Infiltration in vegetated soil: empirical modeling and sensitivity analysis
Authors: Md. Enayet Chowdhury, Mohammad Shariful Islam, Tahsina Alam, Sagar Barua, Md. Safin Shahriar, Humaira Anisa
This study mathematically correlates the infiltration rate of vegetated soil with time and demonstrates how to estimate the amount of infiltration in any type of vegetated soil without conducting a field experiment. Vetiver (Vetiveria zizanioides) was planted in two plots: one with cohesive soil and one with cohesionless soil. The infiltration was measured under field conditions for 26 weeks, using the double-ring infiltrometer method. Sixteen analytical equations were carefully chosen to empirically correlate the infiltration rate with time, using the field data; 4 of the 16 were taken from the currently available empirical infiltration equations. The models’ performances were evaluated, based on four statistical parameters: sum of squares due to errors (SSE), R-square, adjusted R-square, and root-mean-square error (RMSE). The accuracy of each equation was determined by the number of times it ranked first for a single week, as well as its cumulative ranking, based on the statistical parameters. The equations’ sensitivity to the seasons, age of vegetation, and soil type were analyzed, and it was revealed that E3 performed the best of all of the equations. Occasions where other equations outperformed E3 were critically examined, based on the E3 coefficient values, and their performances were validated by data from another study. It was concluded that the performance of the equations may show sensitivity to the type of vegetation, geographic location, and climatic condition.
Published On: March 2021
Publication: Modeling Earth Systems and Environment