Assessing Variability of Infiltration Characteristics and Reliability of Infiltration Models in a Tropical Sub-humid Region of India.

Infiltration process, which plays a paramount role in irrigation and drainage systems design, groundwater recharge and contamination evaluation, flood and drought management etc. is often controlled by several factors, among which land use/land cover (LULC) and soil physical properties are the prime factors. These factors lead to significant spatial variability of infiltration process, which poses a serious challenge for hydrologists and water managers. However, studies analyzing spatial variability and influence of both LULC and soil physical properties are scarce. To this end, grid-based infiltration experiments were carried out in a tropical sub-humid region of India to investigate spatial variability of infiltration characteristics, saturated hydraulic conductivity (K) as well as to evaluate reliability of seven infiltration models in predicting infiltration behaviour and estimating K. Additionally, uncertainty analysis was performed to quantify uncertainties associated with estimated K for different LULC and soils. Results indicated that quasi-steady infiltration rate over the study area vary considerably with a majority of the area falling under 'low' and 'medium' infiltration categories. The infiltration process is greatly influenced by macro-pores and relatively low-permeable layers present at varying depths, typical features of lateritic vadose zones in tropical sub-humid regions, rather than its sole dependence on texture and LULC. Further, the Brutsaert model estimates K with the highest accuracy and least uncertainty followed by Swartzendruber and Horton models. Except the Brutsaert model, other models are sensitive to a particular LULC. Overall, it is inferred that the Brutsaert and Swartzendruber models are robust and more reliable in predicting infiltration behavior and K for the area. Findings of this study including quantification of spatial variability of important soil properties are useful for understanding detailed hydrological processes in the region and thereby, ensuring better planning and management of recurring floods and drought problems of the region.