以华北平原山前至滨海区为研究区, 选择 8 个钻孔, 利用渗透系数反演公式和改进的 Archie 公式对渗透系数进行计算, 同时充分考虑研究区垂向渗透系数影响因素, 基于随机森林算法和时间序列算法对垂向渗透系数的空间 变异性进行研究。研究结果表明: 渗透系数在垂向上的影响因素主要为沉积环境、地层压力、随机误差三大类, ; 在剔除沉积环境和随机误差的影响后, 渗透系数随深度的增加而不断减小, 山前至滨海区渗透系数对地层压力的敏感性呈逐渐递减的趋势。研究结果可为华北平原农业灌溉、污染物迁移、垃圾填埋场选址及深层地下水资源开发等提供科学依据。

The permeability coefficient is the main parameter to characterize permeability, and has strong spatial variability. Taking the piedmont coastal area of North China Plain as the study area, eight boreholes were selected, and the permeability coefficient is calculated by the inversion formula and the improved Archie formula. The spatial variability of vertical permeability coefficient is studies using time series method and random forests algorithm based on fully considering the influencing factors of vertical permeability coefficient in the study area. The result showed that the influencing factors of the vertical permeability coefficient could be divided into three categories: sedimentary environment, formation pressure, and random error; after removing the influence of sediment ary environment and random error, the permeability coefficient decreased with the increase of depth, and the sensitivity of the permeability coefficient to the formation pressure decreased gradually from the piedmont to the coastal area. A scientific basis was provided for understanding agricultural irrigation, pollutant migration, site selection of landfill sites, and development of deep groundwater resources in North China Plain.

国家自然科学基金( 41807231) ; 河北省教育厅重点资助项目( ZD2016038)