Estimating crop stresses, aboveground dry biomass and yield of corn using multi-temporal optical data combined with a radiation use efficiency model

Abstract

Crop descriptors, such as leaf area index, crop cover fraction, and leaf chlorophyll content, can be successfully estimated using appropriate spectral indices from the visible and near infrared spectral regions. However, these indices do not provide estimates of dry biomass, an important indicator of crop productivity. For estimating crop aboveground dry biomass and yield, this study developed an approach to integrate crop stressors and crop descriptors derived from optical remote sensing data with the Monteith radiation use efficiency model. Multi-temporal remote sensing data were acquired by the Compact Airborne Spectrographic Imager and the Landsat-5/7 Thematic Mapper/Enhanced Thematic Mapper Plus (TM/ETM+) sensors to monitor the growth conditions of corn in the 2001 and 2006 growing seasons. The modified triangular vegetation index (MTVI2) derived from the remote sensing data was used to estimate the fraction of absorbed photosynthetically active radiation ( f APAR ). A canopy structure dynamics model was then used to simulate the seasonal variation of f APAR . Crop water stress was estimated from the near and shortwave infrared reflectance of the Landsat images for a dry period in the 2001 growing season. By estimating leaf chlorophyll content using the Transformed Chlorophyll Absorption in Reflectance Index (TCARI) in combination with the Optimized Soil Adjusted Vegetation Index (OSAVI), different levels of nitrogen content could be identified. For the two growing seasons, the aboveground dry biomass and yield were linearly related with the cumulative absorbed photosynthetically active radiation (APAR) using the Monteith radiation use efficiency model. The cumulative APAR accounted for 96% of the corn aboveground dry biomass variability and 72% of the yield variability. Biomass and yield variability were partly explained by the variations in crop water stress intensity, which was dependent on soil texture. The seasonal radiation use efficiency was stable over the 2 years and was about 3.9 g MJ − 1 , with a confidence interval of 0.6 g MJ − 1 at the 95% confidence level. The assimilation of remotely sensed data into the radiation use efficiency model performed well for monitoring dry biomass accumulation and estimating corn yields.