Sun-pointing ground-based FTS measurements of atmospheric water vapour absorption

Water vapour is the most dominant greenhouse gas in the earth's atmosphere, accounting for around 75 Wm^-2 of the absorption in the shortwave. A recent review of the GCMs used in the IPCC-AR4 report showed on average they overestimate the short-wave, and underestimate the long-wave, downward radiation at the surface by 6 Wm^-2, even under clear sky. As such, the correct parameterisation of water vapour within radiative transfer schemes is important in understanding of the earth's radiation budget for today's and future climates.

Infrared water vapour line absorption is fairly well understood, being made up of many thousands of rotational-vibrational transitions as listed in spectral line databases (e.g. HITRAN, GEISA). There is however also a less well understood water vapour continuum, which varies slowly with wavelength and underlies the spectral line absorption throughout the infrared. This continuum has a significant impact on fluxes and heating rates in this region and it is therefore important from a physical point of view to understand its contribution correctly. Present continuum models are limited by the paucity of high quality observations in some spectral regions, most notably the near-infrared, and for a wide range of conditions.

In this work, measurements of atmospheric absorption of solar radiation under clear sky conditions are made over two field campaigns using a calibrated ground-based high-resolution Fourier transform infrared spectrometer (FTS). This work is part of a UK research council funded consortium known as Continuum Absorption at Visible and Infrared wavelengths and its Atmospheric Relevance (CAVIAR). The CAVIAR project brings together theoretical, laboratory, airborne and ground based field work to derive new estimates of the water vapour continuum.

The FTS was calibrated at the National Physical Laboratory (NPL) and is traceable back to the UK's primary radiance standard. One campaign (summer 2008) was based at a UK Met Office observation station in Camborne, UK (N50.2, W5.3 alt. 87m) and the other (summer 2009) at the High Altitude Research Station at Jungfraujoch, Switzerland (N46.3, E7.59 alt. 3450m). The eventual aim of the field campaigns is to detect the continuum under sea level, “moist” atmospheric conditions and high altitude, “dry” conditions. At low altitudes, many water vapour bands are saturated, and continuum absorption is likely only to be detected in the band wings; however at high altitude it is possible to spectrally analyse much closer to the band centres. Details of the state of the atmosphere during the measurements is constrained using ECMWF analyses, in-situ radiosonde launches, ground based GPS water vapour instruments and from dropsonde and in situ measurements from the UK Facility for Airborne Atmospheric Measurements (FAAM) aircraft, which was also making spectroscopic measurements in the infrared.

We firstly present work analysing the consistency of near-infrared water vapour line intensities in recent HITRAN databases, building on earlier work by our groups. We simulate the optical depth of water vapour and five other primary absorbers in the infrared using a line-by-line (LBL) radiation code and fit this in a linear least squares sense to a convolved “pseudo” optical depth from the measurements. Due to instrumental response, we would expect a constant scaling factor to be applied to all water vapour line intensities, and, if our input water profile to the LBL code was correct, this scaling factor would be one. Our results show that HITRAN is generally in good agreement with observations, except in the 8000 cm^-1 to 9500 cm^-1 spectral region where it underestimates water vapour line intensities by up to 20%. This result was observed for all measurements over a variety of days, zenith angles and water vapour column amounts. This underestimation is reduced to about 10% when using new ab-initio calculations of lines in this spectral region by University College London (a CAVIAR partner).

We secondly present our work on absolutely calibrated spectral data. This allows us to spectrally resolve the radiance reaching the surface at high resolution (0.03cm^-1) and put an absolute value on any detected atmospheric continuum signal. We believe this is the first attempt to detect and quantify the water vapour continuum from direct solar measurements over such a broad spectral region in the infrared (500 – 15000 cm^-1). The variation of atmospheric conditions between Camborne and Jungfraujoch will allow us to retrieve continuum signal in both the band wings and band centres respectively. We constrain the aerosol contribution using a ground based sunphotometer, and the FTS is constrained against the UK's primary radiance scale.