Comparative performance assessment of landslide susceptibility models with presence-only,presenceabsence,and pseudo-absence data

The quality of the data for statistical methods plays an important role in landslide susceptibility mapping.How different data types influence the performance of landslide susceptibility maps is worth studying.The goal of this study was to explore the effects of different data types namely,presence-only(PO),presence-absence(PA),and pseudo-absence(PAs) data,on the predictive capability of landslide susceptibility mapping.This was completed by conducting a case study in the landslide-prone Honghe County in the Yunnan Province of China.A total of 428 landslide PO data points were selected.An equivalent number of nonlandslide locations were generated as PA data by random sampling,and 10,000 sites were uniformly selected at random from each region as PAs data.Three landslide susceptibility models,namely the information value model(IVM),logistic regression(LR) model,and maximum entropy(MaxEnt) model,corresponding to the three data types were investigated.Additionally,the area under the receiver operating characteristic curves(ROC-AUC),seven statistical indices(i.e.accuracy,sensibility,falsepositive rate,specificity,precision,Kappa,and Fmeasure),and a landslide density analysis were used to evaluate model performance regarding landslide susceptibility mapping.Our results indicated that the MaxEnt model using PAs data performed the best and had the highest fitness with the highest ROC-AUC values and statistical indices,followed by the IVM model with only landslide data(PO),and the LR model using PA data.Using PAs data avoided the inherent over-predictive shortcomings of PO data by limiting the predicted area of high-landslide susceptibility.Additionally,the random sampling design of landslide PA data increased the uncertainty of landslide susceptibility mapping and influenced the performance of the model.Therefore,our results suggested that the PAs data sampling provided a useful data type in the absence of high-quality data.Finally,we summarized the principles,advantages,and disadvantages of the three data types to assist with model optimization and the improvement of predicted performance and fitness.

Landscape structure and suitable habitat analysis for effective restoration planning in semi-arid mountain forests

Background:Suitable habitat and landscape structure play a pivotal role in the success of forest restoration projects.This study aimed to model the habitat suitability of wild almond(Amygdalus scoparia Spach)using three individual species distribution models(SDMs),i.e.,backpropagation artificial neural network(BP-ANN),maximum entropy(MaxEnt),generalized linear model(GLM),as well as the ensemble technique along with measuring the landscape metrics and analyzing the relationship between the distribution of the suitable habitat of the species in different landform classes in Fars Province,southern Iran.Results:There was no clear difference in the prediction performance of the models.The BP-ANN had the highest accuracy(AUC=0.935 and k=0.757)in modeling habitat suitability of A.scoparia,followed by the ensemble technique,GLM,and MaxEnt models with the AUC values of 0.890,0.887,and 0.777,respectively.The highest discrimination capacity was associated to the BP-ANN model,and the highest reliability was related to the ensemble technique.Moreover,evaluation of variable importance showed that the occurrence of A.scoparia was strongly dependent on climatic variables,particularly isothermality(Bio 3),temperature seasonality(Bio 4),and precipitation of driest quarter(Bio 17).Analysis of the distribution of species habitat in different landform classes revealed that the canyon,mountain top,upland drainage,and hills in valley classes had the highest suitability for the species establishment.Conclusions:Considering the importance of landform in the establishment of plant habitats,the combination of the outputs of the SDMs,landform,and the use of landscape metrics could provide both a clear view of habitat conditions and the possibility of analyzing habitat patches and their relationships that can be very useful in managing the remaining forests in semi-arid regions.The canyon,mountain top,and upland drainage classes were found to be the most important landforms to provide the highest suitable environmental conditions for the establishment of A.scoparia.Therefore,such landforms should be given priority in restoration projects of forest in the study area.