Wenyun's Homepage

Please click here to download GeoSVM, including manual (in folder "doc"). GeoSVM is a species potential distribution predicting system. Here is the abstract of our study.

Abstrat:

Most predictive models of species potential distributions are based on environmental variables, because they are potentially important niche dimensions. Unfortunately, most predictive models suffer from the ^high dimension small sample size ̄ problem!they cannot give satisfactory results when there are only limited specimen data, and cannot handle large number of environment factors. Support Vector Machine (SVM), which is based on the Structural Risk Minimization principle, especially suitable for these kinds of data. Here, we implement a new predictive system, Called GeoSVM, for modeling species potential distributions based on SVM methods. To evaluate the effectiveness of the method, we perform a country-scale case study using 30 species of Rhododendron L. in China, with geographically referenced specimen data and 11 layers of 1km resolution environmental data. Here we report three results. First, using expert evaluation and Receiver Operator Characteristic (ROC) curves, we compare the SVM with the commonly used Genetic Algorithm for Rule-Set Prediction (GARP). Our results show that all GeoSVM predictions are consistently better than GARP. Furthermore, GeoSVM runs much faster than GARP.  

Second, we investigate GeoSVM performance with different numbers of environmental layers and specimen localities. We show that, 1) the more environmental layers available, the better predictions SVM gives; 2) even for species with few records, SVM can produce reasonable potential distribution maps. Therefore, GeoSVM can predict potential distributions of rare species. 

Finally, we use GeoSVM to predict potential distribution of 298 Rhododendron species in China. We use all available Chinese specimens of Rhododendron L. and 83 layers of 1km resolution environmental data. Based on potential distribution of these species, we quantity the spatial patterns of species diversity, endemic species diversity, endangered species diversity, subgenus diversity, and different life styles diversity. These distribution maps of biodiversity not only quantity the biogeography of Rhododendron, but also provide valuable information for conservation, reintroduction, and new species discovery.

Performance:

GeoSVM is designed for high speed prediction of species geological distributions. With a 3.0GHz Intel® processor, 81 environment layers (each of the size 6869*4012), and 1000 specimen data, GeoSVM can produce training data in less than 10 seconds, train SVM model in less than 1 second, and produce a 6869*4012 grid prediction in about 10 minutes. This speed is ideal for handling a lot of large scale experiments.

Technical contacts: Wenyun Zuo or Ni Lao

 

 

 

 

© Wenyun Zuo 10/07/2007