A COMPARISON OF STREAMFLOW GENERATION MODELS FOR RESERVOIR CAPACITY-YIELD ANALYSIS1

ABSTRACT: An analysis of four streamflow generation schemes for the use in the estimation of the required conservation storage for a single reservoir is presented. The comparison of the generating schemes should aid in the selection of an appropriate model type for the reservoir sizing problem. The streamflow generation models are compared using two criteria. The first comparison is between the statistics of the generated streamflow sequences and the corresponding statistics from the historical record. The second evaluation compares the median reservoir size determined by each model with the required storage based on the historical flow sequence. The results of a comparative analysis for monthly streamflow data for the Rzav River in Yugoslavia are presented and discussed. The results indicate that both evaluation criteria are required to discriminate between the various options.     

IS IT REALISTIC TO DEFINE A 100-YEAR DROUGHT FOR WATER MANAGEMENT?1

ABSTRACT: The concept of recurrence interval has been used for years in engineering designs. Can the same concept be applied to the drought analysis? This paper uses the plotting position method to define drought of various recurrence intervals based on stream-flow data. The method of truncation level was applied to the same data to examine the defined drought. Based on the method of truncation level, drought duration and its corresponding flow deficit were investigated. Eighteen flow gage stations from the Scioto River Basin in Ohio were selected for the study. The results show that flows of 100-year droughts using the plotting position method are practically nil. On the other hand, flows of droughts using the truncation method are gradually decreasing with an increase in truncation level, where flows of 95 percent are approximately equal to those of two-year droughts defined by the plotting position. It is also shown that there is a strung correlation between drought duration and deficit.     

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.     

COMPARISON OF PROCESS‐BASED AND ARTIFICIAL NEURAL NETWORK APPROACHES FOR STREAMFLOW MODELING IN AN AGRICULTURAL WATERSHED1

ABSTRACT: The performance of the Soil and Water Assessment Tool (SWAT) and artificial neural network (ANN) models in simulating hydrologic response was assessed in an agricultural watershed in southeastern Pennsylvania. All of the performance evaluation measures including Nash‐Sutcliffe coefficient of efficiency (E) and coefficient of determination (R²) suggest that the ANN monthly predictions were closer to the observed flows than the monthly predictions from the SWAT model. More specifically, monthly streamflow E and R² were 0.54 and 0.57, respectively, for the SWAT model calibration period, and 0.71 and 0.75, respectively, for the ANN model training period. For the validation period, these values were ?0.17 and 0.34 for the SWAT and 0.43 and 0.45 for the ANN model. SWAT model performance was affected by snowmelt events during winter months and by the model's inability to adequately simulate base flows. Even though this and other studies using ANN models suggest that these models provide a viable alternative approach for hydrologic and water quality modeling, ANN models in their current form are not spatially distributed watershed modeling systems. However, considering the promising performance of the simple ANN model, this study suggests that the ANN approach warrants further development to explicitly address the spatial distribution of hydrologic/water quality processes within watersheds.     

Short-term wind speed forecasting based on random forest model combining ensemble empirical mode decomposition and improved harmony search algorithm

ABSTRACT

Wind speed forecasting plays an important role in power grid dispatching management. This article proposes a short-term wind speed forecasting method based on random forest model combining ensemble empirical modal decomposition and improved harmony search algorithm. First, the initial wind speed data set is decomposed into several ensemble empirical mode functions by EEMD, then feature extraction of each sub-modal IMF is performed using fast Fourier transform to solve the cycle of each sub-modal IMF. Next, combining the high-performance parameter optimization ability of the improved harmony search algorithm, two optimal parameters of random forest model, number of decision trees, and number of split features are determined. Finally, the random forest model is used to forecast the processing results of each submodal IMF. The proposed model is applied to the simulation analysis of historical wind data of Chaoyang District, Liaoning Province from April 27, 2015 to May 22, 2015. To illustrate the suitability and superiority of the EEMD-RF-IHS model, three types of models are used for comparison: single models including ANN, SVM, RF; EMD combination models including EMD-ANN, EMD-SVM, EMD-RF; EEMD combination models including EEMD-ANN, EEMD-SVM, EEMD-RF. The analysis results of evaluation indicators show that the proposed model can effectively forecast short-term wind data with high stability and precision, providing a reference for forecasting application in other industry fields.     

HYDROLOGIC EFFECTS OF THE TONKAWA CREEK FLOOD ABATEMENT PROGRAM1

ABSTRACT: Floodwater-retarding impoundments, controlling 68 percent of the drainage area of Tonkawa Creek, a Washita River tributary in southwestern Oklahoma, have reduced the total flow volume about 36 percent over a 5-year period. Analyses showed the reduction occurred primarily in the less-than-2.5-cfs flow range, indicating the base flow regime has been altered. However, channelizing the downstream, mild-sloped, 3.6 miles of Tonkawa Creek that flows across a Washita River terrace increased the flow volume fourfold at the outlet. A double-masscurve analysis of water yield from a 1,127-square-mile Washita basin segment versus an untreated tributary showed the yield has not changed after 25 percent of the tributary area had been treated. Therefore, the flow reduction caused by structures is being offset by increased yields from channelization.     

Hydrograph Separation by Incorporating Climatological Factors: Application to Small Experimental Watersheds 1

Abstract: Evaluating the relative amounts of water moving through the different components of the hydrological cycle is required for precise management and planning of water resources. An important aspect of this evaluation is the partitioning of streamflow into surface (quick flow) and base‐flow components. A prior study evaluated 40 different approaches for hydrograph‐partitioning on a field scale watershed in the Coastal Plain of the Southeastern United States and concluded that the Boughton’s method produced the most consistent and accurate results. However, its accuracy depends upon the proper estimation of: (1) the end of surface runoff, and (2) the fraction factor (α) that is function of many physical and hydrologic characteristics of a watershed. Proper identification of the end of surface runoff was accomplished by using a second derivative approach. Applying this approach to 12 years of separately measured surface and subsurface flow data from a field scale watershed (study area) proved to be accurate for 87% of the time. Estimation of the α value was accomplished in this study using two steps: (1) alpha was fitted to individual hydrographs: and, (2) a regression equation that determines these alpha values based on climatological factors (e.g., rainfall, evapotranspiration) was developed. Using these strategies improved the streamflow partitioning method’s performance significantly.     

Characterizing a Major Urban Stream Restoration Project: Nine Mile Run (Pittsburgh,Pennsylvania, USA)

Urban stream restoration continues to be used as an ecological management tool, despite uncertainty about the long‐term sustainability and resilience of restored systems. Evaluations of restoration success often focus on specific instream indicators, with limited attention to the wider basin or parallel hydrologic and geomorphic process. A comprehensive understanding of urban stream restoration progress is particularly important for comparisons with nonurban sites as urban streams can provide substantial secondary benefits to urban residents. Here, we utilize a wide range of indicators to retrospectively examine the restoration of Nine Mile Run, a multi‐million dollar stream restoration project in eastern Pittsburgh (Pennsylvania, USA). Examination of available continuous hydrological data illustrates the high cost of failures to incorporate the data into planning and adaptive management. For example, persistent extreme flows drive geomorphic degradation threatening to reverse hydrologic connections created by the restoration and impact the improved instream biotic communities. In addition, human activities associated with restoration efforts suggest a positive feedback as the stream restoration has focused effort on the basin beyond the reach. Ultimately, urban stream restoration remains a potentially useful management tool, but continued improvements in post‐project assessment should include examination of a wider range of indicators.     

PREDICTIVE MODEL FOR NITRATE LOAD IN THE BULL RUN WATERSHED,OREGON1

ABSTRACT: Predictive models for nitrate in four streams in the Bull Run Watershed in the Cascade Mountains of Oregon were developed from a record of 17 years of nitrate samples. The models are time series regression models written in terms of Log(nitrate load). The independent variables are logarithm of 14-day mean daily stream discharge, current day's precipitation, logarithm of the previous day's precipitation, total precipitation for the previous seven days, a hydrograph position variable that indicates rising or falling limb, and average maximum air temperature for the preceding 14 days. The models describe annual cycle and seasonable trends and variations in nitrate load, but are unable to describe large day to day variations like those associated with hydrograph peaks.     

Adaptation of Climate Model Projections of Streamflow to Account for Upstream Anthropogenic Impairments

A statistical procedure is developed to adjust natural streamflows simulated by dynamical models in downstream reaches, to account for anthropogenic impairments to flow that are not considered in the model. The resulting normalized downstream flows are appropriate for use in assessments of future anthropogenically impaired flows in downstream reaches. The normalization is applied to assess the potential effects of climate change on future water availability on the Rio Grande at a gage just above the major storage reservoir on the river. Model‐simulated streamflow values were normalized using a statistical parameterization based on two constants that relate observed and simulated flows over a 50‐year historical baseline period (1964–2013). The first normalization constant is a ratio of the means, and the second constant is the ratio of interannual standard deviations between annual gaged and simulated flows. This procedure forces the gaged and simulated flows to have the same mean and variance over the baseline period. The normalization constants can be kept fixed for future flows, which effectively assumes that upstream water management does not change in the future, or projected management changes can be parameterized by adjusting the constants. At the gage considered in this study, the effect of the normalization is to reduce simulated historical flow values by an average of 72% over an ensemble of simulations, indicative of the large fraction of natural flow diverted from the river upstream from the gage. A weak tendency for declining flow emerges upon averaging over a large ensemble, with tremendous variability among the simulations. By the end of the 21st Century the higher‐emission scenarios show more pronounced declines in streamflow.