Submitted to: Journal of Hydrology
Publication Type: Peer reviewed journal
Publication Acceptance Date: 4/17/2001
Publication Date: N/A
Citation: Interpretive Summary: This is a paper which helps scientists understand why measurements from experimental plots vary so much from plot to plot. In particular, we are trying to understand why the amount of surface water runoff, which is made during rains, can be so different from one plot to the next, even when the plots appear for all practical purposes to be the same. To study the problem we used data from an experiment conducted in Missouri in 1981 on 40 plots, which were on the same hillslope and were all treated exactly the same. We evaluated the data using a fairly sophisticated numerical model, which computes the infiltration of water into the soil and the amount that runs off the surface. The answer to our question lies in the fact that all soils contain some natural variability within the plot. Some areas of the plot have higher rates of water intake than others. So during the rain a part of the plot may generate runoff which then moves to a different part of the plot and may, if conditions are right, infiltrate. Differences between 2 plots can occur just due to the way the areas of high intake and low intake is arranged spatially in the plot, even when the average water intake rates on the two plots are exactly the same! The impact of this work is that scientists will better understand the processes they are trying to model, and we will have better ability to know how to use hydrologic and erosion models to solve real world problems, which translates to better management of soil and water resources.
Technical Abstract: It is expected that runoff volumes from plots can be quite variable, but reasons for this variability are poorly understood. These variations can be important not only for understanding the hydrologic system, but also relative to evaluating the effectiveness of infiltration and runoff prediction models. In this study we investigated the sources of variability among 40 replications in a previously reported experiment on fallow plots located on a claypan soil in Missouri, USA. A numerical model was calibrated using data from the experiment and from other published data on the variability of soil properties. The results suggest that approximately 75 percent of the observed variation among the replicated plots could be explained by the spatial variability of Ks, surface storage, and the depth to claypan. Another approximately 20 percent was estimated to be due to variability in rainfall among plots and measurement error. Changes in the relative differences in runoff between plots may be explainable by the modification of the spatial distribution of Ks and surface storage, which occurs during tillage. The introduction of these sources of variability in the model formulation produced a realistic description of the variance of the observed values of runoff, as well as a relatively clear delineation between the explained and unexplained variability. These results can serve as an index of model performance in predicting observed data.