Even while emissions are in decline, sulfur released into the air primarily by coal- and oil-burning power plants continues to acidify streams in the eastern United States, stressing vegetation and harming aquatic life. Watersheds rich in base cations—nutrients that attract and bind acidic molecules—naturally buffer streams against acidification. These watersheds can be identified by their soils, vegetation, and other physical properties. Mitigating stream acidification depends on knowing how much sulfuric acid falls on a landscape over time, but also on accurate predictions of base cation weathering (BCw) rates in watersheds. However, previous models lack the accuracy to predict BCw rates across large regions. Accurate predictions are needed to inform policy and management decisions.Using a machine learning approach, scientists with the U.S. Forest Service Pacific Northwest Research Station developed a model that predicts and maps BCw rates in the southern Appalachian Mountains. The model made better predictions than traditionally used linear regression methods, and confirmed many findings of prior empirical studies. The study also found that BCw rates were influenced by other climatic variables, such as precipitation and temperature. Results are being used to support sulfur critical loads modeling throughout the region.
展开▼