e-journal
A simple method for estimating species abundance from occurrencemaps
1. The issue of how to estimate species abundance from presence/absence maps has attracted much attention.
Several methods have been developed to address this problem. However, those methods either overlook the
structure of spatial autocorrelation of species distribution, thus leading to underestimation, or they demand extra
information besides presence/absencemaps.
2. This study first developed a new method that takes account of spatial autocorrelation and only requires occurrence
maps, without any extra information. This method was further improved by incorporating a correction
factor to it.We used an index defined by joint counts of occupied and unoccupied cells to measure spatial autocorrelation
and to correct the underestimation of the randomplacement model.
3. The performance of ourmethod was compared against four othermajor methods (randomplacement model,
negative binomialmodel,Conlisk et al.’s method and Solow&Smith’s method) using both simulated and empirical
data. The results showed that the performance of our method is comparable with othermethods but requires
less and readily obtained input data, a property important for real applications.
4. We suggest this simple, data-parsimonious method be a useful alternative to the currently available methods
for estimating abundance from occurrence.
Key-words: estimating abundance, occurrence, presence/absencemap, spatial autocorrelation
Tidak ada salinan data
Tidak tersedia versi lain