Investigating sampling accuracy to estimate sediment concentrations in erosion plot tanks

Authors: DAVOOD NIKKAMI

Abstract: Significant errors when sampling from collection tanks installed at the lower end of soil erosion plots may lead researchers toward wrong conclusions. Limited research has been found on sampling accuracy. In this study, a cylindrical sampler is introduced as a new sampling device and its sampling accuracy is investigated and compared with 2 other manual methods, namely bottle and pipette. Three target sediment concentrations were prepared in separate 213.5-L collection tanks as 3 replications with 3 mixing periods. The results using the bottle and pipette in all mixing periods and all target sediment concentrations showed that the concentration of sediment increases with the depth of the tank, which implies that making a uniform sediment concentration in the collection tank for sampling is impossible. The results indicated that the cylindrical sampler had the highest sampling efficiency of 88.68%. Computed sampling efficiencies in tanks of 20, 40, 60, and 80 cm in depth were 34.76%, 43.94%, 54.45%, and 67.21% for the bottle and 34.58%, 43.05%, 49.20%, and 56.42% for the pipette, respectively. There was no significant difference between sampling from the center and from the side of the collection tank and among mixing periods of 1, 2, and 5 min. In order to maximize the accuracy of the sampling, a 10-L bucket was placed in the collection tank to trap the coarse material. Weighing collected coarse material and adding the sampled part improved the accuracy by 9.45%. In fact, using a cylindrical sampler and a bucket together resulted in 98.13% sampling accuracy.

Keywords: Collection tank, cylindrical sampler, erosion plot, runoff, sediment

Full Text: PDF