GRASS (Gonio Radiometric Spectrometer System)
GRASS is a field calibration-validation instrument for the measurement of angular reflectance data. It has been designed and developed at the National Physical Laboratory (NPL), Teddington, UK. The overall objective is to provide quasi-simultaneous, multi-angle, multispectral measurements of Earth surface reflected sunlight to support vicarious calibration of satellite sensors operating in the optical region and the validation of their products. 1) 2)
Quality-controlled long-term observation data sets are of great interest to the user community. They are being used for a variety of applications, among them also for climate modeling studies. To provide confidence in the data from these sources, the data should be referenced to a traceable and internationally accepted standard. With the increased importance of cross-comparison and validation, it is also essential that all instrumentation is traceable to SI (International Standard of Units) in a valid manner. The results of such studies, given to policy makers, ought to provide unequivocal evidence to support national and international legislation.
Generally, satellite instruments measure the light reflected from the Earth's atmosphere and surface, but despite detailed pre-launch calibration campaigns to establish the satellites' performance, post-launch calibration is vital to validate the instruments' performance once in orbit. This is often achieved by using field instruments, which measure the light reflected from `calibration sites', (for example deserts or grassland), that can be viewed by satellite sensors. These field instruments provide the in-situ data to validate the satellite measurements.
Understanding the angular reflectance characteristics of different surfaces is becoming more important as more satellites have the capability of observibg surfaces at multiple angles. Examples of such instruments are: MODIS (on Terra and Aqua missions of NASA) and MISR (on Terra), as well as POLDER (on PARASOL mission of CNES). As the data products from these instruments become more readily available for the investigations into the bidirectional reflectance distributions of various surfaces, it is vital to validate the satellite data using field instrumentation that can also measure at multiple angles.
The goal of the GRASS instrument is to provide a means in validating the BRDF (Bidirectional Reflectance Distribution Function) measurements of airborne and spaceborne sensors. The transportable instrument is required to be robust for quick and easy assembly in remote locations to provide multi-angle in situ data (these turn out to be underflight observations when the required satellite sensor is observing the location simultaneously).
GRASS has been designed to measure the Earth's reflected sunlight over half a hemisphere (Figure 1), at 30º intervals, i.e. at 0º, 30º, 60º, 90º, 120º, 150º, 180º, on a series of seven arms. Six of these arms will have five collecting optics (referred to as a “camera”), and the seventh will have six, to be able to capture the nadir measurement, as well.
This results in 36 cameras and angles within the half of the hemisphere. Each camera consists of a collimating lens or entrance optic and an optical fiber. The fibers from these entrance optics feed to a series of multiplexers that results in one optical output that can be coupled to a spectrometer. The spectrometers that will be initially considered for integration with the Goniometer are those available from the UK NERC/FSF (Natural Environmental Research Council/Field Spectroscopy Facility).
The height of the GRASS structure is designed such that the focus of the nadir-viewing camera is 2 m from the center of the target. The semi-hemispherical structure uses a series of legs that are adjustable, such that the working height can be altered. The application of this design feature is so that a spectral measurement can be taken of the top of a vegetation canopy, for example.
Figure 1: Schematic illustration of the GRASS instrument layout (image credit: NPL)
The arms of the Goniometer can be rotated on the circular base of the mounting structure permitting the measurement of the forward and backward scattered radiation. The positions of the entrance optics on the arms are also moveable so that effectively any five-zenith angles (up to 60º), can be chosen to be captured during each measurement sequence. This allows detailed studies of particular BRDF characteristics, e.g. the hot spot.
A further design feature of GRASS is that the lenses on the end of each of the fibers can be removed, and replaced with a cosine diffuser, and the orientation of the viewing optic rotated such that the entrance optic can then measure the down-welling irradiance. This means that the instrument can measure both the radiance and irradiance at concurrent angles (Figure 2).
Figure 2: Schematic view of the radiance and irradiance orientations of the entrance optics (image credit: NPL)
Figure 3: Photograph of the GRASS structure in the Laboratory (image credit: NPL)
As of 2007, the full structure of GRASS has been completed, as shown in Figure 3. Some initial testing of the angular alignment of the instrument showed that the system was flexing, and therefore the angular orientations had significant errors associated with them. Modifications to the instrument have been made, and the instrument is currently being tested in the laboratory, before being utilized in the field.
The first field experiments with GRASS took place in June 2006 within the national field campaign organized by NCAVEO (Network for the Calibration and Validation of Earth Observation) of the University of Southampton. The experiment was a scoping exercise for the establishment of one or more VALERI (Validation of Land European Remote sensing Instrument) sites in the UK as well as an opportunity to learn and share best practice amongst NCAVEO partners and the wider Earth observation community. Data were being collected from 4 aircraft and 5 satellites to validate and intercompare data from a range of airborne and satellite sensors. GRASS, although not completely built, performed fairly well.
NPL was involved in a couple of aspects of the experiment. To provide traceability to SI, all of the field spectroradiometers used, were compared against the NPL Transfer Standard Absolute Radiance Source (TSARS). Secondly, the goniometer was used to perform multi-angular, multispectral measurements of the main calibration site used for the experiment that will later be compared with data that was acquired by others during the experiment.
The overall objective of campaigns is to evaluate the accuracy of biophysical products, e.g. LAI (Leaf Area Index), and fAPAR (fraction Absorbed Photosynthetically Active Radiation), derived from high-resolution sensors such as CHRIS (Compact High Resolution Imaging Spectrometer) on PROBA (ESA satellite), the imagers on DMC (Disaster Monitoring Constellation), The imager on Ikonos, and others. 3)
1) H. Pegrum, N. Fox, M. Chapman, E. Milton, “Design and testing a new instrument to measure the angular reflectance of terrestrial surfaces,” Proceedings of IGARSS 2006 and 27th Canadian Symposium on Remote Sensing, Denver CO, USA, July 31-Aug. 4, 2006
2) H. M. Pegrum a, N. P. Fox, E. J. Milton, M. Chapman, “Development of the GONIO Radiometric Spectrometer System to conduct multi-angular measurements of terrestrial surfaces,” 10th International Symposium on Physical Measurements and Signatures in Remote Sensing (ISPMSRS'07), March 12-14, 2007, Davos, Switzerland, URL: http://www.commission7.isprs.org/ispmsrs07/P90_Pegrum_Gonio.pdf
The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: "Observation of the Earth and Its Environment: Survey of Missions and Sensors" (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates.
Provided validation for: