bio_17 = precipitation of driest quarter, dred = dredging Village_dist = distance to village road_dist = distance to road The performance of logistic model was excellent (AUC = 0.93; Fig.1). In addition, the overall prediction accuracy was 85.77%. The predicted distributions of Water Onion cover an area of 395.07 km2 or 0.08% of the study area (Fig.2). Suitable areas were located in Mueng Ranong, Kraper, Suksamran, Kruraburi, Takaopha Kapong, Thaimeung districts. Fig.2 The distribution of Water Onion in Thailand by logistic regression 93.72 logistic thresholds (85.77). The logistic value of 0.31 was used for binary classification. The likely distributions for Water Onion cover an area of 126.08 km2 (Fig.3). Fig.3 The distribution of Water Onion in Thailand by MaxEnt The result revealed that the distributions of Water Onion generated by the maximum entropy (MaxEnt) model had overall prediction accuracy of 94%, which was greater than the map generated by the logistic regression (accuracy of 86%). The AUC value of MaxEnt model was 0.97 and its overall prediction accuracy was 93.72%. Among the 17 environment factors, annual precipitation made the highest percentage contribution to spatial distribution model, followed 3.2 Maximum entropy (MaxEnt) The probability of distribution of Water Onion was classified by using the equal training sensitivity and specificity logistic threshold because it provided the highest accuracy among Proceedings of the International Conference on Climate Change, Biodiversity and Ecosystem Services for the 33 Sustainable Development Goals (SDGs): Policy and Practice 27-29 June 2016, Cha-am, Phetchaburi, Thailand
Proceedings of International Conference on Climate Change, Biodiversity and Ecosystem Services for the Sustainable Development Goals : Policy and Practice 27-29 June 2016 at the Sirindhorn International Environmental Park, Cha-am, Phetchaburi, Thailand
To see the actual publication please follow the link above