摘要:

Optimal allocation of water to sequential irrigators is considered with emphasis on downstream salinity damages caused by salt‐concentrating mechanisms inherent in irrigated agriculture. The allocation is achieved by using a dynamic programing framework with decision variables of amount diverted, land irrigated, and amount of irrigation canals lined. This framework is used to investigate a scheme whereby water is purchased and held for downstream use solely as a dilutant. The basis is established in principle for an operational, water rights market where all purchasers are financed by downstream beneficiaries. Several examples are give

ISSN:0043-1397    

DOI:10.1029/WR013i002p00225

年代:1977

数据来源: WILEY

摘要:

Synergistic gains are defined as the gains in benefits due to joint operation of a system of reservoirs in excess of the benefits from optimal individual operation. These gains are a result of both the deterministic (regional) and the stochastic diversity of flows into each of the reservoirs. They are captured by the use of flexible reservoir operating rules which base release decisions for a given reservoir on the state of the entire system, not just on the state of that reservoir. One possible operating rule is formulated and is demonstrated to be highly effective in capturing synergistic gains. A hypothetical design problem of sizing three water supply reservoirs in parallel is solved by three methods. The first method assumes that no synergistic gains are achievable. The second and third methods recognize the existence of such gains. The method recommended here incorporates the operating rule into the design process and makes use of the observation that the system yield is a function of system capacity.

ISSN:0043-1397    

DOI:10.1029/WR013i002p00239

年代:1977

数据来源: WILEY

摘要:

A stochastic dynamic programing model for determining reservoir release rules which maximize expected net benefits subject to reliability constraints on system performance is presented within the framework of Lagrangian duality theory. The approach characterizes optimal release rules under discounting of benefits as being nonstationary. It also indicates when there exists a randomized rule which dominates the optimal deterministic rule. The methodology for obtaining these rules uses dual optimization to provide bounds on the optimal solution and a branch and bound procedure to resolve any duality gap which may occur if the bounds are not equal. The model is applied to a group of hypothetical problems for which nonoptimal policies had previously been found by using chance‐constrained dynamic programing (Askew, 1974b

ISSN:0043-1397    

DOI:10.1029/WR013i002p00247

年代:1977

数据来源: WILEY

摘要:

Standardized measurements of width at gaging stations on high‐gradient stream channels show that a power function relation, indicative of static allometric growth, exists between width and average discharge in the downstream direction. Results are in close agreement with studies by Leopold and Maddock (1953) and confirm their ‘b’ exponent. Workers currently using active channel width measurements as a means of estimating average discharge from ungaged basins should consider 2.0 as a standard exponent of the regression rel

ISSN:0043-1397    

DOI:10.1029/WR013i002p00256

年代:1977

数据来源: WILEY

摘要:

Analysis of an excellent 30‐year record of rainfall and storm flow (545 events) from a 3‐mi2forested watershed in the southern Appalachians was made to determine whether rainfall intensity influences storm flow volume or peak discharge per unit area. For all practical purposes, hourly and minutely rainfall intensities during storms had no effect on storm flow volumes delivered by the basin. Storm rainfallP, antecedent flowI, season (winter or summer), and duration of the rainstormDaccounted for 86.4% of the total variation in the log of storm flow. Addition of maximum 60‐, 30‐, 15‐, and 5‐min intensities raised this to 86.7% (the increase was not significant at the 0.05 level) but reduced the standard error (0.354 in logs) by less than ½%. The same four variables (P,I,S, andD) accounted for 72% of the variation in the log of peak flow. Addition of the intensity variables raised this to 76.7% (increase significant at the 0.01 level). Thus only 4.7% of the total variation in the log of peak flow was attributable to intensity. Analysis by eight storm rainfall classes reinforced the conclusion that intensity added nothing to storm rainfall as a predictor of storm flow regardless of storm magnitude.P,I,S, andDaccounted for 25–65% of the total variation in the log peak flows within storm classes; addition of all intensity variables raised this to 36–85%, but the increases were nonsignificant at the 0.05 level in four of the eight classes. A strong inference is made that neither estimates nor measurements of hourly and minutely rainfall intensity are of practical value in predicting storm flow volume from this forested basin. Furthermore, intensity played a much smaller role in peak discharge generation on this basin than was expected. The hypothesis is offered that rainfall intensity variation is not a dominant variable controlling flood generation in humid region, wild

ISSN:0043-1397    

DOI:10.1029/WR013i002p00259

年代:1977

数据来源: WILEY

摘要:

A 65‐ac (26.3‐ha) cedar‐hemlock catchment in northern Idaho was sprayed with a 5% aqueous emulsion of silicone oil to reduce transpiration. The antitranspirant, Dow Corning XEF‐4‐3561, was applied by helicopter at the rate of 40 gal./ac (375 1/ha). From June 1 to September 15, 1974, streamflow, soil water, and leaf water potential were monitored on the 65‐ac (26.3‐ha) treated watershed and an adjacent 50‐ac (20.2‐ha) control watershed. By using the paired watershed method with regression equations based on 6 years of pretreatment data, predicted and observed streamflow were obtained for the period June 1 to September 15. Differences between predicted and observed streamflow were compared by using the paired ‘t’ test and found significant at the 97.5% confidence level. During a 63‐day period a 12% increase in streamflow resulted from the antitranspirant application. During this same period, soil water storage was greater on the treated watershed than on the control watershed; however, the difference was not statistically significant. Seasonal leaf water potential in cedar indicated that stress was greater on the control watershed than on the treated watershed. Systematic differences in leaf water potential fo

ISSN:0043-1397    

DOI:10.1029/WR013i002p00267

年代:1977

数据来源: WILEY

摘要:

A general method for the economic design of natural resource information systems is presented for a certain class of natural phenomena. The system design is determined by the interaction of the technical input‐output relationship, i.e., the production function, the set of resource constraints, and an economic loss function defined in terms of parameter uncertainty. An application of the proposed method to streamflow network design is presented. Results of this analysis indicate that the method is fairly robust with respect to the assumptions. Observations are made which suggest extensions to flood measurement networks, long‐term precipitation networks, and seismic observation network des

ISSN:0043-1397    

DOI:10.1029/WR013i002p00273

年代:1977

数据来源: WILEY

摘要:

Formulation of two models of long‐term persistence, fast fractional Gaussian noise (ffGn) and the first‐ order autoregressive‐first‐order moving average (Arma (1, 1)) process for a three‐parameter log normal and three‐parameter gamma distribution are given. For the three‐parameter log normal distribution the marginal probability distribution is generated exactly for both models, but the desired autocorrelation functions are distorted. Use of the three‐parameter log normal distribution requires a transformation of the lag one correlation coefficient and the Hurst coefficient for the ffGn model and of the parameters ϕ and θ for the Arma (1, 1) model. Because of the nonlinearity of the three‐parameter log normal transformation, sequences with slightly higher long‐term persistence (largerHor ϕ) must be generated in the logarithmic domain to achieve the desired model properties in the skewed domain. For the gamma distribution, the autocorrelation function is preserved exactly, but the desired marginal probability distribution is distorted. Monte Carlo tests showed that this distortion is evidenced primarily in the loss of the theoretical lower bound of the gamma distribution. This distortion is most extreme for time series having high long‐term persistence (Hor ϕ) and large coefficients of variation. For operational applications, there appeared to be no clear advantage to use of either of

ISSN:0043-1397    

DOI:10.1029/WR013i002p00281

年代:1977

数据来源: WILEY

摘要:

A Bayesian framework for the synthetic generation of annual Streamflows from a multivariate first‐ order autoregressive model is presented. This framework allows the user to include directly the effects of the parameter uncertainties in the evaluation of proposed projects through simulation with the synthetically generated records. The model produces synthetic traces with higher variances than those in the historical records when the records are too short to estimate reliably the ‘time’ values of the parameters of the process. This multivariate model is a natural extension of the model presented by Vicens et al. (

ISSN:0043-1397    

DOI:10.1029/WR013i002p00291

年代:1977

数据来源: WILEY

摘要:

The problem of determining the optimal operation of a montly regulated two‐purpose reservoir is discussed in this paper. A basic assumption is made about the decision‐making structure of the system in the sense that the operation is not assumed to result from a supervisor's choice but from the users' trade‐off. The solution of the problem is found by means of a game theory approach in both the noncooperative and the cooperative situa

ISSN:0043-1397    

DOI:10.1029/WR013i002p00296

年代:1977

数据来源: WILEY