The Remote Sensing Group has focused on the in-flight radiometric calibration of satellite and airborne sensors since the mid 1980s. Dr. Philip Slater developed the reflectance-based approach, where the surface reflectance of the test site is measured during a satellite overpass. Ground-based atmospheric measurements are made simultaneously to characterize scattering by molecules and aerosols. The results of these measurements are used as input to a radiative transfer code to predict the at-sensor radiance. This predicted radiance is compared to the satellite sensor output to determine a calibration coefficient.
Additional methods were developed using well-characterized, airborne radiometers that measure upwelling radiance from test sites at altitudes above most of the atmospheric scattering (radiance-based approach). Other approaches include diffuse skylight measurements for atmospheric characterization, as well as cross-calibration techniques using additional satellite sensors. Many of these approaches were developed and/or enhanced by current and past students of the group.
Satellite calibration field work is supported by instruments developed in house by faculty and students in the group. Examples include a spectropolarimeter for atmospheric characterization, a solar aureole camera, diffuse-to-global irradiance meter, a SWIR spectroradiometer, and a BRF camera.
The group currently uses four sites:
- White Sands Missile Range near Alamogordo, New Mexico
- Lunar Lake Playa, near Ely, Nevada
- Railroad Valley Playa, near Ely, Nevada
- Ivanpah Playa, located on the California-Nevada border southwest of Las Vegas.
These sights have been selected for their uniformly high reflectance and large size, both of which improve the accuracy of the methods used by the group. A mobile laboratory purchased in 1994 alleviates some of the logistical difficulties in working at these remote sites.
While the field experiments provide an excellent diversion from the day-to-day tasks of the office, they would not be possible without the laboratory facility that has been developed. This includes two ultrastable radiometers, designed by group members, that characterize the sources used to calibrate satellite and airborne sensors. The group also measures the bidirectional reflectance of the reflectance standards used in the field and has developed a new technique for preflight calibration of sensors using the sun as a source.
Current research focuses on NASA's Earth Observing System, which is part of the Earth Science Enterprise (ESE). This Enterprise was developed by NASA to develop a better understanding of the processes that affect the earth's climate. Members of the group are on the science teams for ASTER, Landsat-7 ETM+, MODIS and were members of the SeaWiFS Science Team. ASTER and MODIS are due to be launched on the Earth Observing System's Terra platform in July 1998. Landsat 7 was launched on 15 April. The primary objective of the Remote Sensing Group's work is to validate the radiometric accuracy of these sensors and to provide means for validating this accuracy throughout the lifetime of the missions and follow-on missions. This work builds on the distinct, yet interrelated research areas mentioned above.
Work that the group has done recently in support of this project includes using the ultrastable radiometers mentioned above to characterize the calibration sources used for ESE. The group has developed equipment and methods to do in-flight calibration at hyperspectral intervals (that is at greater than 1000 spectral bands rather than the multispectral case of less than 10 bands). The group is also developing the atmospheric correction for ASTER and methods for cross-calibrating the sensors mentioned above.
Work for this project relies heavily on the vicarious calibration field campaigns for which the group has become known. A major difference is that the group now organizes joint campaigns consisting of other ESE scientists. Three such campaigns have been held during the summers of 1996-1998 in south-central Nevada. These campaigns have included scientists from Japan and Canada, as well as teams from the Jet Propulsion Laboratory, EROS Data Center, University of California Santa Barbara, and South Dakota State University. The purpose of these campaigns is to better understand the accuracies of vicarious calibration and to develop a set of consistent methods between groups.
The hope is that through the use of joint campaigns and improvements to existing methods, the radiometric calibration can be known to NIST-traceable standards with uncertainties less than 3%. It is also hoped that sensors of a similar nature, that is spectral and spatial resolution, can be intercompared in orbit to levels better than 1%.