南方红壤区潋水流域降雨侵蚀力时空特征

Temporal and spatial distribution characteristics of rainfall erosivity of the Lianshui basin in the red soil region of South China

降雨侵蚀力(R)反映降雨引起土壤侵蚀的潜在能力,是进行土壤侵蚀预测预报及科学布置水土保持措施的重要依据。本文以南方红壤区典型流域——赣江上游潋水流域2000—2016年4个雨量站的降雨资料为基础,采用月雨量模型计算降雨侵蚀力,并采用统计学方法及Arc GIS空间分析技术对其时空特征进行分析。结果表明:1)流域的多年平均R值为5 899.0 MJ·mm/(hm2·h·a),最大值为10 306.9 MJ·mm/(hm2·h·a)(2015年),最小值为2 387.1 MJ·mm/(hm2·h·a)(2003年),各站R值年际间变化无显著差异。2)研究期内流域各雨量站R值的统计值M均为正数,流域面临着不同程度的水土流失潜在危机。3)R值和降雨量年内变化趋势一致,均表现为单峰型,集中分布在3—8月,约占R值全年的80.92%;最大值出现在6月,约占全年R值的23.8%。4)各站点年均R值分布范围为1 904.12~10 841.48 MJ·mm/(hm2·h·a),空间上表现为从流域的东北部向西南部呈逐渐增加的趋势。潋水流域降雨侵蚀力的年内分布、年际变化特征与降雨量时空分布基本一致,时空特征除与降雨量分布密切相关外,还与降雨格局等因素有关。

[Background] Rainfall erosivity（ R-factor） is one of the most important factors in the universal soil loss equation（ USLE）,which directly reflects the potential of soil erosion caused by rainfalls. Due to the unique topography and climate in the red soil region of South China,rainfall erosion force is three times higher in this region than in other regions of China. Therefore,analyses of R-factor and its spatial and temporal variations will be valuable for understanding the soil erosion processes in the red soil region,as well as for evaluating soil erosion and conducting soil and water conservation practices.[Methods]The R-factor was calculated using a simple algorithm model based on monthly precipitation data from four rainfall stations in the Lianshui basin from 2000 to 2016,followed by analyses of temporal and spatial variations of R-factor via statistical methods and the Arc GIS spatial analysis. The nonparametric Mann-Kendall test（ i. e. the M-K test） was also applied to evaluate the monotonic trend of Rfactor over the study period. [Results]1） Annual R values in the Lianshui basin averaged at 5 899. 0 MJ·mm/（ hm2·h·a） from 2000 to 2016,of which the maximum and minimum values were 10 306. 9 MJ·mm/（ hm2·h·a）（ 2015） and 2 387. 1 MJ·mm/（ hm2·h·a）（ 2003）,respectively. The inter-annual variation of R values was generally moderate in the four rainfall stations,with the coefficients of variation ranging from 0. 37 to 0. 43 and showing no difference in variation among stations. 2） Similar to the intraannual distribution pattern of rainfalls,there was one peak in terms of the intra-annual variation of R values. Specifically,monthly R values distributed mainly between March and August within a year,which accounted for 80. 92% of the annual R value. The maximum and minimum monthly R value were observed in June and December respectively,which accounted for 23. 87% and 1. 86% of the annual R value. 3） The statistical results of M-K tests（ i. e. M values） on R values were all positive（ ranging from0. 31 to 0. 78） in the four rainfall stations over the study period,suggesting a potential increasing trend in the probability of soil erosion in the study region. However,all M values were not statistically significant.4） Annual R values ranged from 1 904. 1 to 10 841. 48 MJ·mm/（ hm2·h·a） in the four rainfall stations,with no sinificant difference among them. The spatial distribution pattern of R values was similar to the spatial pattern of rainfalls in the Lianshui basin,both of which showed a gradually increasing trend from the northeast to the southwest of the basin. [Conclusions]In our study,a model using monthly rainfall data was applied to calculate the rainfall erosion force,and the characteristics of its spatial and temporal variation were analyzed. Our results suggested that both intra-and inter-annual variations of rainfall erosivity in the Lianshui basin were consistent with the temporal and spatial distribution of rainfalls and the precipitation patterns in the region.