Methods in Field Spectroscopy
In most cases, the reflectance of a vegetation canopy or a soil surface is expressed as a ‘reflectance factor’, which is the ratio between the radiant flux reflected from the surface of interest compared to that which would be measured if the surface was replaced with a perfectly white Lambertian reflector (see Nicodemus, 1977 for a more precise definition). In practice, a calibrated white panel is placed horizontally beneath the spectroradiometer to provide an estimate of the spectral irradiance, or a second identical spectroradiometer fitted with a cosine-corrected receptor is used.
 |
 |
A dual-beam GER IRIS MkIV spectroradiometer being used to measure grass reflectance. The area measured is immediately adjacent to the white reference panel. |
A Spectron SE590 spectroradiometer fitted with a cosine-corrected receptor being used to measure spectral irradiance. Using a white panel would have been more difficult and prone to error under these conditions. |
This simplification of the measurement environment provides a practical method to make reflectance measurements in the field but it also introduces a second reflecting surface into the measurement procedure, the spectral and angular properties of which affect the resulting reflectance. Furthermore, unlike brdf, reflectance factors calculated in this way are not solely a property of the surface, but also depend on the conditions of measurement, especially the sky hazyness which affects the proportion of direct-to-diffuse flux.
Using a parasol to measure the proportion of spectral irradiance orginating from the sky, rather than coming directly from the Sun (the direct-to-diffuse ratio). The trolley has two GER1500 spectroradiometers operated in dual-beam configuration.
The high cost of field spectroradiometers means that most measurements are ‘single-beam’, that is, the same instrument is used to measure the spectral radiance of the target and the reference panel. A limitation of single-beam devices is that the assumption of identical illumination conditions for both the reference panel and the target surface is rarely met due to the time lag between scans. Often, this is addressed by performing reference panel scans before and after that from the target, and then interpolating the reference panel radiance at the instant the target is measured. However, this assumes that irradiance is changing in a predictable manner between successive reference panel scans, which is true for smoothly varying variation, such as that due to solar zenith angle, but is unlikely to be true for short-term episodic changes, for example those due to the passage of sub-visual clouds through the direct solar beam. ‘Dual-beam’ methodologies overcome this limitation, employing two sensors to measure the target and the panel simultaneously. Generally speaking, dual-beam instruments produce more precise results than their single-beam counterparts, although this requires that the two sets of detectors are very well matched.
The quality of reflectance data acquired using the single-beam methodology is critically dependent upon the stability of the atmosphere between measurement of the surface and measurement of the reference panel. The time interval between these two measurements must be as short as possible. Furthermore, the reference panel should be positioned in a repeatable manner and kept fixed while measurements are being made. If the quality of the resulting data is not critical, it is possible to hold the reference panel by hand, as in the following video clip. However, note that even in this example, the surface and panel measurements were made in rapid succession.
Click here to play a short video of a group of students from the University of Southampton measuring the reflectance of a grass field using the single-beam, sequential methodology. Conditions at the time were cold and windy, which made it impossible to keep the reference panel (or the camera!) stable.
Solving problems in field spectroscopy...
 |
 |
Modern notebook computer screens can be very difficult to see in bright sunlight... |
- various ad hoc solutions are favoured... |
 |
|
Lightweight step-ladders were essential when measuring spectra from this field of oil-seed rape. |
