New research predicts that changes in mountain snowmelt will shift peak streamflows to much earlier in the year for the vast Colorado River Basin. It will alter reservoir management and irrigation across the entire region. The basin stretches from sea level at the Gulf of California as higher as 14,000 feet in the Rocky Mountains of Colorado. It provides critical water to cities and farmers within the basin and beyond.Significant water is diverted to large population centers. It includes Albuquerque, Denver, Los Angeles, Salt Lake City, San Diego and Santa Fe.
A team has used artificial intelligence and predicted that snowmelt will disappear entirely in some sub-watersheds and large snowpack will lossin others.The team also found that as temperatures continue to rise, higher-elevation areas of the basin are projected to see a large loss of snowpack. The team found distinct variations in how much the seasonality and intensity of future runoff will change.
The study also projects more arid conditions in the Green River Valley near the border of Colorado, Utah and Wyoming. Scientists have also predicted that mountainous areas in Arizona will see significantly decreased soil moisture.One aim of the project was to demonstrate a form of AI called unsupervised machine learning. It accelerates analyzing climate and hydrology data. They have introduced a new tool for sifting through vast data sets to identify key features.
Scientists derived the drought indicators from historical data. It resulted from simulations of future scenarios by several climate models over a 30-year time period.The AI analyzed the simulation. It then pinpointed key sub-watersheds with big expected increases in drought. The AI reduced their size for quick processing, to manage the enormous resulting data sets.