Abstract
Species shift their ranges in response to climate change (CC). However, they may not be able to track optimal conditions as soon as possible, due to limited dispersal ability or habitat fragmentation, caused by land use and land cover change (LULC). This study aimed to explore the combined impacts of CC, LULC and dispersal limitations on the future range dynamics of Quercus acutissima Carruth., Q. variabilis Blume and Q. chenii Nakai, three dominant Cerris oak tree species in warm-temperate and subtropical deciduous forests of China. We used the Maximum Entropy (Maxent) algorithm to predict the suitable habitats for the years 2050 and 2070, under three representative concentration pathways (RCPs). Habitat fragmentation patterns were examined to assess the influence of LULC. Two migration scenarios (full- and partial-migration) were compared to evaluate the effect of dispersal limitations. We found that annual precipitation (AP), minimum temperature in the coldest month (MTCM) and temperature seasonality (TS) play a key role in determining the present distributions of Q. chenii, while AP, MTCM and annual mean temperature (AMT) contribute the most to the distribution models of Q. variabilis and Q. acutissima. For all the three species, LULC will increase the level of habitat fragmentation and lead to the loss of core areas, while limited dispersal ability will restrict the accessibility of future potentially suitable habitats. Under the scenarios of CC and LULC, the suitable areas of Q. chenii will decrease sharply, while those of Q. variabilis in South China will become unsuitable. Our findings highlight the importance of considering dispersal ability, as well as land use and land cover change, for modeling species’ range shifts in the face of global warming. Our study also provides vital information for guiding the management of East Asian Cerris oaks in China; Q. chenii should be listed as a species requiring priority protection, and the threatened habitats of Q. variabilis should be protected to buffer the impacts of CC and LULC.
Highlights
Global climate change (CC) has become one of the major threats to biodiversity and conservation [1,2]
The species distribution models (SDMs) only considering climatic variables showed that annual precipitation (AP), minimum temperature in the coldest month (MTCM) and annual mean temperature (AMT) were the most important variables for predicting the distributions of the two widespread species, Q. variabilis and Q. acutissima (Figure 2c)
Our research found that Q. acutissima and Q. variabilis are mainly affected by AMT, AP and MTCM
Summary
Global climate change (CC) has become one of the major threats to biodiversity and conservation [1,2]. Given that climate has been documented to play a key role in determining largescale species distributions [8,9], ecologists often use species distribution models (SDMs) to quantify the relationships between existing occurrence records and climatic factors through multivariate algorithms [5] These models can generate a prediction for species’ future habitat suitability, providing us with a useful tool for evaluating habitat vulnerability in the face of global warming [10,11]. More researchers have noticed this issue and tried to use various modeling approaches to incorporate migration scenarios into SDMs, for example, future projections for fir species in Southwest China and amphibians in the Himalayas [14,15] These predictions took species’ dispersal abilities into account and improved the accuracy of SDMs, which will guide the conservation of endangered species more reliably than the models only considering full-migration or no-migration scenarios
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
