K P Sudheer

The establishment periods of switchgrass and Miscanthus can be a time window when evapotranspiration, surface runoff generation, and sediment and nutrients losses are quite different from when the grasses become fully established, and this period may result in environmental concerns for large-scale biomass production. The current SWAT model does not simulate the establishment periods of perennial grasses. In this study, we modified the model to simulate these periods by updating the maximum annual leaf area index (LAI) instead of using a static value in the unmodified model. The improved SWAT model provided more realistic simulations of LAI values and yields comparable to observed yields from field plots during the establishment periods of the two perennial grasses. The modification made in the present study enabled the SWAT model to be more suitable for evaluating perennial biomass grass-related scenarios.

In this paper, the issue of nonlinear sensitivity analysis for dimensionality reduction in hydrologic model calibration is discussed, and a novel method to quantify the sensitivity of each parameter that considers the nonlinear relationship in the model is presented. The method is based on computing the absolute variation of the nonlinear function represented by the model in its parameter space. The paper discusses the theoretical background of the method and presents the algorithm. The algorithm employs neural network as a pseudo simulator to reduce the computational burden of the analysis. The proposed approach of sensitivity analysis is illustrated through a case study on a physically based distributed hydrologic model. The results indicate that the method is able to rank the parameters effectively, and the ranking can be interpreted in the context of the physical processes being considered by the model. © 2010 Elsevier Ltd.

Regionalization of model parameters by developing appropriate functional relationship between the parameters and basin characteristics is one of the potential approaches to employ hydrological models in ungauged basins. While this is a widely accepted procedure, the uniqueness of the watersheds and the equifinality of parameters bring lot of uncertainty in the simulations in ungauged basins. This study proposes a method of regionalization based on the probability distribution function of model parameters, which accounts the variability in the catchment characteristics. It is envisaged that the probability distribution function represents the characteristics of the model parameter, and when regionalized the earlier concerns can be addressed appropriately. The method employs probability distribution of parameters, derived from gauged basins, to regionalize by regressing them against the catchment attributes. These regional functions are used to develop the parameter characteristics in ungauged basins based on the catchment attributes. The proposed method is illustrated using soil water assessment tool model for an ungauged basin prediction. For this numerical exercise, eight different watersheds spanning across different climatic settings in the USA are considered. While all the basins considered in this study were gauged, one of them was assumed to be ungauged (pseudo-ungauged) in order to evaluate the effectiveness of the proposed methodology in ungauged basin simulation. The process was repeated by considering representative basins from different climatic and landuse scenarios as pseudo-ungauged. The results of the study indicated that the ensemble simulations in the ungauged basins were closely matching with the observed streamflow. The simulation efficiency varied between 57 and 61 % in ungauged basins. The regional function was able to generate the parameter characteristics that were closely matching with the original probability distribution derived from observed streamflow data.

Implementation of sensitivity analysis (SA) procedures is helpful in calibration of models and also for their transposition to different watersheds. The reported studies on SA of Soil and Water Assessment Tool (SWAT) model were mostly focused on identifying parameters for pruning or modifying during the calibration process. This paper presents a sensitivity and identifiability analysis of model parameters that influence stream flow generation in SWAT. The analysis was focused on evaluating the sensitivity of the parameters in different climatic settings, temporal scales and flow regimes. The global sensitivity analysis (GSA) technique based on classical decomposition of variance, Sobol', was employed in this study. The results of the study indicate that modeled stream flow show varying sensitivity to parameters in different climatic settings. The results also suggest that the identifiability of a parameter for a given watershed is a major concern in calibrating the model for the specific watershed, as it might lead to equifinality of parameters. The SWAT model parameters show varying sensitivity in different years of simulation suggesting the requirement for dynamic updation of parameters during the simulation. The sensitivity of parameters during various flow regimes (low, medium and high flow) is also found to be uneven, which suggests the significance of a multi-criteria approach for the calibration of models. Copyright © 2010 John Wiley & Sons, Ltd.

One of the principal sources of uncertainty in hydrological models is the absence of understanding of the complex physical processes of the hydrological cycle within the system. This leads to uncertainty in input selection and consequently its associated parameters, and hence evaluation of uncertainty in a model becomes important. While there has been considerable interest in developing methods for uncertainty analysis of artificial neural network (ANN) models, most of the methods are relatively complex and/or require assumption about the prior distribution of the uncertain parameters. This paper presents an effective and simple way to perform uncertainty analysis for ANN-based hydrologic model. The method is based on the concept of bootstrap technique and is demonstrated through a case study of the Kolar River basin located in India. The method effectively quantifies uncertainty in the model output and the parameters arising from variation in input data used for calibration. In the current study, the uncertainty due to model architecture and the input vector are not directly considered; they have been minimized during the model calibration. The results from the case study suggest that the sampling variability of the training patterns as well as the initial guess of the parameters of ANN do not have significant impact on the model performance. However, despite good generalization properties for the models developed in this study, most of them fail to capture the hydrograph peak flow characteristics. The proposed method of uncertainty analysis is very efficient, can be easily applied to an ANN-based hydrologic model, and clearly illustrates the strong and weak points of the ANN model developed. Copyright 2007 by the American Geophysical Union.

Predictions in ungauged basins (PUB) are one of the major challenges in hydrology. Physically-based distributed models are ideal for PUB because they can account most of the heterogeneity in the system. Cibin et al.1 proposed a method to derive the probability distribution function (PDF) of the sensitive parameters using a single likelihood (a global measure over the entire ranges of flow) and use them for PUB. Instead of considering a single criterion for deriving the PDF, a multi-criteria approach that can account variation in sensitivity of parameters and model performance in different flow ranges may be a better approach for identifying the parameter characteristics in terms of their PDF. The study proposes a method to minimize the predictive uncertainty of distributed models by deriving the PDF of sensitive parameters based on the Bayesian approach. The method employs Monte Carlo simulations of parameter sets generated by ‘Latin Hypercube Sampling.’ Within the Monte Carlo simulations, those parameter sets that produced reasonably good performance in all ranges of flow are used for estimating multi-criteria index and updating will continue till both (prior and posterior) PDFs converge in successive cycles. These PDFs, which are derived using gauged basin data, are then transferred to hydrologically similar ungauged basins for generating ensembles of simulations. The proposed methodology is illustrated through a case study of a watershed in USA. The Soil and Water Assessment Tool model was considered for the application. The study also discusses a comparison of PUB using a single criterion approach and a multi-criteria approach. It is observed that confidence band for predictions by proposed approach is narrow and the number of cycles required for deriving the PDF is less as compared with the former.

The quantification of stream flow in ungauged basins is one of the most challenging tasks in surface water hydrology due to non-availability of data and system heterogeneity. Even though physics based distributed hydrologic models are considered best suited for the ungauged basins, uncertainty in model simulations, in the absence of any parameter estimations reflecting accurate watershed characteristics, may be high. A successful application of these models in making hydrologic response predictions in ungauged basins requires reducing number of parameters and output uncertainty. The current study proposes a method to minimize the predictive uncertainty of distributed hydrological models by deriving regionalized probability distribution of sensitive parameters of the model. To facilitate stochastic validation of the model in ungauged basins, the derived PDF of the parameters are obtained for a number of gauged basins, the derived PDFs are regionalized and then transferred to the ungauged basin. The method is illustrated through a case study of SWAT model applied to a watershed.

This paper presents a method of constructing prediction interval for artificial neural network (ANN) rainfall runoff models during calibration with a consideration of generating ensemble predictions. A two stage optimization procedure is envisaged in this study for construction of prediction interval for the ANN output. In Stage 1, ANN model is trained with genetic algorithm (GA) to obtain optimal set of weights and biases vector. In Stage 2, possible variability of ANN parameters (obtained in Stage 1) is optimized so as to create an ensemble of models with the consideration of minimum residual variance for the ensemble mean, while ensuring a maximum of the measured data to fall within the estimated prediction interval. The width of the prediction interval is also minimized simultaneously. The method is demonstrated using a real world case study of rainfall runoff data for an Indian basin. The method was able to produce ensembles with a prediction interval (average width) of 26.49m3/s with 97.17% of the total observed data points lying within the interval in validation. One specific advantage of the method is that when ensemble mean value is considered as a forecast, the peak flows are predicted with improved accuracy by this method compared to traditional single point forecasted ANNs. © 2013 Elsevier B.V.

Watershed management in rainfed agricultural areas of arid and semi-arid regions aims at alleviating agricultural drought by employing water conservation measures. Water conservation measures serve as mechanisms for recharging groundwater, and also for surface storage. While the benefits derived from these structures is widely accepted, there is increasing concern regarding their sustainability and efficiency. The application of water conservation measures in upstream reaches contributes to reduced inflows to downstream reaches and structures. Currently watershed management is planned by considering elementary information about the hydrological regime and its associated impacts on the upstream reaches of the watershed. This approach may lead to inefficient water conservation structures, and there is also a significant probability that the structure may be under designed (or over designed). Therefore, careful planning that considers the hydrological changes that the conservation structure may bring in the watershed and the associated benefits/compromise for both upstream and downstream is essential for developing a successful watershed management plan. A simulation-optimization framework for optimal sizing of the water conservation structure considering the dual objectives of improving the benefits in the upstream reaches while maintaining flows in the downstream reaches is proposed in this study. The proposed method is demonstrated for optimal sizing of water conservation structures (check dams in this study) for an experimental watershed in Kondepi Mandal, Andhra Pradesh, India. The results indicate that the check dams of sizes obtained from both traditional method and simulation-optimization method increase moisture availability in the watershed. However, for check dam sizes obtained from the simulation-optimization framework, there is an increase in the flow to downstream reaches compared to the check dam sizes obtained from the traditional methods. The increase in downstream flows obtained by optimizing the check dam heights is at the expense of increased moisture stress days in the non-growing period, thus, not affecting the crop growth or productivity. The results from this study indicate that the proposed simulation-optimization framework can be of assistance in sizing of check dams for sustainable and effective watershed management.

The temporal variability of rainfall in rainfed regions is one of the main factors for their low agricultural productivity. The future climate projections show an increased variability of rainfall, thus further impacting the rainfed agriculture. The change in rainfall pattern is expected to alter the cropping period and making the crop sowing date critical to mitigate crop failure. However, with enhanced temporal variability of rainfall, arriving at an optimal crop sowing date is a challenging task. One of the widely adopted measure to improve the agricultural productivity in the rainfed regions is water harvesting structures (WHS). This study evaluates the ability of the WHS in absorbing the shock of the temporal variability of the rainfall on the agricultural productivity. In addition, the efficacy of the structures in improving the agricultural productivity in the future climate projections is also evaluated. The proposed analysis is performed over Kondepi watershed in Andhra Pradesh, India, where water conservation measures are implemented by Government and Non-Government Organizations. The results of the study show that the WHS can minimize the sensitivity of the agricultural productivity to the crop sowing date. The extended availability of water in WHS resulted in removing the relationship between crop sowing date and crop productivity, thus exhibiting the ability of WHS in dams in absorbing the shock caused by the temporal variability of the rainfall. Further, the agricultural productivity was found to be increasing due to the presence of WHS in both current and future climate conditions.