Effect of Sensor Error on Interpretation of Long‐Term Water‐Level Data

Submersible strain-gage pressure transducers and potentiometer-float systems were used to monitor water-level fluctuations in wells. Use of uncorrected data from six pressure transducers at a site in New Hampshire would have caused an incorrect interpretation of the vertical hydraulic-head distribution. Seven potentiometer-float systems at a site in Nebraska provided data that indicated a slight attenuation of water-level fluctuations, which likely caused some small fluctuations to go unobserved. Digital data loggers were used at both sites to scan the sensors and calculate hourly and daily average values of water levels in the wells. Check measurements using a chalked steel tape were made weekly at the New Hampshire site and monthly at the Nebraska site. Comparisons of recorded data with check measurements indicated values measured by pressure transducers over periods of weeks to months were irregularly larger or smaller than the check measurement. This error was attributed to a drift of the zero reference point of the sensor, and was unpredictable. Values obtained by potentiometer-float systems lagged the water-level change indicated by the check measurements. The lag was attributed to the sum of the mechanical friction of the moving parts of the monitoring system. This error was somewhat constant and predictable. The average absolute error of six submersible pressure transducers that were in operation for 15 months was 0.098 foot. The average absolute error of seven potentiometer-float systems that were operated for 17 months was 0.027 foot.