Different sensing techniques for geomorphometric measurements

Abstract

Solid-state MEMS accelerometers are small and cost-effective new sensors developed within the last decade that have often been used as tilt sensors in non-agricultural environments. In some applications accelerometers can replace electronic clinometers, which are heavily dampened sensors used in common tilt measurements. Our research tested the hypothesis that either sensor can be used as dynamic tilt sensors and accurately map areas with varying topography. Slope is a vector measurement of the rate of change of elevation, and is a soil attribute commonly studied in the geosciences due to the influence it exerts over other soil attributes and crop yield. Mathematical models were used to relate pitch and roll angles of a roving vehicle to field slope gradient and aspect, based on data collected with a dual-axis accelerometer and a dual-axis clinometer in a field test. Benchmark for comparisons were slope values derived from RTK-GPS measurements. Time-domain and spatial-domain filters were applied to the data collected in the field. The coefficient of determination ( r 2), mean absolute error (MAE), and model efficiency (ME) were used to compare slope gradient results. Slope aspect was evaluated on whether it was correctly classified according to cardinal directions. Results suggest that the accelerometer and the clinometer are accurate enough to be used as dynamic tilt sensors. The MAEs of slope gradient were less than 1° when compared to measurements derived from RTK-GPS, for scales similar to Order I soil maps. The correct classification of slope aspect was higher than 70%. Solid-state MEMS accelerometers represent a new and cost-effective technology that can be successfully used in such measurements. Accelerometers outperform clinometers in cost, speed, and simplicity of use, but not in accuracy.