Most attempts to quantify the impact of humanity on nature and bring it to public attention have centred around estimates of extinction rates. Suggestions that these figures have been exaggerated are, in our view, misplaced, but extinction rate estimates do face other problems -inevitable uncertainty, an arguably weak link to economic value, and insensitivity to short-term change. We therefore look here at other large-scale measures of the changing state of nature, focusing on recent analyses of trends in population size, numbers of populations and habitat extent. In spite of being limited by sampling inadequacies, these data provide a sensitive short-term complement to the long-term perspective gained from considering extinction rates that can be linked directly both to economic values and to public concerns. Although further work is needed on extinction rates, we conclude that significant new emphasis should be placed on instituting broader, more systematic monitoring of habitats and populations.For almost a quarter of a century, much of the effort put into quantifying the scale and significance of the impacts of humans on our fellow creatures has centred around assessing rates of species extinction [1][2][3][4][5][6][7][8][9][10][11][12]. This is understandable: ongoing attempts to clone ancient DNA from mammoths [13] and thylacines [14] notwithstanding, extinctions are forever. Estimating rates of human-driven extinction thus addresses a major moral concern. Nevertheless, quantifying extinction is difficult, and extinction rates are relatively insensitive to short-term changes in human impacts. Against this background, it is timely to look at measures of habitat loss, and at newly emerging meta-analyses of changes in populations. These too suffer from undersampling and biases in coverage, but are more sensitive to short-term changes in drivers, and are linked more immediately to economic concerns about the value of ecosystem services.
Concerns about extinction ratesBecause species greatly outnumber field biologists (and most of these study species-poor groups [15,16]), we have reasonably comprehensive information about recent extinction rates for only a handful of more readily studied, species-poor groups (such as birds, large mammals and palms [7]). We have little data with which to infer how representative extinction rates for these taxa are for biodiversity as a whole. Moreover, educated guesswork suggests that biologists have described less than half and perhaps ,10% of all the species on Earth [4,17]. Hence, any extrapolation from extinction rates for well-known groups to estimates of the total number of extinctions per year across all groups is impossible. Instead, conservation scientists usually estimate extinction rates in relative terms as the proportion of species going extinct during a given interval, or as average species life spans [2,6,12].A related difficulty is that, because of the scarcity of fieldworkers, most extinctions (even in well studied groups) go unwitnessed. Therefore, extinction r...