Feature - Forecasting an El Niño a half-century in advance
|
||
|
|
||
The El Niño Southern Oscillation, a natural phenomenon involving the fluctuation of Pacific atmospheric and ocean conditions between warming (El Niño) and cooling (La Niña), occurs roughly every three to six years and greatly affects the climate, especially on the coasts of Peru and Ecuador. Researchers in Peru are looking to grid computing to help them run their climate forecasting models efficiently and find ways to prevent or mitigate the effects of future El Niños. "Global warming may contribute to an increase in frequency and intensity of El Niño, which could lead to serious drought in some areas and excessive rainfall in others," says Richard Miguel of the Numerical Prediction Center of SENAMHI (the national meteorology and hydrology service) in Peru. "It is important for scientists to better understand El Niño and the significant impact it could have on Latin America in the future so that we can be prepared." As a collaborating group in the E-science grid facility for Europe and Latin America (EELA), SENAMHI researchers are working to integrate global and regional climate modeling applications so that scientists can better study and forecast the effects of El Niño. Global models simulate the dynamics of atmosphere and ocean circulation with a coarse resolution of about 100 km (about 60 miles). Regional models use data from the global models to simulate circulation in specific, smaller areas. Using statistical relationships between local climate variables such as temperature and precipitation, and large scale predictors such as pressure fields, these regional models can yield resolutions of about 5-10 km (about 3-to-6 miles). Miguel and his team currently use two small clusters and are in the process of gaining certification to use the EELA grid to run computationally intensive regional models of the effects of global warming according to three scenarios: "pessimistic," "normal" and "optimistic." The models enable them to produce long-term forecasts (30 to 50 years in the future) of the intensity of effects related to El Niño, such as temperature and rainfall. Disaster mitigation experts can then use this information in their planning, Miguel says. |
||
Seamless Although several global modeling applications exist, no application yet seamlessly integrates global and regional components. SENAMHI and fellow EELA collaborators are working to create a seamless application, expected to come online in 2010, and an accompanying Web portal where researchers can run their models on the grid without having to develop their own interfaces. It bears the not-so-seamless name of "Simulation in cascade of numerical models on the grid," and is based on the Climate Atmospheric Models (CAM) and the Weather Regional Forecast (WRF) applications. In a separate effort, SENAMHI is one of five research institutions in Peru coming together to create PeruGRID, a national grid infrastructure expected to become operational later this year. The consortium aims to increase awareness in Peru of the advantages of grid computing technology in the scientific community, industry and government. "Peru is a developing country with many limitations in financial resources for research in science and technology," Miguel says. "It is important for Peru to have a national grid infrastructure because we need to share resources, and the grid will make it possible for our researchers to easily access the computational resources they need." -Amelia Williamson, for iSGTW |
Our Underwriters
Thank to you our underwriters, who have supported us since the transition from International Science Grid This Week (iSGTW) into Science Node in 2015. We are incredibly grateful.
