Baylor Researcher Develops New Method for Detecting Biodiversity Losses

February 2, 2010

Follow us on Twitter: @BaylorUMediaCom

Baylor University researchers along with ecologists from the University of Maryland, Baltimore County have developed a new method that measures the impact of human-caused environmental degradation on environmental biodiversity. The new statistical analysis method is more precise than current methods and has revealed a dramatically lower ecological "tipping point" at which species are threatened.

The new method, called the Threshold Indicator Taxa Analysis (TITAN), is detailed in the British Ecological Society's journal Methods in Ecology and Evolution.

"This new method addresses what we perceived to be problems with existing quantitative approaches for detecting 'thresholds' in the response of organisms to pollutants," said Dr. Ryan King, associate professor of biology at Baylor, who, along with Dr. Matthew Baker, assistant professor of geography and environmental systems at UMBC, developed the method. "Many types of organisms will suddenly decline or disappear once a threshold level of pollution is reached. We were particularly dissatisfied with the lack of sensitivity of existing methods and some of the assumptions required by those methods. This method has proven to be much more sensitive while also being very precise in its estimation of thresholds."

Environmental scientists are increasingly relying on statistical methods for determining thresholds, or "tipping points," when ecological systems are damaged by changes to the environment. Accurately measuring these tipping points is important for protecting threatened species and better understanding how ecosystems respond to major changes such as global warming, coal mine leaching, agricultural pollutants or water-runoff from highly developed areas.

The researchers said a decade-old analysis widely cited by environmental professionals and policymakers suggests that it takes up to 15 percent of solid surfaces like roads or parking lots, or 20 to 30 percent developed land in a given area before local water systems no longer sustain normal aquatic life. King and Baker's new method demonstrates that aquatic life actually shows significant loss of biodiversity with only 1 to 3 percent developed land in a watershed.

"This really surprised us, but after carefully examining the data and testing the method using simulations, it became apparent that these declines were real," King said. "It certainly brings to light a strikingly strong relationship between development and degraded water quality in streams, but the mechanisms are not yet clear."

Media contact: Frank Raczkiewicz, Assistant Vice President of Media Communications, 254-710-1964.