(F.P.J. Francisco)

Broad Spectral Bandpass Hemispherical Field-of-View Solar Radiometer (BBHSR): These systems use an electrically calibrated pyroelectric detector, optical chopping and null- balanced operation. Prototypes were tested during NASA and NOAA field experiments >see Valero et al. 1982, 1983, 1984; Ackerman and Valero 1984, Valero et al., 1989).

Narrow Spectral Bandpass, Narrow Field-of-View Multiple Spectral Channels Infrared Radiance Radiometer (NFOVR): A liquid nitrogen cooled reference black body together with the electrically calibrated pyroelectric detection system provides outstanding accuracy. See Liou et al., 1990.

Broadband Hemispherical Field-of-View Multiple Spectral Channels Infrared Flux Radiometer (HFOVR): Similar to the BBHSR except that this instrument incorporates a long pass interference filter and uses the field of view inversion method. Used to measure infrared net fluxes.

Total-Direct-Diffuse Multiple Spectral Channels Radiometer (TDDR): Measures total, direct, and scattered radiation fields in several spectral channels. A unique optical radiation collecting system is used to collimate and spectrally separate hemispherical radiation. A prototype was tested during a NOAA field experiment, (Valero et al., 1989).

Broad Spectral Bandpass Hemispherical Field-of-View Solar Radiometer (BBHSR) Instrument Concept: The electrically calibrated pyroelectric radiometers differ from other radiometers in the use of optical chopping and null-balanced operation. Optical chopping is made possible by the fast response of the lithium tantalate pyroelectric detector and is effective in eliminating drift. A gold-black coating on the detector gives it a spectral range from the ultraviolet to the far infrared. Null-balanced operation, in turn, renders the resultant measurement insensitive to detector responsivity and amplifier gain and hence to ambient temperature. A laboratory model of the instrument is described by Geist and Blevin, 1973. An airborne model is described by Valero et al., 1982 and 1984.

Instrument Description: The most important design objectives are to:

Objective a) is met by using a lithium tantalate pyroelectric detector. This is a very fast detector which easily permits signal modulation and therefore synchronous amplification and detection. Furthermore, as in the case of the broad spectral bandpass radiometers, synchronous amplification is conducive to a large SNR and the rapid acquisition of data which is important in the case of fast moving clouds and/or aircraft.

Objective b) is met by using both optical chopping and the proper location of spectral filters. Radiation emitted by any component of the system located between the chopper and the detector produces a DC signal and, hence, does not contribute to the output signal of the synchronous amplifier. Provision is therefore made for the placement of filters between the chopper and the detector.

The key to the operation of the radiometer is the generation of an electrical signal synchronized 180 degrees out of phase with the optical chopper. This signal is fed into the gold-black coating on the detector surface which also serves as a resistive heating element; this gives a thermal signal in opposition to that resulting from the chopped optical radiation. The amplitude of the electrical drive is then varied until a null is detected at the output of the synchronous amplifier. At this point, the optical power absorbed by the detector is nominally equal to the electrical drive power. The latter is measured and digitally recorded on computer memory. Objective c) is fulfilled by the use of this detection method since the null detection is independent of circuitry changes and temperature.

The optical system is built around a double elliptico-logarithmic light collection cavity that directs the radiation through the chopper toward the detector assembly. A spectral filter introduced between the chopper and the detector eliminates any unwanted radiation (including infrared radiation) from the optical and other components ahead of the filter. The whole assembly, including preamplifiers, is encapsulated in a vacuum tight container to eliminate the systemıs sensitivity to microphonics and environmental moisture.

Calibration: All our instruments are carefully calibrated in our laboratory and, depending on the particular instrument, energy response calibrations are made at the NOAA Mauna Loa Observatory and from high altitude aircraft like the NASA ER-2. In general, calibration of the instruments are performed in three areas: energy response, zenith angular response, and azimuth angular response. Each of these are discussed below and apply to all the instruments. Special requirements of particular instruments are discussed in the corresponding sections. Spectral response is verified using the facilities of our spectroscopy laboratory.

Energy Response: The only optical calibration required is characterization of the detector to determine the degree of equivalence of electrical and optical heating. The instrumental response will be calibrated in the laboratory using National Institute of Standards and Technology (NIST) traceable standards and standard detectors. NIST developed laboratory versions of electrically calibrated pyroelectric detectors are used. Accuracies of 1% or better are achieved. The detectors are linear within 0.1%.

Zenith Angular Response: The zenith angular or ³cosine² response of the instrument can be made quite accurately. Small deviations are taken care of by angular calibration. Selection of proper diffuser materials eliminates most of the cosine response imperfections. Our experience shows that we can achieve a cosine response accuracy of 0.5%. The calibration of the angular response is done in the laboratory with a well collimated beam of light where the response to radiation entering at various angles is measured.

Azimuthal Angular Response: The azimuthal angular response of the instrument can also be made quite accurately.

Narrow Spectral Bandpass, Narrow Field-of-View Multiple Spectral Channels Infrared Radiance Radiometer (NFOVR): Instrument Concept and Description: The instrument design is directed towards obtaining accurate infrared radiance measurements by careful choice of materials, components, layout, and the best available technology, together with a superior approach to the physical problems of IR measurements.

The most important design objectives include the objectives discussed above (a-c) for the solar radiometers plus an additional key requirement for accurate detection of infrared radiation:

d) The accuracy of the measurements depends on the ability of the instrument to remain calibrated under all foreseeable conditions and for long periods of time. We achieve this by using the electrically calibrated pyroelectric technology (as described for the broad band solar radiometer above) plus, in the case of infrared measurements, a liquid nitrogen cooled blackbody reference that provides the zero radiation signal level. Identical optical elements in the scene and reference optical paths provide a symmetric, self-compensating, balanced infrared detection arrangement.

Before discussing how the design objectives are met, a brief description of the prototype airborne system is in order. The radiometer uses three germanium lenses, an electrically calibrated pyroelectric detector, optical chopping, a null-balanced method of measurement, and a liquid nitrogen cooled blackbody surface as a zero-radiation source. The two channels are identical except that one is a mirror image of the other, and they may differ in the spectral filtering used. The radiation passes through a 25 mm diameter objective lens, reflects off the gold-plated surface of the optical chopper, and is then focused on a 25 mm diameter field lens which images the objective on the detector. During the half-cycle when the optical chopper has rotated out of the path of the upward radiation, a third lens identical to the objective lens permits the detector field to be filled by the cold blackbody surface. This lens is also imaged on the detector.

In the hemispherical solar radiometer the chopper blade blocks the atmospheric radiation from the detector. For the narrow field-of-view infrared radiometer, the chopper blade is not used to block but to reflect the atmospheric radiation into the detector field. The electrical heating of the detector surface occurs during the half-cycle in which the liquid- nitrogen-cooled surface is viewed. In other respects the detection schemes are the same.

Design objective d) is met by using identical objective lenses mounted in the same heat sink for viewing both the field radiation and the liquid-nitrogen-cooled blackbody surface. This insures that the optical path characteristics are the same for the two fields viewed and that radiation emitted by each lens and viewed by the detector is the same. The primary difference between the two optical paths is the presence of the optical chopper when viewing the radiation field to be measured. Radiation emitted by the chopper is then received by the detector along with the radiation from the atmosphere. This chopper emission is relatively small since the gold-plated chopper has an emissivity less than 0.01 at the operating temperature. This small contribution is taken care of as discussed in the following calibration section.

The detector is an absolute device and calibration in the usual sense is not required. The fraction of radiation attenuated by the lenses and the filter can be measured and a resultant loss factor applied to the measured signal. Alternatively, the complete radiometer may be calibrated using either a standard source or a standard detector. The latter approach has proven superior in our experience with the 2-channel aircraft prototype.

Broadband Hemispherical Field-of-View Multiple Spectral Channels Infrared Flux Radiometer (HFOVR): For aircraft operations we have used an infrared version of the broad band solar hemispherical system described above to measure IR Net Fluxes with very high accuracy. We were able to achieve this by reversing the field of view of the radiometer to determine upwelling and downwelling fluxes. To obtain the net flux, one is subtracted from the other eliminating the systematic error introduced by the temperature dependent IR emission from the optical components (Valero et al., 1982).

The calibration procedures for the BBHIR will be the same used for the NFOVR. Our calibration laboratory is equipped with infrared radiation sources suitable for energy response calibration. Angular calibrations for zenith and azimuth will be performed following our standard procedure for hemispherical radiometers. We confirm the spectral response of the system using Fourier transform spectrometers that cover the wavelength range from the ultraviolet to beyond 100 microns.

Total-Direct-Diffuse Multiple Spectral Channels Radiometer (TDDR)
Instrument Concept
We describe here an instrument that we have developed and successfully used for research in the Arctic from high altitude aircraft. We have adapted this basic prototype to produce a radiometric system easily compatible with a ground-based operation.

Normally, optical depths in the Earthıs atmosphere are measured using hand-held or suntracking sunphotometers. However, automatic sun tracking systems are complex and require attention and considerable sophistication to maintain tracking accuracy within the very narrow field of view of the instruments. An alternative approach, that we have adopted, is to render the suntracking requirement unnecessary by using an optical system that is capable of separating the contribution of the direct (parallel) solar beam from the total hemispheric radiative flux. We have achieved this by incorporating an oscillating shadow ring in front of the optical aperture of the hemispherical field-of-view radiometer. In this fashion, at some point during the oscillation cycle, the ring will project a shadow that will exclude the direct solar beam from the field of view of the radiometer; only the scattered or diffuse component, Fd, of the total radiation field will reach the aperture of the optical system. On the other hand, when the oscillating ring is out of the field of view of the radiometer, the total hemispherical radiation field, Ft, is detected. From the values of Fd and Ft the direct solar beam, Fs, is obtained. The values of Fs at different altitudes in the atmosphere are then used to determine optical depths. By incorporating several spectral channels, optical depths as a function of wavelength can be measured. Additionally, the total hemispherical flux and the diffuse radiation field obtained using this technique are fundamental parameters that, through radiative transfer calculations, can be used to determine important quantities as discussed above.

An important consequence of the TDDR method of operation is that the oscillating shadow arm provides a semi-independent set of measurements comprised of flux values determined when different sectors of the hemispherical diffuse radiation field are shadowed. These measurements can be inverted to derive the angular dependence of the scattered field. The shadow arm operation also provides a way to infer the forward scattered flux directly from measurements which, in turn, permits the determination of ³true² optical depths (free of forward scattering error).

Instrument Description
Optics: We employed very simple optical arrangement for the radiometer. The diffuser- light trap arrangement provides a hemispherical field of view with incident radiation being collimated by the high reflectance walls of the exponential-logarithmic cavity. Enough collimation of the radiation is achieved with this design that narrow spectral bandpass interference filters can be used to select desired wavelength regions.

Electronics: The instrument electronics includes five major functional blocks. They are the detectors signal conditioning block, the data processing block, the system controller block, the shadow ring drive and control block, and the data storage block.

The signal detectors are silicon photodiodes operating in the photovoltaic mode and covering the spectral range from about 0.3 to l.l µm. Their signals are converted into electrical voltages by low noise FET input operational amplifiers. Programmable gain amplifiers allow adjustments for dynamic range, and filter circuits condition the signals for analog to digital processing. Data processing units consist of an analog multiplexer circuit, a sample-and-hold circuit, and an analog to digital converter providing a 12-bit resolution output. The shadow ring is driven by a DC motor rotating at a constant speed. A motor controller is used to maintain motor speed. The system controller provides the timing necessary to perform all the systemıs tasks. It sets the shadow ring in motion and steps through the detectorıs outputs, maintaining the proper dynamic range for the analog to digital converter by selecting the proper amplifier gain. It also controls the analog to digital conversion and selectively stores data.

Calibration of the instrument is to be performed for its function in four areas: energy response, zenith angular response, azimuthal angular response (as described in a previous section), and exoatmospheric constants. Here we will only discuss those calibration aspects that are particular to this instrument.

Azimuthal Response: In the TDDR instrument used for Earth atmosphere studies, the azimuthal angular response varies by ~10% in flux due to the configuration of the detector in the instrument (a linear array of silicon photodiodes are used). This is removed in the data analysis by calibrating the azimuthal angular response in the laboratory as described above. With very accurate information on the orientation of the TDDR with respect to the sun, the variation in azimuthal response can be removed from the data.

Exoatmospheric Constants: The measured values of direct solar flux, Fs, outside the atmosphere need to be determined prior to evaluating optical depths at levels within the Earthıs atmosphere. We derive these values from a Langley method calibration at the Mauna Loa Observatory. Preliminary tests from an urban site, where aerosol mass loading was somewhat variable during data acquisition, resulted in extrapolated constants at three radiometer channels having a precision of better than 2%. The conditions at the Mauna Loa Observatory, which lies well above the convective layer, enable us to improve that precision to a few tenths of 1%.