The CLouds and Aerosol Radiative Interaction and Forcing Investigation (CLARIFI) project aims at quantifying the radiative effect of ultraviolet (UV)-absorbing aerosol layers above cloud layers, using space-based instruments and radiative transfer modelling.

Aerosols affect the Earth's albedo and shortwave radiation balance by scattering and absorption of sunlight and observations of aerosols are needed to advance our understanding of the Earth's energy balance. In particular, observations of the interactions between clouds and aerosols are of interest to understand processes of cloud forming and dissolving, which have a very strong and direct relation to the Earth's energy balance.

In the CLARIFI project the radiative effects of absorbing aerosols on underlying clouds are investigated using satellites. For this, the difference between a scene containing clouds and overlying aerosols, as measured by the space-borne spectrometer SCIAMACHY onboard Envisat, and a modelled scene with a similar cloud but without aerosols, is quantified.

Thus far, the project focused on the simulation of the unpolluted cloud scene. Using a model that computes the path of light from the Sun through the atmosphere, the reflectance of a scene with only clouds can be simulated. SCIAMACHY, a space-based radiometer, can measure the complete spectral reflectance of a scene from the ultraviolet range to the near-infrared. Using this model, measurements from SCIAMACHY, and standard cloud parameters retrieval techniques, the spectral reflectance of a scene with clouds can be simulated.

This technique can also be used for scenes with clouds that are polluted with aerosols, if the cloud parameter retrievals are not affected by the presence of the aerosols. UV-absorbing aerosols greatly affect the reflectance of the scene in the UV, but not so much in the near-infrared and this can be used to separate the cloud retrieval from the aerosol retrieval. The standard cloud retrieval technique is adapted to be applied exclusively in the near-infrared, so that the cloud simulation can be done also for polluted clouds. Then the effect of the aerosols on the scene reflectance can be determined from the actual measured reflectance and the reflectance that was simulated for the unpolluted cloud.

For instance, in the South Atlantic Ocean and the west coast of Africa, clouds over the ocean are polluted by smoke from fires on the continent. Smoke is UV-absorbing aerosol that can be identified by high values of SCIAMACHY AAI (Absorbing Aerosol Index). Also the clouds in a MERIS image turn from bright white to greyish, due to the presence of the smoke.

Using the 1064 nm backscatter signal of CALIPSO, a NASA space-based lidar, as a function of altitude it can be observed a clear, mostly closed stratocumulus cloud deck around 1 kilometre altitude. The cloud deck is very shallow and above the clouds there is an extensive aerosol layer from about 1 to 5 kilometre altitude, which is the smoke layer. Above 10 kilometres some cirrus clouds can be identified.

The aerosols clearly overlie the cloud layer, which is a configuration that is known to effectively absorb solar radiation in the UV. This effect is studied comparing the scene reflectance as measured by SCIAMACHY over a test area to the reflectance of a scene with the same clouds, but no aerosols above it. The clean cloud scene reflectance is simulated using a model which not includes most of the gas absorption bands. The difference between the two reflectance spectra is largest in the UV, around 350 nm, and decreases towards larger wavelengths. This is because the smoke strongly absorbs radiation in the UV, reducing the scene reflectance. Towards the near-infrared the aerosol optical thickness decreases and the absorption decreases accordingly. From the differences between the measured polluted cloud reflectances and the simulated unpolluted cloud reflectances, the absorbed energy by the aerosols can then be determined, using measurements of the incoming solar irradiance.

Information on the amount of absorbed energy by aerosols in the presence of clouds is needed, because the absorbed energy heats the atmosphere locally. The heat from the aerosol absorption at high altitudes may increase the atmospheric column stability, while less solar radiation will reach the cloud layer beneath. This could delay the break up of the cloud deck through less available energy and reduced convection. In that case, even less radiation will reach the Earth's surface, because clouds are very effective light scatterers and would have a strong impact on the Earth's energy balance. This interaction between clouds and aerosols can amplify the effect of the light absorption that the aerosols would have by themselves, creating a much stronger signal in the Earth's energy balance.

Presently, the simulation of the unpolluted cloud reflectance is improved through studies of the model simulations and actual unpolluted cloud scenes. The simulations can be improved by incorporating all the features of the spectrum as they are measured by SCIAMACHY. The calibration of SCIAMACHY reflectances is an ongoing issue, so calibration errors should have a minimal impact on the retrieved result. Once these improvements have been applied, the entire data set from SCIAMACHY (currently nine years) can be investigated to find significant changes in cloud cover through aerosol semi-direct effects.