Digital Soil mapping – Discussion with John Triantafilis

A couple of weeks ago, I attended a very interesting seminar at the James Hutton Institute by Dr John Triantafilis. John is a Senior Lecturer in the School of Biological, Earth and Environmental Sciences at University of New South Wales in Australia. He is a soil scientist who has research interests in the area of applying proximal and remote sensing geophysical instruments and in the area of digital soil mapping. He has collected extensive Electro magnetic and soil data sets which have been used to develop terraGIS; which is a web-based digital soil mapping platform that allows farmers and natural resource managers easy access to soil information in a form which is accessible and easy to use. John had come to the James Hutton Institute to share his research experience and to seek collaborative research projects. I thought it was vital that I attended this seminar on many levels. Primarily, digital soil mapping is a huge part of my project, particularly dealing with the concept of how to improve the quality or level of data that we currently have. Although I am looking at DSM in the context of mapping soil properties which can then be used to map soil functions, it is interesting to see the differing approaches and philosophies undertaken by many different scientists in this field of research.

What is a fundamental requirement for effective soil management at the field and district level, as Dr Triantafilis argues is an understanding of the spatial distribution of soil classes and soil properties. It is no secret that the scientific field of Digital Soil Mapping (DSM) (McBratney et al, 2003) has been expanding rapidly, owing in large part to the prohibitive cost of soil data acquisition. DSM uses applied pedometrics, mathematical and statistical models which integrate legacy information collected from soil observation points alongside information contained in correlated environmental variables (e.g. climate, land cover, geology, relief etc.) and remote sensing (e.g. LANDSAT imagery).

Dr Triantafilis looked at the way electromagnetic induction could conduct a current in the soil.  John talked about what related soil properties can have an influence on electromagnetic conductivity – these were the clay content, mineralogy, moisture content and salt concentration with temperature and bulk density also can have an effect. Examples of demonstrated case studies around the Murray Darling Basin in Australia. What I found pretty interesting here was the wide degree of instruments that were available to gather information about the soils. It is also becoming increasingly popular but also extremely expensive to obtain a hold of these instruments. The majority of them are designed in Canada and can be worth in the region of £32,000!


Work John was involved with in one of his projects

Current and future research opportunities were also presented and explored in his presentation, including applications in 3D time-lapse soil moisture imaging and soil salinity mapping. What was concluded in his presentation was the capability of mapping clay content using gamma spectrometry  and EM data as a surrogate for soil properties.

From conclusions of this presentation ,I spent an hour a following afternoon with John while he was still in Aberdeen. It was an interesting meeting with him discussing what I was doing for my own project and the contrasting approaches to what he was working with. It just goes to show you different philosophies and ways of thinking can give interesting new outcomes. Who knows, I might end up working with John in the future!


Author: Grant Campbell

Twitter; @Stato_Grant