6A.4 Comparison of observed cloud properties at the three ARM sites in the Tropical West Pacific and the AMF site in Niamey with their representation in operational models

Tuesday, 7 August 2007: 5:00 PM
Hall A (Cairns Convention Center)
Ewan J. O'Connor, University of Reading, Reading, Berks., United Kingdom; and R. J. Hogan and A. J. Illingworth

The correct representation of clouds is crucial to models used to provide short term weather forecasts and predict future climate change. Such models now typically have one or two prognostic variables to simulate the clouds, such as cloud fraction and ice/liquid water content. Cloudnet was a successful EU project comparing radar and lidar observations at several sites in Europe with the representation of clouds over these sites in a number of operational forecast models. Here, we extend the Cloudnet scheme to the ARM (Atmospheric Radiation Measurement) sites in the Tropical West Pacific (Darwin, Manus, Nauru) and the AMF (ARM Mobile Facility) site in Niamey, Niger.

The evaluation of the NWP model performance will focus on cloud fraction, IWC (ice water content) and LWC (liquid water content). At least 9 months of observations are available for comparison at Niamey and much more for the three TWP sites.

The lidar backscatter and Doppler radar parameters are used to classify the targets as one of the following: i) aerosols, ii) insects, iii) aerosols and insects, iv) ice and supercooled droplets, v) ice, vi) drizzle/rain and cloud droplets, vii) drizzle or rain, viii) cloud droplets only and ix) clear sky, this information can then be used to determine the number of cloud pixels in a model grid box (i.e. derive cloud fraction). Ice water content (IWC) is derived from radar reflectivity (Z) and temperature (T) from the operational model using an empirical IWC-(Z,T) relationship (Hogan et al., 2004). It is assumed that liquid cloud layers have a linear increase of liquid water content (LWC) with height. The LWC of the liquid layer is obtained by determining the cloud base and top of a liquid layer and scaling the assumed linear profile to be consistent with the liquid water path (LWP) measured by a dual-wavelength microwave radiometer (e.g. Albrecht et al., 1990).

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner