QUANTIFYING ERRORS IN INCLINOMETER FIELD MEASUREMENTS

Inclinometers have been used since the 1960s to measure ground deformation. It is a relatively simple matter to test the accuracy of an inclinometer in a laboratory setting; however, it’s more difficult to test the accuracy of an inclinometer in a field setting. To provide a reliable field setting for this purpose, I used data from inclinometer casings that were installed at a site in the Cascade Mountains. Numerous readings had been obtained over many years, and no ground displacement has occurred, providing a large data set to work with. For statistical purposes, I used the average reading as the “true” value for slope displacement and used the standard deviation at each reading depth to determine how accurate the instrument was from one reading to the next. I discovered that the average standard deviation at a specific depth is 0.0038 inches (0.0096 cm). Propagating that error randomly up a 120 foot (36.6 m) casing gives an average cumulative error of 0.07 inches (0.18 cm). This favorably compares to the level of accuracy provided by the manufacturer, of 0.3 inches (0.76 cm) of error / 100 feet (30.5 m). I also found that greater error in readings correlates to casings that have greater “wobbliness” to them, as defined by the change in angle from one reading to the next, and that the accuracy in the readings is affected by variations in personnel. I discovered no apparent correlations between inclinometer accuracy and time of the year, inclination of the casing, and total depth of the inclinometer casing.