Determination of Near-IR Water Vapor Self Continuum from Observations

There has been great progress in recent decades in validating and improving the spectroscopic foundation for atmospheric radiative transfer calculations, resulting in a generally high level of confidence in line-by-line radiative transfer models for clear-sky remote sensing and energy balance applications. One remaining significant uncertainty is the strength of water vapor continuum absorption in the windows between near-infrared water vapor bands. Many radiation codes obtain their water vapor continuum absorption coefficients from the MT_CKD continuum model, which derives its near-infrared values from a water vapor line shape function that has been constrained by continuum measurements in other spectral regions. A number of recent laboratory studies have shown that the near-IR self continuum in the MT_CKD model is too low. However, these laboratory studies disagree by more than an order of magnitude in their assessment of the strength of the near-IR self continuum. At the high end of the range of measured strengths, the self continuum absorbs a significant amount of solar radiation, so it is crucial that other observational studies are performed to establish the actual strength of this absorption source.

We describe here an analysis of measurements from a solar FTS in Lamont, OK, from the Total Carbon Column Observing Network (TCCON). Thirteen periods on individual days in 2012 were identified as being reasonably stable with respect to aerosol optical depth and precipitable water vapor. For each period, the FTS measurements, coincident measurements from a Normal Incidence Multi-Filter Radiometer (NIMFR), and calculations from the Line-By-Line Radiative Transfer Model (LBLRTM) were used to obtain aerosol optical depths in window regions from 8000-20000 cm-1, which were then extended to lower wavenumbers through a generalized Angstrom relationship. In windows between 4000-7000 cm-1, these aerosol optical depths were subtracted from FTS-derived total optical depths, as were LBLRTM optical depths without any self continuum. The self continuum absorption coefficients that were derived from this procedure were at the low end of the recent laboratory studies, but higher than the current version of MT_CKD. The implications of the results of this study with respect to the absorption of solar irradiance will be discussed.