文摘
One of the most popular geophysical methods currently being used to provide information about the spatial variation of soil properties is electromagnetic (EM) induction. This is because EM instruments are easy to set-up, use and mobilise. In addition, the measured apparent electrical conductivity (¦Òa¡ªmS m? 1) has been shown to be useful in mapping soil (e.g. clay) and hydrological (e.g. water holding capacity) properties. However, in most of these studies, the potential to model the stratigraphic nature of soil variation (i.e. horizons) and at discrete depth intervals has not been fully realised. In this research, we describe how a single-frequency and multiple-coil EM instrument (i.e. DUALEM-421) and an inversion algorithm (EM4Soil) can be used to generate a two-dimensional model of the electrical conductivity (¦Ò¡ªmS m? 1). The algorithm was evaluated using ¦Òa acquired across an irrigated agricultural field, in an extensive Quaternary alluvial clay plain northwest of Moree, New South Wales (NSW, Australia). In addition, we conduct inversions of various combinations of ¦Òa data. We do this to compare and contrast the results that can be achieved using the various combinations of commercially available instruments. We compare the inverted models of ¦Ò with measured soil properties such as clay content, cation exchange capacity, and, the electrical conductivity of a one part soil to five part water (EC1:5¡ªdSm? 1) and saturated soil paste (ECp¡ªdSm? 1). In general, ¦Ò is shown to compare favourably with ECp. In terms of identifying an optimal set of ¦Òa data, we found that whilst the Akaike Information Criteria (AIC) indicated that a DUALEM-1 and a DUALEM-421 configuration (using only Hcon) provided the best balance between goodness and fit between ¦Ò and ECp, the best reconstruction of ¦Ò and in terms of a 1:1 relationship with ECp is achieved using all six possible ¦Òa measurements that can be obtained by a DUALEM-421.
