Atmospheric absorption model for dry air and water vapor at microwave frequencies below 100 GHz derived from spaceborne radiometer observations

Abstract

The Liebe and Rosenkranz atmospheric absorption models for dry air and water vapor below 100 GHz are refined based on an analysis of antenna temperature (TA) measurements taken by the Global Precipitation Measurement Microwave Imager (GMI) in the frequency range 10.7 to 89.0 GHz. The GMI TA measurements are compared to the TA predicted by a radiative transfer model (RTM), which incorporates both the atmospheric absorption model and a model for the emission and reflection from a rough‐ocean surface. The inputs for the RTM are the geophysical retrievals of wind speed, columnar water vapor, and columnar cloud liquid water obtained from the satellite radiometer WindSat. The Liebe and Rosenkranz absorption models are adjusted to achieve consistency with the RTM. The vapor continuum is decreased by 3% to 10%, depending on vapor. To accomplish this, the foreign‐broadening part is increased by 10%, and the self‐broadening part is decreased by about 40% at the higher frequencies. In addition, the strength of the water vapor line is increased by 1%, and the shape of the line at low frequencies is modified. The dry air absorption is increased, with the increase being a maximum of 20% at the 89 GHz, the highest frequency considered here. The nonresonant oxygen absorption is increased by about 6%. In addition to the RTM comparisons, our results are supported by a comparison between columnar water vapor retrievals from 12 satellite microwave radiometers and GPS‐retrieved water vapor values.

The Salinity Retrieval Algorithms for the NASA Aquarius Version 5 and SMAP Version 3 Releases

“…This temperature dependence is given by the expression ( 300K /T air ) α . The numerical value of the exponent α is 1.5 in [18] and it is 0.8 in [19]. This means that the change of the absorption coefficient as function of air temperature when going from warm to cold temperatures has effectively been reduced in Version 5 by about 50% compared to prior versions.…”

“…Figure 3 depicts the radiometric impact of the change in the O2 absorption by showing the difference in the correction (TOA TB minus surface TB) between the two absorption models as function of time and latitude. [19], which is used in the Version 5 algorithm and the Wentz Meissner O2 absorption model [18], which had been used in prior versions.…”

“…Prior to Version 5, all Aquarius salinity releases had used the oxygen absorption model of Wentz and Meissner [18], which has been extensively used in retrieval algorithms at higher frequencies. The salinity retrievals of these prior versions showed spurious seasonal salty biases at high N and S latitudes, which were largest the N Pacific during April and May.…”

“…2018, 10, x FOR PEER REVIEW 6 of 26 L-band frequencies, which are far from the oxygen absorption lines, the oxygen absorption is caused almost entirely by non-resonant continuum absorption, which is difficult to determine experimentally. Prior to Version 5, all Aquarius salinity releases had used the oxygen absorption model of Wentz and Meissner [18], which has been extensively used in retrieval algorithms at higher frequencies. The salinity retrievals of these prior versions showed spurious seasonal salty biases at high N and S latitudes, which were largest the N Pacific during April and May.…”

Low Dimensional Embedding of Climate Data for Radio Astronomical Site Testing in the Colombian Andes

“…The main factor contributing to the atmospheric opacity is water vapor, which very efficiently absorbs light in the THz range , Kuhn et al 2002 due to a continuum absorption spectrum formed by collisionally broadened absorption lines of water vapor in this frequency range (Clough et al 1989, Pickett et al 1998, Turner et al 2009). In order to characterize a site according to its atmospheric transparency to THz radiation, it is necessary to retrieve the precipitable water vapor (pwv) profile using remote sensing techniques such as microwave radiometry (Peter & Wentz & Meissner 2016), radiosonde humidity measurements (Liljegren et al 2001, Luini & Riva 2016, or indirectly via models of in situ climatological measurements (Lew & Uscka-Kowalkowska 2016) or GPS-delay studies (Bevis et al 1992, Niell et al 2001, Wang et al 2007). …”