[A-HW28_30PO1] Water and material transport and cycle in watersheds: from headwater to coastal area

[AHW28-P06] Estimation of water balance in a coastal agricultural catchment using SWAT and HYDRUS Model.

*Guangzhe JIN¹, Yuta SHIMIZU², Shin-ichi ONODERA¹, Mitsuyo SAITO³, Kenji MATSUMORI² (1.Hiroshima University Graduate School of Integrated Arts and Sciences, 2.National Agriculture and Food Research Organization, Western Region Agricultural Research Center, 3.Graduate School of Environmental and Life Science, Okayama University)

Keywords:Ikuchi Island, SWAT Model, HYDRUS Model, groundwater recharge rate, water balance

Quantify the rate of ground water recharge and clarify the water balance in watersheds is basic and important for efficient ground water resource management. It is particularly important in regions with little rain which face with the risk of water shortage. However, the rate of aquifer recharge is one of the most difficult factors to evaluate. Especially, the former method of groundwater recharge estimation, are normally subject to large uncertainties and easily to cause errors. Recently, there are several attempting for estimation of groundwater recharge using distributed hydrological models in the world.The Soil and Water Assessment Tool (SWAT) Model is one of a physically based and quasi-distributed continuous time hydrological model used to estimate water budget in previous researches around the world. SWAT Model has been implemented for watershed hydrology related issues such as estimation of surface water flow and groundwater recharge rate. We could more specific testify the groundwater flux combined SWAT Model with HYDRUS Model which is a software package for simulating water, heat, and solute movement in two- and three-dimensional variably saturated media. The objective of this research is to estimate water balance and to clarify the groundwater recharge parameter in an agricultural catchment in the Seto Inland Sea, using the SWAT Model, and to estimate the groundwater flow using the HYDRUS Model. The study site is located on the southern part of Ikuchi Island, which is one of the islands in central Seto Inland Sea. The orange groves cover approximately 50% of the total catchment area. Due to the small annual precipitation (approx.1000mm/y) with large inter-annual variation, Ikuchi Island is facing a risk of water shortage in the serious dry year. As input to SWAT Model, topographic data (10 m grid), soil map (1/25000), land use map (1/25000) and weather information were used to build and calculate the SWAT Model. Evaporation was estimated by the Penman-Monteith method. Simulation time periods is 2000-2013, including warm up period of 2000-2003 and calibration period of 2003-2004. The calibration was conducted using the Sequential Uncertainty Fitting (SUFI2). The reproducibility of daily discharge in calibration period by the model was found to be acceptable (NSE=0.69, RSR=0.56, PBIAS%=18, R²=0.75). Amount of groundwater recharge is accounted as the water discharge into aquifer except the flows which are eventually discharged from aquifer, such as return flows into river and amount of water moving into the vadose zone.The result shows spatial difference in groundwater recharge rate. About 10 times higher groundwater recharge rate was found in middle and downstream areas. While middle and downstream area are indicated the main groundwater recharge area, upstream is small recharge rate due to steep slope. Groundwater recharge shows smaller volume than river discharge, it comprise about 17% of total precipitation in annual average consideration. From the comparison of water balance calculation, it is found that both of river discharge and groundwater recharge fluctuated in high precipitation year of 2011 (1,527mm), low precipitation year of 2005 (781mm) compared to average balance. In high precipitation year, groundwater recharge rate increased about 6 times than in low precipitation year, the increasing of river discharge is at about 2.5 times.Consequently, it was confirmed that spatial and temporal variation of groundwater recharge rate in long term. And we could estimate the long term water balance base on these information. However, it is noted that this result may include some uncertainty and chance to improve. Seat model could not reflect the groundwater flow, simulated with HYDRUS Model on the groundwater flow could provide us with the groundwater data. In the presentation, more detailed data cover long time periods and results testifying groundwater level variation with HYDRUS Model will be displayed