BASIN SCALE WATER MANAGEMENT AND FORECASTING USING ARTIFICIAL NEURAL NETWORKS1

ABSTRACT: Water scarcity in the Sevier River Basin in south‐central Utah has led water managers to seek advanced techniques for identifying optimal forecasting and management measures. To more efficiently use the limited quantity of water in the basin, better methods for control and forecasting are imperative. Basin scale management requires advanced forecasts of the availability of water. Information about long term water availability is important for decision making in terms of how much land to plant and what crops to grow; advanced daily predictions of streamflows and hydraulic characteristics of irrigation canals are of importance for managing water delivery and reservoir releases; and hourly forecasts of flows in tributary streams to account for diurnal fluctuations are vital to more precisely meet the day‐to‐day expectations of downstream farmers. A priori streamflow information and exogenous climate data have been used to predict future streamflows and required reservoir releases at different timescales. Data on snow water equivalent, sea surface temperatures, temperature, total solar radiation, and precipitation are fused by applying artificial neural networks to enhance long term and real time basin scale water management information. This approach has not previously been used in water resources management at the basin‐scale and could be valuable to water users in semi‐arid areas to more efficiently utilize and manage scarce water resources.     

GROUND WATER DROUGHT MANAGEMENT BY A FEEDFORWARD CONTROL METHOD1

ABSTRACT: Management of a regional ground water system to mitigate drought problems at the multi‐layered aquifer system in Collier County, Florida, is the main topic. This paper developed a feedforward control system that consists of system and control equations. The system equation, which forecasts ground water levels using the current measurements, was built based on the Kalman filter algorithm associated with a stochastic time series model. The role of the control equation is to estimate the pumping reduction rate during an anticipated drought. The control equation was built based on the empirical relationship between the change in ground water levels and the corresponding pumping requirement. The control system starts with forecasting one‐month‐ahead ground water head at each control point. The forecasted head is in turn used to calculate the deviation of ground water heads from the monthly target specified by a 2‐in‐10‐year frequency. When the forecasted water level is lower than the target, the control system computes spatially‐varied pumping reduction rates as a recommendation for ground water users. The proposed control system was tested using hypothetical droughts. The simulation result revealed that the estimated pumping reduction rates are highly variable in space, strongly supporting the idea of spatial forecasting and controlling of ground water levels as opposed to a lumped water use restriction method used previously in the model area.     

Comparison of the Performance of Statistical Models in Forecasting Monthly Total Dissolved Solids in the Rio Grande1

Abudu, S., J.P. King, Z. Sheng, 2011. Comparison of the Performance of Statistical Models in Forecasting Monthly Total Dissolved Solids in the Rio Grande. Journal of the American Water Resources Association (JAWRA) 48(1): 10‐23. DOI: 10.1111/j.1752‐1688.2011.00587.x Abstract: This paper presents the application of autoregressive integrated moving average (ARIMA), transfer function‐noise (TFN), and artificial neural networks (ANNs) modeling approaches in forecasting monthly total dissolved solids (TDS) of water in the Rio Grande at El Paso, Texas. Predictability analysis was performed between the precipitation, temperature, streamflow rates at the site, releases from upstream reservoirs, and monthly TDS using cross‐correlation statistical tests. The chi‐square test results indicated that the average monthly temperature and precipitation did not show significant predictability on monthly TDS series. The performances of one‐ to three‐month‐ahead model forecasts for the testing period of 1984‐1994 showed that the TFN model that incorporated the streamflow rates at the site and Caballo Reservoir release improved monthly TDS forecasts slightly better than the ARIMA models. Except for one‐month‐ahead forecasts, the ANN models using the streamflow rates at the site as inputs resulted in no significant improvements over the TFN models at two‐month‐ahead and three‐month‐ahead forecasts. For three‐month‐ahead forecasts, the simple ARIMA showed similar performance compared to all other models. The results of this study suggested that simple deseasonalized ARIMA models could be used in one‐ to three‐month‐ahead TDS forecasting at the study site with a simple, explicit model structure and similar model performance as the TFN and ANN models for better water management in the Basin.     

A High‐Resolution National‐Scale Hydrologic Forecast System from a Global Ensemble Land Surface Model

Warning systems with the ability to predict floods several days in advance have the potential to benefit tens of millions of people. Accordingly, large‐scale streamflow prediction systems such as the Advanced Hydrologic Prediction Service or the Global Flood Awareness System are limited to coarse resolutions. This article presents a method for routing global runoff ensemble forecasts and global historical runoff generated by the European Centre for Medium‐Range Weather Forecasts model using the Routing Application for Parallel computatIon of Discharge to produce high spatial resolution 15‐day stream forecasts, approximate recurrence intervals, and warning points at locations where streamflow is predicted to exceed the recurrence interval thresholds. The processing method involves distributing the computations using computer clusters to facilitate processing of large watersheds with high‐density stream networks. In addition, the Streamflow Prediction Tool web application was developed for visualizing analyzed results at both the regional level and at the reach level of high‐density stream networks. The application formed part of the base hydrologic forecasting service available to the National Flood Interoperability Experiment and can potentially transform the nation's forecast ability by incorporating ensemble predictions at the nearly 2.7 million reaches of the National Hydrography Plus Version 2 Dataset into the national forecasting system.     

GROUND WATER/SURFACE WATER RESPONSES TO GLOBAL CLIMATE SIMULATIONS,SANTA CLARA‐CALLEGUAS BASIN,VENTURA, CALIFORNIA1

ABSTRACT: Climate variations can play an important, if not always crucial, role in successful conjunctive management of ground water and surface water resources. This will require accurate accounting of the links between variations in climate, recharge, and withdrawal from the resource systems, accurate projection or predictions of the climate variations, and accurate simulation of the responses of the resource systems. To assess linkages and predictability of climate influences on conjunctive management, global climate model (GCM) simulated precipitation rates were used to estimate inflows and outflows from a regional ground water model (RGWM) of the coastal aquifers of the Santa Clara‐Calleguas Basin at Ventura, California, for 1950 to 1993. Interannual to interdecadal time scales of the El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) climate variations are imparted to simulated precipitation variations in the Southern California area and are realistically imparted to the simulated ground water level variations through the climate‐driven recharge (and discharge) variations. For example, the simulated average ground water level response at a key observation well in the basin to ENSO variations of tropical Pacific sea surface temperatures is 1.2 m/°C, compared to 0.9 m/°C in observations. This close agreement shows that the GCM‐RGWM combination can translate global scale climate variations into realistic local ground water responses. Probability distributions of simulated ground water level excursions above a local water level threshold for potential seawater intrusion compare well to the corresponding distributions from observations and historical RGWM simulations, demonstrating the combination's potential usefulness for water management and planning. Thus the GCM‐RGWM combination could be used for planning purposes and — when the GCM forecast skills are adequate — for near term predictions.     

Ensemble Streamflow Forecast: A GLUE‐Based Neural Network Approach1

Abstract: While training a Neural Network to model a rainfall‐runoff process, generally two aspects are considered: its capability to be able to describe the complex nature of the processes being modeled and the ability to generalize so that novel samples could be mapped correctly. The general conclusion is that, the smallest size network capable of representing the sample distribution is the best choice, as far as generalization is concerned. Oftentimes input variables are selected a priori in what is called an explanatory data analysis stage and are not part of the actual network training and testing procedures. When they are, the final model will have only a “fixed” type of inputs, lag‐space, and/or network structure. If one of these constituents was to change, one would obtain another equally “optimal” Neural Network. Following Beven and others' generalized likelihood uncertainty estimate approach, a methodology is introduced here that accounts for uncertainties in network structures, types of inputs, and their lag‐space relationships by looking at a population of Neural Networks rather than target in getting a single “optimal” network. It is shown that there is a wide array of networks that provide “similar” results, as seen by a likelihood measure, for different types of inputs, lag‐space, and network size combinations. These equally optimal networks expose the range of uncertainty in streamflow predictions and their expected value results in a better performance than any of the single network predictions.     

INTEGRATED MANAGEMENT OF IN‐FIELD,EDGE‐OF‐FIELD,AND AFTER‐FIELD BUFFERS1

This review summarizes how conservation benefits are maximized when in‐field and edge‐of‐field buffers are integrated with each other and with other conservation practices such as residue management and grade control structures. Buffers improve both surface and subsurface water quality. Soils under permanent buffer vegetation generally have higher organic carbon concentrations, higher infiltration capacities, and more active microbial populations than similar soils under annual cropping. Sediment can be trapped with rather narrow buffers, but extensive buffers are better at transforming dissolved pollutants. Buffers improve surface runoff water quality most efficiently when flows through them are slow, shallow, and diffuse. Vegetative barriers ‐ narrow strips of dense, erect grass ‐ can slow and spread concentrated runoff. Subsurface processing is best on shallow soils that provide increased hydrologic contact between the ground water plume and buffer vegetation. Vegetated ditches and constructed wetlands can act as “after‐field” conservation buffers, processing pollutants that escape from fields. For these buffers to function efficiently, it is critical that in‐field and edge‐of‐field practices limit peak runoff rate and sediment yield in order to maximize contact time with buffer vegetation and minimize the need for cleanout excavation that destroys vegetation and its processing capacity.     

Automated Algorithms for Heuristic Base‐Flow Separation1

Abstract: The subjective nature of graphical base‐flow separation combined with the many applications of base‐flow time series derived from continuous streamflow data, motivates the development and application of automated algorithms for heuristic base‐flow separation. Base‐flow time series derived from gauged streamflow support diverse applications in engineering hydrology, catchment analysis, hydrogeologic investigations, regional low‐flow analysis, and recharge estimation. Whether based on graphical procedures for recession analysis or analytical expressions derived from fundamental equations of ground‐water flow, the variety of base‐flow separation algorithms belies the array of base‐flow definitions and interpretations that variously refer to dominant process, source, flow path, and characteristic response time. Algorithms that are invariant in their consistent – though heuristic – characterization of base‐flow response are particularly useful for interbasin comparisons of low‐flow characteristics and hydrologic regionalization. More adaptable algorithms provide application‐specific flexibility in allocating flow components like interflow to either quickflow or slowflow. Four widely used algorithms that produce consistent base‐flow time series using only gauged streamflow records are compared and contrasted with a complementary heuristic algorithm that incorporates hydrologic judgment explicitly, through manual parameterization. The utility of these inherently subjective algorithms is illustrated through a simple example of flow phase separation in a two‐component end‐member mixing model of dissolved chlorides in the Cuyahoga River.     

Riparian Ground‐Water Flow Patterns Using Flownet Analysis: Evapotranspiration‐Induced Upwelling and Implications for N Removal1

Abstract: Ground‐water flow paths constrain the extent of nitrogen (N) sinks in deep, stratified soils of riparian wetlands. We examined ground‐water flow paths at four forested riparian wetlands in deep, low gradient, stratified deposits subjected to Southern New England’s temperate, humid climate. Mid‐day piezometric heads were recorded during the high water table period in April/May and again in late November at one site. Coupling field data with a two‐dimensional steady‐state ground‐water flow model, flow paths and fluxes were derived to 3 m depths. April/May evapotranspiration (ET) dominated total outflux (44‐100%) while flux to the stream was <10% of total outflux. ET exerted upward ground‐water flux through shallow carbon‐rich soils, increasing opportunities for N transformations and diverting flow from the stream. Dormant season results showed a marked increase in flux to the stream (27% of the total flux). Riparian sites with deep water tables (naturally or because of increased urbanization or other hydrologic modifications) or shallow root zones may not generate ground‐water upwelling to meet evaporative demand, thereby increasing the risk of N movement to streams. As water managers balance issues of water quality with water quantity, they will be faced with decisions regarding riparian management. Further work towards refining our understanding of ET mediation of N and water flux at the catchment scale will serve to inform these decisions.     

Mid‐Range Streamflow Forecasts Based on Climate Modeling – Statistical Correction and Evaluation1

Abstract: Mid‐range streamflow predictions are extremely important for managing water resources. The ability to provide mid‐range (three to six months) streamflow forecasts enables considerable improvements in water resources system operations. The skill and economic value of such forecasts are of great interest. In this research, output from a general circulation model (GCM) is used to generate hydrologic input for mid‐range streamflow forecasts. Statistical procedures including: (1) transformation, (2) correction, (3) observation of ensemble average, (4) improvement of forecast, and (5) forecast skill test are conducted to minimize the error associated with different spatial resolution between the large‐scale GCM and the finer‐scale hydrologic model and to improve forecast skills. The accuracy of a streamflow forecast generated using a hydrologic model forced with GCM output for the basin was evaluated by forecast skill scores associated with the set of streamflow forecast values in a categorical forecast. Despite the generally low forecast skill score exhibited by the climate forecasting approach, precipitation forecast skill clearly improves when a conditional forecast is performed during the East Asia summer monsoon, June through August.     

Evaluation of Ground‐Water and Surface‐Water Exchanges Using Streamflow Difference Analyses1

Abstract: In the karstic lower Flint River Basin, limestone fracturing, jointing, and subsequent dissolution have resulted in the development of extensive secondary permeability and created a system of major conduits that facilitate the exchange of water between the Upper Floridan aquifer and Flint River. Historical streamflow data from U.S. Geological Survey gaging stations located in Albany and Newton, Georgia, were used to quantify ground‐water and surface‐water exchanges within a 55.3 km section of the Flint River. Using data from 2001, we compared estimates of ground‐water flux using a time adjustment method to a water balance equation and found that these independent approaches yielded similar results. The associated error was relatively large during high streamflow when unsteady conditions prevail, but much lower during droughts. Flow reversals were identified by negative streamflow differences and verified with in situ data from temperature sensors placed inside large spring conduits. Long‐term (13 years) analysis showed negative streamflow differentials (i.e., a losing stream condition) coincided with high river stages and indicated that streamflow intrusion into the aquifer could potentially exceed 150 m³/s. Although frequent negative flow differentials were evident, the Flint River was typically a gaining stream and showed a large net increase in flow between the two gages when examined over the period 1989‐2003. Ground‐water contributions to this stream section averaged 2‐42 m³/s with a mean of 13 m³/s. The highest rate of ground‐water discharge to the Flint River occurred during the spring when regional ground‐water levels peaked following heavy winter and spring rains and corresponding rates of evapotranspiration were low. During periods of extreme drought, ground‐water contributions to the Flint River declined.     

AN EVALUATION OF WEEKLY AND MONTHLY TIME SERIES FORECASTS OF MUNICIPAL WATER USE1

ABSTRACT: A cascade model for forecasting municipal water use one week or one month ahead, conditioned on rainfall estimates, is presented and evaluated. The model comprises four components: long term trend, seasonal cycle, autocorrelation and correlation with rainfall. The increased forecast accuracy obtained by the addition of each component is evaluated. The City of Deerfield Beach, Florida, is used as the application example with the calibration period from 1976–1980 and the forecast period the drought year of 1981. Forecast accuracy is measured by the average absolute relative error (AARE, the average absolute value of the difference between actual and forecasted use, divided by the actual use). A benchmark forecast is calculated by assuming that water use for a given week or month in 1981 is the same as the average for the corresponding period from 1976 to 1980. This method produces an AARE of 14.6 percent for one step ahead forecasts of monthly data and 15.8 percent for weekly data. A cascade model using trend, seasonality and autocorrelation produces forecasts with AARE of about 12 percent for both monthly and weekly data while adding a linear relationship of water use and rainfall reduces the AARE to 8 percent in both cases if it is assumed that rainfall is known during the forecast period. Simple rainfall predictions do not increase the forecast accuracy for water use so the major utility of relating water use and rainfall lies in forecasting various possible water use sequences conditioned on sequences of historical rainfall data.     

Development and Operational Testing of a Super‐Ensemble Artificial Intelligence Flood‐Forecast Model for a Pacific Northwest River

Coastal catchments in British Columbia, Canada, experience a complex mixture of rainfall‐ and snowmelt‐driven contributions to flood events. Few operational flood‐forecast models are available in the region. Here, we integrated a number of proven technologies in a novel way to produce a super‐ensemble forecast system for the Englishman River, a flood‐prone stream on Vancouver Island. This three‐day‐ahead modeling system utilizes up to 42 numerical weather prediction model outputs from the North American Ensemble Forecast System, combined with six artificial neural network‐based streamflow models representing various slightly different system conceptualizations, all of which were trained exclusively on historical high‐flow data. As such, the system combines relatively low model development times and costs with the generation of fully probabilistic forecasts reflecting uncertainty in the simulation of both atmospheric and terrestrial hydrologic dynamics. Results from operational testing by British Columbia's flood forecasting agency during the 2013‐2014 storm season suggest that the prediction system is operationally useful and robust.     

Surface Water Seepage Effects on Shallow Ground‐Water Quality Along the Rio Grande in Northern New Mexico1

Abstract: Interactions between surface irrigation water, shallow ground water, and river water may have effects on water quality that are important for both drinking water supplies and the ecological function of rivers and floodplains. We investigated water quality in surface water and ground water, and how water quality is influenced by surface water inputs from an unlined irrigation system in the Alcalde Valley of the Rio Grande in northern New Mexico. From August 2005 to July 2006, we sampled ground water and surface water monthly and analyzed for concentrations of major cations and anions, specific conductance, pH, dissolved oxygen, and water levels. Results indicate that irrigation ditch seepage caused an increase in ground water levels and that the Rio Grande is a gaining stream in this region. Temporal and spatial differences were found in ion concentrations in shallow ground water as it flowed from under the ditch toward the river. Ground‐water ion concentrations were higher when the ditch was not flowing compared with periods during peak irrigation season when the ditch was flowing. Ditch inputs diluted ion concentrations in shallow ground water at well positions near the ditch. Specifically, lower ion concentrations were detected in ground water at well positions located near the ditch and river compared with well positions located in the middle of an agricultural field. Results from this project showed that ditch inputs influenced ion concentrations and were associated with ground‐water recharge. In arid region river valleys, careful consideration should be given to management scenarios that change seepage from irrigation systems, because in some situations reduced seepage could negatively affect ground‐water recharge and water quality.     

Dynamic Modeling of Ground‐Water Quality Using Bayesian Techniques1

Abstract: Water industry experts have been arguing that the traditional techniques are not an accurate means of measuring water contamination. This is mainly because these techniques emphasize neither the stochastic nature of the water contamination process nor the precision and the accuracy of the tested methods used by environmental laboratories. In this work, we describe the development and application of prototype Dynamic Bayesian Networks (DBNs) that model ground‐water quality to determine the impact of chemical contaminants on ground‐water quality in the Salalah area, which is allocated to the south of Oman. We also present a new technique for data pre‐processing because it is needed for the treatment of ground‐water datasets that are used as the data source to learn the probabilities for dynamic decision models. Among more than 20 wells in area, only four wells were selected to be analyzed and the results show that we achieved an acceptable level of efficiency.     

Estimating Hydraulic Properties of Volcanic Aquifers Using Constant‐Rate and Variable‐Rate Aquifer Tests1

Abstract: In recent years the ground‐water demand of the population of the island of Maui, Hawaii, has significantly increased. To ensure prudent management of the ground‐water resources, an improved understanding of ground‐water flow systems is needed. At present, large‐scale estimations of aquifer properties are lacking for Maui. Seven analytical methods using constant‐rate and variable‐rate withdrawals for single wells provide an estimate of hydraulic conductivity and transmissivity for 103 wells in central Maui. Methods based on constant‐rate tests, although not widely used on Maui, offer reasonable estimates. Step‐drawdown tests, which are more abundantly used than other tests, provide similar estimates as constant‐rate tests. A numerical model validates the suitability of analytical solutions for step‐drawdown tests and additionally provides an estimate of storage parameters. The results show that hydraulic conductivity is log‐normally distributed and that for dike‐free volcanic rocks it ranges over several orders of magnitude from 1 to 2,500 m/d. The arithmetic mean, geometric mean, and median values of hydraulic conductivity are respectively 520, 280, and 370 m/d for basalt and 80, 50, and 30 m/d for sediment. A geostatistical approach using ordinary kriging yields a prediction of hydraulic conductivity on a larger scale. Overall, the results are in agreement with values published for other Hawaiian islands.     

FLOOD STAGE FORECASTING WITH SUPPORT VECTOR MACHINES1

ABSTRACT: Machine learning techniques are finding more and more applications in the field of forecasting. A novel regression technique, called Support Vector Machine (SVM), based on the statistical learning theory is explored in this study. SVM is based on the principle of Structural Risk Minimization as opposed to the principle of Empirical Risk Minimization espoused by conventional regression techniques. The flood data at Dhaka, Bangladesh, are used in this study to demonstrate the forecasting capabilities of SVM. The result is compared with that of Artificial Neural Network (ANN) based model for one‐lead day to seven‐lead day forecasting. The improvements in maximum predicted water level errors by SVM over ANN for four‐lead day to seven‐lead day are 9.6 cm, 22.6 cm, 4.9 cm and 15.7 cm, respectively. The result shows that the prediction accuracy of SVM is at least as good as and in some cases (particularly at higher lead days) actually better than that of ANN, yet it offers advantages over many of the limitations of ANN, for example in arriving at ANN's optimal network architecture and choosing useful training set. Thus, SVM appears to be a very promising prediction tool.     

SOIL MOISTURE PREDICTION USING SUPPORT VECTOR MACHINES1

ABSTRACT: Herein, a recently developed methodology, Support Vector Machines (SVMs), is presented and applied to the challenge of soil moisture prediction. Support Vector Machines are derived from statistical learning theory and can be used to predict a quantity forward in time based on training that uses past data, hence providing a statistically sound approach to solving inverse problems. The principal strength of SVMs lies in the fact that they employ Structural Risk Minimization (SRM) instead of Empirical Risk Minimization (ERM). The SVMs formulate a quadratic optimization problem that ensures a global optimum, which makes them superior to traditional learning algorithms such as Artificial Neural Networks (ANNs). The resulting model is sparse and not characterized by the “curse of dimensionality.” Soil moisture distribution and variation is helpful in predicting and understanding various hydrologic processes, including weather changes, energy and moisture fluxes, drought, irrigation scheduling, and rainfall/runoff generation. Soil moisture and meteorological data are used to generate SVM predictions for four and seven days ahead. Predictions show good agreement with actual soil moisture measurements. Results from the SVM modeling are compared with predictions obtained from ANN models and show that SVM models performed better for soil moisture forecasting than ANN models.     

FROZEN SOIL IMPACT ON GROUND WATER‐SURFACE WATER INTERACTION1

ABSTRACT: Ground water and surface water interaction in the prairie pothole region of the United States and Canada is seasonally dominated by the presence of thick, frozen soil layers that affect infiltration. During a spring thaw, the subsoil may remain frozen, preventing infiltration. The impact of the frozen soil layer on the timing of infiltration of depressional‐focused recharge to the ground water is not clearly understood. The objective of this paper is to relate changes in the water table during spring to changes in frost depth and soil water content. A depression and adjacent upland study site were instrumented with CRREL‐type frost tubes, neutron probe access tubes, and ground water monitoring wells. Increases in water table levels in a depression occurred before the frost layer decomposed and infiltrating water quickly formed a recharge mound. Water table responses at the upland site took place as two events. The first event was a gradual rise, probably caused by the lateral dissemination of the recharge mound. The second rise was a rapid rise coinciding with the decomposition of the soil frost layer. Because of the accumulation of surface water in depressions, agricultural practices that remove water from a field can affect water resources management by limiting the addition of water recharge to unconfmed ground water.     

Drought Resilience of the California Central Valley Surface‐Ground‐Water‐Conveyance System1

Abstract: A series of drought simulations were performed for the California Central Valley using computer applications developed by the California Department of Water Resources and historical datasets representing a range of droughts from mild to severe for time periods lasting up to 60 years. Land use, agricultural cropping patterns, and water demand were held fixed at the 2003 level and water supply was decreased by amounts ranging between 25 and 50%, representing light to severe drought types. Impacts were examined for four hydrologic subbasins, the Sacramento Basin, the San Joaquin Basin, the Tulare Basin, and the Eastside Drainage. Results suggest the greatest impacts are in the San Joaquin and Tulare Basins, regions that are heavily irrigated and are presently overdrafted in most years. Regional surface water diversions decrease by as much as 70%. Stream‐to‐aquifer flows and aquifer storage declines were proportional to drought severity. Most significant was the decline in ground water head for the severe drought cases, where results suggest that under these scenarios the water table is unlikely to recover within the 30‐year model‐simulated future. However, the overall response to such droughts is not as severe as anticipated and the Sacramento Basin may act as ground‐water insurance to sustain California during extended dry periods.