Diversity of Tropical Spiders

The spider fauna of the Netherlands is quite well known. Species are added to the national list very slowly and it has been a very long time since an undescribed species was discovered there. In the tropics, it is a very different story, and it is not unusual for collections to include more undescribed species than described ones. With over 40,000 described species, spiders are the world’s seventh largest order. They are ubiquitous and abundant predators, especially in tropical forest ecosystems. One of the challenges of working with such a group is that it is almost impossible to obtain a complete collection where every species from a place is represented. This is because communities consist of a few common species and a very large number of rare ones. It seems like no matter how many hours you spend collecting, you keep finding species you haven’t seen before. This phenomenon can be plotted on a graph, with the amount of sampling effort on one axis and the cumulative number of species on the other. The result is an asymptotic curve. At first, the number of species rises rapidly, but after a while, it starts to flatten. The flatter the curve, the more complete your inventory. But there are typically so many rare species in tropical forests that it is practically impossible to get the curve completely flat. This means that species are always overlooked, no matter how much effort you invest. Fortunately, statistical methods are available to extrapolate the actual number of species. These statistics are driven primarily by the number of rare species. We use the proportion of species in a sequence of rare abundance classes (e.g., species represented in the total sample by only one or two individuals) to extrapolate to the zero abundance class, which is the estimated number of species overlooked by the survey. This plus the observed species richness is the estimated total number of species.
We are sampling from three parks in Vietnam. We also want to know about how different the spider communities are between these three parks. We could simply compare the list of species we found in each park and assess the overlap. But this is not ideal because of the problem of overlooked species. A rare species present in both sites may have been missed in one or both inventories. Again, statistical methods are available to estimate the number of unobserved species in both samples including those that may be shared between two sites. We will use these methods to assess the distinctness of the spider community in different parts of Vietnam.