Engineering topics
Publications and source records attributed to Vanderbilt, V. C..
Investigations of the multispectral reflectance characteristics of crops and soils as measured from laboratory, field, aerial, and satellite sensor systems are reviewed. The relationships of important biological and physical characteristics to the spectral properties of crops and soils are addressed.
During the past six years several thousand reflectance spectra of corn, soybean, and wheat canopies were acquired and analyzed. The relationships of biophysical variables, including leaf area index, percent soil cover, chlorophyll and water content, to the visible and infrared reflectance of canopies are described. The effects on reflectance of cultural, environmental, and stress factors such as planting data, seeding rate, row spacing, cultivar, soil type and nitrogen fertilization are also examined. The conclusions are that several key agronomic variables including leaf area index, development stage and degree of stress are strongly related to spectral reflectance and that it should be possible to estimate these descriptions of crop condition from satellite acquired multispectral data.
Over a two year period, data were collected regarding the canopies of soybeans grown in rows in planter boxes placed on a turntable in an effort to investigate changes in the spectral reflectance factor related to row direction, Sun direction, soil background, and crop development stage. Results demonstrate that the direction of rows in a soybean canopy can affect the reflectance factor of the canopy by as much as 230%. The results for the red spectral region tend to support the validity of canopy reflectance models; results for the infrared region do not.
The linear polarization and reflection of visible light by wheat as a function of sun-view directions, crop development stage, and wavelength were examined. Two-hundred spectra were taken continuously in wave-lengths from 0.45 to 0.72 Micron in 33 view directions using an Exotech model 20C spectroradiometer six meters above two wheat canopies in the boot and fully headed maturity stages. The analysis results show that the amount of linearly polarized light from the wheat canopies is greatest in the blue spectral region and decreases gradually with increasing wavelength. The results also show that the linearly polarized light from the canopies is generally greatest in the azimuth direction of the Sun and tends toward zero as the view direction tends toward the direction of the hot spot or anti-solar point. It is demonstrated that the single, angle of incidence of sunlight on the leaf, explains almost all of the variation of the amount of polarized light with Sun-view direction.
The feasibility of using the principles of computer axial topography (CAT) to quantify the structure of crop canopies was investigated because six variables are needed to describe the position-orientation with time of a small piece of canopy foliage. Several cross sections were cut through the foliage of healthy, green corn and soybean canopies in the dent and full pod development stages, respectively. A photograph of each cross section representing the intersection of a plane with the foliage was enlarged and the air-foliage boundaries delineated by the plane were digitized. A computer program was written and used to reconstruct the cross section of the canopy. The approach used in applying optical computer axial tomography to measuring crop canopy geometry shows promise of being able to provide needed geometric information for input data to canopy reflectance models. The difficulty of using the CAT scanner to measure large canopies of crops like corn is discussed and a solution is proposed involving the measurement of plants one at a time.
The interaction of the solar illumination angle and row azimuth angle of the measured reflectance factor (RF) of soybean canopies was investigated. Diurnal changes of nearly 140% were observed in the red wavelength region when canopies covered 64% of the soil. The amount of shadow observed was a function of the plant geometry and row width. As soil cover approached 100%, the diurnal changes diminished. A function tht describes the solar illumination angle with respect to the row azimuth explained most of the diurnal variation in the measured RF. Variation in near infrared response was much less and did not appear to be as strongly related to Sun-row angle interactions. The near infrared/red ratio was highly sensitive to Sun angle-row direction interactions, whereas the greenness function, utilizing all four spectral bands, was not.
Reflectances were calculated from measurements at four wavelength bands through eight view azimuth and seven view zenith directions, for various solar zenith and azimuth angles over portions of three days, in an experimental characterization of a soybean field by means of its reflectances and physical and agronomic attributes. Results indicate that the distribution of reflectance from a soybean field is a function of the solar illumination and viewing geometry, wavelength, and row direction, as well as the state of canopy development. Shadows between rows were found to affect visible wavelength band reflectance to a greater extent than near-IR reflectance. A model describing reflectance variation as a function of projected solar and viewing angles is proposed, which approximates the visible wavelength band reflectance variations of a canopy with a well-defined row structure.
(Previously announced in STAR as N82-23600)
A model for the amount of linearly polarized light reflected by the shiny leaves of grain crops is based on the morphological and phenological characteristics of the plant canopy and upon the Fresnel equations which describe the light reflection process at the smooth boundary separating two dielectrics. The theory used demonstrates that, potentially, measurements of the linearly polarized light from a crop canopy may be used as an additional feature to discriminate between crops such as wheat and barley, two crops which are so spectrally similar that they are misclassified with unacceptable frequency. Examination of the model suggests that, potentially, satellite polarization measurements may be used to monitor crop development stage, leaf water content, leaf area index, hail damage, and certain plant diseases. The information content of these measurements is needed to evaluate the proposed polarization sensor for the satellite-borne multispectral resource sampler.
There are no author-identified significant results in this report.
It is pointed out that in order to develop the full potential of multispectral measurements acquired from satellite or aircraft sensors to monitor, map, and inventory agricultural resources, increased knowledge and understanding of the spectral properties of crops and soils are needed. The present state of knowledge is reviewed, emphasizing current investigations of the multispectral reflectance characteristics of crops and soils as measured from laboratory, field, aerial, and satellite sensor systems. The relationships of important biological and physical characteristics to their spectral properties of crops and soils are discussed. Future research needs are also indicated.
Oblique viewing sensors are to be launched in the late 1980's on the Multispectral Resource Sampler, being developed by the U.S. Since the resulting oblique measurements will need to be better understood, the reflectance response with a view angle of wheat, excluding atmospheric effects but simulating the response of a multispectral scanner, is analyzed. Spectra were taken continuously in the wavelength range of 0.45 to 2.4 microns at more than 1200 view/illumination directions with a 20 C spectral radiometer, and data were acquired six meters above four wheat canopies, each at a different stage of growth. The study shows that the canopy reflective response is a function of the illumination angle, the scanner view angle, and the wavelength; and the variation is greater at low solar variations.
Sensors to remotely measure the linear polarization of ground scenes have been proposed for the Multispectral Resource Sampler (MRS), a satellite sensor system proposed to complement the Thematic Mapper. At present justification for a sensor on MRS to measure scene polarization is limited. This paper discusses a model for the amount of linearly polarized light reflected by the shiny leaves of such crops as wheat, corn, and sorghum. The theory demonstrates that, potentially, measurements of the linearly polarized light from a crop canopy may be used as an additional feature to discriminate between crops. Examination of the model suggests that, potentially, satellite polarization measurements may be used to monitor crop development stage, leaf water content, leaf area index, hail damage, and certain plant diseases. The model adds to the understanding of the potential information content of scene polarization measurements acquired by future satellite sensor systems such as MRS.
A multiband radiometer for field research with 8 bands between 0.4 and 12.5 micrometers is described. The data acquisition system will record the results from the radiometer, a precision radiation thermometer, and ancillary sources. The radiometer and data handling systems will be adaptable to helicopter, truck, to tripod platforms; the system will also be suitable for portable hand-held operation. The general characteristics of this system are that it will be (1) inexpensive to acquire, maintain, and operate, (2) simple to calibrate, (3) complete with data handling hardware and software, and (4) well-documented for use by researchers.
The interception of solar power by the canopy is investigated as a function of solar zenith angle (time), component of the canopy, and depth into the canopy. The projected foliage area, cumulative leaf area, and view factors within the canopy are examined as a function of the same parameters. Two systems are proposed that are capable of describing the geometrical aspects of a vegetative canopy and of operation in an automatic mode. Either system would provide sufficient data to yield a numerical map of the foliage area in the canopy. Both systems would involve the collection of large data sets in a short time period using minimal manpower.