Uncertainties associated with aerosol emissions in terms of their intensity and distribution pattern, atmospheric processes, and optical properties, represent a significant part of the uncertainty associated with the quantification of the impact of aerosols on climate and air quality in regional and global models. Lidar assimilation in global aerosol-climate models is an active area of research at many forecasting centres and research institutes. Assimilation systems used range from variational to ensemble methods, variables assimilated are aerosol extinction and backscatter coefficients or lidar raw signals (by using customized forward models). Applications range from aerosol global forecasts, to volcanic ash detection and regional air quality.

Data assimilation techniques are implemented to decrease these uncertainties, constraining models with available information from observations in order to make a best estimate of the state of the atmosphere. The short-range forecasts from such systems have the potential to be useful for the calibration/validation (Cal/Val) of new satellite data as they provide a stable reference for inter-comparison between products from different satellites. In particular, the use of a forecast model minimises errors due to temporal differences when comparing two different observational datasets.

This Cal/Val technique has been found to be useful for satellite observations sensitive to temperature and humidity, since the short-range forecasts are highly accurate for these variables, and this has been explored further within the GAIA-CLIM project. However, for aerosol products the short-range forecasts are not yet accurate enough to be able to identify more than gross errors in the satellite observations.

Further improvements to the aerosol data assimilation systems are needed, particularly in the area of bias correction, before aerosol forecasts can be used as a reference for satellite Cal/Val. This is a long-term goal, however, and in the short-term direct comparisons between aerosol observations should continue to be carried out for the Cal/Val of new satellite products.

Aerosol lidar data can also be used to constrain uncertain model processes in global aerosol-climate models. Satellite-borne lidar data can be effectively assimilated to improve model skill but, currently, aerosol lidar data assimilation experiments are mainly involving lidar attenuated backscatter, which is a non-quantitative optical property of aerosol. Ground based lidar networks can in addition provide quantitative measurements of aerosol backscatter and extinction coefficients. However, a limited number of aerosol lidar data assimilation experiments have been performed, preventing us from assessing the effective impact of assimilating continuous satellite lidar data and whether the current state of the lidar technology fulfils the modellers’ needs.