Projection of the potential distribution of suitable habitats for Siberian crane (Grus leucogeranus) in the middle and lower reaches of the Yangtze River basin

Abstract

Introduction: The middle and lower reaches of the Yangtze River basin (hereafter MLYRB) are one of the most important wintering grounds for waterbirds along the East Asian-Australasian Flyway (EAAF). Climate change and human activities have greatly altered the habitats of waterbirds in this region.Methods: The potential distribution of suitable habitats for the Siberian crane (Grus leucogeranus) in the MLYRB was projected using Maximum Entropy Models (MaxEnt) and three Global Climate Models (GCMs). Additionally, estimations for temperature and precipitation before 2060 were made using CMIP6 multi-model and multi-scenario data.Results: 1) the MaxEnt model was highly applicable (AUC = 0.939) for analyzing the suitable habitat distribution and climatic suitability of Siberian cranes in the study area; 2) Precipitation of the driest quarter and altitude were the main factors affecting the potential suitable habitat distribution of Siberian cranes, accounting for 40% of the total contribution rate each; 3) The climatically suitable areas for the distribution of suitable habitats of Siberian cranes in the study area was mainly concentrated in parts of Poyang Lake, Dongting Lake, Taihu Lake, and the mainstream of the Yangtze River; 4) In the BCC-CSM2-MR, CanESM5, and CNRM-CM6-1 models, the suitable habitat area for Siberian cranes is expected to decrease under different SSP scenarios from 2021 to 2060 compared to 1970–2000.Discussion: The BCC-CSM2-MR model’s SSP2-4.5 scenario shows the most significant decline in the suitable habitat area for Siberian cranes, with a maximum decrease of approximately 35.7%, followed by a maximum decrease of about 26.2% and 16.4% under the CNRM-CM6-1 and CanESM5 models’ SSP2-4.5 scenario, respectively. In general, the suitable habitat areas for Siberian cranes are projected to decline, indicating the need for comprehensive evaluation and uncertainty research using more models.