Planning priority conservation areas for biodiversity under climate change in topographically complex areas: A case study in Sichuan province, China

with future conservation networks change may add additional challenges. To reduce the impacts of uncertainties, the existing nature reserves were also taken as a protection target so that the conservation network could meet the requirements of both current conditions and future scenarios. The goals for the percentage of nature reserves were defined as follows: 90% for national nature reserves, 80% for provincial nature reserves, and 70% for other levels. Planning priority conservation areas primarily aims to reach the highest ecological benefits from minimum cost. It was desired to avoid in urban, farmland, and road net, while empha- sized the effect of topographic complexity. We calculated two different conservation costs in each planning unit. One is a combination of land use and the effect of topography, and the other one is land use only. For combination cost, we set the weight of 1 for two types, as shown in Eq (2). Table 1 displays land use as conservation costs within this study. C ¼ Clanduse þ ð1 AÞ � 100 ð2Þ Where, C is a total of conservation cost, Clanduse is land use only, and A is the adaptation index (Eq (1)). Additionally, Sichuan province was divided into 25,181 planning units (20 km2 each). The conservation species information, nature reserves information and cost information were cal- culated in each planning unit. For the parameters in the model, we set to run 100 times to cal- culate the irreplaceability score for each planning unit, which represents the number of counts

selected in the optimal solution. Once the score was more than 90, this unit was selected as pri- ority conservation areas. To optimize the spatial distribution, Boundary Length Modifier (BLM) was used to determine optimum value, which was set 0.005 based on the Marxan Good Practice Guide for a compact spatial distribution. Besides the BLM value, other parameters were set the default values in the Marxan model. 3. Results 3.1. Predicting changes in species distributions Fig 4A–4C summarized how the sensitive species distributions would shift based on the IPCC 5 climate model data. After firstly excluding the construction and farmland land, the number of species in each land unit was calculated in three periods. In detail, areas of the highest spe- cies richness (equal and more 10) comprised the least area under current climatic conditions (3% of the study area), while mainly distributed in and around Chengdu plain. As climate change, these areas were more abundant (10% and 11% of the study area). Spatially, the areas of biodiversity richness would concentrate on topographic complexity under current and future climatic conditions, which located in the transition from plain to plateau. Additionally, future biodiversity distributions also demonstrated that current nature reserves did not meet Table 1. The conservation costs. Land use type Cost Land use Cost Forest 5 Sand 40 Shrub 5 Reservoir 50 Grass 5 Barren land 50 Other forest 15 Paddy 65 Lake 20 Settlements 90 Marsh 30 Construction land 90 Snow 30 Urban 100 farmland 40 Others 80

https://doi.org/10.1371/journal.pone.0243425.t001 PLOS ONE Planning priority conservation areas for biodiversity under climate change in topographically complex areas PLOS ONE | https://doi.org/10.1371/journal.pone.0243425 December 23, 2020 7 / 15