Water Supply Forecasts Research Articles

The use of linear programing techniques for estimating the benefits from increased accuracy of water supply forecasts

Probabilistic linear programing techniques and Bayesian statistics are combined in this paper by utilizing an example from the area of irrigated agriculture and water supply forecasts to assess the value of increased forecast accuracy to decision units. The case studied was confined to a two‐period analysis involving (1) a crop planting period and (2) a growing and harvesting period. The model is developed in terms of Bayesian analysis and demonstrates how linear programing can be applied to rather complex decision making problems involving uncertainty. Testing of the model involved the use of an IBM computer program available at the University of Michigan. For various assumptions as to supplemental water supply the model showed a net benefit to irrigators of about $6/acre for a reduction in uncertainty of 33% resulting from the introduction and improvement of water supply forecasts.

Water Supply Forecast for the Western United States

Water Supply Forecast for the Western United States

Water Supply Forecast for the Western United States

Water Supply Forecast for the Western United States

Water Supply Forecast for the Western United States

Water Supply Forecast for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Water Supply Forecasts for the Western United States

Runoff from rain and snow

The basic principles of the idealized hydrologic cycle are reviewed with emphasis on storage and movement of water in the soil. A distinction is made between ground‐water runoff and overland runoff in terms of storage and lag, expressed as accumulated deviations from uniform flow over a period of several years. These functions are presented for the period 1920 through 1945 for three rivers in central Oregon: John Day, Deschutes, and Metolius, which exemplify minimum, moderate, and maximum effects, respectively, of storage and lag. The Metolius River is shown to have extremely small fluctuations in discharge from year to year, ascribed to a great portion of its flow being derived from ground‐water runoff; and its response to fluctuations in annual precipitation lags behind that of the John Day River by about five years. Specific techniques and measures for improving seasonal water supply forecasts are suggested.

Accuracy of 1942 Oregon water-supply forecasts

Stream-flow throughout Oregon in 1942 exceeded the amount forecast except in one small area on the east slope of the Cascades near Bend. This behavior agreed with the distribution of precipitation during April to May. Since all forecasts are based on expectation of normal temperature and precipitation during runoff, the effect of State-wide, April to May precipitation 60 per cent, above normal was notable. Precipitation by districts, expressed as per cent above normal, followed Southeastern, 59; South-central, 86; North-central, 19; Columbia River, 43; Wallowa Mountains, 79; Blue Mountains, 65; Southern, 75; Willamette Valley, 65.