estimate the spatial and temporal dynamics of CO₂ flux in rangelands. Vegetation indices derived from remotely sensed data, such as the Normalized Difference Vegetation Index (NDVI), have been correlated with CO₂ fluxes.

Return to Top

Remote Sensing and Ecosystem Performance
Remote Sensing Models to Quantify Performance Anomalies in Grasslands and Steppes

Climate, pest infestations, fire, and poor land management are stressors that can change ecosystem dynamics. As systems become stressed and degraded, they become more susceptible to the effects of surface water runoff and erosion. The vegetative composition of grasslands and shrub steppes changes when native species are unable to thrive or if invasive species become prolific. This research seeks to quantify ecosystem changes at a regional scale, both over a growing season and in near real-time. We have developed a regional monitoring approach that is able to identify problematic areas and provide land managers with locations in need of best management practices and other amelioration efforts. Our method integrates multi-sensor and multi-temporal spatial data in a performance model that calculates anomalies between expected and actual performance. Inputs to the model include individual land cover types, mean vegetation indices, and several soil and climate data sets. Outputs include ecological performance maps that show areas with higher or lower performance values than expected.

Remote Sensing Models to Quantify Performance Anomalies in Grasslands and Steppes

Remote Sensing and Modeling Carbon Fluxes
Flux Tower Data
             

The Bowen ratio/energy balance technique (BREB) was used to continuously monitor CO₂ at a sagebrush steppe site near Dubois, Idaho. Mathematical modeling of CO₂ fluxes was done to fill gaps resulting from missing or aberrant data. Thus, daily integrals across the entire growing season were quantified. Results of this research suggested that a combination of BREB measurements and modeling techniques can be used to estimate CO₂ fluxes on important rangeland ecosystems.

Sagebrush-steppe BREB flux tower near Dubois, Idaho (N. Saliendra) Growing Season CO2 Fluxes in a Sagebrush-Steppe Ecosystem in Idaho: Bowen Ratio/Energy balance Measurements and Modeling

As a 'Proof of Concept' to correlate remote sensing and flux tower data, a strong relationship between AVHRR NDVI (1 km resolution) and daytime CO₂ fluxes was demonstrated at the Dubois site, and was found to be fairly consistent across four growing seasons (1996-1999) during both drought and wet years.

Calibration of Remotely Sensed, Coarse Resolution NDVI to CO₂ Fluxes in a Sagebrush-Steppe Ecosystem


Return to Top

Remote Sensing and Modeling Carbon Fluxes
Deriving Gross Primary Productivity and Ecosystem Respiration from Flux Tower Data

To improve the utility of flux tower data, daytime net ecosystem CO₂ exchange was partitioned into fluxes due to respiration and gross primary productivity (GPP). This was done by Tagir Gilmanov using complex, light-curve equation fitting. The resulting separation of CO₂ flux data provided very useful information on these fundamental processes, which are difficult to quantify with classical measurement techniques. This technique was first applied to data collected from a flux tower in Central Asia.

A Method to Estimate Gross Primary Productivity from Bowen Ratio-Energy Balance CO₂ Flux Measurements and Construction of Predictive Relationships between NDVI and CO₂ Flux

and later to flux towers in the United States:

Gross Primary Production and Light Response Parameters of Four Southern Plains Ecosystems Estimated using Long-Term CO₂-Flux Tower Measurements


Return to Top

Remote Sensing and Modeling Carbon Fluxes
Scaling Up CO₂ Fluxes: Central Asia
             

Remotely sensed vegetation indices provide a temporally and spatially dynamic variable that is a key for the mapping of CO₂ fluxes. NDVI, which was found to be a surrogate for photosynthetic potential, was expected to be more strongly correlated to GPP than the daytime net CO₂ fluxes as demonstrated in the study on a sagebrush steppe at Dubois, Idaho.

The correlation between 1-km NDVI and flux tower data, together with the quantification of ecosystem respiration and GPP, was first employed in measurements obtained from Central Asia. LDRCT has been operating Bowen ratio energy-balance (BREB) flux towers in the Central Asian nations of Turkmenistan, Uzbekistan, and Kazakhstan (one flux tower per country) since 1998.

Since the collapse of the Soviet empire, large-scale abandonment of economically and ecologically unsustainable wheat farms has occurred on the steppes of northern Kazakhstan, providing an opportunity to study carbon flux (and potentially carbon sequestration) on these abandoned croplands. Using flux tower data combined with historic SPOT vegetation data sets, the GPP, respiration (R), and net ecosystem exchange (NEE) was mapped for every ten days throughout the 2000 growing season. This allowed the reconstruction of the temporal dynamics of CO2 flux components throughout the growing season for any rangeland location in the Kazakh Steppe and provided useful insights into the forcing factors of rangeland carbon dynamics.

Flux tower locations and Kazakhstan Steppe in Central Asia

For further International activities please visit http://edcintl.cr.usgs.gov/.

Remote Sensing and Geographic Information System in Rangelands of Northern Kazakhstan: Quantification and Mapping of Seasonal CO₂ Fluxes from Bowen Ratio-Energy Balance Measurements

Gross Primary Productivity of the True Steppe in Central Asia in Relation to NDVI: Scaling up CO₂ Flux Measurements

Scaling Up of Carbon Fluxes to Predict the Carbon Dynamics in Kazakhstan Central Asia


Return to Top

Remote Sensing and Modeling Carbon Fluxes
Rangeland CO₂ Flux in the Western United States
             

The USDA Agricultural Research Service (ARS) has been monitoring rangeland CO₂ fluxes across the western United States since 1996 using a network of flux towers. Based on the success of the Central Asian scaling-up of flux tower measurements, the GL CRSP intiated a collaborative effort among USAID, ARS, South Dakota State University, University of California at Davis, Utah State University, and EROS to spatially extrapolate CO₂ fluxes from the USDA ARS Rangeland (AgriFlux) Network and selected towers from the AmeriFlux Network. This synergistic use of flux tower data sets for validation and alogithm development focused on improving scaling up algorithm accuracy and robustness in both Central Asia and the Western U.S. This project, the "ARS-CRSP CO₂ Flux Scaling Project", was supported by funding from USAID-GL CRSP, USDA ARS Agriflux participants, USDA Global Change Program, and the USGS.

This project has demonstrated the complementary nature of domestic and international projects: Central Asian and North American data sets have been pooled together to develop more robust CO₂ flux algorithms, and this work may lead to monitoring of rangeland fluxes throughout the Northern Hemisphere, and ultimately, worldwide.

Gross Primary Production and Light Response Parameters of Four Southern Plains Ecosystems Estimated Using Long-Term CO₂ Flux Tower Measurements

Scaling-Up Tower CO₂ Fluxes in Semi-Arid Grasslands of the Great Plains Using Spectral Vegetation Indices and Phenomenological Modeling Poster

Land Remote Sensing Models Quantify Ecosystem Carbon Dynamics: Land Cover Change, Management, and Climate Impacts

Modeling and Mapping of Carbon Fluxes in Rangeland Ecosystems

Quantifying Regional Range Condition for Erosion and Carbon Modeling

Adaptive Rule-Based Piece-Wise Regression Models for Estimating Regional Net Ecosystem Exchange in Grassland and Shrubland Ecoregions Using Regional and Flux Tower Data

Evaluation of the Empirical Piecewise Regression Model in Simulating GPP in the Northern Great Plains



Return to Top

Remote Sensing and Modeling Carbon Fluxes
USGS Earth Resources Observation & Science (EROS)/NOAA Carbon Flux Tower

The USGS Earth Resources Observation & Science (EROS) and the National Oceanic and Atmospheric Administration (NOAA) in collaboration with South Dakota EPSCoR Center for Biocomplexity Studies, South Dakota State University (SDSU), South Dakota School of Mines and Technology (SDSM&T), and Augustana College have begun work with a carbon flux tower in eastern South Dakota. The tower is part of the paired plot analysis being conducted on two separate grazed grassland sites in South Dakota. The work stems from the proposal, Quantification and scaling-up of the coupled biogeochemical cycles of carbon and water in grassland ecosystems of South Dakota: Synthesis of flux tower measurements, modeling, GIS, and remote sensing, funded through the South Dakota EPSCoR Center for Biocomplexity Studies. The study seeks to quantitatively characterize diurnal, seasonal and year-to-year dynamics of carbon and water fluxes in grazed grasslands of South Dakota using eddy covariance flux tower measurements, refine the use of remote sensing to quantify these fluxes, and quantitatively characterize the carbon pools using on-site biomass measurements and application of carbon isotope discrimination techniques in pastures at different range conditions classes. A paired-plot analysis will be used to evaluate environmental driving factors and quantify magnitudes and variability of CO₂ and H₂O fluxes by comparison measurements at the two sites.

Those involved in the proposal are as follows:

• Tagir G. Gilmanov, Principal Investigator, Associate Professor, Department of Biology and Microbiology, South Dakota State University, Brookings, SD.
• Bruce K. Wylie, Principal Scientist, SAIC, USGS Earth Resources Observation & Science (EROS), Sioux Falls, SD.
• Alexander J. Smart, Range Scientist/Assistant Professor, Department of Animal and Range Sciences, South Dakota State University, Brookings, SD.
• Lee Alexander Vierling, Assistant Professor, Institute of Atmospheric Sciences, South Dakota School of Mines and Technology, Rapid City, SD.
• Patrick R. Zimmerman, Institute of Atmospheric Science (IAS), Director and Department Chair, South Dakota School of Mines and Technology, Rapid City, SD.
• Tilden Meyers, Atmospheric Scientist, NOAA/Atmospheric Turbulence and Diffusion Division, Oak Ridge, TN.
• Roger N. Gates, Range Extension Specialist/Assistant Professor, Animal and Range Science Department, South Dakota State University, SDSU West River Ag Center, Rapid City, SD.
• Lan Xu, Research Associate, Department of Animal and Range Sciences, South Dakota State University, Brookings, SD.
• Patricia Johnson, Professor of Range Science, Department of Animal and Range Sciences, South Dakota State University, Brookings, SD.
• Steven Lee Matzner, Assistant Professor, Augustana College, Sioux Falls, SD.
• Ruth Anne F. Doyle, Environmental Analyst, SAIC, USGS Earth Resources Observation & Science (EROS), Sioux Falls, SD.

USGS Earth Resources Observation & Science (EROS)/NOAA Carbon Flux Tower	Data from the tower is reflected in the above graph.

Return to Top

Remote Sensing and Modeling Carbon Fluxes
Additional Instrumentation

A Soil Climate Analysis Network (SCAN) tower has been placed at the USGS Earth Resources Observation & Science (EROS) Campus by the USDA Natural Resources Conservation Service. It monitors environmental variables and soil moisture at numerous depths. This data can be downloaded at ftp://ftp.wcc.nrcs.usda.gov/data/scan/sd/2072.



















Return to Top

U.S. Department of the Interior | U.S. Geological Survey
URL: http://eros.usgs.gov/
Page Contact Information: custserv@usgs.gov
Page Last Modified: August 22, 2006