浙闽樱桃地理分布模拟及气候限制因子分析

Modeling the Geographical Distribution Pattern and Climatic Limited Factors of Cerasus schneideriana

为了解浙闽樱桃(Cerasus schneideriana)地理分布特征及与气候限制因子之间的关系,基于DIVA-GIS平台获取实际地理分布点的气候资料,利用BIOCLIM模型预测当下适生区范围以及预测其未来潜在分布。结果表明,浙闽樱桃实际分布区覆盖浙江、福建、安徽、江西、广西及湖南6省,浙闽交界的山区是自然分布的核心区,浙皖交界是其分布的北界。未来气候变化情境(CCM3)下,浙闽樱桃的潜在分布区概率将增大,且有北扩的趋势。主成分分析(PCA)表明,年降水量(bio12)、最湿季降雨量(bio16)、最暖季降雨量(bio18)、温度季节变化方差(bio4)是影响浙闽樱桃当下适生区的气候限制因子,频率直方图进一步确定他们的适宜范围分别为1503~2003 mm、604~951 mm、528~791 mm和601~872(标准差*100)。ROC曲线检验表明BIOCLIM对浙闽樱桃分布预测的精度很高(AUC=0.998),结果是可信的。因此,浙闽樱桃适合生长在我国北亚热带温暖湿润区,水热因子是造成其分布格局的主导气候限制因子。

To understand the relationship between geographical distribution pattern of Cerasus schneideriana and its climatic limited factors,the DIVA-GIS software and BIOCLIM model were applied to predict the present and future potential spatial areas on the basis of actual distribution points.The results showed that the actual distributions of C.schneideriana covered6provinces of east China,including Zhejiang,Fujian,Anhui,Jiangxi,Guangxi and Hunan.The natural distribution core area located in mountain areas of Zhejiang-Fujian border,and the north boundary arrived at Zhejiang-Anhui border.Under future climate change situation(CCM3),the probability of potential areas of C.schneideriana would increase,and had a tendency to the north expansion.The principal component analysis(PCA)indicated that the annual precipitation(bio12),precipitation of the driest quarter(bio6),precipitation of the warmest quarter(bio18)and temperature seasonality(bio4)were the dominant factors for geographical distribution of C.schneideriana.The frequency histograms further showed the factor's optimum range were1503~2003mm,604~951mm,528~791mm and601~872(SD*100),respectively.Evaluation by the ROC curve proved the BIOCLIM model predicted the distribution accurately(AUC=0.998)and reliably.Therefore,these revealed that C.schneideriana were suited to live in north subtropical region of China,with warm and humid condition,water and heat were the two key climatic factors to the distribution pattern.