This paper in press at Conservation Letters by Haines et al. presents a novel method for assessing conservation actions. There’s been quite a bit of work done in the past decade, particularly by NGOs, to develop methods to assess whether their actions have actually succeeded; this work was spear-headed in particular by Nick Salafsky and his Foundations of Success. This paper suggests that many of conservation biggest problems can be monitored with spatial datasets and proposes using the Human Footprint as a basis for such monitoring. The Human Footprint is, in essence, a collection of spatial datasets that holistically represent the collective anthropogenic impact on the land. In their paper, Haines et al. suggest that by tracking these spatial datasets through time in a paired way — conservation action site randomly paired with a control — we can get a better handle on whether the particular action was successful. The nice thing about the paper is how clear-eyed it is about what is and is not possible using this approach:
The human footprint is a spatially explicit approach to conservation planning that may serve as an effective visual medium to public audiences and stakeholders worldwide by simplifying the presentation of complex information.
(This is always the last, best resort for spatial analysts: even if the model isn’t perfect, it’s a great communication tool. ) But they also warn:
Spatial data rarely produce a complete picture of what negative impacts are occurring because human footprint data are not well-suited to track anthropogenic impacts that lack a spatial signature…[e.g.] the spread of some chemical pollutants, invasive species, diseases, and impacts of poaching…
Although I have to disagree partially with these particulars — presence of roads is often a very good correlative of poaching — their main point is an important one to consider. How well does a spatial model of human influence catch these hidden factors? A few years ago I did an informal (and sadly never completed) analysis of invasive plants and the Human Footprint and found that they were actually fairly well correlated. You could also argue that disease may be higher amongst individuals that are negatively impacted by the presence of humans. There’s plenty of opportunity here for further exploration.