When a signal transmitted by the global positioning system (GPS) and received by a low-Earth orbiter (LEO) passes through the Earth's atmosphere its phase and amplitude are affected in ways that are characteristic of the index of refraction of the propagating medium. By applying certain assumptions on the variability of the index of refraction of the propagating media (e.g. spherical symmetry in the locality of the occultation), phase change measurements between the transmitter and the receiver yield refractivity profiles in the ionosphere (~60-1000 km) and neutral atmosphere (0-50 km). The refractivity, in turn, yields electron density in the ionosphere, and temperature and pressure in the neutral atmosphere. In the lower troposphere, where water vapor contribution to refractivity is appreciable, independent knowledge of the temperature can be used to solve for water vapor abundance.
The radio occultation technique has a 30 year tradition in NASA's planetary program and has been a part of the planetary exploration programs to Venus, Mars and the outer planets. However, the application of the technique to sense the Earth's atmosphere using GPS, first suggested by Yunck et al. , was tested for the first time with the launch of the GPS/MET mission on April 3, 1995. GPS/MET is an experiment managed by the University Corporation of Atmospheric Research (UCAR) [Ware et al., 1995] and it consists of a 2 kg GPS receiver piggybacked on the MicroLab I satellite which has a circular orbit of 730 km altitude and 60 deg. inclination. The GPS receiver is a space qualified TurboRogue [Meehan et al., 1992] capable of tracking up to 8 GPS satellites simultaneously at both frequencies transmitted by GPS. Under an optimal mode of operation, the GPS receiving antenna boresight is pointed in the negative velocity direction of the LEO and provides 100-120 globally distributed setting occultations per day. Tens of thousands of occultations have been collected over the past two years and can be used to assess the accuracy and potential benefit of the GPS radio occultations.
The basic observable for each occultation is the phase change between the transmitter and the receiver as the signal descends through the ionosphere and the neutral atmosphere. After removal of geometrical effects due to the motion of the satellites and proper calibration of the transmitter and receiver clocks, the extra phase change induced by the atmosphere can be isolated. Excess atmospheric Doppler shift is then derived. This extra Doppler shift can be used to derive the atmospheric induced bending as a function of the asymptote miss distance, a. [see figure above defining these terms]. Assuming a spherically symmetric atmosphere, the relation between the bending and extra Doppler shift induced by the atmosphere is given by
where f is the operating frequency, c is the speed of light, v⃗t and v⃗r are the transmitter and receiver's velocity respectively, and k^t and k^r are the unit vectors in the direction of the transmitted and received signal respectively, k^ is the unit vector in the direction of the straight line connecting the transmitter to the receiver.
The spherical symmetry assumption can also be used to relate the signal's bending to the medium's index of refraction,n, [Born and Wolf, 1980] via the relation
where a= nr and r is the radius at the tangent point [see figure above ]. This integral equation can then be inverted by using an Abel integral transform given by
The refractivity, N, is related to atmospheric quantities via
where Pis total pressure (mbar), T is temperature (K), Pw is water vapor partial pressure (mbar), Ne is electron density (m-3), f is operating frequency (Hz), ρ is density, R is the gas constant, m is the gas effective molecular weight, h is height, g is gravitational acceleration.
When the signal is passing through the ionosphere (tangent point height > 60 km), use of a single GPS frequency is sufficient to estimate the bending to be used in Eq. (3). Moreover, the first two terms on the right hand side of Eq. (4) are negligible, therefore, knowledge of the index of refraction leads directly to electron density.
When the signal is going through both the neutral atmosphere and the ionosphere (tangent point height < 60 km), a linear combination of the two bending angles, associated with the two GPS frequencies, is used to isolate the neutralatmospheric bending and its refractivity profile is derived by use of Eq. (3) [ Vorob'ev and Krasil'nikova, 1994]. In the stratosphere and the region of the troposphere where temperature is colder than ~250K, the water vapor term in Eq. (4) is negligible. Therefore, knowledge of refractivity yields the density of the medium by use of the ideal gas law (Eq. 5). The density in turn yields the pressure by assuming hydrostatic equilibrium (Eq. 6) and a boundary condition at some height. Applying the gas law once more, knowledge of density and pressure yields the temperature. In the troposphere, at height where the temperature is larger than 250K, the water vapor term in Eq. (4) becomes significant and it is more efficient to solve for water vapor given some independent knowledge of temperature [ Kursinski et al., 1995].
Details about vertical and horizontal resolution of the technique, and refractivity, temperature, pressure, water vapor or electron density accuracies as a function of height, are discussed by Kursinski et al., 1997. Here we highlight some of these GPS occultation features.
Due to the nature of the measurement, which is a pencil-like beam of the electromagnetic signal probing the atmosphere, the technique has a much higher vertical and across-beam resolution than horizontal (i.e. along the beam). The vertical resolution of the technique is essentially set by the physical width of the beam where geometrical optics is applicable. This scale is set by the Fresnel diameter which, in vacuum, is given by
where λ is the signal's wavelength, RGPS and RLEO are the distances of the tangent point to the GPS and LEO respectively. For a LEO, Dvacuum is ~1.5 km. In the presence of a medium, due to bending induced on the signal, the Fresnel diameter is ~0.5 near the surface and approaches 1.5 km above 20 km altitude where bending becomes small. When the signal encounters sharp gradients in refractivity due to either water vapor layers near the surface or sharp electron density changes at the bottom of the ionosphere, the Fresnel diameter shrinks to ~200 meters.
A horizontal resolution scale is set by the length of the beam inside a layer with a Fresnel diameter thickness. This length is 160-280 km for a Fresnel diameter of 0.5-1.5 km.
In the ionosphere, the vertical scale is still set by the Fresnel diameter; however, the horizontal scale can extend several thousands of kilometers due to the large vertical extent and scale height of the ionosphere. These features of the ionosphere allow one to use tomographic approaches in order to combine information from neighboring occultations to solve for horizontal and vertical structure [ Hajj et al., 1994].
Under ideal conditions, when a LEO tracking GPS has a 360o field of view of the Earth's horizon, about 750 occultations per LEO per day can be obtained. However, side-looking occultations (GPS-LEO link > 45o from velocity or anti-velocity of LEO) sweep across a large horizontal region, and the spherical symmetry assumption becomes inaccurate. Discarding side-looking occultations, one LEO provides up to 500 occultations per day.
A high inclination LEO provides a set of occultations that covers the globe fairly uniformly. This feature is particularly advantageous when comparing LEO-GPS occultation coverage to that obtained from balloon launched radiosondes. A total of about 800 radiosondes are launched each 12 hours from sites around the world. The vast majority of these sites are over the northern hemisphere continents, particularly Europe and North America. This creates the need for high resolution temperature/pressure/water vapor profiles in the southern hemisphere and over the oceans. The contribution of radio occultation retrievals to climate and weather modeling should be particularly important in these regions. (Global data provided by spaceborne nadir sounders average over large-3-7 km-vertical distances.)
When compared to infrared spaceborne sounders, the radio occultation technique has the advantage of being an "all-weather" system. Namely, it is insensitive to aerosols, cloud or rain due to the relatively large GPS wavelengths.
Unlike other techniques such as radiosonde or microwave sounders, where instruments need constant calibration, the GPS radio occultation provides a self calibrating system, as will be discussed in more detail below. The long term stability inherent in radio occultation make this an excellent system to keep an accurate record of climate changes.
The main observable used in an occultation geometry is the phase change between the transmitter and the receiver as the occulting signal descends through the atmosphere. This phase change is due to (1) the relative motion of the LEO with respect to the GPS, (2) clock drifts of the GPS and LEO and (3) delay induced by the atmosphere. In order to derive the excess atmospheric Doppler shift, one must remove the contribution of the first two effects.
Accurate knowledge of the GPS orbits comes from an overall solution involving all 24 GPS satellites and a global network of ground receivers. The LEO orbit is determined by use of other links tracking the non-occulting GPS satellites.
When the occultation is mostly radial (i.e. GPS-LEO link has no horizontal motion out of the occultation plane), the occultation link descends through the ionosphere and stratosphere at a rate of about 3 km/sec; thus, crossing a Fresnel diameter in about 0.5 seconds. However, in order to investigate sub-Fresnel structure (by examining the diffraction pattern of the received signal's phase and amplitude) and for other purposes (such as eliminating different signals caused by atmospheric multipath in the lower troposphere) the occulting data is taken at a rate of 50 Hz. In order to calibrate the LEO clock, one more GPS transmitter is tracked by the LEO at the same high rate (labeled as link 2 on the figure). In addition, in order to calibrate the GPS clocks, a ground receiver tracks both GPS satellites at 1 Hz (labeled as links 3 and 4). One can interpolate the lower rate GPS clock solutions to 50 Hz, due to the greater clock stability (of order 10^12 sec/sec, as opposed to 10^9 sec/sec for the LEO clock), and the smoothness of the DoD Selective Availability dithering.
Knowing the position of all four participants (i.e. two GPS satellites, one LEO and one ground receiver), and modeling various physical effects such as light travel time, the three spaceborne clocks can be solved for w.r.t. to the ground clock. The net result of the calibration is the excess phase due to the atmosphere as a function of time.
The radio occultation technique holds the promise of providing atmospheric sounding which combines the high-resolution vertical profiling characteristic of radiosondes and the global coverage usually provided by passive remote sounders. The technique depends on the proven ability to measure the time delay of transmitted radio signals with high precision and stability as they traverse the atmosphere. This ability to determine time differentials very precisely and to provide an observational data base with long-term stability is enhanced by the use of GPS. Also, the use of radio frequencies means that cloud and aerosol particles have little effect on retrieved quantities. Although measurements are sensitive to free electrons in the ionosphere, the use of dual frequencies can remove their effects. Sensitivity to very high concentrations of water vapor complicates the retrievals near the Earth's surface, but also holds forth the promise that key boundary layer structure may be retrieved.
The ability of radio occultation to provide high-vertical-resolution profiles has been demonstrated repeatedly in planetary missions. GPS occultations provide key measurements in four areas of concern to the scientific community:dynamics in the vicinity of the tropopause, measurements of climate change, atmospheric structure in the troposphere,and the calibration of IR sensors.
The tropopause is the transition region between the dynamically controlled thermal structure in the troposphere below and radiative control in the middle atmosphere above. It is a region of active scientific interest and a difficult region to characterize via traditional remote sensing. Early results of GPS/MET ( Kursinski et al., 1995) suggest that it can measure temperature as a function of pressure to an accuracy of 1K, or less, with a vertical resolution on the order of 1 km at and near the tropopause. This suggests that the technique may have a special role to play in this altitude regime. Here we discuss some of the scientific issues related to dynamics associated with the tropopause, explain why GPS occultations are important in addressing these issues in the context of the present network of operational remote and in situ sensors.
Understanding the exchange processes between the stratosphere and the troposphere is essential in understanding chemical and dynamic processes in both regions. Such an understanding is difficult to obtain because vertical transport across the tropopause depends on irreversible mixing associated with smaller scale processes. This interaction tends to be difficult to monitor and depends strongly on the local temperature and water vapor.
Observing System Simulation Studies (OSSEs) have shown that GCM assimilation systems are extremely sensitive to wind and geopotential inputs at upper tropospheric jet levels. The GPS measurements promise to provide accurate measurements of the 200 mb geopotential on a global basis and, if sufficient measurement density is achieved, should have a positive impact on global assimilation systems.
Accurate knowledge of the vertical and horizontal temperature structure near the tropopause may also greatly aid in planning air flight routes. Air traffic usually takes advantage of the strong jet streams associated with the tropopause, which can be estimated by examining horizontal temperature gradients in this region. High accuracy temperature measurements with high vertical resolution are needed on a global scale to determine and predict the location and strength of these jet streams. Furthermore, areas of intense clear air turbulence can be sensed by high vertical resolution temperature data because clear air turbulence leaves a distinctive structure in the temperature field.
Analysis of GPS/MET data suggests that occultation data may play an important role in assimilation models where the available data is sparse. For instance, we have noticed significant disagreements between the ECMWF global model and GPS/MET temperatures, especially near the tropopause, in the southeastern Pacific. There are typically ~60 radiosonde flights daily in the southern hemisphere, and consequently the ECMWF model relies heavily upon TOVS observations to infer tropopause temperatures in such regions. Accurate temperatures from GPS, at high southern latitudes, should improve the determination of the structure and evolution of the southern polar vortex during its annual cycle, and the characterization of the ozone hole formation and the subsequent dispersal of ozone-poor air over the southern hemisphere.
Dynamically, the tropopause represents the boundary between high potential vorticity (PV) air in the stratosphere and relatively low PV air in the troposphere. As such, the structure and topography of the tropopause is dynamically coupled to the troposphere below and therefore, as implied by the invertibility principle for potential vorticity, a detailed knowledge of tropopause topography can be utilized to infer the dynamical structure of the troposphere below (Hoskins et al., 1985). GPS observations provide the static portion of the PV and, with sufficient sampling density, should provide useful information on horizontal gradients and therefore the wind field and the relative vorticity. This implies that GPS observations near the tropopause may prove important in this regard for model initialization and assimilation. As cyclogenesis is often associated with high PV stratospheric air pushing downward into the troposphere depressing the tropopause, interest has been expressed by a number of researchers in exploring how GPS data may be used in the study of cyclogenesis particularly in regions such as the Northern Atlantic storm track where high vertical resolution soundings are sparse.
The GPS occultation technique is particularly powerful in retrieving temperatures with good accuracy, vertical resolution, and adequate spatial coverage near the tropopause. The ~200 km "line-of-sight" spatial averaging is comparable to or smaller than the horizontal scale of interest. Together with temperature retrievals by nadir viewing microwave radiometers, occultation temperature retrievals in this region should provide a powerful combination for climate studies of the Earth's lower atmosphere ( Spencer & Christy, 1990). Such a combination would dramatically increase our ability to study dynamics in the vicinity of the tropopause beyond the current network of passive infrared instruments and microwave radiometers.
Although the current infrared and microwave instruments on operational weather satellites offer better horizontal resolution than GPS occultations, they have difficulty measuring temperature in the vicinity of the tropopause with sufficient vertical resolution. First, any temperature profile retrieved from a radiometer measurement must be deconvolved from a set of radiances by an associated set of weighting functions typically having half widths of one half to a full scaleheight. Second, the comparatively sharp change in the lapse rate and the cold temperatures near the tropopause pose difficulties in the retrieval process. The results of comparisons between radiosondes and retrievals from satellite radiometric data consistently show increased errors in the retrieved temperature near the tropopause. For instance, temperature retrievals from the High-resolution Infrared Sounder (HIRS/2) and the Microwave Sounding Unit (MSU) show errors of ~2.5 K near the tropopause while only ~1.5 K in the mid-troposphere ( Reale et al., 1988). In fact, for extreme cases in the tropics, simulations show that the coldest retrieved temperature can be more than 10 K warmer than the actual temperature ( Yates et al., 1989).
Temperature measurements by radiosondes are themselves subject to systematic errors at tropopause heights. Because of the low atmospheric density at these altitudes, the physical temperature of the radiosonde thermistor is actually determined by a balance between conductive heat transfer with the air temperature, absorbed long-wave and solar radiation and the long-wave radiation emission of the sensor. This makes the measured temperature sensitive to a variety of factors such as cloud cover, cloud top temperatures, solar zenith angle, surface temperature, vertical temperature structure and vertical distribution of aerosols, ozone, water vapor and carbon dioxide ( Finger and Schmidlin 1991; McMillin et al. 1988). Present estimates indicate that the magnitude of these errors may be of order 1 to 3 K (Ahnert 1991; Schmidlin 1991; McMillin et al. 1988).
The occultation technique is well suited to tracking changes in regional and global climate. First, no long time scale drifts in calibration are associated with the occultation technique. For example, no degradation in accuracy was experienced when Voyager acquired radio occultation observations of Neptune and of Jupiter although ten years elapsed between them. In the GPS case, because the occulted satellite can be simultaneously viewed by other GPS receivers, the instabilities of the signal source during the observation can be measured and removed by the "double differencing" technique. By contrast, calibration drifts present a significant obstacle in detecting climate change with operational remote sensors. Second, the relatively long horizontal averaging interval of the limb sounding geometry is advantageous in this context because it inherently produces a spatial average of the quantity being measured. For instance, occultations effectively smooth over small scale internal gravity waves. Finally, with the increased accuracy of occultation temperature retrievals, the limiting factor in detecting long term climate trends and in determining climate averages becomes the natural variability of the climate instead of errors in the measurements themselves. Consequently, fewer measurements are required in determining averages and secular trends. Below we discuss some of the climate related issues which might be addressed by future GPS occultation measurements.
Fundamental toward understanding the response of the atmosphere to changes in its constituents is the balance between incoming short-wave/warming and outgoing long-wave/cooling radiation. In particular, temperatures in the upper troposphere/tropopause region are of tremendous interest because they determine the ability of the atmosphere to cool. Changes in concentrations of atmospheric radiative forcing constituents, such as aerosol and cloud particles and greenhouse gases, including water vapor, force the atmosphere to alter its temperature structure in this region in order to maintain an energy balance with incoming solar radiation. Major volcanic eruptions inject a large amount of aerosol material into the lower stratosphere, significantly altering the radiative forcing of the atmosphere in this region. Detailed monitoring of the evolution of the atmospheric thermal structure following these events can only be done on a global scale with remote sensing and can only be done remotely with wavelengths sufficiently long to be unaffected by the enhanced aerosols concentrations. As discussed earlier, initial results suggest that occultation temperature retrievals are most accurate in this cooling layer.
The reaction of high cirrus cloud formation to anthropogenic increases in greenhouse gas concentrations has been proposed as a climate feedback mechanism. If global warming displaces a given cloud layer to a higher and colder region of the atmosphere, then the colder cloud will emit less radiation forcing the troposphere to warm to compensate for this decrease (Houghton et al., 1990; referred to as IPCC hereafter). Accurate high vertical resolution temperature reconstructions in the tropopause region where cirrus clouds exist will assist in our understanding of the conditions under which they form and evolve. Currently, the changes in this altitude region associated with anthropogenic increases in greenhouse gases are poorly understood as indicated by the variability of GCM predictions ( IPCC). This present uncertainty in understanding combined with the importance of this altitude regime in maintaining radiative energy balance illustrates the importance of accurate, high resolution monitoring in this regime.
Tracking the geopotential height of constant pressure surfaces over time spans of decades allows one to detect an integrated expansion of the lower atmosphere. Looking for a systematic elevation of constant pressure surfaces over times spans of years would be a useful technique in searching for secular warming of the Earth's atmosphere. Thus far we have focused on temperature and water vapor as the only parameters of interest produced by the occultation technique; however, very accurate reconstructions of pressure as a function of geopotential height will also be provided wherever accurate temperature retrievals are produced. Since pressure is a vertical integral of the atmospheric mass density, an elevation of a surface of constant pressure would indicate that the mass of gas below that particular level expanded, which can only arise when the integrated temperature of that mass of air increases (Gary, 1992). Therefore, if the troposphere warms in response to the anthropogenically increased greenhouse gas forcing as is generally predicted, the average pressure scale height across the troposphere will increase, causing the height of a given pressure level to rise. The largest inflation should occur in the vicinity of the tropopause because the heights of pressure levels there represent the integrated effect of the warmer temperatures below. A simple order of magnitude estimate indicates that a 2 K increase (a magnitude often discussed in climate change simulations) will produce about a 70 m increase in height. This is equivalent to a change in pressure at fixed altitude of approximately 1%. We have not yet assessed the accuracy of pressure retrievals from GPS/MET, but our theoretical predictions indicate that pressure errors should be ~0.1%. In fact, the theorized sensitivity implies changes may be detected more easily than with other data, and that the sensitivity may be used to distinguish between and evaluate the climatic change predictions of different models, depending on the ability to separate the climate change signal from synoptic, diurnal, seasonal, and other climatic variability.
The change in surface temperatures due to an increase in greenhouse gas densities is generally predicted to be largest and may therefore become apparent first at high latitudes (IPCC). Occultation temperature retrievals at high latitude winter conditions are an obvious example of how the radio occultation observations can complement data from passive sounders. Accurate temperature measurements by passive remote sensing techniques under these conditions are difficult because of very cold temperatures, common near-surface thermal inversions, and the presence of ice clouds which limit IR soundings to the cloud tops. In contrast, the radio occultation technique has an advantage with colder temperatures because the air is dry, relatively dense, and scale heights are relatively small, all of which result in accurate temperature retrieval.
Despite the fundamental role played by water in weather and climate, an adequate climatology of atmospheric water vapor does not exist (Starr and Melfi, 1991). Radiosondes presently provide most of the high vertical resolution profiles of humidity with a highly in-homogeneous, land-biased spatial distribution. In the lower troposphere, GPS observations should deliver hundred meter to one km vertical resolution with global coverage and 100-200 km horizontal averaging. This vertical resolution lies between that of radiosondes and current satellite remote sounding instruments and should yield a significant improvement in the vertical scales observable globally. In addition, horizontal averaging produces profiles that are more representative climatologically than point measurements. Finally, insensitivity to particulates allows GPS occultations to measure humidity structure representative of the entire range of climatological variation by retrieving water vapor profiles within and below clouds.
The GPS water vapor retrieval accuracies estimated by Kursinski et al. (1995a) at low latitudes compare favorably with goals of 5 and 10% established by Starr and Melfi for the boundary layer and overlying troposphere respectively, and are generally conservative. Climatological boundary layer humidities may in fact be retrieved at the 1.5% level. Changes in water vapor abundance at low latitudes in the lower and mid-troposphere have been identified as reliable indicators of modeled climate change (Santer et al., 1990; Schlesinger et al., 1990). These are precisely the regions of greatest accuracy for GPS occultations.
GPS occultations present the only method of probing the troposphere in a limb sounding geometry routinely. Typically, tropospheric temperature retrievals from limb sounders are complicated by the presence of aerosols and clouds in the troposphere. The GPS L-band dual frequencies, however, are well away from atmospheric absorption lines and the effect of particle extinction will be negligible with an extreme upper bound of ~1% change in the total water refractivity(Kursinski et al.,1995). On the other hand, the permanent dipole moment of water vapor makes it a large contributor to atmospheric refractivity in the lower troposphere at microwave wavelengths. This is particularly true at low latitudes where the air is warm and absolute humidity levels are high. By using a priori knowledge of the temperatures in an individual profile, the water vapor profile can be retrieved. Our studies suggest that, in warm conditions, it should be possible to produce high vertical resolution profiles of water vapor accurate to better than 5 % in the boundary layer and 10% in the lower half of the troposphere (Kursinski et al., 1995). Again, a strength of the GPS occultation technique is its high vertical resolution in the troposphere. Vertical resolution in the troposphere can range from ~1 km down to 100 meters where the refractivity gradients are large. Under conditions where horizontal gradients are small, the vertical resolution can be pushed beyond the Fresnel diffraction limit by sampling and subsequently inverting the diffraction pattern.
Although occultations involve long horizontal path lengths through the atmosphere, this horizontal scale is consistent with the expected vertical resolution. Lindzen and Fox-Rabinovitz (1989) have pointed out that horizontal and vertical resolution are strongly coupled and must therefore be self-consistent in both atmospheric models and observational systems. Their work indicates that vertical resolution is inadequate in virtually all large scale models and observing systems. With GPS/MET, the ratio of vertical to horizontal scale resolution is typically of ~1:160, which is in good agreement with the constraints set by Lindzen and Fox-Rabinovitz (1989).
Retrievals of water vapor in tropical and subtropical latitudes are of interest because water vapor is responsible for most of the transfer of heat and energy in the tropics, where most of the Earth's insolation lies. Much of the insolation is directly transformed into the latent heat of evaporation of water from the ocean surface. The moist air is then advected in a Hadley circulation, in which it eventually rises and rains out. The details of the evaporation/ precipitation cycle in the tropics are not well understood, and consequently tracking the quantity and movements of water vapor is essential to characterizing heat transfer at low latitudes. Unfortunately, because of the limited number of radiosondes flights conducted in the tropics, only sparse and unevenly sampled information on water vapor concentrations exists. A GPS occultation data set would greatly augment the capacity for studying water vapor transport and the hydrological cycle. The impact of low level water vapor measurements in the low latitudes should have a dramatic impact on important assimilated parameters such as water vapor transport calculations.
As we extend the GPS retrievals to or near the surface, the height of the marine boundary layer should be determined with great accuracy as have been found in simulations ( Kursinski et al., 1997). The reason is simply the very sharp and large contrast in refractivity between the moist air below and relatively dry air above which results in dramatic changes in the amplitude and phase of the signal as the signal path crosses this region. This is of interest for weather and climate characterization and is extremely difficult to recover accurately with nadir-viewing sounders. Knowledge of this height can also potentially greatly improve the accuracy of nadir sounding retrievals of water vapor distribution of the same region.
Under conditions when it is difficult to separate the wet and dry contributions to the refractivity of the lower troposphere, the refractivity profiles themselves should provide a very accurate representation of the sampled volume. Although refractivity is not currently considered a variable of climatological interest, the accuracy available with GPS occultation retrievals could cause it to become so.
The anticipated accuracy of the radio occultation temperature retrievals, their long term stability, and the physical simplicity of the observable are key factors which suggest these observations may be useful as a calibration tool to aid other remote sensing observations. First of all, frequency and time differences can be determined much more accurately than intensity. Secondly, the use of well-known frequencies generated by highly precise clocks makes the physics of probing the atmosphere at all levels far simpler. In contrast, passive IR observations observe intensity as a function of frequency, a physically more complex relationship. For passive IR observations, at each frequency the observed intensity is the result of emission/absorption and subsequent radiative transfer through the intervening media. Both the emission and radiative transfer are functions of pressure and temperature dependent line-shapes and the densities of the emitting, absorbing and scattering constituents in the atmosphere.