Measurement Errors in Tipping Bucket Rain Gauges under Different Rainfall Intensities and their implication to Hydrologic Models

Abstract

Measurements from tipping bucket rain gauges (TBRs) consist of systematic and random errors as an effect of external factors, such as mechanical limitations, wind effects, evaporation losses, and rainfall intensity. Two different models of TBRs, viz. ISCO-674 and TR-525 (Texas Instr., Inc.), being used in Ohio's Upper Big Walnut Creek Watershed, were calibrated in the lab to quantify measurement errors at different rainfall intensities. A range of rainfall intensities (12.5 to 230 mm-hr-1) was simulated for each TBR using a pre-calibrated peristaltic pump mechanism. The instantaneous and cumulative values of simulated rainfall were measured at 1-min intervals. Actual and measured rainfall at each intensity was compared. Both TBR measurements showed a significant deviation from the actual rainfall rates with increasing underestimation error at higher intensities (>50.8 mm-hr-1) and slight overestimation at lower intensities (<25.4 mm-hr-1). Model TR-525 showed an earlier and larger deviation (up to 20%) as compared to ISCO-674 (up to 13%) over the range of intensities. These findings are being used to correct precipitation data being collected by both TBRs and test the effect of these corrections on the outputs of hydrologic models, such as SWAT and DRAINMOD.