Water levels dropping in some major rivers as global climate changes

By Kevin E. Trenberth and Aiguo Dai

Sunshine is delicious, rain is refreshing, wind braces us up, snow is exhilarating; there is really no such thing as bad weather, only different kinds of good weather.
—John Ruskin

A key scientific question is how the global hydrological cycle changes over time, especially with global warming. A particular focus of our work at the National Center for Atmospheric Research (NCAR) has been how precipitation changes as the climate changes and changes in extremes, including risk of flooding and drought. We have therefore put together the most comprehensive dataset on river discharge into the ocean for comparison with precipitation changes over each river basin as checks on both the precipitation and river flow data.

Accordingly, we have recently analyzed the flows of 925 of the planet’s largest rivers from 1948 to 2004, combining actual measurements with computer-based streamflow models to fill in data gaps. The rivers in the study drain water from every major landmass except Antarctica and Greenland and account for 73 percent of the world’s total streamflow.

The results show significant changes in about one-third of the world’s largest rivers. Of those, rivers with decreased flow outnumbered those with increased flow by a ratio of about 2.5 to 1.

Overall, the study found that from 1948 to 2004 annual freshwater discharge into the Pacific Ocean fell by about 6 percent, or 526 cubic kilometers—approximately the same volume of water that flows out of the Mississippi River each year. The annual flow into the Indian Ocean dropped by about 3 percent, or 140 cubic kilometers. In contrast, annual river discharge into the Arctic Ocean rose about 10 percent, or 460 cubic kilometers. In the United States, the Columbia River’s flow declined by about 14 percent during the 1948–2004 study period, largely because of reduced precipitation and higher water usage in the West. The Mississippi River, however, has increased by 22 percent over the same period because of greater precipitation across the Midwest since 1948.

Several of the rivers channeling less water serve large populations, including the Yellow River in northern China, the Ganges in India, the Niger in West Africa and the Colorado in the southwestern United States. Some rivers, such as the Brahmaputra in South Asia and the Yangtze in China, have shown stable or increasing flows. But they could lose volume in future decades with the gradual disappearance of the Himalayan glaciers feeding them. In contrast, greater streamflow occurs over sparsely populated areas near the Arctic Ocean, where snow and ice are rapidly melting. Reduced runoff increases the pressure on freshwater resources in much of the world, especially with more demand for water as population increases. Freshwater being a vital resource, the downward trends are a great concern.

Many factors can affect river discharge, including dams and the diversion of water for agriculture and industry. However, by comparing the precipitation changes with the river discharge changes, we can sort out the climate component of the changes. It appears that the reduced flows in many cases are related to global climate change, which is altering precipitation patterns and increasing the rate of evaporation. The results are consistent with our previous research by showing widespread drying and increased drought over many land areas.

This study raises wider ecological and climate concerns. Discharge from the world’s great rivers results in deposits of dissolved nutrients and minerals into the oceans. The freshwater flow also affects global ocean circulation patterns, which are driven by changes in salinity and temperature and which play a vital role in regulating the world’s climate. Although the recent changes in the freshwater discharge are relatively small and may only have impacts around major river mouths, the freshwater balance in the global oceans needs to be monitored for any long-term changes.

Our new results conflict with some earlier results that had been published based on many fewer rivers, and we have utilized much improved methods of dealing with missing data in both time and space (such as where river gauges do not exist) that are based on modeling the river flow from the observed precipitation amounts and changes in clouds, etc.

Our studies on other aspects of the hydrological cycle suggest that net changes in surface evaporation are fairly modest and a much larger percentage change occurs in the water-holding capacity as atmospheric temperatures increase (4 percent per degree Fahrenheit) with global warming. Therefore, where moisture supply is not limited (such as over oceans), a consequence is increased water vapor in the atmosphere, which feeds storms and thus leads to more intense precipitation; increased water vapor, heavier rains and stronger storms are already observed to be happening. However, the disparity between modestly enhanced evaporation and heavier rains means decreases in frequency of precipitation, longer dry spells and enhanced droughts. In contrast, over land in summer and tropical continents, moisture is limited in supply and conditions tend to be either hot and dry or cool and wet. Generally, with more moisture, wet areas get wetter and dry areas get drier. Our new study shows that this is what has happened over the past 30 to 50 years.

One particular extra note is the rather startling change in both river discharge into the oceans and land precipitation following the eruption of Mount Pinatubo in 1991. Indeed, there was an unprecedented drop in land precipitation and runoff and widespread drought as precipitation shifted to be more over oceans and then evaporation faltered, providing lessons for possible geoengineering.

We believe that it is important to understand not only changes in mean precipitation but also the intensity, frequency, duration and type (snow versus rain), and this also applies to the storms that bring precipitation. Understanding these profound consequences of climate change is especially important for water managers. As climate change inevitably continues in coming decades, we are likely to see greater impacts on many rivers and water resources that society has come to rely on.

The results were published May 15, 2009, in the American Meteorological Society’s “Journal of Climate.”¹ The research was supported by the National Science Foundation, the National Center for Atmospheric Research’s sponsor.