Time To Start Believing The Weathermen

Water Supply In Western United States On The Verge Of Rapid Transformation

By Rok Roškar

By 2 pm on an afternoon in July, the first drops began to fall and clouds clashed overhead, but my climbing partner and I were already safe from the inevitable and violent thunderstorm. Setting off in the crisp hours before sunrise, we climbed a pillar on the Tofana di Rozes massif, the largest of silent vertical landscapes in this part of the northeastern Italian Dolomites. Our early start was not an accident: we cheated. The night before, my mobile phone buzzed with a message from my dad, a meteorologist. He advised us that we should expect severe thunderstorms by early afternoon. By 2pm, to be exact.

I don't always follow my father's advice about the weather, combining the skepticism toward meteorological prediction common to millions of modern-day city dwellers with the knee-jerk impulse to always negate my parents' advice. The weathermen never get it quite right, anyway. Regardless, I grudgingly allow weather forecasts to dictate the course of my days. Truth is, short-term weather prediction has improved dramatically over the years and weather professionals are able to predict local phenomena with astounding accuracy.

A difference of a degree in temperature is hardly felt on one's skin, but it may represent the critical boundary between rain and snow, and routine weather prediction can often robustly sit on one side of this line or the other. Perhaps even more incredible is the ability of a climatologist to tell me what the likely temperature was in the Dolomites 500 years ago, and what it will likely be 150 years from now. Surprisingly, determining the details of the fate awaiting the rivers in those charming alpine valleys over the next decade or two turns out to be a very different (and difficult) problem.

Scientists worldwide are scrambling to address this issue of climate change effects on our waters. In a recent study, a team led by Tim Barnett from the Scripps Institution of Oceanography in La Jolla, California, linked trends observed in hydrological patterns of the western United States in the past five decades to human activity. Published in a recent issue of Science, their report "Human-Induced Changes in the Hydrology of the Western United States” represents one of the first successful attempts to pin down the elusive connection between human activity and changes in regional hydrologic patterns. If the trends continue, a significant transformation of the region's water resources is imminent, they conclude. Even the skeptics who never allow the weatherman to keep them indoors will want to carefully consider this particular forecast.

The study analyzed historic hydrologic data from four major river drainages (Columbia, Colorado, the Great Basin, and California) of the Western U.S. in the context of global climate change. The researchers used global climate models calibrated on observational data to constrain the possible sequence of events with and without human environmental influence. The key finding was that changes in hydrology over the past several decades are impossible to reproduce without a generous sprinkling of human-produced greenhouse gases.

Such findings at first glance simply reiterate the notion that humankind is affecting the world's climate. Temperatures will increase and oceans will rise. Tornadoes may be stronger, droughts longer. These scenarios pervade the entire spectrum of public discourse, from political rallies to car commercials.

But, global climate predictions are based on climate models that reduce Earth's mind-boggling geographic complexity to featureless chunks of land, ocean, and atmosphere, in order to make calculations viable. Representation of individual river systems is virtually impossible at this scale. The results are typically long-term (fifty years or more) global temperature and coarse continental weather predictions. While gloomy, the forecast appears distantly removed from you and me, framed in the global context without clear local implications for supply of crucial resources, such as water. As far as policy makers are concerned, this just isn't good enough yet.

Alarmingly, although "climate change” is deeply rooted in our collective consciousness, "there are no accepted standards on how to incorporate climate change into evaluations of hydrologic system performance,” explains Andy Wood, a senior scientist specializing in regional hydrologic modeling at 3Tier, a Seattle-based renewable energy consulting firm, and an affiliate professor of civil and environmental engineering at the University of Washington. This means that when a water management system is expanded or a municipal drought plan updated, technical assessments are usually based on historic climate and streamflow records, without adjustment for the potential effects of climate change on those records. Slowly, however, government agencies are beginning to pay attention.

The City of Portland water bureau was one of the first government water management institutions to consult about climate impact on water supply in 2001. However, even if the indicators of change are clear, the path to action is far from clear-cut because "in most cases, there is no simple way to account for climate change,” explains Philip Mote, a climatologist at the University of Washington's Climate Impacts Group. When it comes to policy makers, addressing climate change is difficult because it is long-term process whose impacts are spread over decades (in contrast to running length of political mandates) and its future trend is based on sometimes wildly diverging model predictions.

Firmly establishing the connections between long-term large-scale climate change and local hydrology is where researchers like Wood and collaborators, who participated in the study of hydrology in the western U.S., are blazing a new path. They are gradually establishing a robust link between climate and local weather, between long-term global prediction and medium-term regional forecast at the level of individual river drainages. Predicting a river's flow is severely complicated as it depends not only on the quantity of precipitation, but also on its form (rain or snow) and subsequent interaction with the land's surface. Snow is essentially a natural short-term water reservoir; it locks up the liquid in crystals until the springtime rise in temperature releases it as runoff to the river basins below. It is a natural timing mechanism, modulating river flow throughout the year.

Our rhythm of life in the valleys and basins below the snow-capped peaks is of course set to the timing of this snowmelt cycle. Higher temperatures in the crucial winter months mean earlier snowmelt (or more rain instead of snow) and correspondingly earlier increase in seasonal river flow. For the past fifty years, this timing has been steadily pushed to earlier times in the year, by up to 1.7 days per decade. The danger is that these rates will accelerate along with global temperature increases. In some areas, such changes may lead to more crop cycles in a given season. However, in many food-producing regions, there is "a higher likelihood of water supply cutoff before the crops have matured,” explains Richard Slaughter of Richard Slaughter Associates from Boise, Idaho. This, in turn, will force farmers to opt for crops that are less dependent on late August and September irrigation, or even, as Wood speculates, "water will be sold by farmers to the cities.”

When it comes to water, human impact goes beyond climate change. In the case study of Portland, researchers found that by the 2020s, changes in the water supply caused by an evolving climate will account for only one third of total strain on the city's water resources. Changes in human population patterns are projected to account for the rest. Direct and indirect impacts of human activity on the most precious and currently abundant natural resource will soon force our perceptions of water to change drastically. Consider such forecasts more carefully than the Weather Channel bulletins.

Rok Roškar is pursuing a Ph.D. in astronomy at the University of Washington.

Images:

Top: Modeling regional hydrology in the context of global climate change requires models on several spatial scales. First, the global models are calibrated and their output used to inform regional modeling. At the final level of refinement, individual watershed and terrain information is required for accurate analysis. Source: Andrew Wood

Bottom: The combined effects of climate change on Portland's water supply will account for only roughly 30% of total change in the city's water resources. Source: Philip Mote and University of Washington Climate Impacts Group