Evaluation of model-derived dust vertical profiles
An essential step in dust forecast is the model evaluation. This process consists in the comparison of the model results with the available observations on different temporal and spatial scales. It facilitates the understanding of the model’s capabilities, limitations, and appropriateness for the purpose for which it was developed. The first problem for the evaluation of dust models is the scarcity of routine observations intended for dust monitoring. By the moment, sun-photometric measurements and satellite retrievals are used for the SDS-WAS joint evaluation exercise. However, to date, the only parameter that is routinely evaluated is the dust optical depth, which integrates the contribution of the entire vertical column.
Ground- and satellite-borne lidar and last generation of ceilometers are the only tools capable of inquiring about the vertical profiles of aerosol-related variables. Therefore, information provided by them may potentially be used to evaluate the vertical component of the dust fields.
Some work has already been done in this direction, as in Binietoglou et al. (2015), where dust concentration forecast by different numerical models is compared with lidar retrievals of dust mass vertical profiles obtained through the use of the Lidar/Radiometer Inversion Code (LIRIC) algorithm. However, retrieval of mass vertical profiles has limitations and now it has been opted to establish a routine comparison of extinction vertical profiles.
Up to now, University of Lille, State Meteorological Agency of Spain, University of Granada and Barcelona Supercomputing Center contribute to the initiative, which is expected to extended soon to other institutions.
Two examples of compared profiles:
- Dakar, Senegal, 13 Jan 2016. BSC-DREAM8b model run by the Barcelona Supercomputing Center and lidar operated by the University of Lille (left)
- Santa Cruz de Tenerife, Spain, 23 Feb 2016. BSC-DREAMb model and ceilometer operated by the Izana Atmospheric Reseach Center, AEMET (right)
Access to raw data: