Groundwater-based irrigation has dramatically expanded over the past decades. It has important implications for terrestrial water, energy fluxes, and food production, as well as local to regional climates. However, irrigation water use is hard to monitor at large scales due to various constraints, including the high cost of metering equipment installation and maintenance, privacy issues, and the presence of illegal or unregistered wells. This study estimates irrigation water amounts using machine learning to integrate in situ pumping records, remote sensing products, and climate data in the Kansas High Plains. We use a random forest regression to estimate the annual irrigation water amount at a reprojected spatial resolution of 6 km based on various data, including remotely sensed vegetation indices and evapotranspiration (ET), land cover, near-surface meteorological forcing, and a satellite-derived irrigation map. In addition, we assess the value of ECOSTRESS ET products for irrigation water use estimation and compare with the baseline results by using MODIS ET. The random forest regression model can capture the temporal and spatial variability of irrigation amounts with a satisfactory accuracy (R2 = 0.82). It performs reasonably well when it is calibrated on the western portion of the study area and tested on the eastern portion that receives more rain than the western one, suggesting its potential transferability to other regions. ECSOTRESS ET and MODIS ET yield a similar irrigation estimation accuracy.
Better understanding the variabilities in crop yield and production is critical to assessing the vulnerability and resilience of food production systems. Both environmental (climatic and edaphic) conditions and management factors affect the variabilities of crop yield. In this study, we conducted a comprehensive data-driven analysis in the U.S. Corn Belt to understand and model how rainfed corn yield is affected by climate variability and extremes, soil properties (soil available water capacity, soil organic matter), and management practices (planting date and fertilizer applications). Exploratory data analyses revealed that corn yield responds non-linearly to temperature, while the negative vapor pressure deficit (VPD) effect on corn yield is monotonic and more prominent. Higher mean yield and inter-annual yield variability are found associated with high soil available water capacity, while lower inter-annual yield variability is associated with high soil organic matter (SOM). We also identified region-dependent relationships between planting date and yield and a strong correlation between planting date and the April weather condition (temperature and rainfall). Next, we built machine learning models using the random forest and LASSO algorithms, respectively, to predict corn yield with all climatic, soil properties, and management factors. The random forest model achieved a high prediction accuracy for annual yield at county level as early as in July (R2 = 0.781) and outperformed LASSO. The gained insights from this study lead to improved understanding of how corn yield responds to climate variability and projected change in the U.S. Corn Belt and globally.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.