Clouds/Aerosols 1

Within the ESA Climate Change Initiative (CCI) project Aerosol_cci (Phase 1: 2010 –2014; Phase 2: 2014-2017) intensive work has been conducted to improve algorithms for the retrieval  of aerosol information from European sensors ATSR-2 (ERS-2), AATSR (3 algorithms), MERIS (3 algorithms), synergetic AATSR/SCIAMACHY, GOMOS (all on ENVISAT), PARASOL and OMI (EOS-Aura) (both part of NASA’s A-Train). Whereas OMI and GOMOS were used to derive absorbing aerosol index and stratospheric extinction profiles, respectively, Aerosol Optical Depth (AOD) and Ångström coefficient were retrieved from the other sensors. The cooperation between the project partners, including both retrieval teams and independent validation teams, has resulted in a strong improvement of most algorithms. In particular the AATSR retrieved AOD is qualitatively similar to that from MODIS, usually taken as the standard, MISR and SeaWiFS. This conclusion has been reached by several different ways of validation of the L2 and L3 products. Using AERONET sun photometer data as the common ground-truth both ‘traditional’ statistical techniques and a ‘scoring’ technique based on spatial and temporal correlations were applied. Quantitatively, the limited AATSR swath width of 512km results in a smaller amount of data. Nevertheless, the assimilation of AATSR-retrieved AOD, together with MODIS data, contributes to improving ECMWF / MACC climate model results. In addition to the multi-spectral AOD, and thus the Ångström Exponent, also a per-pixel uncertainty is provided and validated. By the end of Aerosol_cci Phase 1 the ATSR algorithms have been applied to both ATSR-2 and AATSR resulting in an AOD time series of 17 years dating back to 1995.

In phase 2 this work is continued with a focus on the further improvement of the ATSR algorithms as well as those for the other instruments and algorithms, mentioned above, which in phase 1 were considered less mature. The first efforts are on the further characterization of the uncertainties and on better understanding of the cloud screening in the various algorithms. Other efforts will focus on surface treatment and possible improvement of aerosol models used in the retrieval. A yearly re-processing of the full 17-year global ATSR-2/AATSR data set is planned to evaluate the effect of different changes and to monitor further improvement.

As a new additional focus in phase 2 it is envisaged to produce a full-mission dataset of dust AOD from IASI with four different algorithms, which are based on different retrieval techniques. A major task within the project is the first inter-comparison of those IASI dust retrieval algorithms on the basis of a large set of observations. For this purpose one year of IASI observations (2013) over the major dust belt of the Northern hemisphere, including the Northern Atlantic Ocean, the Sahara desert, the Arabian Peninsula as well as the Central Asian desert regions, is consistently processed with all four algorithms and similar retrieval output (visible and infrared AOD, AOD uncertainty, retrieval quality, cloud flags) is generated in order to facilitate the comparison of results.

The retrieval inter-comparison, called Round Robin exercise, consists of an analysis of the different sensitivities of the four algorithms to dust and environmental conditions.  The retrieval methods are based on different retrieval strategies such as look-up tables, optimal estimation and singular value decomposition. The sensitivity analysis will reveal the major uncertainties of infrared dust remote sensing from space as well as specific strengths and weaknesses of the different retrieval approaches under varying environmental conditions and can be used to select the best-suited approach for specific conditions.

The Round Robin exercise includes the evaluation of retrieval results from the four different algorithms with external data. AERONET sun photometers are used for evaluation as well as observations from the German SALTRACE campaign over the tropical Atlantic Ocean in summer 2013. Evaluation of subsets, for example subdivided by atmospheric moisture or surface characteristics, will thus allow for an improved understanding of the feasibility of hyperspectral infrared dust remote sensing with different approaches under varying conditions.

The presentation will summarize the concept and status of the Aerosol_cci project in both phases and discuss in particular the achievements regarding the 17 year ATSR time series and the 1 year IASI round robin exercise.

AerGom is a retrieval algorithm recently developed for the GOMOS instrument onboard Envisat as an alternative to the GOPR official algorithm. A main objective of the AerGom project was the need to improve the stratospheric aerosol extinction product for which GOPR is known to present quite poor performances, especially out of the reference wavelength of 500 nm. However, AerGom retrieves extinction profiles for particulate matter and gaseous species concentrations as well.

The AerGom algorithm has been chosen and confirmed (during the Phase II of the project) as main algorithm for the development of stratospheric aerosol data records in the framework of Aerosol_CCI (Climate Change Initiative), what reinforced the interest to further improve the algorithm.

This presentation shows the current issues identified for AerGom and the latest developments and improvements brought to this algorithm. These aspects and the current performances will be illustrated using the most recents datasets.

Stratospheric aerosol is known for its impact on climate. Recent research found that especially the volcanic aerosol in the lower stratosphere at mid and high latitudes has been underestimated regarding its impact on radiative forcing due to a lack of measurements in this region (Ridley, 2014). The stratospheric sulfate aerosol is significantly influenced by volcanic eruptions. In the last decade the stratospheric aerosol burden has increased likely due to several smaller volcanic eruptions. However, the pathways of sulfur from these eruptions into the stratosphere are uncertain.

Polar orbiting infrared limb sounders, such as Envisat MIPAS, are particularly suited for measuring and tracing aerosol in the upper troposphere and stratosphere. They measure vertical profiles and provide global coverage at day- and nighttime during all seasons. Also, due to the limb geometry they are highly sensitive towards aerosol and clouds. For the discrimination between ice clouds and aerosol (especially in the troposphere) there are several methods available for IR nadir instruments. Here, we present a new method for IR limb measurements to detect aerosol and to discriminate it from ice clouds. The new method was tested for MIPAS measurements and we confirmed the MIPAS aerosol detections by comparing with other instruments. We also analysed the MIPAS aerosol detection sensitivity and the accuracy of the aerosol altitudes.

We applied the new aerosol detection method to the 10 years of Envisat MIPAS measurements. The time series of the MIPAS aerosol measurements is dominated by volcanic sulfate aerosol. However, also volcanic ash (Griessbach, 2014), mineral dust, bush fires and non-ice polar stratospheric clouds can be identified in the MIPAS aerosol data. The high sensitivity and the global coverage of the MIPAS measurements allows us to trace single volcanic eruptions in the horizontal as well as in the vertical on a daily basis for several months. The MIPAS aerosol data clearly shows how smaller volcanic eruptions at high latitudes (e.g. the Sarychev eruption in 2009) contribute to the stratospheric sulfate aerosol layer in the tropics.

Griessbach, S., Hoffmann, L., Spang, R., and Riese, M.: Volcanic ash detection with infrared limb sounding: MIPAS observations and radiative transfer simulations, Atmos. Meas. Tech., 7, 1487–1507, doi:10.5194/amt-7-1487-2014, 2014.

Ridley, D. A., Solomon, S., Barnes, J. E., Burlakov, V. D., Deshler, T., Dolgii, S. I., Herber, A. B., Nagai, T., Neely, III, R. R., Nevzorov, A. V., Ritter, C., Sakai, T., Santer, B. D., Sato, M., Schmidt, A., Uchino, O., and Vernier, J. P.: Total volcanic stratospheric aerosol optical depths and implications for global climate change, Geophys. Res. Lett., 41, 7763–7769, doi:10.1002/2014GL061541, 2014.

GOMOS (Global Ozone Monitoring by Occultation of Stars) is a stellar occultation
instrument, combining four spectrometers and two fast photometers, that flew on board the
European platform ENVISAT from 2002 to 2012. Polar mesospheric clouds (PMCs), that form
during summer in the polar upper mesosphere, could be detected using the photometers' signals.
Their main properties (occurrence frequency, peak altitude, radiance) have been retrieved from
2002 to 2010, leading to a 16-summer (in both hemispheres) database of more than 21000 clouds.

PMCs are very sensitive to changes in their environment. That makes them important
tracers for the complex mechanisms that control the summer mesopause region. A better
understanding of the microphysical processes going on in this atmospheric region is essential in order to
model of their growth, their transport mechanisms and their lifetime. To that purpose, the particle
size distribution is an important parameter.

This presentation will be focused on PMC particle sizes retrieved from GOMOS spectral
observations in the northern hemisphere. The retrieval method will be explained, and results based
on the obtained 8-year dataset will be described and compared to PMC particle sizes derived from
the measurements of other instruments.

Smoke aerosols affect clouds both through aerosol-cloud interactions and aerosol-radiation interactions, the latter formerly known as the semi-direct effect. By absorption of sunlight, aerosols can heat the atmosphere locally, and reduce the amount of radiation received at the surface and the top-of-atmosphere (TOA). This changes the atmospheric lapse rate and column stability, influencing convection and cloud forming processes. Aerosol-cloud interactions are notoriously difficult to isolate in remote sensing measurements, as cloud doplets and aerosols have to be discriminated in mixed layers. Aerosol-radiation interaction measurements, on the other hand, are as challenging as aerosol-cloud interactions, even though the cloud and aerosol layers can be physically separated. From space, measurements of aerosols in cloud scenes have been very difficult, since cloud screening is often applied before aerosol parameters wae retrieved. Some very promising new techniques have been developed recently to distinguish aerosols from clouds using active measurements, polarised measurements, or hyperspectral measurements. Using shortwave hyperspectral measurements, the absorption by aerosols, which is large in the ultraviolet (UV), can be distinguished from the scattering by cloud droplets, retrieved in the shortwave infrared (SWIR). 

Interestingly, this can be used directly to quantify the aerosol direct effect (DRE) of small smoke particles that drift over clouds, without retrieving aerosol parameters first. By simulating the aerosol-free cloud scene, and comparing with the measured aerosol-cloud scene, the DRE can be determined with very high accuracy. Application of this method to SCIAMACHY measurements over the southeast Atlantic Ocean during the African monsoon dry season (June-Sept.) in 2006-2009, revealed aerosol DRE over clouds up to 128 +/- 8 Wm-2 instantly and a monthly average of about 35 Wm-2 [De Graaf et al, 2012]. This is much larger than previously estimated with climate models, that simulate a monthly mean aerosol DRE up to about 6 Wm-2. A first attempt to reconcile simulations with observations failed to identify the processes in the models that are responsible for the underestimation of the DRE, but cloud brightness is most likely the most critical parameter [De Graaf et al, 2014]. This is now under investigation. 

Here, we will present aerosol DRE over clouds from combined OMI and MODIS hyperspectral measurements. Since SCIAMACHY was lost in 2012, new measurements from OMI and MODIS can help to continue the observation of aerosol absorption over clouds from space. Each instrument by itself does not provide enough information on both aerosols and clouds, but OMI gives detailed information of UV aerosol absorption, while MODIS broadband channels provide cloud information from the SWIR range of the spectrum. OMI and MODIS are flying in formation in the A-Train constellation, providing observations about 7 minutes after one another. This creates uncertainties in the observed scene, especially in scenes where convection is strong and cloud parameters change rapidly. However, OMI and MODIS overlap at MODIS shortest wavelength band, 469 nm, which can be used to test the matching of the spectra. Furthermore, MODIS provides cloud products at 1x1 km resolution, and better, which can be used to test and improve the cloud retrieval algorithm that was developed for the much larger SCIAMACHY and OMI pixels. 

Application of this unique hyperspectral method to OMI and MODIS is used to prepare for TROPOMI, which will provide information on both the UV and the SWIR. If successful, TROPOMI will provide aerosol DRE over clouds from one instrument with an unprecedented accuracy and unprecedented spatial resolution. 

We will introduce the newly developed algorithm for the combined OMI-MODIS measurements, and present measurements of aerosol DRE over the southeast Atlantic Ocean. 

De Graaf, M., L.G. Tilstra, P. Wang and P. Stammes, Retrieval of the aerosol direct radiative effect over clouds from space-borne spectrometry, J. Geophys. Res. 117, D07207, doi: 10.1029/2011JD017160, 2012;

De Graaf M., N. Bellouin, L.G. Tilstra, J. Haywood, P. Stammes, Aerosol direct radiative effect of smoke over clouds over the southeast Atlantic Ocean from 2006 to 2009, Geophys. Res. Lett., 2014, doi: 10.1002/2014GL061103

The ATSR Dual View (ADV) and ATSR Single View (ASV) aerosol retrieval algorithms have been developed for use with the Along Track Scanning Radiometer ATSR-2 which flew on the ERS-2 satellite (1995-2003) and the Advanced ATSR (AATSR) which flew on the ENVISAT satellite (2002-2012) (both from the European Space Agency, ESA). The ATSR instruments provide two views: one near-nadir and the other at 55 degrees forward. Each view provides the radiances at the top of the atmosphere (TOA) in 7 wavebands from the visible (VIS) to the thermal infrared (TIR). The TIR wavebands are mainly used for cloud detection, in combination with a number of tests including shorter wavelengths. The VIS and NIR (near-infrared) wavebands are used for aerosol retrieval. ADV is applied over land where the ratio of both views in the 1.6 µm band are used to account for effects of the land surface reflectance on the radiation at the top of the atmosphere (TOA), assuming that the effect of aerosols on the TOA reflectance is small. The dual view algorithm provides information on the aerosol properties over land with a quality which is similar to that of other satellites which are commonly used as the ‘standard’ for aerosol retrieval (in particular MODIS, MISR). However, ATSR has a smaller swath resulting in less coverage than MODIS. Over ocean only one of the views is used (forward) and also here the quality of the data is competitive with that of other instruments. The advantage of the ATSR-2/AATSR combination is that it provides a time series of 17 years, longer than any other of the currently available quality products (MODIS, MISR, SeaWiFS). This time series will be further expanded with those from the Sea and Land Surface Temperature Radiometer (SLSTR) on the ESA/EU GMES Sentinel-3 mission which is planned to be launched in the autumn of 2015. The AOD time series is one of the Aerosol ECVs.

The Dual View algorithm provides aerosol data on a global scale with a default resolution of 10x10km² (L2) and an aggregate product on 1^o x1^o (L3). Optional, a 1x1 km² retrieval products is available over smaller areas for specific studies. Since for the retrieval of AOD no prior knowledge is needed on surface properties, the surface reflectance can be independently retrieved using the AOD for atmospheric correction.

For the retrieval of cloud properties, the SACURA algorithm has been implemented in the ADV/ASV aerosol retrieval suite. Cloud properties retrieved from AATSR data are cloud fraction, cloud optical thickness, cloud top height, cloud droplet effective radius, liquid water path.

In the presentation an overview will be presented of the aerosol and cloud remote sensing activities and applications at FMI and UHEL. The application of ADV/ASV to produce and AOD ECV for 17 years, with applications over different locations across the world, shows the different temporal variations of the AOD between 1995 and 2012. Spatial variations will be highlighted. The simultaneous retrieval of aerosol and cloud properties provides information on aerosol-cloud interaction.

ACKNOWLEDGEMENTS

Work presented in this contribution is supported by the ESA-ESRIN projects Aerosol_cci Pahses 1&2 (project AO/1-6207/09/I-LG), STSE-ALANIS Atmosphere-Land-Interaction Study, Theme 3 Aerosols (Contract no 4200023053/10/I-LG), and Globemission 1&2 (ESA-ESRIN Data Users Element (DUE), project AO/1-6721/11/I-NB), the EU projects PEGASOS (EU FP7 ENV.2010.1.1.2-1), Marco Polo (EU FP7 Grant agreement no  606953), BACCHUS (EU-FP7 Grant agreement no  603445), MACC II, EU-FP7 Grant agreement no: 283576) and MACC III (EU-H2020 Grant Agreement-633080-MACC-III ) and the Nessling Foundation. This work was supported by the Centre of Excellence in Atmospheric Science funded by the Finnish Academy of Sciences Excellence (project no.  272041), the CRAICC (Cryosphere-atmosphere interactions in a changing Arctic climate), which is part of the Top-level Research Initiative (TRI) of the joint Nordic research and innovation initiative.