Examination of bias of dry weather spot-flow rates compared to the mean flow rates of four periods in three mountain streams with different characteristics: verification of the usage of spot-flow rates to estimate mean discharge loads and implications for appropriate sampling strategies

抄録

Our previous study showed a close relationship between spot-flow rate (Q) and periodical (30, 14 and 8 days) mean Q by correlation regression analysis for 64 spot Q readings in 2011–2016. The results suggested that spot Q could be used instead of periodical mean Q in discharge load (L)–flow rate (Q) equations with exponential coefficients that are close to unity. This finding supports the estimation of periodical mean L, even when low frequency samplings to know the concentrations (C) of Chemical Oxygen Demand, Total Organic Carbon, Total Nitrogen, Total Phosphorus and D-SiO₂ are conducted. Moreover, changes in the correlation factor and the slope of the regression could be used to estimate the most appropriate sampling frequency. Here, the validity of using the spot Q instead of the periodical (30, 14, 8 and 4 days) mean Q is examined using datasets obtained for the Maekawa River (2011–2015) and the Kurosano River (2015–2017) in addition to the pre-reported Okura River (2011–2016). All three rivers are forestry rivers in Miyagi Prefecture, Japan. Despite the marked variations in differences in their catchment and river characteristics, regression analyses suggested that spot Q is well suited for estimating the periodical mean Q.