boosted regression tree

Remote sensing data provide a high ability to represent habitat characteristics and use in species distribution models. The purpose of this study is to determine the most important of remote sensing predictors, including climatic indices (precipitation and temperature), primary and secondary topographic indices (elevation, slope, direction, Topographic Position Index (TPI) and Terrain Roughness Index (TRI) and remote sensing indices (Global Environmental Monitoring Index (GEMI), Leaf Area Index (LAI), Modified Normalized Difference Water Index (MNDWI), Modified Simple Ratio Index (MSR), Normalized Burn Ratio Index (NBR) and Visible Atmospherically Resistant Index (VARI)) using two species distribution models (Boosted Regression Tree and Random Forest) to predict the presence of J. excelsa in Khalkhal County of Ardabil province and northern part of Zanjan province using SAHM software. The evaluation of prediction models using AUC chart (Area under curve) showed that it is at an excellent level for both the BRT model (0.991) and the RF model (0.974). The most important affecting habitat desirability based on the BRT method include annual precipitation, slope, digital elevation model, temperature, GEMI index and TRI index variables respectively. The most important variables affecting habitat desirability based on the RF method, respectively, include annual precipitation, digital elevation model, GEMI index, slope, VARI index, temperature, MSR index, TRI index, TPI index, NBR index, MNDWI index, LAI index and aspect. The region mapped in the study as suitable habitats’ for the species could be used in the planning strategies with the aim of evaluating the susceptible habitats, the possibility of conservation, reproduction and breeding. Considering that the modeling method choice is the main source of variability in predictions and choosing the best prediction model is not simple, therefore, it is suggested to use a combination of these models instead of relying on the outputs of a single model.

Present and future distribution status of important forest species needs to be studied as a prerequisite for the need of conservation programs. This thus can play an undeniable role in tackling global climate change, land use change and air pollution. The main purpose of this study was to determine the distribution of oriental beech (Fagus orientalis Lipsky) under the influence of topographic factors in forests of northern Iran. In so doing, the present distribution status of this species in the 51 watershed of Mazandaran province was determined. Also, the relationship between the main topographic factors (slope, direction, elevation) and sub-components factors (topographic moisture, surface curvature and cross-section curvature) was analyzed and modeled using boosted regression tree (BRT) model and R software package. Under the curve area statistics (AUC) and actual threshold-related skill statistics (TSS) were used to evaluate the model. Findings showed that the AUC and TSS were 0.82 and 0.63, respectively. This indicates that the model was able to predict the distribution of this species. The results also showed that altitude and slope percentages had the most influence on the distribution of this species and the distribution was less affected by secondary topographic variables such as topographic moisture, cross-section curvature and surface curvature. This species is more likely to occur in slopes of 20 to 60 percent, in the northern and eastern directions, and at altitudes of 500 to 1500 meters above sea level. The habitats located at this topographic location are more likely to be affected by future climate change scenarios. Therefore planning and implementing conservation plans for this species in northern forests of Iran seems necessary.

Identifying the flood susceptible areas is a vital and substantial element of disaster management to control and mitigate injuries of the natural hazards. The purpose of this study was to identify the important variables in creating flood areas and to present the potential hazard of flood in Golestan National Park (GNP) using machine learning techniques including random forest (RF), boosted regression tree (BRT) and maximum entropy (ME) models. In order to achieve these purposes, firstly, factors were determined by reviewing the relevant sources, and the databases were created by sorting out these factors. Finally, Flood risk modeling was done using machine learning techniques and the accuracy assessment were determined using the ROC method and real data recorded in nature. The results of the models showed the importance of elevation, distance from the river and transit road, moisture and maximum temperature variables in the event of flood hazard. So that the results of the BRT showed role elevation variable to 38.9%, mean temperature 19/2 %, Rainfall 13/6 % and distance from the rivers 13% and the results of ME showed role elevation, mean temperature and distance of road variables respectively 35.7, 22/4 and 13.8%. The results of the accuracy assessment models using 30% of the data that were not included in the modeling the ROC value showed BRT and RF model with 0.99 values, and the proper accuracy ME was with value of 0.89. Therefore, the maps obtained can be used to manage the flood crisis in the area