Sampling accuracy of runoff and sediment from erosion plot tanks using bucket

Using erosion plots is a way to research on soil erosion and conservation that with different dimensions, which are constructed in the natural arenas or in laboratories. Mistakes and false conclusions are common task that should be avoided when sampling runoff and sediment from the tanks of these plots. A few researches have been done on the ways of sampling from these tanks and reducing the errors. In this study, the accuracy of sampling was investigated by placing a bucket in the tank and using bottle sampling. For this purpose, two 220 liter tanks were used. In the first tank located at the upper level, concentrations of sediment were prepared for five, 10 and 20 grL-1 and the mixture of runoff and sediment was transferred to the second tank. Coarse particles were trapped by a bucket of twenty liters placed underneath the importer tube in the second tank. The contents of the bucket were weighed in the laboratory after drying. Also, sampling of the second tank contents was carried out after mixing the solution and using one-liter bottles in three replications per each concentration. Sampling was also done without placing the bucket as a witness. The significance of the effect of concentrations on the error value was defined by analysis of variance through a random block design scheme. Results demonstrated that there was a significant difference between the concentrations obtained in the experiment with bucket and without bucket at a level of one percent. Without using bucket, the amount of errors were 63.13, 78.40 and 73.97 percent for 5, 10 and 20 gr.L-1 concentrations, respectively and with using bucket they were 15.90, 15.07 and 4.83 percent. The mean errors in all concentrations were 71.83 and 11.93 percent without and with using buckets, respectively.