北温带地区小叶锦鸡儿的地理分布模拟

Geographical Distribution Simulation of Caraganamicrophylla in the North Temperate Zone

摘要:

在北温带地区收集了536个小叶锦鸡儿(Caragana microphylla)的地理分布数据,利用WorldClim V2.0提供的19个气候变量,通过主成分分析和5种模型(最大熵模型、随机森林模型、广义增强模型、人工神经网络模型和决策树模型)分析了小叶锦鸡儿的地理分布及其与气候因子的关系,并模拟了其分布格局。结果显示:(1)小叶锦鸡儿的地理分布主要受温度和降水的日、季节变化以及生长季的温湿度控制;(2)最大熵模型对小叶锦鸡儿分布的模拟结果最好,随机森林模型和广义增强模型次之(AUC＞0.9);(3)模型准确模拟出小叶锦鸡儿在甘肃省中部、宁夏回族自治区和内蒙古自治区中部和南部以及蒙古国中部和东部等地区的分布概率较大,未来应重点关注以上区域小叶锦鸡儿的分布动态。本研究可为小叶锦鸡儿在未来气候变化下的地理分布预测和灌丛化草原生态系统评估提供数据和模型支持。

Abstract:

The presence data of 536 Caragana microphylla were collected in the north temperate zone. By making use of 19 climate variables provided by Worldclim V2.0, principal component analysis, and five species distribution models such as Maximum Entropy Model, Random Forest, Generalized Boosting Model, Artificial Neural Network, and Classification Tree Analysis, the geographical distribution of Caragana microphylla and its relationship with climate factors were simulated. The results show that: (1) the geographical distribution of Caragana microphylla was mainly controlled by the diurnal and seasonal changes of temperature and precipitation as well as the growing-season temperature and humidity;(2) the Maximum Entropy Model was the best to simulate the distribution of Caragana microphylla,followed by the Random Forest and the Generalized Boosting Model (AUC＞0.9); (3) the model accurately simulated the distribution of Caragana microphylla,which had a large probability in the middle of Gansu Province, the middle and south of Ningxia Hui Autonomous Region, and Inner Mongolia Autonomous Region of China as well as the middleand east of Mongolia. Our results suggest that more attention should be paid on the distribution dynamics of Caragana microphylla in the above regions in the future. This study can provide data and model support for the geographical distribution prediction of Caragana microphylla under future climate change and the assessment of shrubbery grassland ecosystem.