The Texas District of the Water Resources Division of the U.S. Geological Survey has collected and analyzed hydrologic data since 1953 to define the effects of systems of floodwater-retarding structures on downstream water and sediment yield. The district project includes 11 study areas ranging from 18 to 80 square miles in size and from 0 to 67 in percent of study area controlled by floodwaterretarding structures. The 11 study areas are within that part of Texas where the west-to-east average annual runoff ranges from about 2 to 7 inches. This report presents results of analyses, development of methodolgy, and results of application of methods for defining the downstream effects of systems of floodwaterretarding structures.

Annual inflow to and outflow from the system of floodwater-retarding reservoirs in seven of the 11 study areas were found to be related by the equation: O=0.98/ 0.68, where O is annual outflow, in inches, and / is annual net inflow, in inches. Transmission loss of structure outflow to the downstream study-area stream-gaging station was determined and compared with the transmission loss of natural flood flow between tandem stream-gaging stations on Denton Creek, a tributary to Elm Fork Trinity River above Dallas.

Trap efficiency of most floodwater-retarding structures was found by the U.S. Soil Conservation Service to be about 97 percent. Downstream increases in suspended-sediment concentration in the outflow were found to be large in a study area with mostly silt and clay sediments, but even a large increase in suspendedsediment concentration did not represent a significant quantitative pickup of sediment by the outflow water.

Water consumption in floodwater-retarding reservoirs from the combined actions of evaporation, evapotranspiration, and seepage was found to be as much as twice the average annual consumption attributable to evaporation alone. Average annual consumption in reservoirs in the seven study areas analyzed ranged from 1.57 inches of equivalent runoff in the easternmost study area, where annual runoff averaged 6.96 inches, to 0.77 inch of equivalent runoff in the westernmost study area, where the average annual runoff was 2.35 inches. The effect of consumption on downstream flow is partially offset by rainfall on pool surface. Studies covering as much as 15 years of streamflow record at the stream-gaging stations that gage outflow from the Deep and Honey Creek study areas indicated no increase in base flow.

Multiple-linear-regression techniques were used in developing methodology to determine reservoir consumption in seven study areas. The physical and climatic fnctors influencing consumption were grouped as variables in regard to their relative effect on the actions of evaporation, evapotranspiration, and seepage. The resulting generalized equation was then used in synthesizing the consumptive effects of a planned system of 162 floodwater-retarding reservoirs controlling 26 percent of a 1,660-square-mile drainage basin upstream from a major water-supply reservoir. The analyses were based on the assumption that all water consumed at the floodwater-retarding reservoirs would have reached the downstream watersupply reservoir. Water-sediment discharge relationships were derived for the runoff into the structures as well as for the runoff through and below the structures. A mathematical response model of the floodwater-retarding reservoir systems and the entire drainage basin was computer programed to yield monthly water and sediment inflow to the water-supply reservoir.

Results of the response model showed that with full development, depletion of annual yield to the large reservoir would be as much as 10 percent in the early years; but after the permanent pools of the floodwater-retarding structures had mostly filled with sediment, depletion of annual yield would be generally less than 1 percent. The depletion of yield to Garza-Little Elm Reservoir during the 39-year synthesized period of study was estimated as 296,800 acre-feet out of 18,256,000 acre-feet total yield. During the same period, the floodwater-retarding structures were estimated to have kept 19,700 acre-feet of sediment from being deposited in the reservoir.

"Firm"- or "critical"-yield studies were made of the large reservoir on the basis of two sets of conditions : with floodwater-retarding structures in the drainage basin, and without such structures. Results of the firm-yield studies indicated that with full development, annual firm yield would be initially reduced by 10 percent. After 30 or more years, when the permanent pools of the floodwaterretarding reservoirs would be mostly filled with sediment, the firm yield would be almost the same with or without the upstream development.