Development, evaluation and interpretation of sediment rating curves for a Japanese small mountainous reforested watershed

Extensive reforestation using Cypress, Pinus and Cedar has widely taken place in Japan since early 1960s in attempt to fulfill national wood demands and control soil erosion. In 2004 one small headwater mountainous reforested watershed encompasses 4.8 ha was established in Mie Prefecture in Japan to monitor the hydrological response. Frequently and automatically collected values of discharge and suspended sediment concentration were examined for the main outlet and five other stations, spanning from June 2004 to July 2006. In the present study, the different functional linear and non-linear sediment rating curves analytical methods were employed in the investigation of ordinary and some transformed flow discharge–suspended sediment concentration relationships for the study watershed using from 48 to 162 simultaneous discharge–sediment records (2004–2006). According to the results of statistical analyses using different criteria and out of many types of regression functions and data transformation, power rating curve by least square regression on fourth root transformed flow discharge and suspended sediment concentration data performed well. Contrary to what is oft-reported, the best fitted rating curves to the entire data collected for each individual station excessively overestimated the suspended sediment concentration in the study area by 113–430%. The temporal and magnitude stratification of flow discharge and sediment data, flow components separation as well as employing bias correction factor did not improve the relationship, while better estimates were obtained when power regression was applied to the fourth root transformed data separated based on their locations on rising and falling limbs of hydrographs. Results also showed a complex suspended sediment concentration and flow discharge relationship in different subwatersheds of the study reforested watershed reflecting the effect of different physical local characteristics, sediment availability, contribution of various hydrologic cycle components, and subtle variation of soil hydrophobicity on runoff generation and consequently sediment supply.