Stuart F. Biggar

Reflectance- and radiance-based methods for the in-flight absolute calibration of multispectral sensors

Abstract Variations reported in the in-flight absolute radiometric calibration of the Coastal Zone Color Scanner (CZCS) and the Thematic Mapper (TM) on Landsat 4 are reviewed. At short wavelengths these sensors exhibited a gradual reduction in response, while in the midinfrared the TM showed oscillatory variations, according to the results of TM internal calibration. The methodology and results are presented for five reflectance-based calibrations of the Landsat 5 TM at White Sands, NM, in the period July 1984 to November 1985. These show a ±2.8% standard deviation (1 σ) for the six solar-reflective bands. Analysis and preliminary results of a second, independent calibration method based on radiance measurements from a helicopter at White Sands indicate that this is potentially an accurate method for corroborating the results from the reflectance-based method.

Vicarious Radiometric Calibrations of EOS Sensors

Abstract Four methods for the in-flight radiometric calibration and cross calibration of multispectral imaging sensors are described. Three make use of ground-based reflectance, irradiance, and radiance measurements in conjunction with atmospheric measurements and one compares calibrations between sensors. Error budgets for these methods are presented and their validation is discussed by reference to SPOT and TM results and shown to meet the EOS requirements in the solar-reflective range.

Field calibration of reference reflectance panels

Abstract The measurement of radiation reflected from a surface must be accompanied by a near-simultaneous measurement of radiation reflected from a reference panel in order to calculate a bidirectional reflectance factor for the surface. Adequate calibration of the reference panel is necessary to assure valid reflectance-factor data. A procedure is described by which a reference panel can be calibrated with the sun as the irradiance source, with the component due to diffuse flux from the atmosphere subtracted from the total irradiance. Furthermore, the radiometer that is used for field measurements is also used as the calibration instrument. The reference panels are compared with a pressed polytetrafluoroethylene (halon) standard. The advantages of this procedure over conventional laboratory calibration methods are, first, that the irradiance and viewing geometry is the same as is used in field measurements and, second, that the needed equipment is available, or can be constructed, at most field research laboratories, including the press necessary to prepare the halon standard. A disadvantage of the method is that cloud-free sky conditions are required during the measurement period. The accuracy of the method is estimated to be 1%. Calibration results are given for four reference panels.

Uncertainties in the in-flight calibration of sensors with reference to measured ground sites in the 0.4-1.1 μm range

Abstract This article describes the error sources for three in-flight sensor calibration methods used by the Remote Sensing Group of the Optical Sciences Center at the University of Arizona. The three methods are the reflectance-, improved reflectance-, and radiance-based methods, which all reference the earth-atmosphere system. The sources of error or uncertainty for each method are discussed, and an estimate of the percent uncertainty associated with each source is made for conditions similar to those actually used for calibrations at White Sands, New Mexico. The results of in-flight calibrations are compared to those of the on-board lamp calibration system for a SPOT HRV camera.

A generalized approach to the vicarious calibration of multiple Earth observation sensors using hyperspectral data

Abstract The paper describes a new methodology that uses spatially extensive hyperspectral imagery as reference data to carry out vicarious radiometric calibrations for multiple satellite sensors. The methodology has been validated using data from a campaign at the Railroad Valley playa test site in Nevada in June 1998. The proof of concept has been further tested based on data acquisition campaigns at the Newell County rangeland test site in Alberta in August and October 1998. The rangeland test site in the Newell County region of Alberta is tested for its suitability as a calibration test site for satellite sensor systems. All three campaigns included ground-based measurements, satellite imagery, and airborne hyperspectral data. The airborne hyperspectral sensor data were acquired using the Airborne Visible and Infrared Imaging Spectrometer (AVIRIS) at Railroad Valley and the Compact Airborne Spectrographic Imager (casi) in all three campaigns. This paper describes the formulation and implementation of the new methodology, and radiometric calibration monitoring results obtained for five different sensors: NOAA-14 AVHRR, OrbView-2 SeaWiFS, SPOT-4 VGT, SPOT-1/2 HRV, and Landsat-5 TM. The results indicate that the nominal on-orbit radiometric calibrations of all the satellite sensors fit within their predicted uncertainties. The combination of both lower-reflectance and higher-reflectance test sites is shown to improve the quality of the calibration monitoring results. In particular, the combined QUASAR monitoring results obtained from the three airborne data acquisition days at the two test sites, encompassing five satellite sensors and a total of 40 spectral band cases, yield a correlation between QUASAR-based and nominal TOA radiances characterized by y=1.026x−1.26, and r2=0.990. Temporal extensions of QUASAR data sets to calibrate satellite sensors imaging the test site one or more days away from the airborne data acquisition day yield mixed results.

Evaluation of the Applicability of Solar and Lamp Radiometric Calibrations of a Precision Sun Photometer Operating Between 300 and 1025 nm

Over a period of 3 years a precision Sun photometer (SPM) operating between 300 and 1025 nm was calibrated four times at three different high-mountain sites in Switzerland, Germany, and the United States by means of the Langley-plot technique. We found that for atmospheric window wavelengths the total error (2varsigma-statistical plus systematic errors) of the calibration constants V(0) (lambda), the SPM voltage in the absence of any attenuating atmosphere, can be kept below 1.6% in the UV-A and blue, 0.9% in the mid-visible, and 0.6% in the near-infrared spectral region. For SPM channels within strong water-vapor or ozone absorption bands a modified Langley-plot technique was used to determine V(0) (lambda) with a lower accuracy. Within the same period of time, we calibrated the SPM five times using irradiance standard lamps in the optical labs of the Physikalisch-Meteorologisches Observatorium Davos and World Radiation Center, Switzerland, and of the Remote Sensing Group of the Optical Sciences Center, University of Arizona, Tucson, Arizona. The lab calibration method requires knowledge of the extraterrestrial spectral irradiance. When we refer the standard lamp results to the World Radiation Center extraterrestrial solar irradiance spectrum, they agree with the Langley results within 2% at 6 of 13 SPM wavelengths. The largest disagreement (4.4%) is found for the channel centered at 610 nm. The results of these intercomparisons change significantly when the lamp results are referred to two different extraterrestrial solar irradiance spectra that have become recently available.

Landsat-8 Operational Land Imager Radiometric Calibration and Stability

The Landsat-8 Operational Land Imager (OLI) was radiometrically calibrated prior to launch in terms of spectral radiance, using an integrating sphere source traceable to National Institute of Standards and Technology (NIST) standards of spectral irradiance. It was calibrated on-orbit in terms of reflectance using diffusers characterized prior to launch using NIST traceable standards. The radiance calibration was performed with an uncertainty of ~3%; the reflectance calibration to an uncertainty of ~2%. On-orbit, multiple calibration techniques indicate that the sensor has been stable to better than 0.3% to date, with the exception of the shortest wavelength band, which has degraded about 1.0%. A transfer to orbit experiment conducted using the OLI’s heliostat-illuminated diffuser suggests that some bands increased in sensitivity on transition to orbit by as much as 5%, with an uncertainty of ~2.5%. On-orbit comparisons to other instruments and vicarious calibration techniques show the radiance (without a transfer to orbit adjustment), and reflectance calibrations generally agree with other instruments and ground measurements to within the uncertainties. Calibration coefficients are provided with the data products to convert to either radiance or reflectance units.

Improved evaluation of optical depth components from langley plot data

Abstract A simple, iterative procedure to determine the optical depth components of the extinction optical depth measured by a solar radiometer is presented. Simulated data show that the iterative procedure improves the determination of the exponent of a Junge law particle size distribution. The determination of the optical depth due to aerosol scattering is improved as compared to a method which uses only two points from the extinction data. The iterative method was used to determine spectral optical depth components for 11–13 June 1988 during the MAC III experiment.

Cross comparison of EO-1 sensors and other Earth resources sensors to Landsat-7 ETM+ using Railroad Valley Playa

The Remote Sensing Group at the University of Arizona has used ground-based test sites for the vicarious calibration of airborne and satellite-based sensors, of which the Railroad Valley Playa in north central Nevada has played a key role. This work presents a cross comparison of five satellite-based sensors that all imaged this playa on July 16, 2001. These sensors include the Advanced Land Imager and Hyperion on the Earth Observer-1 platform, the Landsat-7 Enhanced Thematic Mapper Plus (ETM+), Terras Moderate Resolution Imaging Spectroradiometer, and Space Imagings Ikonos. The approach atmospherically corrects the ETM+ data to derive surface reflectance for a 1 km /spl times/ 1 km area of the playa and then uses these reflectances to determine a hyperspectral at-sensor radiance for each of the sensors taking into account the changes in solar zenith angle due to any temporal differences in the overpass times as well as differences in the view angles between the sensors. Results show that all of the sensors agree with ETM+ to within 10% in the solar reflective for bands not affected by atmospheric absorption. ETM+, MODIS, and ALI agree in all bands to better than 4.4% with better agreement in the visible and near infrared. Poorer agreement between Hyperion and other sensors appears to be due partially to poorer signal to noise ratio in the narrowband Hyperion datasets.

Vicarious radiometric calibration of EO-1 sensors by reference to high-reflectance ground targets

The Remote Sensing Group at the University of Arizona has been using ground targets for the in-flight vicarious calibration of airborne and satellite sensors since the early 1980s. Targets such as Railroad Valley Playa in north central Nevada and White Sands Missile Range in New Mexico have proven to be useful for this work. This paper presents the results from multiple vicarious calibration experiments at a variety of sites for two of the Earth Observing 1 (EO-1) optical sensors. The Advanced Land Imager (ALI) and Hyperion sensors operate in the visible and shortwave infrared portions of the spectrum. The ground sample distance of about 30 m works well for vicarious calibration and allows an easy comparison to legacy sensors such as the Enhanced Thematic Mapper Plus (ETM+) which has the same 30-m ground sample distance. The approach used in this work is to measure the surface reflectance and atmospheric properties during the sensor image acquisition. These data are used as input to a radiative transfer code which computes the top of atmosphere spectral radiance. This predicted radiance is compared to the radiance from the image of the site. Results show that the preflight calibrations of ALI and Hyperion are probably not consistent with in-flight performance of the instruments. New calibration coefficients adopted in December 2001 improve the comparison to the vicarious predictions.