Intra- and inter-storm oxygen-18 and deuterium variations of rain, throughfall, and stemflow, and two-component hydrograph separation in a small forested catchment in Japan

Abstract

Inter- and intra-storm oxygen-18 variations in rain, throughfall, and stemflow were measured to estimate accurate “new” water inputs to the watershed within a 0.84-ha watershed planted with 80-year-old Cryptomeria japonica and Chamaecyparis obutusa artificial forest at the Hitachi Ohta experimental watershed in Japan. In addition two-component hydrograph separation using oxygen-18 concentrations was conducted with four types of “new” water concentration: (1) incremental volume-weighted mean concentrations of rainfall; (2) constant concentrations of rainfall; (3) incremental volume-weighted mean concentrations of throughfall; and (4) constant concentrations of throughfall. Six storms from June to October 2000 were investigated. In the 26 July, 24 September, and 29 October storms, old-water percentages at the hydrograph peak were more than 56.4%, 66.6%, and 80.4%, respectively. In the 24 September and 29 October storms, the old-water contribution estimated by incremental volume-weighted mean concentrations of throughfall was the greatest. On the other hand, this was the smallest after the 26 July storm. Thus, by estimating the “new” water inputs more accurately, the “old” water contribution could be either large or small. In addition, there was a relative difference of about 5–10% between old water percentages calculated from the four types of “new” water. The uncertainty associated with the analysis of oxygen-18 concentrations was up to 4.8%. Thus the difference in estimates of “new” water affected the “old” water percentage more than the uncertainty of the two-component model itself.