Shifting irrigation patterns in response to policy, technology, and water availability: Mapping three decades of irrigation across the US High Plains Aquifer with Landsat satellite data
Abstract
Irrigated lands are dynamic. Understanding how irrigated areas change over time is vital to effectively manage limited agricultural water resources, but long-term, high-resolution, and spatially-explicit datasets are rare. The High Plains Aquifer (HPA) in the central United States is one of the largest and most stressed aquifer systems in the world. It supports a $20 billion economy, but groundwater use is unsustainable over much of the aquifer. Uncertainty over the location of irrigated areas and their interannual variability limit the ability to fully quantify, evaluate, model, and manage water use in this region. Emerging cloud computing tools like Google Earth Engine (GEE) make it possible to leverage the full Landsat record to monitor regional systems like the HPA with high spatial and temporal resolution over multiple decades. Challenges remain, however, to develop irrigation classification methods that are robust to a wide range of climate, crops, and evolving management, along with missing data.
Here, we address these challenges to produce annual, moderately high resolution (30 m) irrigation maps from 1984-2017 over the aquifer using random forest classification. Leveraging GEE's extensive data catalog, we combined Landsat imagery with climate and environmental covariables to create a single random forest classifier, filling any gaps in the satellite record with probability-based estimates of irrigation status derived from the 34-year time series. A novel Neighborhood Greenness Index contributed to a 91% map accuracy across years. Spatial analysis of irrigated area through time identified regions of stable, expanding, and declining irrigated area. Given declining aquifer conditions, we estimate that up to 24% of irrigated area may be lost this century. The map dataset is one of the longest, most spatially-refined record of where and when irrigation occurs in the world. It is freely available for stakeholders, managers, and researchers to inform future policies and management, as well as for use in hydrology, agronomy, and climate models.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H34H..06D
- Keywords:
-
- 0402 Agricultural systems;
- BIOGEOSCIENCESDE: 0496 Water quality;
- BIOGEOSCIENCESDE: 1834 Human impacts;
- HYDROLOGYDE: 1842 Irrigation;
- HYDROLOGY