Relationships between vegetation indices and root zone soil moisture under maize and soybean canopies in the US Corn Belt: a comparative study using a close-range sensing approach

Abstract

Understanding the relationships between root zone soil moisture and vegetation spectral signals will enhance our ability to manage water resources and monitor drought-related stress in vegetation. In this article, the relationships between vegetation indices (VIs) and in situ soil moisture under maize and soybean canopies were analysed using close-range reflectance data acquired at a rainfed cropland site in the US Corn Belt. Because of the deep rooting depths of maize plants, maize-based VIs exhibited significant correlations with soil moisture at a depth of 100 cm (P < 0.01) and kept soil moisture memory for a long period of time (45 days). Among the VIs applied to maize, the chrolophyll red-edge index (CI_red-edge) correlated best with the concurrent soil moisture at 100 cm depth (P < 0.01) for up to 20 day lag periods. The same index showed a significant correlation with soil moisture at a 50 cm depth for lag periods from 10 (P < 0.05) to 60 days (P < 0.01). VIs applied to soybean resulted in statistically significant correlations with soil moisture at the shallower 10 and 25 cm depths, and the correlation coefficients declined with increasing depths. As opposed to maize, soybean held a shorter soil moisture memory as the correlations for all VIs versus soil moisture at 10 cm depth were strongest for the 5 day lag period. Wide dynamic range VI and normalized difference VI performed better in characterizing soil moisture at the 10 and 25 cm depths under soybean canopies when compared with enhanced VI and CI_red-edge.

Acknowledgements

This research was supported by the US Department of Agriculture's Federal Crop Insurance Corporation (FCIC) through the Risk Management Agency (RMA) under USDA partnership (02-IE-0831-0208) with the National Drought Mitigation Center, University of Nebraska-Lincoln, and the US Geological Survey (USGS). We gratefully acknowledge the Center for Advanced Land Management Information Technologies (CALMIT) for providing support, facilities, and equipment, and the High Plains Regional Climate Center (HPRCC) for providing soil moisture and other meteorological data. We also would like to thank Dr Anatoly Gitelson, Eric Hunt, and Todd Schimelfening for many insightful discussions and suggestions. We express our sincere appreciations to the anonymous reviewers for their helpful comments and suggestions that substantially improved the quality of this article.