A Methodology for Selection of Habitat Management Areas for Amphibians and Reptiles Considering Soil Loss

토양유실을 고려한 양서파충류의 서식지 관리지역 선정방법

Abstract

As disaster risk and climate change volatility increase, there are more efforts to adapt to disasters such as forest fires, floods, and landslides. Most of the research, however, is about influence of human activities on disaster and there is few research on disaster adaptation for species. Previous studies focusing on biodiversity in selecting conservation areas have not addressed threats of disaster in the habitats for species. The natural disasters sometimes play role of drivers of ecological successions in the long run, but they might cause serious problems for the conservation of vulnerable species which are endangered. The purpose of this study is to determine whether soil loss (SL) is effective in selecting habitat management areas for amphibians and reptiles. RUSLE model was used to calculate soil loss (SL) and the distribution of each species (SD) was computed with MaxEnt model to find out the biodiversity index. In order to select the habitat management area, we estimated the different results depending if value of soil loss was applied or not by using MARXAN, a conservation priority selection tool. With using MARXAN, conservation goals can be achieved according to the scenario objectives, and the study has been made to meet the minimum habitat area. Finally, the results are expressed in two; 1) the result of soil loss and biodiversity with MATRIX method and 2) the result of regional difference calculated with MARXAN conservation prioritization considering soil loss. The first result indicates that the area with high soil loss and low species diversity have lower conservation values and thus can be managed as natural disturbances. In the area where soil loss is high and species diversity is also high, it becomes where a disaster mitigation action should be taken for the species. According to the conservation priorities of the second result, higher effectiveness of conservation was obtained with fewer area when it considered SL in addition to SD, compared to when considered only biodiversity. When the SL was not taken into consideration, forest area with high distribution of species were important, but when SL considered, the agricultural area or downstream of the river were represented to be a major part of habitats. If more species data or disaster parameters other than soil loss are added as variables later, it could contribute as a reference material for decision-making to achieve various purposes.

Figure 1. Study flow

Figure 2. Study site

Figure 3. Boundary Length Modifier (BLM) optimization

Figure 4. SL and Stream order

Figure 5. Species Diversity (Calculated with SD)

Figure 6. MATRIX result and areas of MARXAN results

Figure 7. Box plot graph of SL and Species Diversity

Table 1. Support practice factor (P factor) for each land use (Notification No. 2015-138 of the Ministry of Environment)

Table 2. Environment variables list applied to MaxEnt model

Table 3. Relative contribution of the environmental variables to the MaxEnt model (Rana dybowskii)

Table 4. Conservation goal value for each species