That is from the abstract. From the paper: We asked how much of the Earth’s land surface is already “biotically compromised” (exceeds the boundaries of 10% lo

That is from the abstract. From the paper: We asked how much of the Earth’s land surface is already “biotically compromised” (exceeds the boundaries of 10% loss of abundance or 20% loss of species). This measure (10% abundance or 20% species) is used throughout the full paper and the 23-page supplementary material. The general area of quantitative biodiversity study is in its infancy. It is not as well-developed as other study areas within Earth system science. For example, there are no universally-agreed-upon variables that must be monitored, like temperature, sea level, etc., that exist for other areas. So biodiversity-related measurements tend to be side-effects of other measurements for other purposes. For instance, habitat loss could be indirectly measured by classifiers run on satellite-produced ground imagery. Here's some information on NASA's biodiversity research program: http://cce.nasa.gov/cce/biodiversity.htm http://cce.nasa.gov/cce/biodiversity.htm