Abstract：The loess hilly region is characterized topographically by steep slopes and highly dissected terrain, with sharp ridges and deep gullies. Population growth and growing food demand led to increasing pressure on the limited agricultural land, which caused conversion of grassland into agricultural land in the steep-hill slope; there is a growing need to assess conservation effectiveness of agriculture and grassland especially in semiarid loess hilly region. The study has been carried out in the Anjiagou small watershed, Dingxi County, Gansu province. At plot scale, an experiment investigated upon soil erosion of cultivated land and grassland on gentle-hill slope (10-20°) was performed over a period of 1986-1999. The results showed that the magnitude of soil erosion of cultivated land increases with rising of the slope, and there are significant differences among three slope plots; the annual soil erosion amount from plots of grassland varies with slope too, however, there is not a significance difference between plots. A considerable variation in soil erosion among plots indicates that cultivated land differed significantly in response to slope; there is significant difference in runoff and soil sediment between the cultivated land and grassland. Grassland provided good protection against surface runoff (which was consistently less than 2% of rainfall) and soil erosion (less than 27 g·m-2 per year), cultivated land caused twice increase in runoff and 14 times increase in soil erosion. The heavy soil erosion occurred in plots of cultivated land were mainly contributed to a few storm or rainfall events, therefore, there are not direct correlation between yearly runoff, soil erosion and yearly precipitation (total precipitation and runoff-generation rainfall) , whereas, in the plots of grassland, there is a strong correlation between annual runoff, soil erosion with runoff-generation rainfall and yearly precipitation, respectively. The relative contributions to surface runoff and soil erosion among rainfall variable were compared, examined that the precipitation ( P), maximum 30 minutes rainfall intensity ( I30 ) and PI30 are major factors and have positive correlation with runoff and soil erosion of rainfall events. The compound index (PI30) and I30 was used o quantify the relation of rainfall versus runoff and erosion, which can be predicted the runoff and erosion yield by the regression equation. The results of the study led us to the following conclusion: all plots of grassland were significantly lower in water and soil loss and more efficient in soil conservation compared with cultivated land, suggesting that reclamation of grassland performed much worse in conserving soil and water resource, overall, the slope did affect soil erosion of cultivated land. Some of their primary tillage in the fall greatly increases the erosion risk on sloping fields. To reduce such risks from fall tillage, some form of residue management should be applied, and tillage should be performed along the slope contours. In particular, the grassland acts as a buffer against fluctuations in runoff and soil erosion associated with rainfall events, agriculture use of cleared land does result in increased erosion, this study has demonstrated the major effects of slope on soil erosion and has found clear evidence that the rate of soil erosion and runoff has increased with slope. Most of these changes have occurred on land which is marginally suitable for cultivation, increasing the risk of soil erosion, accelerated water loss and further land degradation. As a result of the study, it is strongly recommended to protect the grassland, and the cultivated within steep-hill slope be limited to cultivation at a plot scale, considerable protection is offered to both the soil and water resource by grassland.
黄志霖1,2；傅伯杰1；陈利顶1. 黄土丘陵区不同坡度、土地利用类型与降水变化的水土流失分异[J]. 中国水土保持科学, 2005, 3(4): 11-18.
HUANG Zhi-Lin-；Fu-Ba-Jie-；Chen-Li-Ding. Differentiation of soil erosion by different slope, land use pattern and variation of precipitation in loess hilly region. SSWCC, 2005, 3(4): 11-18.
2005, Vol. 3 
