Accuracy of runoff measurements from small agricultural watersheds
Abstract
The purpose of this study was to examine current techniques of measuring runoff from small agricultural watershed and to develop tools to estimate the impact of those techniques on the validity of the resulting data. Preliminary studies showed that the common practice of forcing runoff through a flume results in significant ponding upstream of the flume, affecting movement of both water and associated sediment. Earlier studies showed errors up to 25% for hydrologic measurements and 50% for sediment movement estimates. A theoretical model of this effect was developed using a flood routing technique for water movement and a distorted plug approach for sedimentation, based on predicting the outflow shape of an inflow plug distorted by mixing with the ponded volume. With this shape, settling theory predicted deposition of sediment within the plug. The initial theory modeled this outflow plug shape based on a combination of jet and open channel flow theory, and was discarded when initial results proved some of its basic assumptions to be false. A second model was hypothesized, assuming the shape of the outflow plug as a beta distribution uniquely defined by a combination of minimum, maximum, mean, and standard deviation. This was tested against experimental data collected on a small-scale version of a watershed. The experiment examined the impact of different flow rates, pond shapes, time scales, and timing of the inlet plug. The theory predicted hydrograph distortion very well. The shapes of the outflow plugs were very similar to beta distributions, the means could be closely approximated by the theoretical mean detention time, and the maxima of the outflow plugs were accurately estimated as the maximum outflow times. The study found no good estimate of the minimum time for the plug to reach the outlet but showed a correlation to the inlet velocity and pond shape, and hypothesized that relative pond volume would also be important. The outflow plug standard deviation was correlated to a dimensionless mixing parameter based on the inlet and outlet flow energies, the pond volume, and the pond length, but no direct relationship between mixing parameter and standard deviation was developed.
Degree
Ph.D.
Advisors
Monke, Purdue University.
Subject Area
Agricultural engineering|Agronomy|Hydrology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.