Soil organic carbon and water content effects on remote crop residue cover estimation
Guy
Serbin, USDA-ARS Hydrology and Remote Sensing Laboratory, guy.serbin@gmail.com
(Presenting)
Craig
S. T.
Daughtry, USDA-ARS Hydrology and Remote Sensing Laboratory, craig.daughtry@ars.usda.gov
E.
Raymond
Hunt Jr., USDA-ARS Hydrology and Remote Sensing Laboratory, ray.hunt@ars.usda.gov
Gregory
W.
McCarty, USDA-ARS Hydrology and Remote Sensing Laboratory, greg.mccarty@ars.usda.gov
Paul
C.
Doraiswamy, USDA-ARS Hydrology and Remote Sensing Laboratory, paul.doraiswamy@ars.usda.gov
Conservation tillage (CT) systems help protect the soil and environment, and improve net farm profitability. CT methods leave increased amounts of crop residue cover (CRC) on the soil surface, minimizing soil erosion and evaporation. CT uses less fuel, disturbs soil less, and requires less fertilizer, reducing greenhouse gas emissions. CRC degradation sequesters carbon into the soil. An efficient verification method is needed for government incentive and carbon credit programs. Remote sensing allows for the rapid and noninvasive assessment of CRC. A remote sensing index used is the Cellulose Absorption Index (CAI). Reflectance spectra of bare soils and crop residues spectrally mix in a linear fashion. Thus, calibration for soil and residue spectral properties can improve CRC estimation accuracy. Soil spectral properties are dependent on soil mineralogy, grain size distribution, soil organic carbon (SOC) content, and water content. In the U.S. Corn Belt, soil spectra were primarily affected by SOC, which darkened soils with increase in content. SOC correlated well with CAI band reflectance values (r2 > 0.84, power regression) within a specific soil series. Spectral interactions between SOC and soil water content were also significant. Water content affected both indices for bare soils. Low SOC soils (< 4.5%) showed the greatest effects of water content on spectral index value; high SOC soils were minimally affected. We surmise that CRC estimation in a given region will be most effective when utilizing remote sensing data in conjunction with SOC and surface water content maps.