Mapping habitat suitability for Asiatic black bear and red panda in Makalu Barun National Park of Nepal from Maxent and GARP models

Abstract

Habitat evaluation is essential for managing wildlife populations and formulating conservation policies. With the rise of innovative powerful statistical techniques in partnership with Remote Sensing, GIS and GPS techniques, spatially explicit species distribution modeling (SDM) has rapidly grown in conservation biology. These models can help us to study habitat suitability at the scale of the species range, and are particularly useful for examining the overlapping habitat between sympatric species. Species presence points collected through field GPS observations, in conjunction with 13 different topographic, vegetation related, anthropogenic, and bioclimatic variables, as well as a land cover map with seven classification categories created by support vector machine (SVM) were used to implement Maxent and GARP ecological niche models. With the resulting ecological niche models, the suitable habitat for asiatic black bear (Ursus thibetanus) and red panda (Ailurus fulgens) in Nepal Makalu Barun National Park (MBNP) was predicted. All of the predictor variables were extracted from freely available remote sensing and publicly shared government data resources. The modeled results were validated by using an independent dataset. Analysis of the regularized training gain showed that the three most important environmental variables for habitat suitability were distance to settlement, elevation, and mean annual temperature. The habitat suitability modeling accuracy, characterized by the mean area under curve, was moderate for both species when GARP was used (0.791 for black bear and 0.786 for red panda), but was moderate for black bear (0.857), and high for red panda (0.920) when Maxent was used. The suitable habitat estimated by Maxent for black bear and red panda was 716 km2 and 343 km2 respectively, while the suitable area determined by GARP was 1074 km2 and 714 km2 respectively. Maxent predicted that the overlapping area was 83% of the red panda habitat and 40% of the black bear habitat, while GARP estimated 88% of the red panda habitat and 58% of the black bear habitat overlapped. The results of land cover exhibited that barren land covered the highest percentage of area in MBNP (36.0%) followed by forest (32.6%). Of the suitable habitat, both models indicated forest as the most preferred land cover for both species (63.7% for black bear and 61.6% for red panda from Maxent; 59.9% black bear and 58.8% for red panda from GARP). Maxent outperformed GARP in terms of habitat suitability modeling. The black bear showed higher habitat selectivity than red panda. We suggest that proper management should be given to the overlapping habitats in the buffer zone. For remote and inaccessible regions, the proposed methods are promising tools for wildlife management and conservation, deserving further popularization.

Prediction of Suitable Distribution of a Critically Endangered Plant Glyptostrobus pensilis

“…In this study, MaxEnt 3.4.1 (https://biodiversityinformatics.amnh.org/open_source/ maxent/, last accessed on 1 July 2021) was used to simulate the potential suitable habitat areas of G. pensilis under the LIG, the LGM, MH, Current and four scenarios in the future (RCP2.6-2050s RCP2.6-2070s, RCP8.5-2050s and RCP8.5-2070s) [17,46]. In order to make the probability of occurrence close to a normal distribution, 75% of the data were selected for model training, the remaining data were used for model testing and other values were the default values [18][19][20][21]. The EMNeval [41,47] package in R (Cran) version 3.6.1 [48] was used to optimize the MaxEnt model and the regularization multiplier (RM) was set to 0.5-4, with intervals of 0.5 each and a total of 8 regulated frequency multipliers.…”

“…The maximum entropy (MaxEnt) [16] model is the most commonly used prediction model for the potential distribution area of species; it has a high prediction accuracy and good versatility [17]. This model is widely used to predict the distribution of invasive species [18], endangered animals and plants [19][20][21], natural disasters [22] and pests and diseases [23,24]. It can also be used for Wildfire Risk Assessment and Zoning [25].…”

Understanding Species–Habitat Associations: A Case Study with the World’s Bears

“…A wider landscape study, involving eight countries in the Hindu Kush mountains, employed presence points from multiple sources and explored the effects of 19 bioclimatic variables and 5 topographic or land use variables in a Maxent model; this study mapped the region of suitable habitat but did not provide specific information about what was considered most suitable, except in terms of elevation (1500-3000 m) [125]. Two Maxent models in a national park in Nepal, using the exact same presence data but employing different variables obtained similar range maps, but one used only variables related to topography and land cover [122], whereas the other used primarily bioclimatic variables and concluded that mean temperature was the strongest predictor of Asiatic black bear presence [126]. Given the wide latitudinal range of Asiatic black bears, from the tropics to the subarctic, it would seem doubtful that temperature per se significantly impacts where they occur, although the temperature was likely coarsely related to habitat features on the ground, especially sources of food.…”

Buffalo on the Edge: Factors affecting Historical Distribution and Restoration of <em>Bison bison </em>in the Western Cordillera, North America

“…MaxEnt has been widely used to estimate habitat suitability and thus predict the geographic distribution of a variety of species [29], including both European and North American bison [30,31]. The output of MaxEnt modelling provides an estimated probability of species occurrence based on environmental variables, and we join others in referring to this as synonymous to habitat suitability [29,32,33].…”