An even grayer Seattle from global warming?

For those harboring the guilty hope that global warming will transform Seattle into a sun lovers' paradise on par with the Côte d'Azur, meteorologist Cliff Mass has some bad news: It might actually get cloudier.

Mass and his colleagues at the University of Washington recently completed the most detailed computer simulation ever conducted of the region's future weather. Among the surprises was a big boost in cloud cover in March, April and May.

"The spring is going to be gunkier — if you believe this — under global warming," he said.

The model also predicts that the number of summer days when temperatures soar into the 90s will more than triple before the end of the century, if greenhouse-gas emissions from cars and industry continue unabated.

And the hopes of some water managers appear to be dashed by the finding that catastrophic losses of winter snowpack will not be offset by more summer thunderstorms.

"We're not going to make up the precipitation in the summer months," Mass said.

Local governments may need to rethink their water-supply projections, said Doug Howell, of King County's Department of Natural Resources and Parks. "We had been predicting that this region would get warmer and wetter."

Mass cautioned that the analysis is only a first cut using a high-resolution forecasting model to tease out geographic details about the way warming would change Northwest weather. Future studies will attempt to root out glitches and make the scenario even more accurate.

But even the trial run reveals nuances that previous global-warming forecasts were not powerful enough to see.

"It's going to warm up pretty much everywhere, but some places are going to warm up a lot more than others — and that has all kinds of implications," Mass said.

While earlier forecasts predicted significant declines in snowpack, the new analysis shows that the most intense warming will occur on the windward slopes of the Cascade and Olympic mountains, with the biggest spikes coming in the spring. By 2090, mountain temperatures could rise 10 degrees or more, with snowpacks reduced to about 20 percent of current levels.

That's a much grimmer outlook than earlier forecasts, Howell said.

But the new work also suggests Western Washington won't warm as rapidly or as much as the east side of the state. By 2050, temperatures in the Puget Sound region will rise about 2 degrees, while some parts of the Columbia Basin could be up to 5 degrees hotter. In 2090, much of the basin will be up to 8 degrees warmer, according to the model.

"This is not good for Eastern Washington farms," Mass said.

Fine-tuning projections

Earlier forecasts relied primarily on global climate models, which give a planetary view of the way temperatures will rise as global warming continues. But those models lack any detail about the mountains and inland waters that play such an important role in local weather.

So, using a global model as a starting point, Mass fine-tuned those projections with a high-resolution regional model that can distinguish topographical features down to a scale of a few miles.

"If you're going to play the game around here, you've got to have the resolution to see local terrain," he said.

Even with the university's enormous data-processing capacity, it took two months of continuous computer runs to simulate each decade into the future. The researchers also factored in things such as changes in soil temperature, which can affect weather.

The project was funded by King County, Seattle City Light, the U.S. Environmental Protection Agency, the National Oceanic and Atmospheric Administration and the UW's Climate Impacts Group.

The surprising jump in springtime clouds appears to be caused by warmer inland temperatures, which cause air to rise and draw in cooler, moist air from the ocean.

The extreme heating on mountain slopes is probably the result of a feedback loop that kicks in as snow melts, Mass said. While snow reflects sunlight, bare ground absorbs it. So as snow disappears, the ground gets warmer, which accelerates the melting and leads to more heat absorption.

Total precipitation doesn't change much across the region, according to the model. But without a snowpack to hold moisture and slowly release it in spring, the region could be facing pronounced dry spells.

"When the precipitation falls as rain, unless you have a massive reservoir to recover it all, it just runs off," said Bob Burns, deputy director of the King County Department of Natural Resources and Parks.

Many caveats remain

While the new analysis is an exciting effort to unravel local effects of climate change, it also comes with many caveats, said Brad Colman, meteorologist-in-charge for the National Weather Service in Seattle.

For example, the global model that provides the input for the regional climate model has a tendency to overestimate the amount of cold air pouring into the region from the east, because it lacks a realistic approximation of the Rocky Mountains. That can compound errors in the local climate scenarios, Colman said.

"It's really pushing the limits of what we can do with numerical prediction."

Which is exactly what Mass and his group hope to do.

They're working to rid the global model of its unrealistic cold spells and to better analyze their results so far. Next, they want to figure out what would happen locally if global greenhouse-gas emissions are reduced.

The initial runs were based on the assumption that emissions will keep rising at current rates.

"Unfortunately, that's probably a pretty good bet," Mass said. "But if people come to their senses and start burning less fossil fuels, we want to see what that would do."

Sandi Doughton: 206-464-2491 or sdoughton@seattletimes.com