NOAA KLM User's Guide

Section 7.4

Blue line drawn across page to separate text from document title.
Introduction Page, NOAA KLM TOC, Acronyms
Previous Section, Next Section

7.4 SBUV/2

The objective of the calibration program is to achieve a state-of-the-art calibration of the SBUV/2 instrument including response to spectral radiance, response to spectral irradiance, ratio of radiance-to-irradiance responses, and the temperature dependence of response. Continuity of this calibration is to be maintained throughout the series of operational instruments.

7.4.1 SBUV/2 Calibration Overview

Each SBUV/2 instrument consists of a nadir-viewing double monochromator of the Ebert-Fastie type, and a cloud cover radiometer (CCR, a filter photometer). A photomultiplier tube detects light exiting the monochromator. They are designed to measure the ratio of backscattered ultraviolet solar radiance to solar irradiance. The SBUV/2 instruments can measure from 152.71 to 443.35 nm in steps of 0.07 nm with a bandpass of 1.1 nm. During normal operation, the backscattered radiance from the nadir is measured at 12 near-UV wavelengths from 252 to 340 nm about 100 times per orbit on the sunlit portion. Light coming into the instrument is first depolarized to remove instrument sensitivity to the polarized backscattered radiances. The CCR makes measurements at 379 nm with a bandpass of 3 nm, and is used to detect scene reflectivity changes during a scan. A diffuser plate is periodically deployed to measure the solar irradiance at the same 12 wavelengths. The instrument is also equipped with a Mercury calibration lamp to monitor both the wavelength scale of the monochromator grating and any degradation of the diffuser.

SBUV/2 instruments use three overlapping gain ranges to measure the solar and Earth radiation which varies over six orders of magnitude. Gain ranges 1 and 2 originate from the photomultiplier anode, while gain range 3 originates from the cathode which allows direct measurement of the instrument's gain. This design differs from SBUV where the anode was the source of all output ranges. For SBUV/2, measurements from the Earth and sun are output on different electronic gain ranges at shorter wavelengths and high solar zenith angles, therefore the instrument's electronic gain is not always canceled out in the albedo ratio. The photomultiplier gain (called the interrange ratio, IRR) must be monitored and carefully evaluated over time for accurate long term measurements.

The objective of the calibration program is to achieve a state-of-the-art calibration of the SBUV/2 instrument including response to spectral radiance, response to spectral irradiance, ratio of radiance to irradiance responses, and the temperature dependence and linearity of the response. Continuity of this calibration is to be maintained throughout the series of operational instruments. Ozone profiles and total amounts are derived from the ratio of the observed backscattered Earth spectral radiance to the incoming solar spectral irradiance. This ratio is proportional to the Earth's geometrical albedo. The only difference in the radiance and irradiance observations is the instrument diffuser used to make the solar irradiance measurement; the remaining optical components are identical. Therefore, the principle changes in the instrument that result in a change in the measured albedo over time are the diffuser reflectivity and the gain range ratios.

7.4.2 Component Calibration

The spectral efficiency of each optical component of the SBUV/2 instrument is measured by the suppliers in the course of their acceptance testing. This demonstrates that specifications have been met. Also, it allows prediction of the eventual spectral transmittance of the system. If there is a need, the component efficiency tests can be repeated. Monitor (witness sample) mirrors accompany the instrument optics throughout the program and will be periodically tested for spectral reflectivity.

At the subsystem level of assembly, additional procedures and test are carried out that support the final calibration. The depolarizer is tested by measuring the residual polarization of initially polarized light after it has passed through the depolarizer. System linearity and dynamic range tests are done before the calibration lamp and diffuser assemblies are installed. Wavelength calibration, bandpass, instrument profile and spectral resolution tests are done with the calibration lamp installed, but before the diffuser assembly is installed. After integration of the complete instrument, but before formal verification testing begins, a series of checks are run during which final adjustments are made. During this process, the optical alignment is measured and adjusted if necessary, and all field-of-view (FOV) related tests performed, since the same fixture is used for all FOV related tests. FOV tests include: size, shape, and uniformity of both monochromator and CCR (cloud cover radiometer or photometer) fields and their coincidence. Out-of-band rejection and stray light tests are conducted at this stage. Finally, the first radiance/irradiance response tests are run in air, establishing a baseline to which subsequent tests can be compared.

During environmental testing, after the vibration sequence, the radiance/ irradiance response in air is measured and the alignment and FOV checks are made. The definitive radiometric calibration is done just prior to T/V testing. During T/V, the spectral radiometric calibration is extended to the required temperature plateaus and down to 160 nm using the Vacuum Test Fixture (VTF). The calibration transfer standard sources are returned to NIST for recalibration as necessary to allow the closest possible interpolation between standardizations. During T/V testing, the monochromator wavelength scale change with temperature, if any, is measured.

7.4.3 Radiance and Irradiance Calibration

In calibrating the Earth Viewing (radiance) mode of operation, the target seen by the instrument is a calibrated reflectance diffuser illuminated by collimated light from and NIST calibrated source of spectral irradiance. The absolute value of the radiance of the test diffuser is a function of the source irradiance, the efficiencies of the intervening components and their geometries.

In calibrating the radiance-to-irradiance ratio, the set-up is fixed and the test diffuser and the instrument diffuser are alternately placed in the light path. the filter detector monitor measures relative irradiance in the incoming beam in three broad wavelength bands centered around 180, 250, and 300 nm. If the filter detector monitor and the instrument indicate a decrease in signal from all light sources, the windows and collimating mirrors are checked for efficiency. By swinging the filter detector monitor around so that it views the diffuser, a relative efficiency measurement of either diffuser can be obtained. In this way, any changes in the diffusers can be detected during testing. The filter detector monitor uses a quartz windowed bi-alkali photodiode, the type used for the detectors. It is used for relative monitoring measurements only.

The instrument is designed to respond in proportion to the spectral radiance, (mW-m -2-sr-1-Angstrom-1) incident upon its entrance slit. The area of the entrance slit, the FOV defined by the field stop, and the bandpass of the monochromator determine the radiant flux, Φ(W), which enters the system. The system transmission efficiency determines the radiance flux at the photocathode. The quantum efficiency and gain of the detector system determine the final response to a given input. To calibrate this system, standard sources of spectral irradiance that will produce a certain spectral irradiance (mW-m-2-Angstrom-1) at a given distance under carefully controlled conditions are available.

The approach taken is to use a standard (NIST secondary) spectral irradiance to illuminate a diffuser whose bidirectional reflectance distribution function (BRDF) is accurately known. A collimating mirror is used to achieve a uniformity of irradiance and of angle of illumination at the diffuser. Thus, at given angles of view, the spectral radiance of the diffuser is known, and the instrument response to this radiance is measured. The problem is that all three elements (window, collimating mirror, and diffuser) have been interposed between the standard source and each introduces changes into the transfer process. The window and collimating mirror can be measured for spectral transmission and reflectance to within 3% to 1% errors. This is done before and after each major calibration sequence. In addition, the filter detector monitor will indicate any general change in collimator or diffuser efficiency. Extensive experiments and analysis have demonstrated that radiometric calibration are precise to better than 1% (2σ). The absolute accuracy of the BRDF values are 3%. Non-linearities, wavelength errors, and other instrument factors have uncertainties that are less than 1%.

The test diffuser efficiency is measured by direct comparison to a NIST calibrated standard. This is done in the calibration test fixture using the instrument as the detector. A mask in the collimated beam with a 2.5 cm (1 in) aperture will restrict illumination to a small area of the test diffuser. The NIST standard diffuser is moved in front of the test diffuser so that its efficiency is mapped over its surface.

With the standards of irradiance, the collimator and the test diffuser, the instrument is presented with a diffuse source of known radiance against which its response can be calibrated. To calibrate the irradiance mode of operation, the test diffuser is moved out of the way as the instrument diffuser is deployed. In effect, this compares the efficiencies of the two diffusers, since their interchange is the only difference in the two measurements.

The result of the radiometric calibration is a function that assigns a calibration constant, KE, for irradiance to each instrument wavelength, λ, given by:

K sub E ( lambda)= {E sub{dk}(lambda)} over {C C sub E (lambda)}

where Edk is the test fixture irradiance, and CCE is the corrected count rate or instrument signal. Similarly, the calibration constant for radiance, KL, is given by:

K sub L ( lambda)= {L sub{dk}(lambda)} over {C C sub L (lambda)}

where Edk is the test fixture radiance and CCL is the corrected count rate or instrument signal.

Integrating spheres are also used to calibrate the SBUV/2 instruments. Comparison of integrating sphere and diffuser calibrations has shown agreement of about 1% for Space Shuttle SBUV (SSBUV) and SBUV/2.

The SBUV/2 instrument has four viewing modes (nadir Earth, solar diffuser, lamp diffuser, and lamp direct), but only two are used for radiometric calibration. In orbit, the instrument has a nadir view of the Earth in radiance mode. During ground testing, the sensor module views an illuminated target. In the irradiance mode, the sensor module deploys a reflective diffuser, which is illuminated by the Sun during a portion of the orbit. In ground test, the diffuser is illuminated by a collimated source. The instrument has two major grating modes for collecting radiance and irradiance data. The first is the step scan mode when 12 channels are scanned from 252 nm to 340 nm in 32 seconds. The second is the continuous scan (160 nm to 405 nm) mode used for solar irradiance measurements and wavelength calibration. So, there are 4 sets of calibration constants, namely: radiance discrete mode, irradiance discrete mode, radiance continuous mode, and irradiance continuous mode.

7.4.4 Additional Instrument Calibration

The spectrometric testing and calibration of the monochromator is done as a subsystem in that the diffuser and depolarizer at the front end are installed. The instrument entrance slit is illuminated via an integrating sphere by various intense line sources and the line positions versus grating positions are recorded. One of the sources is the in-flight Mercury calibration lamp. The integrating sphere fills the optical extent of the monochromator for these tests.

Further calibration information is obtained on the diffuser and gain ranges. The brightness of the diffuser varies with wavelength and illumination angle of the source. A series of measurements of FEL lamp and a Mercury Pen Ray lamp are made over the range of expected solar illumination and viewing angles to characterize the diffuser goniometry. The interrange ratios are determined by the ratio of counts in two gain ranges when both ranges have valid data. They are observed to vary as a function of wavelength.

Out-of-band errors are measured pre-flight by viewing the diffuse sky or a diffuser illuminated by the sun at the earth's surface. Tests on current instruments result in larger responses in the region below 290 nm than the expected signals would produce. Tests for additional characterization of the out-of-band response are planned.

Additional measurements to estimate signal to noise, stray light, out-of-field response, alignment, and stability are made. Full details of each instruments calibrations are presented in the Specification and Compliance Calibration Data Books (see Ball (1991)) prepared by the instrument contractor.

7.4.5 In Orbit Calibration

Once in orbit, the SBUV/2 responses are carefully monitored over time. Hilsenrath et al. (1995) have shown that measured albedo relative to the "day 1" albedo over time is a function of true change in Earth radiance, diffuser degradation, and interrange ratio (PMT gain). Since the primary measurements are ratios, changes in many of the components will cancel. The diffuser plate is used only for the solar irradiance measurements. Accurate characterization of its degradation is required to maintain calibration. This is accomplished by using an on-board Hg calibration lamp. The lamp is alternately viewed directly and indirectly by illuminating the diffuser. Diffuser degradation is monitored at six mercury lines spanning 185 to 405 nm.

The wavelengths associated with the grating positions have been observed to drift. Measurements of the calibration lamp at the six mercury lines, and continuous scans of the solar spectrum (via the diffuser) are used to estimate this drift. Absolute wavelength accuracy can be determined to about 0.02 nm.

The three gain ranges evolve differently over time, and must be periodically renormalize to each other. This is accomplished by using measurements for which two ranges are valid. The magnitude of the interrange ratios (gain range 2 versus gain range 3) for the SBUV/2 on NOAA-11 decreased by about 15% over its 5 years of operation.


Previous Section Top of Page Next Section