The Colorado River Basin Water Census: What USGS Learned, 2015

The National Water Census

As competition for water has grown, the USGS has developed a National Water Census — an accounting of the Nation's water.

• Its 2007 science strategy made it a priority for 2007–17.
• The SECURE Water Act of 2009 authorized a national water availability and use assessment.
• In 2012 the Department of the Interior's WaterSMART initiative funded the start of the census.

The census treats water availability as a water budget: inflows, outflows and changes in storage, nationally and regionally. It has two parts:

• Topical studies develop nationwide methods for single budget components, such as streamflow at ungaged sites, evapotranspiration, and water use for unconventional oil and gas.
• Geographic focus area studies look at one area, usually a river basin, where local stakeholders help pick the least-understood parts of the budget. Each runs about 3 years, with about three under way at a time.

The parts of a simple water budget for a watershed. USGS, modified from Healy (2007).

The Colorado River Basin study

The Colorado River Basin, with the Delaware and the Apalachicola-Chattahoochee-Flint basins, was chosen for the first round because of water shortages and possible conflicts over supply. Surface water there was already well monitored, so stakeholders and the USGS chose four components to improve:

1. evapotranspiration;
2. snowpack behavior;
3. water use;
4. how much groundwater feeds streamflow.

The Colorado River Basin; the Upper Basin is outlined in black. USGS.

Evapotranspiration

Evapotranspiration (ET) — water given off by plants plus water evaporating from soil and open water — is a major part of the budget. The USGS built the SSEBop model (Operational Simplified Surface Energy Balance), which combines land-surface temperature from satellites with local weather data (radiation, air temperature, humidity, wind, pressure) to estimate actual ET at each satellite pass, filling the gaps between passes with potential ET from weather data.

The estimates were checked against flux towers, basin water balances and field reports. The study used the 100-meter estimates to measure irrigation water use field by field.

Evapotranspiration in part of the Willcox irrigation district, Arizona: (A) annual, and (B) on July 6, 2013, the day of a satellite pass. Blue and green are highest; circles mark alfalfa fields. USGS.

Snowpack

In the mountains of the West, most precipitation falls as snow, and seasonal snowpacks act as huge natural reservoirs for drinking water, irrigation, industry, energy and ecosystems.

• Earlier melt. Peak snowmelt in Colorado now comes 2–3 weeks earlier than in the late 1970s — traced to less snowfall, warm spring air, and windblown dust darkening the snow.
• Sublimation. The USGS developed ways to measure snow turning straight to vapor. Continuous measurements show a seasonal loss of 2–30 percent of annual snow water equivalent, depending on elevation, slope direction, temperature and wind.
• Model check. The National Weather Service's SNODAS snow model did well in forests but less well above tree line, where wind moves snow around.

Wind-blown snow in the Rocky Mountains. David Clow, USGS.

A weather and continuous sublimation monitoring station in the basin. David Clow, USGS.

Water use

The USGS defines water use as people's interaction with the water cycle: water withdrawn from a source and delivered for public supply, domestic, commercial, industrial, irrigation, livestock, mining, aquaculture, hydroelectric and thermoelectric power, and wastewater returns. National compilations have appeared every 5 years since 1950.

For the Colorado study, withdrawals were compiled by watershed (8-digit hydrologic units) rather than by county and state, including interbasin transfers out of the basin for cities, industry and irrigation. To show trends, data were recompiled for 1985–2010 at 5-year steps. Water use at this scale is not planned to continue.

The Imperial Dam diverts Colorado River water on the Arizona–California border near Yuma. Bureau of Reclamation.

Groundwater that feeds streams

Management in the basin has focused on surface water, but groundwater discharge sustains much of the streamflow in the Upper Colorado River Basin — and droughts and climate change make the link more important.

The USGS used the specific conductance (electrical conductivity) of stream water to separate out the groundwater share:

• at 14 continuously monitored streams, 21–58 percent of annual flow was groundwater;
• a second method for sites with only occasional measurements, applied at 229 sites, found groundwater supplies on average 48 percent of streamflow in the Upper Basin, ranging from 12 to 92 percent, and more in wetter, mountainous uplands;
• a SPARROW model calibrated on those sites is estimating groundwater discharge for more than 10,000 stream reaches.

Data

The study's data and publications are on the National Water Census portal.

Sources

Based on Breton W. Bruce, David W. Clow, Molly A. Maupin, Matthew P. Miller, Gabriel B. Senay, Graham A. Sexstone and David D. Susong, "U.S. Geological Survey National Water Census: Colorado River Basin Geographic Focus Area Study," U.S. Geological Survey fact sheet (2015); a work of the United States government in the public domain. Figures and photographs are taken from the fact sheet's PDF; its map of groundwater discharge, drawn on an Esri base map, is left out.