Construction and Calibration of Large-Scale Rainfall Simulators

Abstract

Large-scale rainfall simulation offers valuable insight into the effectiveness of erosion control practice. While small and intermediate-scale apparatuses can analyze the effects of splash and sheet erosion on a slope, large-scale plots allow the formation of rills, which better represent real applications on long slopes such as highway embankments. Construction and calibration methods for large-scale rainfall simulators are standardized by ASTM D6459-19. As part of Phase II of “Evaluation of ALDOT Erosion Control Practices using Rainfall Simulation on Various Soil Types and Slope Gradients,” twelve ASTM D6459-19 rainfall simulators were constructed, calibrated, and used to obtain soil erodibility factors in bare-soil testing. Efficient construction practices for ASTM D6459-19 rainfall simulators include using pre-made tanks for catchment basins, using lumber for plot borders, and creating a portable rainfall simulation system with a manifold for flow distribution. These techniques along with others included in this work can reduce the cost of constructing large-scale rainfall simulators, which allows for increased ability to test on various slopes and soils. Calibration analysis was performed to evaluate direct measurement of runoff for intensity calibration. The results indicate that the runoff volume method produced statistically different results than the ASTM D6459-19-recommended rainfall gauge method. The runoff from the impermeably covered plot was 27.8% to 32.8% less than the rainfall gauges predicted. Furthermore, there was not enough evidence to conclude that 20 and 6 observation rainfall gauge setups were statistically different. Photography was investigated for determining raindrop characteristics including size and velocity. Results indicated that the photography method represented a greater proportion of raindrops smaller than 1.68 mm diameter than the ASTM D6459-19 flour pan method. The photography method yielded a raindrop erosivity factor 32.8% less than the flour pan method for the center of the plot. Raindrop velocity results of the photography method yielded statistically different velocities than theoretically predicted based on drop size and fall height. The photographically determined velocities were calculated as the distance traveled by a raindrop over the time the camera captured the photograph, called the shutter speed, and the photographs yielded 25.8% lower velocities than predicted on average. The twelve new rainfall simulators at AU-SRF were calibrated and tested using ASTM D6459-19 methods. Using the same sprinkler design as the original AU-SRF rainfall simulator, the theoretical rainfall erosivity factor (R) with target intensities was 148.5. The first three bare soil control tests on the new rainfall simulator plots yielded soil erodibility factors (K) of 0.18, 0.06, and 0.02 for ASTM sand, loam, and clay, respectively. This thesis includes documentation and analysis of rainfall simulator construction, calibration methods, and control testing. The continuity of rainfall simulator methodology is critical to the continued precise evaluation of erosion control practices at Auburn University- Sediment Research Facility and other facilities.