Annual Irrigation Dynamics in the U.S. Northern High Plains Derived from Landsat Satellite Data. Issue 18 (25th September 2017)

• Record Type: Journal Article
• Authors: Deines, Jillian M.; Kendall, Anthony D.; Hyndman, David W.
• Is Part Of: Geophysical research letters. Volume 44:Issue 18(2017)
• Page Start: 9350
• Page End: 9360
• Publication Date: 2017-09-25
• Subjects: irrigation -- water use -- Google Earth Engine -- agriculture -- high plain aquifer -- Landsat; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1002/2017GL074071
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: Sustainable management of agricultural water resources requires improved understanding of irrigation patterns in space and time. We produced annual, high‐resolution (30 m) irrigation maps for 1999–2016 by combining all available Landsat satellite imagery with climate and soil covariables in Google Earth Engine. Random forest classification had accuracies from 92 to 100% and generally agreed with county statistics ( r 2 = 0.88–0.96). Two novel indices that integrate plant greenness and moisture information show promise for improving satellite classification of irrigation. We found considerable interannual variability in irrigation location and extent, including a near doubling between 2002 and 2016. Statistical modeling suggested that precipitation and commodity price influenced irrigated extent through time. High prices incentivized expansion to increase crop yield and profit, but dry years required greater irrigation intensity, thus reducing area in this supply‐limited region. Data sets produced with this approach can improve water sustainability by providing consistent, spatially explicit tracking of irrigation dynamics over time. Plain Language Summary: Irrigated agriculture is the world's largest consumer of global freshwater. In order to effectively use limited water supplies, managers need to understand when and where irrigation occurs. We fill this knowledge gap by using satellite images to produce annual maps of irrigation for 1999–2016 in a large, economically important agricultural region in the central United States that must manage its water supply for multiple users. We then used these maps to study changes in irrigation over time. We were surprised to find that the total area and individual locations of irrigated fields changed substantially from year to year. Our analysis suggests that farmers expanded irrigation when crop prices were high to increase crop yield and profit. We initially expected to also see increases in drought years to compensate for lack of rainfall, but instead, we found the opposite: irrigated area decreased in dry years. Looking closer, we realized that this happened because farmers had to irrigate more heavily over each field, which reduced the number of fields they could irrigate due to limited water supply. These irrigation maps consistently track irrigation over time and are freely available for others to use to help manage water sustainably and meet food needs. Key Points: Sustainable management of agricultural water resources requires improved understanding of dynamic irrigation patterns We produced high‐resolution irrigation maps for 18 years using the full annual Landsat record and advances in cloud computing Irrigated area had considerable spatiotemporal variability driven by changes in commodity prices, precipitation, and water accessibility