New technology, old attitude

The CSIRO is trapped in the old paradigm of Australian soils: “Measuring carbon in soils is increasingly important world-wide due to its potential conversion to the greenhouse gas, carbon dioxide.” Soils are sources, not sinks, according to the CSIRO soil organic matter research team.

“Soil carbon is stored in a number of fractions with widely varying chemistry and stability. These soil carbon fractions can be converted to the greenhouse gas carbon dioxide,” it says.

These statements are made in a press release about electronic microscopy, a “new, fast and inexpensive technique for measuring carbon in soils. The CSIRO's technique will help predict the carbon status of any region in Australia.”

“Using mid-infrared (MIR) spectroscopy, the CSIRO team has been able to generate a spectrum of any soil similar to a 'fingerprint'. Such spectra contain a picture of all the various minerals and organic carbon fractions in the soil.

“When this 'fingerprint' is combined with previous measurements of carbon fractions across a range of soil types and analysed using a complex mathematical process, the amount of carbon and its allocation to carbon fractions can be predicted easily for additional soils.”

The team describes the MIR technique in a paper recently published in the Australian Journal of Soil Research. Janik LJ, Skjemstad JO, Shepherd KD, Spouncer LR. 2007. The Prediction of Soil Carbon Fractions Using Mid-Infrared-Partial Least Square Analysis. In: Australian Journal of Soil Research. 45(2): 73-81.

“To use MIR spectra for carbon modelling, spectra from a large test set of calibration soil samples are collected and then combined with previously determined allocations of carbon to the soil fractions for each of the calibration soils.” Let’s hope they don’t fall into the ‘data gaps’ that damaged the original work for the NCAS.

“The combined data is then analysed using a complex mathematical process called 'partial least-squares (PLS) analysis'. The model from this process can then be used to easily predict the amount of carbon in its various forms for unknown soils.”

The soundness of this science should be assured against distortion caused by the choice of calibration soil samples. Otherwise the value of all that brilliant technology is zero.