Rainfall intensity-dependent error characteristics of tipping-bucket rain gauges and their impact on rainfall estimation in the Soksacheon watershed

This study investigated the rainfall intensity-dependent error characteristics of tipping-bucket rain gauges (TBRGs) and evaluated their impacts on watershed rainfall estimation using observed rainfall data. Repeated calibration experiments were conducted under five rainfall intensity conditions (10, 20, 30, 50, and 100 mm/h) to quantify relative error and measurement uncertainty. The results showed that the magnitude of the relative error increased with rainfall intensity, changing from near zero (0.01%) at 10 mm/h to 7.66% (i.e., −7.66%) at 100 mm/h, while the expanded uncertainty remained relatively stable (0.215–0.220 mm). A nonlinear rainfall intensity–error correction function was derived from the experimental results, yielding a coefficient of determination ( R 2 ) of 0.961. The proposed correction function was applied to five years (2021–2025) of rainfall observations collected in the Soksacheon watershed, Korea. Annual rainfall increases after correction ranged from 0.16% to 0.76%, reflecting the fact that approximately 83–90% of total rainfall occurred under low-intensity conditions (≤20 mm/h). In contrast, analysis of a heavy rainfall event in August 2022 revealed a maximum observed rainfall intensity of 72 mm/h and an increase in corrected rainfall of up to 2.09%. Time-series analysis further showed that observed and corrected cumulative rainfall remained nearly parallel during low-intensity rainfall periods, whereas divergence increased only during high-intensity rainfall conditions exceeding 20 mm/h. The results indicate, within the adopted correction framework, that rainfall intensity-dependent measurement errors have a limited influence on annual rainfall totals but can become more significant during extreme rainfall events. This study provides a quantitative assessment of the interaction between TBRG measurement errors and watershed rainfall characteristics and quantifies the potential effects of applying a laboratory-derived correction relationship to hydrological and flood-related applications.