Crop productivity estimation by integrating multisensor satellite, in situ, and eddy covariance data into...

Area of Science:

• Agricultural Science
• Remote Sensing
• Ecosystem Modeling

Background:

• Accurate carbon budget estimates require integrating eddy covariance (EC) flux-tower data with remote sensing and ecosystem models.
• Light Use Efficiency (LUE) models offer a method to estimate primary productivity based on radiation absorption and biomass conversion.

Purpose of the Study:

• To estimate primary productivity for major Indian cropping systems using a remote sensing-driven LUE model.
• To compare the model's performance against ground-based EC tower observations and a global GPP product.

Main Methods:

• Utilized multi-temporal Sentinel-2 and Landsat 8 satellite data for crop rotations in Saharanpur, India.
• Estimated monthly photosynthetically active radiation (PAR) and fraction of absorbed PAR (fAPAR) using satellite-derived vegetation indices and insolation data.
• Developed spatial LUE maps by down-regulating maximum LUE with water and temperature stress factors derived from LSWI and EC tower data.

Main Results:

• The LUE model showed higher gross primary productivity (GPP) for sugarcane-wheat systems (C4 crops) than rice-wheat systems (C3 crops).
• Modeled GPP for sugarcane-wheat systems closely matched observed EC tower GPP (Index of Agreement = 0.93).
• The regionally calibrated LUE model demonstrated superior performance in capturing GPP over cropland ecosystems compared to the MODIS GPP product.

Conclusions:

• Remote sensing-based LUE models, when regionally calibrated, can accurately estimate GPP in cropland ecosystems.
• The study highlights the importance of crop type (C4 vs. C3) in determining photosynthetic efficiency and GPP.
• Integration of EC tower data and remote sensing provides a robust approach for regional carbon budget assessments.