[A-HW21] Surface and subsurface hydrologic models: Technical advances and applications for water management

[AHW21-03] A novel synthetic analysis technique of solute movement in catchments without groundwater concentration records - study of a tropical complex hard rock aquifer.

*Baibaswata Bhaduri¹, Sekhar Muddu^1,2, Laurent Ruiz^2,3,4 (1.Civil Engineering Department, Indian Institute of Science, Bangalore, India., 2.Indo-French Cell for Water Sciences, Indian Institute of Science, Bangalore, India., 3.IRD, CNRS, UPS, UMR GET, Toulouse, France., 4.INRAE, AGROCAMPUS OUEST, UMR SAS, Rennes, France.)

Keywords:Critical Zone, Hard Rock Aquifers, Groundwater Recession Analysis, Synthetic Black Box Catchments, Nitrate Travel Time.

Determination of residence time of groundwater and flushing rate of solutes dissolved in it is still a big challenge in critical zone science, especially with complex, agriculture dominated fractured hard-rock aquifer systems. Berambadi (84 km²), a part of the Kabini Critical Zone Observatory of Karnataka, India is one such catchment with a very big well network systematically pumped to extract groundwater for irrigation. The stream being dry and the flow along different transects being very anthropogenically driven, it’s really difficult to constrain Berambadi with proper boundary conditions and heterogeneous hydrological properties using a distributed groundwater flow model. However, Boussinesq equation of groundwater flow can be converted into a simple water table fluctuation model based on linear reservoir concepts. We adapted this kind of 1d analysis technique for Berambadi and applied it to 150 extensively monitored wells to determine the recession behavior of individual wells. Recession (mathematically the inverse of detention time) was found out to be a stronger parameter than draft or recharge factor from the sensitivity analysis performed. Recession was also found to be varying with mean groundwater level over the period of fluctuations - a logarithmic decay with depth was observed which was in accordance with the proposed decrease in fracture density of the hard rock aquifer with depth. Since recession is a function of hydraulic conductivity and drainable porosity, we checked different variations of those properties that reproduce similar recession behavior and made corresponding synthetic homogeneous box catchments in FEFLOW. We then determined the concentration evolution and travel time of agricultural nitrates by running synthetic mass transport simulations on those virtual catchments. Berambadi having no proper long-term nitrate concentration time series measured in its wells yet, and the past fertilizer input also being impossible to be quantified with certainty, this kind of back-and-forth analysis arguably gives us the most decent estimate of the nitrate flushing rates.