K. P. Sudheer

Review: Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions

Over the past 15 years, artificial neural networks (ANNs) have been used increasingly for prediction and forecasting in water resources and environmental engineering. However, despite this high level of research activity, methods for developing ANN models are not yet well established. In this paper, the steps in the development of ANN models are outlined and taxonomies of approaches are introduced for each of these steps. In order to obtain a snapshot of current practice, ANN development methods are assessed based on these taxonomies for 210 journal papers that were published from 1999 to 2007 and focus on the prediction of water resource variables in river systems. The results obtained indicate that the vast majority of studies focus on flow prediction, with very few applications to water quality. Methods used for determining model inputs, appropriate data subsets and the best model structure are generally obtained in an ad-hoc fashion and require further attention. Although multilayer perceptrons are still the most popular model architecture, other model architectures are also used extensively. In relation to model calibration, gradient based methods are used almost exclusively. In conclusion, despite a significant amount of research activity on the use of ANNs for prediction and forecasting of water resources variables in river systems, little of this is focused on methodological issues. Consequently, there is still a need for the development of robust ANN model development approaches.

Planning groundwater development in coastal aquifers

Abstract Abstract This study examines an approach for planning groundwater development in coastal aquifers. The seawater intrusion is controlled through a series of barrier extraction wells. The multi-objective management problem is cast as a nonlinear, nonconvex combinatorial model and is solved using a coupled simulation–optimization approach. A density-dependent groundwater flow and transport model, SEAWAT is used for simulating the dynamics of seawater intrusion. The Simulated Annealing algorithm is used for solving the optimization problem. The idea of replacing the SEAWAT model with a trained artificial neural network (ANN) to manage the computational burden within practical time frames on a desktop computer is explored. The utility of the study is demonstrated through a trade-off curve between prioritizing groundwater development and controlling seawater intrusion at desired levels.

Quantification of the predictive uncertainty of artificial neural network based river flow forecast models

The meaningful quantification of uncertainty in hydrological model outputs is a challenging task since complete knowledge about the hydrologic system is still lacking. Owing to the nonlinearity and complexity associated with the hydrological processes, Artificial neural network (ANN) based models have gained lot of attention for its effectiveness in function approximation characteristics. However, only a few studies have been reported for assessment of uncertainty associated with ANN outputs. This study uses a simple method for quantifying predictive uncertainty of ANN model output through first order Taylor series expansion. The first order partial differential equations of non-linear function approximated by the ANN with respect to weights and biases of the ANN model are derived. A bootstrap technique is employed in estimating the values of the mean and the standard deviation of ANN parameters, and is used to quantify the predictive uncertainty. The method is demonstrated through the case study of Upper White watershed located in the United States. The quantitative assessment of uncertainty is carried out with two measures such as percentage of coverage and average width. In order to show the magnitude of uncertainty in different flow domains, the values are statistically categorized into low-, medium- and high-flow series. The results suggest that the uncertainty bounds of ANN outputs can be effectively quantified using the proposed method. It is observed that the level of uncertainty is directly proportional to the magnitude of the flow and hence varies along time. A comparison of the uncertainty assessment shows that the proposed method effectively quantifies the uncertainty than bootstrap method.

Digital image processing for determining drop sizes from irrigation spray nozzles

Abstract All the existing methods for measuring drop sizes produced by a sprinkler nozzle are either cumbersome, expensive or time consuming. Moreover, none could quantitatively express the relationship between drop size distribution and sprinkler head parameters viz. operating pressure and nozzle size. In the present study digital image processing technique has been applied to determine the drop size distribution from an irrigation spray nozzle. Image processing is the technique of automating and integrating a wide range of processes used for the human vision perception. The present study revealed that image processing technique can be successfully implemented for drop size measurement accurately. Being a novel technique, the method has some limitations for adaptation. These limitations can be very well contained through further research.

Application of a pseudo simulator to evaluate the sensitivity of parameters in complex watershed models

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.

Deficit irrigation management for rice using crop growth simulation model in an optimization framework

Optimization of irrigation water is an important issue in agricultural production for maximizing the return from the limited water availability. The current study proposes a simulation–optimization framework for developing optimal irrigation schedules for rice crop (Oryza sativa) under water deficit conditions. The framework utilizes a rice crop growth simulation model to identify the critical periods of growth that are highly sensitive to the reduction in final crop yield, and a genetic algorithm based optimizer develops the optimal water allocations during the crop growing period. The model ORYZA2000, which is employed as the crop growth simulation model, is calibrated and validated using field experimental data prior to incorporating in the proposed framework. The proposed framework was applied to a real world case study of a command area in southern India, and it was found that significant improvement in total yield can be achieved by the model compared to other water saving irrigation methods. The results of the study were highly encouraging and suggest that by employing a calibrated crop growth model combined with an optimization algorithm can lead to achieve maximum water use efficiency.

Design of Water Distribution Network for Equitable Supply

Water shortage is experienced in different parts of the world in different magnitude. In certain countries, water deficit is a regular phenomenon and in some other countries it happens for a short duration, due to failure of any component in the system. Shortage of water at source can be best tackled by distributing the available water equally among the consumers. This paper deals with the design of water distribution network capable of equitable supply during shortage in addition to the satisfactory performance under non-deficit condition. Performance of a typical water distribution network, with shortage of water at source is illustrated in detail. Head dependent outflow analysis with extended period simulation, is used to determine the actual supply from each node to consumers. Relationship between duration of supply and volume available at source as well as supply from each node are established for understanding the behaviour of network under low supply situation. A term “inequity” which is the maximum difference in supply demand ratio among different consumers is presented. This is based on the actual performance of the network instead of surrogate measures, generally used for reliability. It is illustrated that the maximum “inequity” in supply in a network during the entire duration of supply can be estimated with single analysis. Design of a water distribution network, duly considering equity in addition to the cost minimization and minimum head requirement is presented. Genetic Algorithm is used for solving this multi objective problem. The solution technique is illustrated using two benchmark problems, namely two loop network and Hanoi network. Results show that considerable improvement in equitable supply can be achieved with additional investment on pipes above the least cost solution. Hence it is better to design networks duly considering deficit condition for better reliability. It is also illustrated that it will be difficult to improve equity beyond a limit for a given network, through selection of different pipe diameters.

Dynamic integration of land use changes in a hydrologic assessment of a rapidly developing Indian catchment.

Rapid land use and land-cover changes strongly affect water resources. Particularly in regions that experience seasonal water scarcity, land use scenario assessments provide a valuable basis for the evaluation of possible future water shortages. The objective of this study is to dynamically integrate land use model projections with a hydrologic model to analyze potential future impacts of land use change on the water resources of a rapidly developing catchment upstream of Pune, India. For the first time projections from the urban growth and land use change model SLEUTH are employed as a dynamic input to the hydrologic model SWAT. By this means, impacts of land use changes on the water balance components are assessed for the near future (2009-2028) employing four different climate conditions (baseline, IPCC A1B, dry, wet). The land use change modeling results in an increase of urban area by +23.1% at the fringes of Pune and by +12.2% in the upper catchment, whereas agricultural land (-14.0% and -0.3%, respectively) and semi-natural area (-9.1% and -11.9%, respectively) decrease between 2009 and 2028. Under baseline climate conditions, these land use changes induce seasonal changes in the water balance components. Water yield particularly increases at the onset of monsoon (up to +11.0mm per month) due to increased impervious area, whereas evapotranspiration decreases in the dry season (up to -15.1mm per month) as a result of the loss of irrigated agricultural area. As the projections are made for the near future (2009-2028) land use change impacts are similar under IPCC A1B climate conditions. Only if more extreme dry years occur, an exacerbation of the land use change impacts can be expected. Particularly in rapidly changing environments an implementation of both dynamic land use change and climate change seems favorable to assess seasonal and gradual changes in the water balance.

Indian Summer Monsoon Rainfall: Implications of Contrasting Trends in the Spatial Variability of Means and Extremes.

India’s agricultural output, economy, and societal well-being are strappingly dependent on the stability of summer monsoon rainfall, its variability and extremes. Spatial aggregate of intensity and frequency of extreme rainfall events over Central India are significantly increasing, while at local scale they are spatially non-uniform with increasing spatial variability. The reasons behind such increase in spatial variability of extremes are poorly understood and the trends in mean monsoon rainfall have been greatly overlooked. Here, by using multi-decadal gridded daily rainfall data over entire India, we show that the trend in spatial variability of mean monsoon rainfall is decreasing as exactly opposite to that of extremes. The spatial variability of extremes is attributed to the spatial variability of the convective rainfall component. Contrarily, the decrease in spatial variability of the mean rainfall over India poses a pertinent research question on the applicability of large scale inter-basin water transfer by river inter-linking to address the spatial variability of available water in India. We found a significant decrease in the monsoon rainfall over major water surplus river basins in India. Hydrological simulations using a Variable Infiltration Capacity (VIC) model also revealed that the water yield in surplus river basins is decreasing but it is increasing in deficit basins. These findings contradict the traditional notion of dry areas becoming drier and wet areas becoming wetter in response to climate change in India. This result also calls for a re-evaluation of planning for river inter-linking to supply water from surplus to deficit river basins.

Artificial Neural Network Approach for Mapping Contrasting Tillage Practices

Abstract: Tillage information is crucial for environmental modeling as it directly affects evapotranspiration, infiltration, runoff, carbon sequestration, and soil losses due to wind and water erosion from agricultural fields. However, collecting this information can be time consuming and costly. Remote sensing approaches are promising for rapid collection of tillage information on individual fields over large areas. Numerous regression-based models are available to derive tillage information from remote sensing data. However, these models require information about the complex nature of underlying watershed characteristics and processes. Unlike regression-based models, Artificial Neural Network (ANN) provides an efficient alternative to map complex nonlinear relationships between an input and output datasets without requiring a detailed knowledge of underlying physical relationships. Limited or no information currently exist quantifying ability of ANN models to identify contrasting tillage practices from remote sensing data. In this study, a set of Landsat TM-based ANN models was developed to identify contrasting tillage practices in the Texas High Plains. Observed tillage data from Moore and Ochiltree Counties were used to develop and evaluate the models, respectively. The overall classification accuracy for the 15 models developed with the Moore County dataset varied from 74% to 91%. Statistical evaluation of these models against the Ochiltree County dataset produced results with an overall classification accuracy varied from 66% to 80%. The ANN models based on TM band 5 or