Assessing the field-scale crop water condition over an intensive agricultural plain using UAV-based thermal and multispectral imagery

The ever-increasing food demand globally exerts a pressing need to develop efficient water resource management to meet crop water requirements. While satellite remote sensing has proven invaluable for assessing basin-scale crop water conditions (CWC), foreseeable challenges still exist to evaluate field-scale water status in crop-intensive regions. In this study, we present a systematic framework to assess the field-scale CWC using an unmanned aerial vehicle (UAV) based thermal and multispectral imageries at two (upstream and downstream) instrumented sites of an agro-intensive Ganga basin Critical Zone Observatory (CZO), India. We first estimated the land surface temperature (LST) by incorporating emissivity and atmospheric radiance components. We then used the LST and normalised difference vegetation index (NDVI) to compute the vegetation temperature condition index (VTCI) for each site as a proxy for field scale CWC. Our results reveal a significant correlation between aerial LST and in-situ radiometer observations with an absolute difference of ≤±1K. The aerial LST estimation also showed high accuracy and consistently low error (≤3K) in predicting synchronous ground-based surface temperature with a slightly warm bias of 1–2 K. While the variability in mean VTCI across both windows is minimal, the downstream window reveals a sharp increase in drier CWC from 22 % to 51 % in comparison to the upstream region (20–27 %). Further, the field-scale VTCI demonstrated a significant positive correlation (r = 0.6 to 0.81, Kendall’s τ = 0.5 to 0.82) to concurrent soil moisture conditions during all acquisitions. Our study highlights the potential of UAV remote sensing for assessing of high-resolution CWC and designing strategies for regulating agricultural water resources and improving water use efficiency.