University of Twente Student Theses

Login
This website will be unavailable due to maintenance December 1st between 8:00 and 12:00 CET.

Improving Groundwater Model Reliability by Coupling Unsaturated and Saturated Models: A case study of Sardon catchment, Spain

Berhe, Ermias Tseggai (2010) Improving Groundwater Model Reliability by Coupling Unsaturated and Saturated Models: A case study of Sardon catchment, Spain.

[img] PDF
1MB
Abstract:Groundwater flow models are capable to represent and predict the regional flow system in the saturated zone; however, they suffer with problem of non-uniqueness due to the combination of uncertainty in the parameter and fluxes estimation, model assumptions, model development. Fullytransient models are more reliable solution than steady state models because of their spatial and temporal fluxes input. The accurate representation of vadose zone flow processes (fluxes) in groundwater models enables to simulate the effect of near- and sub-surface hydrologic processes which improve their prediction. The main objective of this study was to improve the reliability of groundwater model by coupling unsaturated and saturated zone models. The developed unsaturated zone model called pyEARTH-2D, provide spatially distributed recharge. The methodology was composed of; (i) subdivision of Sardon catchment into soil zones based on soil hydraulic parameters; (ii) preparation of input data for the distributed recharge model; (iii) coupling of pyEARTH-2D and MODFLOW using PEST. The proposed methodology was tested on Sardon catchment, Spain. The subdivision of catchment in to zone was a pre-requisite for the application of the distributed recharge model which calculate recharge at cell-by-cell basis. Based on the fracture outcrop map the Sardon catchment was subdivided into four zones (massive outcrop, fractured outcrop, regolith/soil cover, and valley/drainage). The soil hydraulic parameters derived from the laboratory analysis of collected soil samples and field tests were assigned to each zone. Input of daily driving forces (rainfall and potential evapotranspiration) and state variables (soil moisture and hydraulic head. The upscaling of PET and interception loss to catchment scale was done based on classified vegetation map into three attributes: Q.ilex, Q.pyrenaica, and grass/shrub. The semi-distributed recharge model pyEARTH-2D was applied to estimate of spatially distributed recharge. The simulated heads by coupled pyEARTH-MODFLOW show trend more similar to the observed heads than the simulated heads by standard (non-coupled) MODFLOW. The water balances of the two, show that the latter model considers high recharge to simulate heads while the former model simulates minimal recharges which closely show the real groundwater conditions. The main objective of this study was to improve groundwater model reliability by coupling pyEARTH- 2D recharge model with MODFLOW using PEST. The RMSE and the similar trend of simulated heads by coupled pyEARTH-MODFLOW show the improvement of MODFLOW simulation due to the accurate recharge input from pyEARTH-2D. The coupling and calibration of pyEARTH-2D using PEST with optimized parameter field capacity show good improvement in the output of the model, therefore optimization of the other parameters hydraulic conductivity, porosity, wilting point etc could improve the performance and the output of the models.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/92334
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page