Credit: Earth as Art

In 2003, the RSG built a ground-based active radiometer to measure backscatter surface reflectance in order to validate on-orbit lidar retroreflectance measurements made by GLAS. The design goals of the reflectometer were driven by the geometry and illumination configuration of spaceborne lidar systems. Additionally, it was desired that the instrument be portable enough to sample a test site with one person, have a low production cost, and a quick build time. The basic design for the reflectometer developed by the RSG is a laser source with a filtered silicon detector and a fold mirror to allow measurement of the retroreflectance. The active source for current lidar sensors (GLAS and CALIOP) is a Nd:YAG laser with output wavelengths of 1064 and 532 nm, thus it was desired that the reflectometer design incorporate similar laser sources. An initial design of the reflectometer implemented a commercially available pulsed wave (PW) laser with selectable output at either, or both, wavelengths. This approach was deemed to be too unstable to allow accurate measurements and after several evaluations in the field the pulsed laser source was abandoned. In the subsequent design two separate continuous wave (CW) lasers were implemented to overcome this problem. The downside to this approach is that the system now requires the lasers be physically switched to operate the system at both wavelengths. However, the CW lasers are small, stable, rugged and less expensive than most pulsed lasers.

The laser is expanded through a very simple beam expander to allow sufficient spatial sampling of the ground (i.e. spot is larger than surface features such as cracks, ridges, etc). This beam expander is a Galilean telescope operating in reverse with a negative primary lens and positive secondary. This results in the beam being expanded to nearly 35 times the original beam size, which provides a spot size on the ground approximately 4 cm in diameter. No spatial filtering of the beam is attempted within the beam expander in order to simplify alignment and make the system more robust. This is not an issue since the reflectometer is designed to operate in a ratioing mode to retrieve reflectance. Thus, as long as the reference source is reasonably homogeneous over the illumination area, any variations in the beam should cancel. The design of the system attempted to reduce chromatic aberrations between the 532 nm and 1064 nm wavelengths while keeping the cost of lenses to a minimum.

Figure 1 shows a schematic of the basic layout of the reflectometer. The expanded beam is turned 90 degrees by an elliptically shaped fold mirror in order to direct the beam to the target. The light reflected from the target then travels through a hole in the center of the elliptical mirror that leads to the end of the detector tube assembly. This tube acts as a baffle to ensure that incoming laser light will not travel to the detector without being reflected off of the target. It also serves as an aperture for the simple radiometer, defining the field of view to be slightly larger than the spot illuminated by the laser. The detector itself is a silicon detector with an op-amp assembly that includes gain electronics. There is a changeable 3 nm bandpass interference filter in front of the detector to improve rejection of out-of-band wavelengths, due to light that might reach the detector from sources other than the laser. The entire reflectometer, excluding the laser source, is enclosed in a half-cylinder cover to reduce further the possibility of stray light from entering the system. Figure 2 shows an external view of the reflectometer with the laser source and entrance aperture in the foreground and the detector assembly on top. The amplifier is powered by a tracking power supply and data are taken using a Fluke Hydra datalogger which also provides the analog to digital conversion. The data can be viewed in real time on a laptop and are written to a file where the voltage data can later be converted to reflectance. Data are taken as rapidly as the datalogger allows, which corresponds to approximately one sample every 2 seconds. A gas generator is used to supply power for remote field operation. The reflectometer, generator, and data acquisition hardware are placed on a cart for increased mobility.

The RSG retrieval of reflectance relies on referencing spectrometer data over the surface of interest to those from a panel of known reflectance. The reflectometer takes the same approach. A nominal 99% Spectralon sample is used as the reference. The panel sits on a tray which is designed so that samples are taken at the same spot on the panel each time. After reference measurements of the Spectralon are made, the instrument is removed from the tray and placed over the target surface. Once the target surface is measured, another reference measurement is taken. Each surface type, both the reference panel and ground, are sampled for approximately 2-3 minutes. The target surface reflectance is then determined by ratioing the averaged ground measurements to that of the Spectralon reference, which is an average of the two contiguous reference measurements. This ratio is multiplied by the reflectance factor of the Spectralon reference, that has been determined in the laboratory, in order to attain the reflectance of the target surface.