The Use of Proximal Soil Sensor Data Fusion and Digital Soil Mapping for Precision Agriculture

Wenjun Ji,Viacheslav I. Adamchuk,Songchao Chen,Asim Biswas,Maxime Leclerc,Raphael Viscarra Rossel

2017-01-01

Abstract:Proximal soil sensing (PSS) is a promising approach when it comes to detailed characterization of spatial soil heterogeneity. Since none of existing PSS systems can measure all soil information needed for implementation precision agriculture, sensor data fusion can provide a reasonable al- ternative to characterize the complexity of soils. In this study, we fused the data measured using a gamma-ray sensor, an apparent electrical conductivity (ECa) sensor, and a commercial Veris MSP3 platform including a optical sensor measuring soil reflectance at 660 nm and 940 nm, a soil ECa sensor and a pH sensor, with the addition of topography for the prediction of several soil properties, i.e. soil organic matter, pH, buffer pH, phosphorus, potassium, calcium, magnesium, aluminum. A total of 65 sampling locations were selected from a 38.5 ha field in Ontario, Canada. Among them, 35 locations were selected by a random stratified sampling strategy. The stratification grid was 1 ha. Sampling was prohibited in areas near the field boundaries and within a safety margin from the drainage system. 20 locations were selected using a neighbourhood search approach , a spatial data integration strategy. These two sample datasets were used as the calibration dataset to build the model between soil properties and readings from different proximal soil sensors. The remaining 10 sensing locations were used as an independent validation dataset. Partial least square regression (PLSR) was performed on the data from each individual sensor and different sensor com- binations (sensor data fusion). For most soil properties, predictions based on sensor data fusion were better than those based on the output of individual sensors. By fusing the data from all of the proximal soil sensors, more properties can be predicted simultaneously (R2>0.5, and RPD>1.50). After choosing the optimal sensor combination for each soil property, different digital soil mapping methods, including support vector machines (SVM), random forest (RF), multivariate adaptive regression splines (MARS), regression trees (RT) and back-propagation artificial neural network (BP-ANN) were used to estimate variograms and pursue regression kriging. High resolution maps were thus interpolated with the most successful methods. The performance of the two different sampling strategies was compared by the prediction accuracy from the validation samples. We thus conclude that proximal soil sensor fusion paired with the digital soil mapping method is a promising way to offer the essential soil information needed for precision agriculture.