基于MaxEnt的农田恶性杂草雀麦的潜在分布预测

雀麦是中国农田一种恶性杂草,危害作物生长和发育。为控制其进一步扩散蔓延,根据目前雀麦在全球的分布记录和气象数据,利用最大熵模型Max Ent和Arc GIS对雀麦在中国和全球的潜在适生区域进行预测和分析。预测结果表明:雀麦在全球的适生区为非洲北部、欧洲的中南部、亚洲西部和东部及中部零星地区、北美中西部及东部部分地区、日本及朝鲜南部地区;在中国主要在黄淮海平原、长江中下游平原、成都平原、陕西中南部、山西南部、宁夏及甘肃南部、贵州中北部、新疆西南零星地区、西藏西南部及青海中部地区分布;中风险潜在分布区域的范围在高风险区的基础上进一步扩大。环境变量中对雀麦发生影响较大的是最冷季度平均温度、年平均气温、温度季节性变化标准差、最冷月最低温、最干季度平均温度。本研究的AUC值为0.943,模拟预测准确性极好,模拟结果可以用于雀麦的适生区预测研究。

Prediction of Potential Distribution of Japanese Brome (Bromus japonicus) Based on MaxEnt

Japanese Brome is a serious weed in farmland, which damages the growth and development of crops. In order to control the spread of Japanese Brome, MaxEnt maximum entropy model and ArcGIS were used to analyze and predict its potential geographical distribution in China and the rest of the world according to its distribution records and the meteorological data. The results showed that areas of western Africa, south central Europe, western and eastern and parts of middle Asia, mid-western and parts of eastern North America had high invasion risk of Japanese Brome. In China, areas of Huang-huai-hai plain, plain in the middle and lower reaches of Yangtze River, Chengdu plain, south central Shaanxi, southern Shanxi, southern Ningxia and Gansu, north central Guizhou, parts of southwest Xinjiang, southwest Tibet and central Qinghai had high invasion risk of Japanese Brome. The scope of secondary risky areas was expanded further on the basis of the high-risky area. The important environmental factors influencing the occurrence of Japanese Brome were mean temperature of the coldest quarter, the annual average temperature, temperature seasonal variation standard deviation, the lowest temperature in the coldest month and the mean temperature of the driest quarter. In this study, the AUC value was 0.943, which indicated that the accuracy of the simulation forecast was good, and the simulation results could be used for the prediction research of the suitable distribution areas of Japanese Brome.