中国北方农牧交错带是一个典型的受气候和人类活动共同影响的敏感区域，存在严重的土壤风蚀和土地退化问题。风蚀模型是目前获得区域风蚀模数的最有效方法之一。利用修正风蚀方程（RWEQ模型）和风蚀预报系统（WEPS模型）对北方农牧交错带2000—2012年潜在风蚀进行评估。结果表明：两个模型模拟得到的多年平均潜在风蚀量不同，但空间分布、年际减少趋势和季节分布等特征基本相似；风速、土壤湿度和土地利用变化对土壤风蚀均有影响。RWEQ模型（R²=0.45，P<0.01）和WEPS模型（R²=0.57，P<0.01）中实测值与预测值具有较好的相关性。WEPS模型（NSC=0.54）纳什系数较RWEQ模型（NSC=0.27）高。RWEQ模型和WEPS模型均能客观预测北方农牧交错带土壤风蚀情况，WEPS模型预测精度较好。

The agro-pastoral ecotone in northern China， a typical region sensitive to interaction between climate and human beings， has always been undergoing wind erosion and soil degradation. At present， the wind erosion model is one of the most effective methods to obtain the regional potential wind erosion. In this study， the regional versions of Revised Wind Erosion Equation （RWEQ） and Wind Erosion Prediction System （WEPS） were used to evaluate the potential wind erosion at the agro-pastoral ecotone in northern China during 2000-2012. The results demonstrate that the magnitudes of average potential wind erosion were different while the spatial distribution， interannual decreasing trend and seasonal distribution of potential wind erosion were similar between the RWEQ and WEPS. It is found that wind speed， soil moisture and land use change have an impact on soil wind erosion by the analysis of the spatial correlation between wind speed， soil moisture and wind erosion modulus， and the influence of land use change on fluctuation of wind erosion amount. The observed wind erosion data are closely related to the values predicted by RWEQ （R²=0.45， P<0.01） and WEPS （R²=0.57， P<0.01）. The Nash （Sutcliffe efficiency coefficient） of WEPS model （NSC=0.54） is higher than that of RWEQ model （NSC=0.27）. Both RWEQ model and WEPS model can objectively predict the soil wind erosion of the APEC， and the prediction accuracy of WEPS model is better.