Principal Investigator: James R. Wallace (Georgia Institute of Technology)
Principal Investigator: Unal A. Sorman (Georgia Institute of Technology)
Sponsor: GWRI
Start Date: 1971-07-01; Completion Date: 1972-08-01;
Keywords:
Description:
The purpose of this study was to obtain a better understanding of the temporal and spatial variability of thunderstorm rainfall and to develop a digital model for the stochastic simulation of thunderstorm rainfall for the Southeast Coastal Plain areas.
In the present study rainfall cells were thoroughly analyzed from rainfall data made available by the Agricultural Research Service, U.S. Department of Agriculture. The data source was a dense network of raingages located over the Little River Experimental watershed near Tifton, Georgia. The study area is monitored by a network of 55 digital type raingages covering a 250 square mile area. Statistical properties and frequency distributions of cell characteristics, such as cell duration, size, spatial and temporal distribution of rainfall intensity, cell movement, and number and orientation of cells, were analyzed. A conceptual model of thunderstorm rainfall was formulated from the observed storm cell characteristics and the parameters of the model were evaluated from isolated thunderstorms which occurred in the summer of 1967. The model was coded for a digital computer and a number of rainfall events were generated by the dynamic model, which is based on the stochastic generation of rainfall patterns from thunderstorm cells.
These simulated precipitation sequences preserved certain of the fundamental statistics of the historical thunderstorm rainfall records. The model was validated by comparing synthetic precipitation events with events observed on Little River watershed during the summers of 1968, 1969, and 1970. Rainfall characteristics which are considered to be representative of the most important features of thunderstorm rainfall were analyzed during the model validation. For the validation of the model a number of statistical rainfall parameters determined from simulated rainfall, such as frequency distribution of the maximum amount of rainfall, maximum accumulated rainfall versus duration of rainfall at the maximum rainfall raingage, and maximum ten-minute rainfall intensity, were compared with the 1968, 1969, and 1970 historical data. In addition, relation ships between correlation coefficients and spacing between the first and the second maximum rainfall gages, as well as time lag of rainfall between them, were derived graphically and a comparison was made between the simulated and observed results. The performance of the model was considered to be successful on the basis of comparisons made between the observed and simulated rainfall characteristics.
This study has lead to the development of a body of knowledge on the characteristics of summer thunderstorm rainfall in the Coastal Plain of Georgia. The size, movement, and intensity of rainfall thunderstorm cells has been measured, and a stochastic model has been developed which will generate precipitation patterns like those observed by the raingage network. The dependence of the characteristics of individual cells on the location, movement, and size of other cells already existing in the same general area needs additional study. Such a study will require a network of gages covering an area larger than that available for the current study or a study using radar measurements in combination with a raingage network.
It is suggested that the simulation model can be used in conjunction with watershed models for generation of synthetic streamflows and that the knowledge gained through this study will aid in the efficient use of water resources throughout the Coastal Plain areas.