Short-term Prediction Research
and Transition Center

Remote Sensing Techniques

AIRS Retrievals and Products

AIRS Version 5.0 Profile Data

For the past two years, the SPoRT Center has been actively pursuing the use of remotely-sensed atmospheric thermodynamic profiles from the Atmospheric Infrared Sounder (AIRS) for use in regional numerical weather prediction. Recent conference papers by Chou et al. (2006; 2007) have shown that use of quality control features from the EOS Science Team can lead to improvements in analyses used as first guess fields for numerical models. These initial fields can, in turn, lead to improvements in sensible predicted variables such as temperature, mixing ratio, mean sea level pressure, and precipitation.

AIRS flies aboard the EOS polar-orbiting Aqua satellite with an early afternoon equator crossing time. It is a cross-track scanning infrared spectrometer/radiometer with 2378 spectral channels between 3.7 and 15.4 µm (650 and 2675 cm-1). Due to its hyperspectral nature, AIRS can provide near-rawinsonde-quality atmospheric thermodynamic profiles with the ability to resolve some small-scale vertical features in both temperature and water vapor. AIRS footprints coincide with footprints from the Advance Microwave Sounder (AMSU) allowing microwave data to be used in the retrieval process for cloud clearing purposes and producing a uniform distribution of AIRS retrievals in both clear and partly cloudy scenes at a spatial resolution of approximately 50 km. The superior vertical resolution and sounding accuracy make the instrument very appealing as a complement to rawinsonde measurements in data sparse regions.

Even though the infrared sensor on AIRS is coupled with a microwave sensor (the Advanced Microwave Sounding Unit; AMSU to assist with cloud clearing, clouds are still disruptive to the retrieval process. A key component of the Version 5.0 of the EOS Science Team AIRS profile data is level-by-level error characteristics for temperature and moisture (Susskind, personal communication). These error characteristics allow the selection of a specific pressure level within the retrieval above which quality data is contained. Prudent use of the quality indicators allows data above clouds to be assimilated into the system rather than completely disregarding entire profiles that may contain low, broken clouds. Currently, SPoRT uses the quality control procedures designated by the EOS Science Team for temperature profiles and applies them to both temperature and moisture profiles. In the future, SPoRT will conduct experiments for quality control explicitly on the moisture profiles.

SPoRT is using soundings from the support product of the Version 5.0 EOS Science Team algorithm, which contain 54 vertical levels between the 1013.25 and 100 hPa. Although these new version profiles have not yet been validated, it is expected that the relative validation errors will be similar to (or better than) those presented for the version 4.0 data because of improvements to the radiative transfer algorithm (Strow, personal communication). For Version 4.0, the errors are within 1K for 1-km layers for temperature and 10-15% RH in 2-km layers for water vapor (compared with globally-collocated rawinsondes) with the lowest errors occurring in the mid-troposphere for clear-sky cases over water (Tobin et al. 2006; Divakarla et al. 2006).

SPoRT plans to conduct a series of model runs using AIRS profiles in a near-real-time framework throughout the spring and summer of 2007 to fully assess impact on both analyses and forecasts (link to Chou’s modeling section).

Infrared image of cloud locations from GOES-11 (left) and maximum pressure level of quality data as determined by EOS Science Team Version 5.0 quality indicators (right) for 07-08 UTC 8 September 2006.

AIRS Retrievals and Products

SPoRT has recently begun producing near-realtime products directly from Atmospheric Infrared Sounder (AIRS) measurements. The instrument, which measures 2378 channels from 3.7-15.4 µm, measures radiances capable of being utilized to provide a far more detailed view of the vertical structure of the atmosphere. This high spectral resolution, as well as its heritage for future instruments, has been utilized to both adapt classic and create new techniques. These measurements are received from the UW direct broadcast and processed locally.

The CO2 sorting technique is implemented to provide the percentage of bands within the 15 µm CO2 absorption continuum which are uncontaminated by clouds. This methodology, adapted from Holz et al. 2006 and described in McCarty et al. 2007, is used for the purpose of determining the channels in an AIRS instantaneous field of view which can be assimilated in an analysis system. The figure below is a visual schematic to explain the CO2 sorting technique. Red and blue lines show the Weighting Function for a selection of AIRS Bands. Column 1 represents a clear IFOV. Column 2 represents a cloudy IFOV with a low cloud. Column 3 represents a cloudy IFOV with a high cloud.

The CO2 sorting technique is used to determine those channels, represented by the red weighting functions, in the second IFOV that are not affected by clouds. This technique is also directly implemented tied to the AIRS radiance assimilation research being performed at SPoRT. The CO2 slicing technique is being utilized to produce cloud top pressure and effective cloud fraction products. These products provide a measurement of cloud height and cloud coverage in the atmosphere. The technique has been adapted to utilize the hyperspectral nature of the AIRS instrument, as described in McCarty and Jedlovec 2006. The increased spectral resolution of the instrument allows for a more detailed investigation of the vertical structure of the atmosphere. This trait is utilized to produce more accurate cloud height assessments.