Limitations of interpreting orbiting satellite-based fire detection for crop residue burning: Has rice straw burning in northwest India decreased in recent years?

In the northwest region of India, large-scale rice straw burning (RSB) takes place after the harvest every year. For policymaking purposes, the year-to-year variation of RSB fires has been studied extensively using the ‘fire detection counts’ (FDC) derived from orbiting satellite-borne sensors such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS). However, because RSB fires are typically extinguished within an hour, orbiting satellites represent only a small fraction of the daily total active fires. Many fires can also go undetected due to cloud cover, further compromising the reliability of the FDC. In this study, we demonstrate the limitations of interpreting the FDC for the post-harvest burning of rice straw, the most extensively burned crop residue in the region. We investigated the effects of cloud cover, estimated the number of fires obscured by clouds using Gaussian fitting, and explored potential fires occurring outside the satellite’s observation times. This analysis was based on FDC data from MODIS and VIIRS, aerosol data from MODIS, and smoke data from the Indian INSAT-3D geostationary satellite. Our findings indicate that while the relatively low FDC in the year 2019 can be explained by cloud shielding, the rapid decrease in the FDC after 2022 cannot be attributed solely to cloud effects or mitigation policies. In particular, in 2024, the higher number of INSAT-3D smoke pixels at hours outside the orbiting satellite overpass (13:30 local time) suggests that burning occurs more frequently outside the orbit passage. These findings are expected to help researchers and policymakers better interpret the year-to-year variation in the orbiting satellite-based observed FDC when monitoring the effectiveness of RSB mitigation policies.