土地沙化和沙尘暴是中国北方农牧交错带严重的生态环境问题。在地理信息系统的支持下,论文建立风蚀自然因子栅格数据库;选取二连浩特—张家口为典型样带,综合运用模糊聚类技术、 专家知识以及神经网络技术,构建区域风蚀危险度评估模型,分析内蒙古自治区风蚀危险性空间分布格局。结果显示:综合应用模糊聚类以及神经网络技术可以有效地对风蚀危险度进行评价,内蒙古自治区风蚀危险性呈现"西高东低、 中间过渡"的总体趋势,这与该区域的自然、 气候与土地利用背景较吻合,其中,风蚀极险型区域主要分布在阿拉善高原的西北部,面积6.9×104 km2;强险型土壤风蚀地区主要位于阴山山脉周围、 穿越巴丹吉林沙漠、 浑善达克沙地与锡林郭勒盟高原西北部,面积4.3×104 km2;危险型风蚀区域大致位于极险和强险之间,面积是1.5×105 km2;轻险型土壤风蚀区域主要位于内蒙古的中部,面积为7.5×105 km2;风蚀无险型区域主要位于大兴安岭周围,面积1.4×105 km2。
One of the most serious environmental problems in Farming-Pastoral Ecotone of Northern China is land desertification and sand storm. With the support of geographic information system, this paper establishes a natural factor database and takes Erenhot-Zhangjiakou as a typical transect to build regional wind erosion hazard assessment model using fuzzy clustering, expert knowledge and neural network technique, and then analyzes the spatial distribution of soil wind erosion hazard in Inner Mongolia; finally, the natural environment of spatial distribution of soil wind erosion hazard and land use background is studied according to land use type of the study area. Result shows: the integrated application of Fuzzy Clustering and Neural Network is an effective way to assess wind erosion hazard; the trend of wind erosion hazard in Inner Mongolia is increasing from east to west, which is consistent with the nature, land use and climate condition in the study area. The zone of severe hazard is mainly in the northwest Alashan with an area of 6.9×104 km2; the zone of intense hazard is mainly located around the Yinshan Mountains, northwest plateau of the Xilin Gol League, through the Badain Jaran Desert and Hunshandake, the area is 4.3×104 km2; the zone of moderate hazard is roughly located between the severe hazard and intense hazard, the area is 1.5×105 km2; the zone of slight hazard is mainly in the central Inner Mongolia, the area is 7.5×105 km2; and the zone of no hazard is located around Daxinganling with an area of 1.4×105 km2.
[1] 陈渭南, 董光荣, 董治宝. 中国北方土壤风蚀问题的研究进展和趋势[J]. 地球科学进展, 1994, 9(5): 6-12.
[2] 胡云锋, 刘纪远, 庄大方. 土壤风力侵蚀研究现状与进展[J]. 地理科学进展, 2003, 22(3): 288-295.
[3] 王涛, 朱震达. 中国沙漠化研究[J]. 中国生态农业学报, 2001, 9(2): 7-12.
[4] 刘连友, 王建华, 李小雁, 等.耕作土壤可蚀性颗粒的风洞模拟测定[J]. 科学通报, 1998, 43(15): 1663-1666.
[5] 刘玉璋, 董光荣, 李长治.影响土壤风蚀主要因素的风洞试验研究[J]. 中国沙漠,1992,12(4): 41-49.
[6] 胡云锋, 刘纪远, 庄大方, 等. 风蚀土壤剖面137Cs的分布及侵蚀速率的估算[J]. 科学通报, 2005, 50(9): 933-937.
[7] 刘纪远, 齐永青, 师华定, 等. 蒙古高原塔里亚特-锡林郭勒样带土壤风蚀速率的137Cs示踪分析[J]. 科学通报, 2007, 52(23): 2785-2791.
[8] 齐永青, 刘纪远, 师华定, 等. 蒙古高原北部典型草原区土壤风蚀的137Cs示踪法研究[J]. 科学通报, 2008, 53(9): 1070-1076.
[9] 王涛, 吴薇, 王熙章.沙质荒漠化的遥感监测与评估——以中国北方沙质荒漠化区内的实践为例[J]. 第四纪研究,1998(2): 108-118.
[10] 胡云锋,刘纪远,庄大方,等.土地利用动态与风力侵蚀动态对比研究——以内蒙古自治区为例[J]. 地理科学进展,2003,22(6): 541-550.
[11] 熊利亚,李海萍,庄大方.应用MODIS数据研究沙尘信息定量化方法探讨[J]. 地理科学进展, 2002,21(4): 327-332.
[12] 延昊,王长耀,牛铮,等.应用遥感技术估算东亚沙尘携沙量[J]. 水土保持学报, 2002, 16(1): 120-123.
[13] 师华定, 高庆先, 庄大方, 等. 基于径向基函数神经网络(RBFN)的内蒙古土壤风蚀危险度评价[J]. 环境科学研究, 2008, 21(5): 129-133.
[14] Shi H D, Liu J Y, Zhuang D F, et al. Using the RBFN model and GIS technique to assess wind erosion hazard of Inner Mongolia, China[J]. Land Degradation & Development, 2007,18: 413-422.
[15] 师华定, 高庆先, 齐永青, 等. 蒙古高原土壤风蚀危险度的FCM模糊聚类研究[J]. 自然资源学报, 2009, 24(5): 881-889.
[16] 汤懋苍, 高晓清. 蒙古高原地面风、 压、 温场的基本特征[J]. 气象学报,1996,54(1): 121-127.
[17] Fryrear D W, Saleh A, Bilbro J D. Revised Wind Erosion Equation [M]. Washington D C: USDA, 1998.
[18] 中国科学院自然区划工作委员会. 中国综合自然区划: 初稿[M]. 北京: 科学出版社, 1959.
[19] 刘明光.中国自然地理图集[M]. 北京: 中国地图出版社, 1998.
[20] 于剑.论模糊c均值算法的模糊指标[J]. 计算机学报, 2003, 26(8): 968-973.
[21] Li B L, English E, Zhu A X. Development of knowledge for predictive mapping using a fuzzy c-means classification//Liu M L, Chen G Q, Ying M S. Fuzzy Logic, Soft Computing and Computational Intelligence. Beijing: Tsinghua University Press, 2005: 1205-1209.
[22] 水利部.土壤侵蚀分级分类标准SL190-96[S]. 北京: 中国水利水电出版社,1998.
