Thanks to the Science section of the New York Times, for the description of the research as well as for the name of the measurement:
In a paper published in PLoS One on Wednesday, Dr. Mountrakis and Sheng Yang, a graduate student, tried slicing deforestation a different way. Using satellite maps, they calculated the average distance to the nearest forest from any point in the continental United States in 1992 versus 2001. Between these years, they found, distance to the nearest forest increased by one-third of a mile.
This new metric, which the researchers named “forest attrition distance,” reflects a particular type of forest loss: the removal of isolated forest patches. When these patches are lost (a process the authors refer to as attrition), adjacent forests become farther apart, potentially affecting biodiversity, soil erosion, local climate and other conditions.
The authors calculated the change in total forest cover from 1992 to 2001, and found a loss of 3 percent or 35,000 square miles, approximately the size of Maine. Over the same time period, Dr. Mountrakis said, forest attrition distance increased by 14 percent, a contrast he called striking.
The difference in the magnitude of these two metrics has to do with the fact that forest attrition distance takes geographic distribution into account. Two forests can each lose 25 percent of their tree cover, but have very different forest attrition distances — and different ecological outcomes — depending on the pattern of tree removal. A high forest attrition distance means tree loss has occurred in complete swaths away from other trees. A low forest attrition distance means tree loss has occurred in patches dispersed among other trees…
Read the whole article here.