Global Soil Partnership

Australia


The Soil and Landscape Science Group (SLSG) at Curtin University in Perth, Australia

 Contact(s): Raphael Viscarra Rossel

Laboratory Capacity

The SLSG conducts theoretical, methodological and applied research to improve understanding of soil processes and the drivers of soil and landscape variability at different spatial and temporal scales. In addition, the group’s research aims tounderstand better and predict how soil functions respond to climate and anthropogenic change. The specific research interests are in soil carbon sequestration science, the preservation of biodiversity, nutrient cycling, and food production.

The SLSG has expertise in developing new methods for soil sensing, spectroscopy, digital mapping, and monitoring and in the development and application of state-of-the-art statistical, chemometric, machine learning and mechanistic models for prediction. Professor Viscarra Rossel brings more than 20-years’ experience in soil spectroscopy, spectroscopic modelling, development of regional, national and global spectral libraries, and in the application of soil spectroscopy for digital soil mapping and modelling. He developed the stand-alone executable software “ParLeS” for chemometric analysis of spectroscopic data, which he gives away for free upon email request. The software is used for teaching and spectroscopic applications in medical, pharmaceutical, engineering, agronomic and soil science research.

The group is now developing web-based applications to enable soil spectroscopy globally and for the common good. The SLSG laboratory infrastructure consists of a laboratory for sample processing, preparation and automated grinding, instrumentation for soil organic and inorganic carbon analyses, and a soil sensing laboratory with a benchtop FT-IR DRIFT/ATR near-infrared and mid-infrared spectrometer, a portable visible–near-infrared spectrometer, a range of miniaturised visible and near-infrared spectrometers, a laser-induced breakdown spectrometer and field equipment. The group also has computing infrastructure and workstations optimised for deep learning with multi-core CPU and GPU

Areas of Interest

The group provides short courses on fundamental soil spectroscopy and spectroscopic modelling with modern chemometrics and machine learning methods. They supervise undergraduate and postgraduate students, host visiting scientists, and collaborate with researchers worldwide.

The Soil Science Group at the University of Sydney

The Soil Science Group at the University of Sydney provides research leadership in digital soil mapping, proximal soil sensing and soil chemometrics.

The University of Sydney started its work on soil spectroscopy in 1998 as means of gathering and analyzing high-resolution soil data, which has necessitated the development of a set of field-sensing technologies as opposed to more costly laboratory measurement ones. An important principle that emerged and motivated our work is that many cheap lower precision but spatially dense geo-located field observations are more useful for spatial management than a few very precise measurements of the same overall cost.

Contact(s): Alex McBratney, Budiman Minasny and Wartini Ng

Laboratory Capacity

Technologies used in The Soil Science Group at the University of Sydney's work includes the development of diffuse reflectance spectral methods, in the ultra-violet, visible, near- and mid- infrared portions of the electromagnetic spectrum.  Extensive facilities on spectroscopy, include a suite of dedicated laboratory (vis-NIR/MIR) spectrophotometer for soil, FTIR with Hyperion 3000 microscope, Raman Spectrometer, and Spero-QT Infrared Microscope. We have extensive field equipment on proximal sensing, include electromagnetic induction instruments, gamma radiometer, on-the-go electrical resistivity sensing system, field NIR spectrophotometers, field portable XRF, NIR Spectro-penetrometer, on the-go pH sensor, crop yield and protein monitors. We have a collection of soil spectral libraries (NIR, MIR) covering all soils from NSW.

The groups research focuses on developing novel soil chemometric procedures, including using machine learning models, spectra and model fusion, deep learning, etc. and have a soil spectral inference system that enables transforming a spectrum into a multitude of soil properties. In addition to basic soil properties, we examine spectroscopy as a tool to screen for soil contaminants, microplastics, and provenance. This group has published the first textbook on digital soil spectroscopy and chemometrics Soil Spectral Inference with R (Springer, 2021) and have conducted training courses since 2015.

The group has active collaborations around the world, including the Kellogg Soil Survey Laboratory, USA; ICRAF, Kenya; University of Sao Paolo, Brazil; Soil Health Institute, USA; University of Nebraska, University of Wisconsin, both USA; CSIRO; Rural Development Agency in Korea, Indonesian Agricultural Research Development Agency, Indian Institute of Technology Kharagpur, Cocoa Board of Papua New Guinea, etc.

Areas of Interest

The group is ineterested in joint project application with different countries, insitiutions, and research groups. etc; 2. host visiting students or researchers; 3. write joint publications; 4,organizing training courses


 

Advanced Soil Spectroscopy laboratories and Institutes are highly welcome to join SoilSpecNet (SSN) and any soil laboratories and institutes are encouraged to collaborate with listed SSN labs and institutes for the capacity development of soil spectroscopy, by sending an email to [email protected] and [email protected].