Estimating Irrigation Water Consumption Using Machine Learning and Remote Sensing Data in Kansas High Plains

Abstract

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.

Estimating irrigation water use from remotely sensed evapotranspiration data: Accuracy and uncertainties across spatial scales

“…While many past studies have sought to estimate irrigation water use using satellite-based ET data and other hydrological variables such as soil moisture (Brocca et al, 2018;Dari et al, 2020;Ketchum et al, 2023), these estimates have typically been evaluated against aggregated statistics or synthetic model estimates of water use. Other studies use statistical or machine learning approaches to relate ET to observed water use, but these approaches are limited in terms of their applicability outside of the model training region (Filippelli et al, 2022;Majumdar et al, 2022;Wei et al, 2022). As a result, there is a lack of knowledge about how effectively ET data can be translated into irrigation water withdrawals and applications across different spatial scales, from an individual field to a region, which are relevant to regulatory and management purposes.…”

Combining Remotely Sensed Evapotranspiration and an Agroecosystem Model to Estimate Center‐Pivot Irrigation Water Use at High Spatio‐Temporal Resolution

“…However, these studies ignored the water unavailable for crop use, such as runoff, drainage, and deep percolation, and also had coarse temporal and/or spatial resolutions. Some researchers estimated irrigation water use using machine learning but required the detailed irrigation water use data sets, such as the in situ pumping records (Wei et al., 2022).…”

A Hierarchical Temporal Scale Framework for Data‐Driven Reservoir Release Modeling

“…Anthropogenic activities, such as reservoir operation (Biemans et al., 2011; Döll et al., 2009; Haddeland et al., 2006; Singh & Basu, 2022; Zeng & Ren, 2022; Y. Zhao et al., 2021), urbanization (Li et al., 2020; Oudin et al., 2018), and large‐scale irrigation (Condon & Maxwell, 2019; Ferguson & Maxwell, 2011; Siebert et al., 2010; S. Wei et al., 2022), have become increasingly important or even dominant driving forces of hydrologic processes in many watersheds over the world. In these watersheds, the streamflow observed at gauging stations represents the interaction between hydrologic and anthropogenic driving forces, rather than the “natural” or “unregulated” flows simulated in hydrologic models (Blair & Buytaert, 2016; Clark et al., 2015).…”