Jb. Reeves et Jas. Van Kessel, Investigations into near infrared analysis as an alternative to traditional procedures in manure nitrogen and carbon mineralisation studies, J NEAR IN S, 7(3), 1999, pp. 195-212
The objective of this investigation was to determine the feasibility of usi
ng near infrared (NIR) reflectance spectroscopy to determine nitrogen and c
arbon mineralisation rates and extents in dairy manures. Ground soil sample
s from studies in which manure was incubated with soil for various time per
iods (0 to 112 d) were scanned in the near infrared (1100 to 2498 nm) using
a scanning monochromator equipped with a rotating sample cup. Control samp
les (soil alone) were also scanned and tested. Calibrations for ammonia, ni
trate, total N (ammonia and nitrate) contents and carbon evolved as carbon
dioxide were developed. Calibration values for ammonia and nitrate were det
ermined on potassium chloride soil extracts using a Technicon autoanalyser.
Carbon dioxide values were determined by titrating evolved carbon dioxide
that was trapped using potassium hydroxide solution. All calibrations were
developed using Partial Least Squares regression analysis. Results using 59
3 samples from several mineralisation experiments have shown that, while NI
R reflectance spectroscopy calibrations for ammonia, nitrate, total N and e
volved carbon dioxide may be possible, their development will require great
care with respect to the samples to be covered by the calibration. While i
t appears that the source of the manure, with respect to the diet of the an
imal(s), was not a serious consideration, manure storage conditions prior t
o incubation had a strong influence on calibration results. In addition, se
parate calibrations appear to be needed for control and non-control samples
. In summary, results presented here, indicate that NIR reflectance spectro
scopy may have potential for analysing manure-amended and non-amended soils
from mineralisation experiments, but further work is needed to determine i
ts true potential.