农牧交错带西段景观结构和空间异质性分析

农牧交错地区是生态环境问题突出的地区。运用景观学理论，选取典型区域通过具体的景观特征指数分析了西北农牧交错景观的景观结构和景观空间异质性。结果表明地形地貌和人类作用是西北农牧交错区景观异质性的主导因素，地貌越破碎人类作用越强，景观也越破碎，斑块不规则性越高，景观多样性越高。进一步的景观样带分析显示，研究区景观结构呈一定的空间梯度变化，由南向北景观破碎度减小，斑块形状的不规则性降低。

The Landscape Structure and Heterogeneity of the West Part of Ecotone Between Agriculture and Animal Husbandry

The transitional region between cropland in semi\|humid regions and grassland (pasture) in semi\|arid regions, is an important ecological barrier, which prevents the desert from extending into the plains of eastern China. It is also a region with severe environmental problems such as desertification, vegetation degradation and soil salinization due to its natural fragility enhanced by human activities. These problems directly affect the environment of central and eastern China. In this paper we analyze the landscape structure and spatial heterogeneity in selected districts within the transitional region, and study the relationship between landscape structure, natural features and human activities. By applying remote sensing and GIS techniques, which are efficient ways of tracing long\|term inapparent environmental changes, we analyzed landscape structure and spatial heterogeneity based on a series of land\|use/land\|cover data extracted from Thematic Mapper and Spot images. Five landscape districts were chosen for analysis: 1. a forest\|shrub\|steppe district in the Helan Mountains, 2. an irrigated cropland district in the Yinchuan plains, 3. a desert\|steppe district in the Yanchi, Tongxin and Xiangshan Hills, 4. a nonirrigated farmland\|steppe district in southern Ningxia, and 5. a forest\|meadow district in the Liupan Mountains. Landscape metrics were calculated using Fragstats 3.0, and many indices are included in the soft such as patch density, edge density, landscape shape index, etc.Patch density ( PD ), edge density ( ED ), and landscape shape index ( LSI ) are lower in district 4 than in district 3 due to the loess topography of district 4, even though both districts are dominated by grassland and cropland and have similar human activities. Due to the loess topography of the forest\|meadow district of the Liupan Mountains, PD, ED and LSI are greater here than in the similar forest\|shrub\|steppe district of the Helan Mountains. In the Helan district, the landscape structure is affected by large elevation differences and less human activities, causing the contagion index to be higher than in the Liupan district. In the irrigated cropland district of the Yinchuan plains, which is greatly influenced by human activities, PD is higher than distract 1 and 3, while ED and LSI are lower than distract 4 and 5. The human activities are the main influence. Using GIS and Ningxia statistical data, we selected four natural factors (temperature, precipitation, altitude, and slope) and four human activity factors (population density, carrying capacity, irrigation channel density, and road density) to determine their influence on these five distracts. Precipitation and slope not only determine landscape structure but also the human manner in these distracts. VAGUE In addition, comparing PD and LSI with precipitation and slope along a landscape transect, the regular gradient change along the transect reflected the relation between landscape indices and natural factors and human activities. In the fragmentized loess area when the precipitation is higher, the nonirrigated agriculture is more developed, so the PD and LSI are higher. In the Yinchuan plain, the precipitation is less, but the flat land, human activities are drastic, also the PD and LSI are higher. Natural features including geology, topography, hydrology and soil, and human activity were considered to be the determining factors of landscape spatial heterogeneity. Through comparing natural features and some features of human activity with the landscape indices, we found that the determining factors in different landscape districts are not the same. In arid and semi arid area Human activity are limited by precipitation and slope, then influence landscape structure and spatial heterogeneity.