Employing integrated electromagnetic induction sensors for soil characterization under different land use conditions

Abstract

This study was conducted to explore the possibility of predicting soil moisture content (SMC) in podzolic soils in western Newfoundland using multi-coil (MC) and multi-frequency (MF) electromagnetic induction (EMI) sensors. Two studies were conducted to assess the effectiveness of apparent electrical conductivity (ECa) from both MC and MF EMI sensors in characterizing SMC. The first study focused on employing MF-EMI in characterizing SMC under three different land uses in western Newfoundland’s podzolic soils. The second study assessed effectiveness of ECa obtained from the MC and MF-EMI sensors to maximize the representation of SMC taking into consideration organic matter, bulk density and texture using a statistical and geostatistical analysis. Geostatistical analysis from the first study revealed that cokriging of SMC with densely collected ECa provided an improved characterization accuracy of soil moisture variability across the different land use conditions. The second study found that multiple linear regression (MLR) models were effective in representing SMC variations. Additionally, MC-EMI sensors provided better predictions of SMC than MF-EMI sensors. The findings from this study demonstrate that EMI has the potential to provide an accurate and robust technique for predicting soil moisture in boreal podzolic soils. Furthermore, it is worth noting that the surveys were performed during the wet period, given that MF–EMI is more reliable for ECa variability in wet soils than in dry soils.