ECOLOGICAL APPLICATIONS OF NEAR-INFRARED REFLECTANCE SPECTROSCOPY A TOOL FOR RAPID, COST-EFFECTIVE PREDICTION OF THE COMPOSITION OF PLANT AND ANIMAL-TISSUES AND ASPECTS OF ANIMAL PERFORMANCE
Wj. Foley et al., ECOLOGICAL APPLICATIONS OF NEAR-INFRARED REFLECTANCE SPECTROSCOPY A TOOL FOR RAPID, COST-EFFECTIVE PREDICTION OF THE COMPOSITION OF PLANT AND ANIMAL-TISSUES AND ASPECTS OF ANIMAL PERFORMANCE, Oecologia, 116(3), 1998, pp. 293-305
Many ecological studies rely heavily on chemical analysis of plant and
animal tissues. Often, there is limited time and money to perform all
the required analyses and this can result in less than ideal sampling
schemes and poor levels of replication. Near infrared reflectance spe
ctroscopy (NIRS) can relieve these constraints because it can provide
quick, non-destructive and quantitative analyses of an enormous range
of organic constituents of plant and animal tissues. Near infrared spe
ctra depend on the number and type of C-H, N-H and O-H bonds in the ma
terial being analyzed. The spectral features are then combined with re
liable compositional or functional analyses of the material in a predi
ctive statistical model. This model is then used to predict the compos
ition of new or unknown samples. NIRS can be used to analyze some spec
ific elements (indirectly - e.g., N as protein) or well-defined compou
nds (e.g., starch) or more complex, poorly defined attributes of subst
ances (e.g., fiber, animal food intake) have also been successfully mo
deled with NIRS technology. The accuracy and precision of the referenc
e values for the calibration data set in part determines the quality o
f the predictions made by NIRS. However, NIRS analyses are often more
precise than standard laboratory assays. The use of NIRS is not restri
cted to the simple determination of quantities of known compounds, but
can also be used to discriminate between complex mixtures and to iden
tify important compounds affecting attributes of interest. Near infrar
ed reflectance spectroscopy is widely accepted for compositional and f
unctional analyses in agriculture and manufacturing but its utility ha
s not yet been recognized by the majority of ecologists conducting sim
ilar analyses. This paper aims to stimulate interest in NIRS and to il
lustrate some of the enormous variety of uses to which it can be put.
We emphasize that care must be taken in the calibration stage to preve
nt propagation of poor analytical work through NIRS, but, used properl
y, NIRS offers ecologists enormous analytical power.