Prediction of the Potential Spatial Distribution of Paeonia obovata in Korean peninsula under SSP scenarios

This study carried out predicting the future geographical distribution of Paeonia obovata using MaxEnt (Maximum Entropy Model) under the SSP scenarios. Paeonia obovata, which is considered as a critically endangered species (CR) on the Korean Red List, is considered to be vulnerable to climate change. We aim to identify the factors affecting the habitat distribution and assess the climate change impact every 20 years in 2020s (2021~2040), 2040s (2041~2060), 2060s (2061~2080), and 2080s (2081~2100) with the base periods (2010~2019).

Maxent requires occurrence data of species and their environmental variables for input data. The occurrence data of Paeonia obovata were obtained from GBIF (Global Biodiversity Information Facility). Bioclimatic variables(bio1~bio19), soil variables (top soil texture, soil series, effective soil depth, drainage classes, and soil erosion class), and topographical variables (DEM, aspect, hillshade, and slope) were used as the environmental variables.

Based on the results obtained through the initial model, variables whose contribution was less than 1 were removed. Subsequently, variables with high Pearson correlation (|r|>0.8) were removed. As a result, it was found that bio8 (Mean Temperature of Wettest Quarter), bio13 (Precipitation of Wettest Month), bio15 (Precipitation Seasonality), bio19 (Precipitation of Coldest Quarter), top soil texture, soil series, and slope were significantly contributed to the distribution of Paeonia obovata. The percent contribution was high in bio8 (56.1%) and soil series (35%) and permutation importance was bio8 (32.4%), soil series (23.6%), bio15 (19.6%), and bio 19 (15.3%), respectively.

In order to understand the impact of climate change, suitability was classified into 5 grades (LL; 0-0.2, LM; 0.2-0.4, MM; 0.4-0.6, MH; 0.6-0.8, HH; 0.8-1) using an equal interval approach. As a result of comparing the area of HH (very high suitability area), the average decreasing rate was 0.81% under SSP1-2.6 scenario, whereas it decreased to 0.99% under SSP5-8.8 scenario. Therefore, the suitable habitat area decreased at a faster rate under the SSP5-8.5 than under the SSP1-2.6 scenario.

These results highlighted that climate change can have a significant impact on the potential spatial distribution of Paeonia obovata.

※ This work was supported by Korea Environment Industry &Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime.", funded by Korea Ministry of Environment (MOE) (No. 202201344001)