Variability of Runoff and Soil Loss from Fallow Experimental Plots

摘要: The amount of unexplained variation in runoff and soil loss studies often limits the interpretation data. In this study, variability from 40 essentially uniform experimental plots was examined for 25 natural rainfall events occurring during a 155d period. Plots had been maintained uniformly prior 3 yr were kept fallow with periodic cultivation study Except producing low loss, event coefficients relatively constant both at about 20%. Differences among varied event. Only minor amounts observed could be attributed to any several measured plot properties, differences expressed by did not persist or subsequent observations site. huge shows that replications treatments are needed confidently estimate mean comparison purposes effects factors having on may difficult detect statistically. fact most is indicates important processes currently understood. Additional Index Words: erosion, spatial variability, simulation models. Wendt, R.C., E.E. Alberts, A.T. Hjelmfelt, Jr. 1986. Variability plots. Soil Sci. Soc. Am. J. 50:730-736. P PLOTS have long used examine management under simulated rainfall. where replicated, experimenters noted large replicates (Browning et al., 1948; Bryan, 1981; Gard Van Doren, 1949; Johnson 1984; Mueller Nyhan Simanton Renard, 1982). Sources obvious, although such as tillage induced (Gard 1949), formation breakup debris dams (Simanton 1982), combined influence raindrop size, surface water films, aggregate stability (Bryan, 1981) suggested potential sources. magnitude treated alike concern interpreting results experiments design. With variation, treatment means purposes. considered identical also development mathematical models simulating loss. Because abstractions, conditions which model 1 Contribution Watershed Research Unit, Agricultural Service, USDA, 207 Business Loop 70 E, Columbia, MO 65203. Received 5 July 1985. 2 Scientists Hydraulic Engineer, respectively, USDA-ARS, MO. parameters gives an indication importance unaccounted variables judge performance. we major affecting constant. Although data reported earlier (Hjelmfelt Burwell, 1984), they subjected further analysis variability. METHODS conducted located near Kingdom City, MO, arranged illustrated Fig. 1. Each 3.2-m wide, 27.4-m long, oriented parallel 3.5% slope. Adjacent separated 2.13-m wide border strip. site Mexico silt loam (fine, montmorillonitic, mesic Udollic Ochraqualfs), has slowly permeable layer illuvial clay (claypan) beginning depths 0.2 0.3 m. Runoff associated sediment collected two volumetrically calibrated tanks series base each plot. first tank collects up depth equivalent 6 mm after overflow channeled through multislot divisor conveys one-ninth second tank. Total capacity 150 runoff. After event, sampled determine content. Rainfall adjacent using single recording raingauge. no available document over area reported, 15 nonrecording gauges grid Configuration