Rapid non-destructive techniques for assessing crop growth rates and nitrogen status
The broad aims of the study were to grow crops under different levels of nitrogen fertilizer, so that a range of crop growth rates and leaf chlorophyll concentrations could be made available for an investigation into the feasibility of monitoring the crop's growth and colour changes, using the spectral properties of the individual leaves and the leaf canopy. The literature concerned with the interaction of radiation with both single leaves and whole canopies was reviewed. In addition a few other promising techniques capable of revealing crop condition non-destructively were also mentioned. Specifically the aims were (a) to develop a robust optical technique for assaying chlorophyll content of plant leaves which would be able to compensate for factors which interfere with the independent absorption of leaf pigments and (b) to quantify leaf pigment concentrations of a crop independently from crop biomass. A standard laboratory spectrophotometer was interfaced to a microcomputer and its cell compartment modified for acquiring reflectance and absorption spectra of intact leaves. For crop canopy reflectance measurements a portable three-channel band-pass radiometer employing wide-band interference filters, silicon photodiode detectors, and integrated sample/hold circuits was designed and constructed. The instrument has the ability to provide simultaneous signal outputs of all three sensors for facilitating the measurement of crop canopy reflectance under conditions of fluctuating incoming radiation, as might occur on cloudy days. In addition to the band-pass radiometer a low cost system employing an inexpensive monochromator for obtaining reflectance spectra of crops in the visible/near-infrared wavelength range was constructed and tested. In this system the rapidly varying analogue output of the monochromator is converted to audio-frequencies for recording on an audio-cassette tape recorder. Demodulation and retrieval of the original monochromator output for computer processing is carried out back in the laboratory. Multiple linear regression of selected features of intact leaf absorption and reflectance spectra and their first order and second order derivatives yielded equations able to compensate for the non-pigment differences of leaves of maize, barley, french dwarf beans, sweet pepper, sunflower, and ornamental tobacco, and accurately predict leaf chlorophyll content. Some of the reported techniques for quantifying crop growth and results of crop reflectance behaviour were confirmed. Leaf pigment concentrations of small plots of barley could be assessed independently of crop biomass using oblique radiometer views of the crop canopy to eliminate the very large influence of the soil background reflectance.