Mapping species abundance by a spatial zero-inflated Poisson model: a case study in the Wadden Sea, the Netherlands
Lyashevska, O.; Brus, D.J.; Van der Meer, J. (2016). Mapping species abundance by a spatial zero-inflated Poisson model: a case study in the Wadden Sea, the Netherlands. Ecol. Evol. 6(2): 532-543. dx.doi.org/10.1002/ece3.1880
In: Ecology and Evolution. John Wiley & Sons: Chichester. ISSN 2045-7758; e-ISSN 2045-7758, meer
|   |
| Author keywords |
Benthic species; count data; generalized linear spatial modeling; spatial correlation |
| Auteurs | | Top |
- Lyashevska, O., meer
- Brus, D.J.
- Van der Meer, J., meer
|
|
|
| Abstract |
The objective of the study was to provide a general procedure for mapping spe-cies abundance when data are zero-in?ated and spatially correlated counts. Thebivalve species Macoma balthica was observed on a 5009500 m grid in theDutch part of the Wadden Sea. In total, 66% of the 3451 counts were zeros. Azero-in?ated Poisson mixture model was used to relate counts to environmentalcovariates. Two models were considered, one with relatively fewer covariates(model “small”) than the other (model “large”). The models contained twoprocesses: a Bernoulli (species prevalence) and a Poisson (species intensity,when the Bernoulli process predicts presence). The model was used to makepredictions for sites where only environmental data are available. Predictedprevalences and intensities show that the model “small” predicts lower meanprevalence and higher mean intensity, than the model “large”. Yet, the productof prevalence and intensity, which might be called the unconditional intensity,is very similar. Cross-validation showed that the model “small” performedslightly better, but the difference was small. The proposed methodology mightbe generally applicable, but is computer intensive. |
|
