You are here

Taking the Andes Mountains into Account

Imagine something as massive and towering as the Andes Mountains not having an effect on its surroundings. When it comes to computer simulations that help predict the weather, traditional global models operate poorly represent the mountain chain, which is more than 4,000 miles long, 300 miles wide and 13,000 feet high.

Until now, that is.

A computer model in the making at the Pratt School of Engineering for the past seven years is changing the way weather and climate forecasters will be able to gauge the impact of different variables effecting weather patterns at both the local level and globally—simultaneously. The new weather simulation, known as the Ocean Land Atmosphere Model (OLAM), was developed by Pratt’s Robert Walko, senior research scientist.

Roni Avissar and David Medvigy“When looking at a specific region, traditional global climate models have not been very helpful,” said Roni Avissar, professor of civil engineering. “The model that we developed is an attempt to improve our ability to study specific regions. While global climate issues are important, what is less understood are the real implications of global trends, such as warming, on specific areas of the planet.”

Earlier this year, the Duke team made OLAM – which is composed of more than 100,000 lines of computer code — available to climate scientists around the world. One of the first fruits of this model is an analysis of how much the phenomenon known as El Niño affects the weather – particularly rainfall – in the Amazon River basin, one of the largest ecosystems in the world. About one-fifth of all the water flowing into the world’s oceans from rivers passes through the Amazon River basin.

El Niños, which occur every two to seven years, usually form over the Pacific Ocean west of South America as a result of temperature fluctuations on the ocean’s surface. Depending on where they occur on land, El Niños are known to cause floods or droughts, and as such, represent one of the largest influences on global weather patterns.

Avissar and his post-doctoral fellow David Medvigy were interested in determining how the interaction of an El Niño and the Andes Mountains might affect the weather of the Amazon River basin. This knowledge would also be important in creating more accurate predictions of how global warming might affect the basin in the future.

Existing data that relies on observation on the ground and satellite imaging suggest that the basin receives less rainfall than average during El Niño seasons. “However, the existing computer models actually showed the opposite effect,” Medvigy explained.

The Duke team employed OLAM to settle this inconsistency. Its strength is that while it can model climate globally, it can also hone in on specific areas at the same time. In this way, the model illustrates how global factors impact a region and how the unique characteristics of a region in turn influence global weather patterns.

Because traditional climate models are geared to predict events on a grander scale, their “resolution” is not fine enough to account for such factors as the Andes. Amazingly enough, the mountain chain is too small to be factored into the models’ processing.

“We found that models at resolutions coarser that 100 kilometers showed large increases in precipitation during El Niño events, while our model with resolutions finer than that showed the opposite – less rainfall,” Medvigy said.

“Our results show that the actual physical barrier of the Andes is the most important factor influencing the ramifications of El Niño on the Amazon basin,” Medvigy continued. “This result held true even after controlling for all other possible influences.”

The OLAM simulation could take into account the role of the mountain chain in preventing the warm and moist air characteristic of an El Niño from reaching the Amazon. Because the Andes “don’t exist” in traditional simulations, the model assumes that the warm and moist air travels freely over the Amazon basin, where it falls to earth in the form of rain.

Avissar said that their research into what happens in the Amazon basin is not only important because it is such an immense ecosystem, but that understanding regional impacts help predict what may happen in the future.

Future studies of the basin using OLAM are focusing on the impact of vegetation on the climate and how such man-made actions as deforestation will affect weather patterns.

“Weather and climate impacts every aspect of peoples’ lives worldwide, and now we have a better tool for predicting when the next hurricane may strike or what the effects of a global two-degree temperature rise will have on a specific region,” Avissar said. “Being able to explain what happens now will allow us to make better predictions in the future.”

The results of the Duke analysis were published in August in Geophysical Research Letters, an American Geophysical Union journal. The National Science Foundation supported the work.