Dominant climatic factors of Quercus mongolica geographical distribution and their thresholds

Abstract

Relationship between plant species distribution and climate is always one of the most important issues in the ecology study. Plant species distribution at regional scale is largely controlled by climate. The dominant climatic factors influencing plant species distribution can be obtained by the relationship between climate and plant species distribution. And the dominant climatic factors are very important for revealing the reason of the formation of plant species distribution, determining the appropriate distribution zone of plant species and evaluating the responses of plant species to climate change. Quercus mongolica is the dominant tree of deciduous broad-leaved forest, conifer-broadleaved mixed forest in the temperate region of China. In order to provide the scientific support for Q. mongolica forest management and countermeasures to cope with future climate change in China, we have to make sure what the dominant climatic factors of its distribution are. But studies in this area are so few that restrict the understanding of the response to climate change and Q. mongolica forest scientific operation. Thus, the potential distribution of Q. mongolica in China and its relationship with climatic factors were studied in this paper. Based on the geographical distribution data of Q. mongolica and six climatic variables potentially controlling plant functional types'distribution including annual precipitation (P), annual range of monthly mean temperature (DTY), temperature of the coldest month (Tc), temperature of the warmest month (Tw), accumulated temperature of not less than 5°C(GDD5) and moisture index (MI), and their contribution to geographical distribution of Q. mongolica from the maximum entropy (Maxent) model. The area under the ROC curve (AUC) was used to evaluate the accuracy of model prediction. The AUC value reached 0. 932, and the high value indicated predictive accuracy of model achieved the "Excellent"level. Combined with the contribution of each climatic factor as mentioned above, four dominant climatic factors controlling Q. mongolica distribution were determined, including annual precipitation (P), annual range of monthly mean temperature (DTY), temperature of the warmest month (Tw) and accumulated temperature of not less than 5°C(GDD5). Thus, a relationship between the geographical distribution of Q. mongolica and dominant climatic variables was reconstructed. The thresholds of four dominant climatic factors were given by the relationship between the geographical distribution probability of Q. mongolica and dominant climatic factors, i. e. 330mm≤ P≤910mm, DTY≥29°C, 1200°C·d≤GDD5≤3500°C·d, 17°C≤Tw ≤26°C. Then the potential distribution of Q. mongolica in China was given by the ArcGIS spatial analysis technique. The potential distribution area of Q. mongolica included Heilongjiang province, Jilin province, Liaoning province, eastern Inner Mongolia autonomous region, Beijing, Tianjin, north-central of Hebei province, central part of Shandong province, most parts of Shanxi province and northeast of Shaanxi province. The total area covered 1002 points of the 1013 distributed points of the actual distribution of Q. mongolica. The simulation accuracy was more than 98%, indicating that the simulated distribution of Q. mongolica was very coincidence with the actual distribution. The results might provide important basis for making scientific management of Q. mongolica forest and developing the countermeasures to deal with future climatic change.