Numerically modelling groundwater in an arid area with ANN‐generated dynamic boundary conditions

Abstract

Abstract:Groundwater is sensitive to the climate change and agricultural activities in arid and semi-arid areas. Over the past several decades, human activities, such as groundwater extraction for irrigation, have resulted in aquifer overdraft and disrupted the natural equilibrium in these areas. Regional groundwater simulation is important to determine appropriate groundwater management policies, and numerical simulation has become the most popular method. However, most groundwater models were developed with static boundary conditions. In this research, the Minqin oasis, an arid region located in northwest China, was selected as the study area. An artificial neural network (ANN) was developed to simulate effects of weather conditions, agricultural activities and surface water on groundwater level in a dynamic boundary of the domain. Subsequently, a groundwater numerical model, named ANN-FEFLOW model, was developed, with a dynamic boundary condition defined by the ANN model. The verifying results showed that the model has higher precision, with a root mean square error (RMSE) of 0Ð71 m, relative error (RE) of 17Ð96% and R 2 of 0Ð84 relative to the great groundwater change. Furthermore, the groundwater model has higher precision than the conventional groundwater model with static boundary condition, particularly in the area near the dynamic boundary. This study demonstrated that dynamic boundaries can improve the precision of the regional groundwater model in an arid area and that ANN can provide higher accuracy prediction capability for groundwater levels with dynamic boundary.

Sustainability of groundwater usage in northern China: dependence on palaeowaters and effects on water quality, quantity and ecosystem health

“…The development of groundwater facilitated enormous increases in agricultural production in the North China Plain and provinces in the north‐central regions (e.g. Shanxi and Shaanxi) and allowed the expansion of agriculture to marginal, arid environments in northwest China, such as Inner Mongolia and Gansu Province, where large agricultural ‘oases’ have developed on the basis of groundwater pumping (Cui and Shao, ; Ma et al , ; Chen et al , ; Wei et al , ; Huo et al , ). In the last decade, several serious groundwater resource problems have emerged in northern China, including overexploitation and deteriorating quality, threatening both agricultural and domestic water supplies.…”

Use of Nested Flow Models and Interpolation Techniques for Science‐Based Management of the Sheyenne National Grassland, North Dakota, USA

“…Triana et al (2010) modeled stream‐aquifer interactions with MODFLOW/MT3DMS and ANNs and extrapolated ANNs beyond the finite‐difference model domain of the Lower Arkansas River. Huo et al (2011) developed an ANN‐finite‐element model, in which transient recharge, irrigation, and groundwater abstraction were obtained with ANNs and implemented as transient boundary conditions in the finite‐element model (FEFLOW) for the Minqin oasis in China. Demissie et al (2009) improved predictive capability of a transient MODFLOW model by reducing residuals between observed and modeled groundwater heads using ANNs, IDW, and support vector machines, that is a variation of ANNs with Gaussian radial basis function (RBF) (Haykin 1999).…”

Assessing Underground Water Exchange Between Regions Using GRACE Data

“…One of the important reasons is that the statistic of groundwater resources is generally based on administrative units. Practically, the area of administrative region does not always coincide with the hydrogeology units (Huo et al, 2011). Therefore, it is of great significance to assess the underground water exchange between adjacent regions for water resources management.…”