Base flow

ABSTRACT: The performance of two popular watershed scale simu-lation models – HSPF and SWAT – were evaluated for simulating the hydrology of the 5,568 km2 Iroquois River watershed in Illinois and Indiana. This large, tile drained agricultural watershed pro-vides distinctly ...

This paper presents the findings of a study aimed at decomposing a flow hydrograph into different segments based on physical concepts in a catchment, and modelling different segments using different technique viz. conceptual and artificial neural networks (ANNs). An integrated modelling framework is proposed capable of modelling infiltration, base flow, evapotranspiration, soil moisture accounting, and certain segments of the decomposed flow hydrograph using conceptual techniques and the complex, non-linear, and dynamic rainfall-runoff process using ANN technique. Specifically, five different multi-layer perceptron (MLP) and two self-organizing map (SOM) models have been developed. The rainfall and streamflow data derived from the Kentucky River catchment were employed to test the proposed methodology and develop all the models. The performance of all the models was evaluated using seven different standard statistical measures. The results obtained in this study indicate that (a) the rainfall-runoff relationship in a large catchment consists of at least three or four different mappings corresponding to different dynamics of the underlying physical processes, (b) an integrated approach that models the different segments of the decomposed flow hydrograph using different techniques is better than a single ANN in modelling the complex, dynamic, non-linear, and fragmented rainfall runoff process, (c) a simple model based on the concept of flow recession is better than an ANN to model the falling limb of a flow hydrograph, and (d) decomposing a flow hydrograph into the different segments corresponding to the different dynamics based on the physical concepts is better than using the soft decomposition employed using SOM.

The field hydrology model DRAINMOD integrated with Arc Hydro in geographical information system (GIS) framework (Arc Hydro–DRAINMOD) was used to simulate the hydrological response of a coastal watershed in southeast Sweden. Arc Hydro–DRAINMOD uses a distributed approach to route water from each field edge to the watershed outlet. In the framework the Arc Hydro data model was used to describe the stream network in the watershed and to connect the individual simulated DRAINMOD-field outflow time series from each plot using Arc Hydro schema-links features, which were summed at Arc Hydro schema-nodes features along the stream network to generate the stream network flow. Hydrology data collected during six periods between 2003 and 2008 were used to test Arc Hydro–DRAINMOD and its performance was evaluated by considering uncertainties in model inputs using generalized likelihood uncertainty estimation (GLUE). The GLUE estimates obtained (uncertainty bands 5% and 95%) agreed satisfactorily with measured monthly discharges. The percentage of time in which the observed discharges were bracketed by the uncertainty bands was 88% in calibration periods and 75% in validation periods. Although monthly time step simulations showed good agreement with observed discharges during the two main discharge events in spring, the contradictory daily time step results indicate that the watershed response simulations on a daily basis need to be improved. The uncertainty analysis showed that in periods of higher discharge, such as spring periods, the uncertainty in prediction was higher. It is important to note that these uncertainty estimations using the GLUE procedure include the uncertainties in measured discharge values, model inputs, boundary conditions and model structures. It was estimated that stream baseflow represented 42% of the total watershed discharge, but further research is needed to confirm this. These results show that the new Arc Hydro–DRAINMOD framework is applicable for predicting discharge from artificially drained watersheds in southeast Sweden. Copyright © 2010 John Wiley & Sons, Ltd.

A plane wall-jet flow is numerically investigated and compared to experiments. The measured base flow is matched to a boundary-layer solution developing from a coupled Blasius boundary layer and Blasius shear layer. Linear stability analysis is performed, revealing high instability of two-dimensional eigenmodes and non-modal streaks. The nonlinear stage of laminar-flow breakdown is studied with three-dimensional direct numerical simulations and experimentally visualized. In the direct numerical simulation, an investigation of the nonlinear interaction between two-dimensional waves and streaks is made. The role of subharmonic waves and pairing of vortex rollers is also investigated. It is demonstrated that the streaks play an important role in the breakdown process, where their growth is transformed from algebraic to exponential as they become part of the secondary instability of the two-dimensional waves. In the presence of streaks, pairing is suppressed and breakdown to turbulence is enhanced.

Anthropogenic activities have exerted increasingly large-scale influences on terrestrial ecological systems from the past century, primarily through agriculture; however, the impact of such changes on the hydrologic cycle is poorly understood. As one of the important land use (LU) in the coastal Dogo Plain of the Seto Inland Sea, Japan, paddy fields have been decreasing with the increase in urbanization in recent decades. As the main source of water in the Dogo Plain, groundwater plays an important role in providing people with fresh water and contributing to stream base flow. The purpose of this study is to analyze the water resource and evaluate the effect of LU change on groundwater table fluctuation in this coastal plain. Firstly, the observations of groundwater table and the investigation of water balance were carried out in this alluvial plain. Then, a distributed four-block three-layer water balance model was employed to analyze the groundwater table fluctuation with response to the change of paddy field area. Moreover, the role of paddy field in recharging groundwater in the basin has been clarified. Results show that groundwater table depends not only on rainfall and discharge from rivers, but also on irrigation water and topology of the study area. The net groundwater recharge was positive in irrigation periods whereas that in non-irrigation periods was nearly equal to zero or negative. The results of this study would be helpful to the urban development policy and land use planning decision.