不同环境类型对华北落叶松分布的潜在影响

植物分布-环境因子之间的关系是生态学、生物地理学研究中一个核心问题。采用最大熵物种分布模型，选取气候、土壤和地形3类环境类型中33个因子，对气候、气候-土壤、气候-地形、气候-土壤-地形4种环境组合对华北落叶松在河北省、山西省和内蒙古自治区的分布预测进行建模和检验。利用ArcGIS空间统计，划分华北落叶松分布适宜区，并制作分布适宜性等级图。结果表明，不同环境类型组合对华北落叶松分布影响的训练集和检验集的AUC值在0.965-0.983之间，均达到极准确的精度水平。在影响华北落叶松分布的主导因子上，气候、气候-土壤两种环境类型组合均为最热月的最高温度、温度的年较差和季节性温度变异，累计贡献率均达到74%以上，而气候-地形、气候-土壤-地形两种环境类型组合中海拔和坡度的影响最大，分别为48.8%和51.8%。在影响华北落叶松适生区（中、高、极高适宜区）面积上，气候、气候-地形两种环境类型组合差异不大，分别为102583km²和100698 km²，而气候-土壤、气候-土壤-地形两种环境类型组合影响下的华北落叶松适生区面积出现显著下降分别为57134 km²和66754 km²。最大熵模型能够很好的预测华北落叶松分布，地形因子能明显改变单一气候因子对华北落叶松分布的预测结果，虽然土壤因子对落叶松分布格局的影响不大，但在适宜性、尤其是中等以上适宜区分布上，其影响作用显著。以上结论可为华北地区的生态修复和落叶松可持续经营提供依据。

Potential impacts of environmental types on geographical distribution of Larix principis-rupprechtii

The relationship between plant distribution and environmental factors is a core issue in ecology and biogeographic research. Maximum entropy models were used for predicting potential distribution suitability for L. principis-rupprechtii, in Hebei, Shanxi and Inner Mongolia province. Thirty-three environmental factors were chosen from climate, soil and terrain. The climate, climate-soil, climate-terrain and climate-soil-terrain four environmental combinations were modeled and tested. Suitable area of the distribution of L. principis-rupprechtii was divided and mapped by ArcGIS spatial statistics. The results showed that the maximum entropy models were successful at discriminating between suitable and unsuitable habitat at the local level for all environmental combination types, and the Area under curve (AUC) values were from 0.965 to 0.983 for calibration and test data with the excellent effect. In the dominant factors affecting the distribution of L. principis-rupprechtii, the highest temperature of the hottest month, the annual temperature difference and seasonal temperature variation were important, and the cumulative contribution rate was above 74% in the environmental types of climate only and climate-soil. The effects of altitude and slope were the highest in the environmental types of climate-terrain and climate-soil-terrain, with 48.8% and 51.8% respectively. In the area affecting the suitable distribution of L. principis-rupprechtii (middle, high, and extremely high-adapted areas), the environmental types of climate and climate-terrain had little difference, with 102583 km² and 100698 km², respectively. While, the environmental types of climate-soil and climate-soil-terrain, under the influence of the suitable area was significantly reduced to 57134 km² and 66754 km², respectively. The maximum entropy model could reliably simulate the potential distribution area of L. principis-rupprechtii. The terrain factors can significantly change the prediction results of single climatic factor on the distribution of L. principis-rupprechtii. Although the soil factors have little effect on the distribution pattern of larch, the influence is significant in the suitability, especially the distribution of suitable areas above medium. All results could be as a reference for ecology restoration and sustainable management in north China.