Monitoring Elephant Populations using GIS

From Barnes et al., 1997 and Michelmore et al., 1994

 

One of the many things that distinguishes elephants is their impressive ivory tusks, which involves them heavily in the debate surrounding the ivory trade. One of the arguments for opening up the trade is that elephants are not as scarce as we think. Getting accurate elephant populations, then, is an increasingly important, yet difficult, task. Especially in areas like central Gabon where thick rainforest is the predominant type of vegetation, counting and keeping track of elephants can be quite difficult. Roads are scarce and badly maintained, river travel is expensive, air counts are useless because of the dense cover, and foot counts are too slow to make significant progress. One study, as highlighted in the Barnes et al. paper, uses GIS to create density maps that will give wildlife managers a better estimate of an actual elephant populations than any of the aforementioned counting methods.

One way to estimate elephant population numbers is through dung counts, which is how population data were obtained for this study. Using the line-transect method, researchers counted dung piles, and each pile was aged. These data were analyzed to produce a population estimate with regards to distance from the roads. Using a GIS, a gradient map was produced like the one seen here. Each contour line represents a 4km interval distance from roads and rivers. The thick white bands represent the roads or rivers where no elephants are found. By combining dung count data with this spatial data, managers can obtain a better idea of not only how many elephants inhabit the area, but also where they are concentrated. The area of available habitat for elephants increases dramatically as distance from the road also increases, and as expected, elephant density increases as well (figure from Michelmore et al., 1994). Use of GIS enables scientists to calculate forest area available for habitat and then make much more accurate population estimates than other counting methods. Because elephant densities are least around roads, the road counts present inaccurate data, and air counts are also unrealistic representations of actual population counts because of the high density of the rainforest cover.

This study could have heavy implications on the way densities are calculated, especially for animals that are difficult to count accurately. By creating spatial models of population densities, wildlife managers can predict population counts more accurately without utilizing nearly as much manpower or time. More educated management decisions can be derived from more accurate population counts, and these decisions could lead to greater sustainability of wildlife populations.