PROBABILITIES OF MONTHLY FLOW ON AN EPHEMERAL STREAM William Dvoranchika Lucien Ducksteina Chester C. Kisiel INTRODUCTION In order to generate synthetic sequences of streamflows on aridland channels for subsequent evaluation of proposed water resource projects, it is necessary to study the structure of historical sequences of flow. Given the 58record of monthly flows on Rillito Creek at Tucson, Arizona, the objective of the research was to determine its stochastic properties so as to develop a model of the historic flows. The stochastic properties of interest included (a) statistical moments (arithmetic mean, variance and skewness) over each of the twelve months over the period of record, (b) frequency of dryness (number of months with zero flows), (c) crosscorrelation between flows over adjacent months, and (d) state transition probabilities for flows in a specified range in one month given that flows were in a specified range in a previous month. This study was intent on a critical evaluation of the value of the lag -one autoregressive model (Fiering, 1967) for modeling of ephemeral flow. The findings of this previously unreported study led to consideration of event -based models (Kisiel, et al., 1971; Denny, et al., 1971).

Objective and Subjective Analysis of Transition Probabilities of Monthly Flow on an Ephemeral Stream

Convergence of benefit differences

The use of discrete conditional dependency matrices as input to stochastic decision models is examined. Some of the problems and initial assumptions involved with the construction of the above mentioned matrices are discussed. Covered in considerable detail is the transform used to relate the gamma space with the normal space. A new transform is introduced that should produce reasonable results when the record of stream flow (data) has a highly skewed distribution. Finally, the possibility of using the matrices to provide realistic inputs to a stochastic dynamic program is discussed.

Deme Michael Clainos

Papers

Input specifications to stochastic decision models