Investigating Sediment Suspension and Sampling Accuracy in Erosion Plot Tanks
2004
Niknami, Davood | Arab Khedri, Mahmood | Razmjoo, Peiman | Ahrar, Mohammad
Soil loss and erosion processes are usually measured and evaluated in soil erosion plots. For this purpose, runoff and sediment are sampled from the tanks at the end of these plots. Although these plots and their accouterments are constructed precisely in right places, significant errors on samplings lead the researchers to wrong decisions in plot projects. There are limited researches on sampling accuracy from these tanks which are accompanied with sampling errors. In this research that was executed in the Soil Conservation and Watershed Management Research Institute (SCWMRI) in 2003, the accuracy of sampling by bottle, pipette, and cylindrical sampler made in SCWMRI was investigated. Three sediment concentration of 1.87, 4.68, and 9.36 gr/l were prepared in 213.5 liter plot tanks with three mixing periods of 1, 2, and 5 minutes with wooden shovel. The samples were taken from the center and side of the tanks and from the depths of 20, 40, 60, and 80 centimeter from the water surface by bottles and pipette and the whole depth by cylindrical sampler in 3 replicates. The cylindrical sampler showed least error of 2.83% (without using a small bucket in the main tank with 3 replicates) and 1.96% (when a small bucket was used) compared to other methods. Sampling with bottle and pipette illustrated that the concentration of sediment increases and becomes more accurate with the depth of the tank. The results show that the most accurate results belong to the last 20 centimeter i.e. 80 centimeter from the water surface. Computed concentration errors of the depth of 20, 40, 60, and 80 centimeter from the water surface for bottle sampling were 65.63, 56.13, 45.63, and 32.73 percent and for pipette sampling were 65.48, 57.02, 50.88, and 43.67 percent respectively. Also, at the level of 1%, there were no significant difference between sampling from center and side of the tanks, except for cylindrical sampler, and between mixing periods of 1, 2, and 5 minutes. For maximizing the accuracy of sampling a 10 liter bucket was placed under tank tube to collect the coarse sands. Weighting collected coarse particles and adding the sampled part increased the accuracy by 10.02 percent. Finally, for correcting the concentrations, computed by bottle sampling in plot projects, sampling by bottles were executed in 1.87, 4.68, and 9.36 gr/l concentrations and from the depth of 20 cm of tank water surface. Results showed 3.04 as correction factor.
Показать больше [+] Меньше [-]