Dynamics-aerosol-chemistry-cloud interactions in West Africa
The LMD + LATMOS contribution

Project summary
Cloud-aerosol interactions and their effects on radiation and precipitation are one of the largest uncertainties in our understanding of climate. Aerosols originate from many natural and anthropogenic sources, directly or through chemical processes in the atmosphere, creating complex feedbacks with potential socio-economic impacts. Large efforts are currently being made at many European centres to incorporate cloud-aerosol interactions into operational models and to better monitor key parameters from space. Yet, observational data to validate these are very limited.

The DACCIWA project proposes a major field campaign in southern West Africa involving three European research aircrafts, two super-sites, new space observations and numerical models operating at different resolutions and levels of complexity in order to generate a benchmark dataset that comprises the full chain of processes from surface-based emissions to climate impacts. West Africa is ideal for such a study due to its rich mix of aerosols and clouds, its rapid population and economic growth, and its strong dependence on the notoriously difficult to predict West African monsoon.

The main objectives of DACCIWA are a comprehensive assessment of the atmospheric composition of the region, impacts on human and ecosystem health, the two-way coupling between aerosols, clouds and precipitation, key meteorological processes, impacts on radiative forcing and the regional scale circulation, and socio-economic implications of changes in anthropogenic emissions, land use and climate. The DACCIWA team includes the necessary range of expertise and interdisciplinarity to tackle this challenging problem. DACCIWA will build on the scientific and logistical foundations laid by the EU FP6 programme AMMA and collaborate closely with operational centres to ensure maximum impact. The resulting improved monitoring and modelling capabilities and reduced uncertainties will be a key element in future planning and policy-making.

  • Animations of BADC obs and WRF model (May to July 2014):
  • DACCIWA informations: In french: News in Ivory Coast about a meeting in Abidjan, October 18, 2018 (from 19'35): 20h de RTI1

    UPMC (=LATMOS+LMD) team publications:

    1. Deroubaix, A., Menut, L., Flamant, C., Brito, J., Denjean, C., Dreiling, V., Fink, A., Jambert, C., Kalthoff, N., Knippertz, P., Ladkin, R., Mailler, S., Maranan, M., Pacifico, F., Piguet, B., Siour, G., and Turquety, S.: Diurnal cycle of coastal anthropogenic pollutant transport over southern West Africa during the DACCIWA campaign, Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-766, in review, 2018.
    2. Flamant, C., Deroubaix, A., Chazette, P., Brito, J., Gaetani, M., Knippertz, P., Fink, A. H., de Coetlogon, G., Menut, L., Colomb, A., Denjean, C., Meynadier, R., Rosenberg, P., Dupuy, R., Schwarzenboeck, A., and Totems, J., 2018: Aerosol distribution in the northern Gulf of Guinea: local anthropogenic sources, long-range transport and the role of coastal shallow circulations, Atmos. Chem. Phys., 18, 12363-12389, https://www.atmos-chem-phys.net/18/12363/2018/
    3. Menut, L., Flamant, C., Turquety, S., Deroubaix, A., Chazette, P., and Meynadier, R.: Impact of biomass burning on pollutant surface concentrations in megacities of the Gulf of Guinea , Atmos. Chem. Phys., 18, 2687-2707, https://doi.org/10.5194/acp-18-2687-2018, 2018. ,
    4. Deroubaix, A., Flamant, C., Menut, L., Siour, G., Mailler, S., Turquety, S., Briant, R., Khvorostyanov, D., and Crumeyrolle, S., 2017, Interactions of Atmospheric Gases and Aerosols with the Monsoon Dynamics over the Sudano-Guinean region during AMMA, Atmos. Chem. Phys., 18, 445-465, https://doi.org/10.5194/acp-18-445-2018, 2018.

    UPMC is involved in the following work-packages:
    WP Title Coordination Deliverable
    4.1 Planning, preparation and execution of the field campaign C.Flamant M24 (November 2015)
    4.2 Evaluation of existing model datasets B.Vogel M36 (November 2016)
    3.2 Model assessment and validation L.Menut M42 (May 2017)
    3.4 Evaluation of impacts of present day emissions on radiative forcing and regional public an ecosystem health in SWA M.Evans M54 (May 2018)
    4.4 Process model studies of cloud-aerosol interactions H.Coe M54 (May 2018)
    4.5 Regional modelling of clouds and aerosol interactions across the SWA J.Marsham M54 (May 2018)
    3.5 Forecasts of impacts of future emissions on atmospheric composition, radiative forcing, regional public and ecosystem health in SWA I.Bey M56 (July 2018)

    WP3 Atmospheric chemistry: (UPMC 18 PM)
    Main deliverables:

    • D3.1) Campaign: Summary document for chemistry campaign to inform case study selection and data users [month 24]
    • D3.2) Model evaluation: Report on evaluation of existing models in SWA [month 36]
    • D3.3) Campaign data: Quality controlled dataset for gas and aerosol phase compounds from field campaign delivered to data centres [month 36]
    • D3.4) Present emissions: Report on the present impact of SWA emissions on regional public and ecosystem health [month 48]
    • D3.5) Future emissions: Report on the future impact of SWA emissions on regional public and ecosystem health [month 54]
    Work package objectives:
    The objective of WP3 is to advance the scientific understanding of atmospheric chemistry in the context of a rapid economic and population growth in the SWA region. In terms of the broader science objectives listed in section B1.1.2, WP3 will focus on O1 and contribute to O2 and O9. The specific objectives of WP3 are to:
    • observe the gas and aerosol phase composition of the atmosphere in SWA and to deliver a high-quality dataset of this composition to the appropriate data centres;
    • assess the performance of models in simulating the composition of the atmosphere over SWA with a focus on ozone and aerosol;
    • assess the impact of atmospheric composition on non-urban air-quality, human and ecosystem health, radiative forcing and climate in particular that of pollutants such as ozone and aerosols;
    • estimate the implications of future (2030,2050) changes in regional anthropogenic emissions on human and ecosystem health and on climate relevant species.

    Task 3.2: Model assessment and validation (UPS, KIT, UPMC, ETHZ, UNIVYORK, ECMWF)
    Coord: UPMC, L.Menut
    Deliverable: M42 (May 2017)
    Work at UPMC: 6 PM
    Simulations of chemical composition for SWA will be conducted with LES (Meso-NH run by UPS), regional (COSMO run by KIT, CHIMERE run by UPMC and COSMO-ART/M7 run by ETZH) and global models (GEOS-Chem run by UNIVYORK, ECHAM6-HAMMOZ run by ETHZ, ECMWF operational data). The models will be forced with chemical emissions from international dataset e.g. those developed for ACCMIP and IPCC that are available through the Global Emission InitiAtive (GEIA). All WP models, whether high resolution LES, regional or global will be evaluated against the DACCIWA observational dataset, both in-situ datasets and from satellites (IASI). Differences between the models will be evaluated and attributed to different model characteristics (spatial resolution, chemical complexity, emissions characteristics). Sensitivity studies will be used to attribute model failures to processes. Notably the role of uncertainty in emissions estimates will be investigated. This will lead to the production of a report (D3.2) on the fidelity of models to reproduce the observations of atmospheric composition over SWA. Particular emphasis will be placed on the role of model spatial resolution in determining the fate emissions in the region. The venting and loss of pollutants from the boundary layer is primarily achieved by thermals and clouds in the SWA convective boundary layer, particularly in the region of the sea-breeze convergence. These processes are not explicitly resolved in global models. A careful comparison of the models at different scales will be conducted to evaluate the impact of sub-grid scale mixing of precursors on the ozone production efficiency, aerosol production and chemical reactivity. Particular attention will be paid to the comparison between LES simulations for which the thermals and clouds are resolved processes and global models (UNIVYORK, UPS, ETHZ) where they are highly parameterized. By focusing on these processes at different resolutions, the upscaling between LES models running at resolutions of 100s of meters (UPS), to regional models run at resolutions of kilometres (KIT, ETHZ), to global models run a tens to hundreds of kilometres (UNIVYORK, ETHZ) (Milestone 9).

    Task 3.4: Evaluation of impacts of present day emissions on radiative forcing and regional public an ecosystem health in SWA (UPMC, KIT, ETHZ, UNIVYORK, ECMWF)
    Coord: Univ York, Evans
    Deliverable: M54 (May 2018)
    Work at UPMC: 6 PM
    The impact of pollutants on regional radiative forcing and on present day health and agriculture will be achieved from the simulations undertaken in Task 3.2 at regional and global scales (UPMC, KIT, ETHZ, UNIVYORK, ECMWF). These simulations will be used to derive deposition maps for ozone and acids (ecosystem and agricultural impacts), air pollution exposure (human health) and the impacts inferred. For the current atmosphere, the exposure will be studied under different meteorological and emissions conditions characteristic of the studied region and sampled during the campaign. This task will benefit from the perturbation experiments undertaken in Task 3.2, which will define the sensitivity of exposure to changes in emissions and hence provide insights into the source-receptor relationships at play in the area. Updated emission inventories from WP2 will be used. The scientific knowledge gained in this WP will be described in a report (D3.4, Milestone 9) and integrated into a policy brief delivered by WP8.

    Task 3.5 Forecasts of impacts of future emissions on atmospheric composition, radiative forcing, regional public and ecosystem health in SWA (UPMC, KIT, ETHZ, UNIVYORK)
    Coord: ETHZ, I.Bey
    Deliverable: M56 (July 2018)
    Work at UPMC: 6 PM
    Using the suite of models available in the project, forecasts will be made using future emissions datasets published by the international GEIA-ACCENT effort (ACCMIP, MACCity, RCPs, Assamoi-Liousse, etc.) and from WP2. These simulations will then be processed in the same way as in Task 3.4 to provide information on the future regional radiative forcing, and regional public and ecosystem health. The scientific knowledge gained in this WP will be described in a report (D3.5, Milestone 9) and integrated into a policy briefing delivered by WP8.

    WP4 Cloud-aerosols interactions: (UPMC 18 PM)
    Main deliverables:

    • D4.1) Campaign: Report on evaluation of existing model products [month 24]
    • D4.2) Model evaluation: Initial summary report of IOPs cloud and aerosol measurements for use by data analysis and interpretation teams [month 36]
    • D4.3) Aerosol and cloud data: Quality controlled dataset for aerosol and cloud from field campaign [month 36]
    • D4.4) Aerosol influences on cloud properties: Report on the quantitative analysis of key influences on cloud properties, focusing on aerosol interactions [month 48]
    • D4.5) Model performance: Report on the performance of the coupled cloud-aerosol model output, tested against observations [month 54]
    Work package objectives:
    The main objective of WP4 is to quantify cloud microphysical and dynamical behaviour across the SWA, with particular emphasis on the influence the variety of aerosol sources across the region have on stratus, cumulus and congestus clouds and their transition. In terms of the broader science objectives listed in section B1.1.2, WP4 will focus on O3 and contribute to O4-O7. The specific objectives of WP4 are to:
    • quantify the two-way coupling between aerosols and cloud and raindrops, focusing on the distribution and characteristics of CCN, their impact on cloud characteristics and the removal of aerosol by precipitation;
    • determine stratus and cumulus cloud micro- and macro-physical properties across the SWA;
    • assess the skill of regional models in predicting regional cloud behaviour;
    • improve the predictive capability for cloud-aerosol interactions across the region.
    The main purpose of WP4 is to provide a comprehensive understanding of the development and maintenance of stratus and cumulus clouds across the SWA and the extent to which their properties are influenced by the wide range of aerosols typical of the region. The region is data sparse and as a result much of the improved knowledge in this WP will be built on the analysis and interpretation of the data collected during the observational campaign in June 2015 and the multi-scale modelling effort from LES to global models. The work will be split in four tasks.

    Task 4.1: Planning, preparation and execution of the field campaign (M7-24; UNIVMAN, UBP, UPMC, DLR, UNIVLEEDS)
    Coord: UPMC, C.Flamant
    Work at UPMC: 5 PM
    Prior to the intensive campaign, the measurement teams (mainly UNIVMAN, UBP, UPMC, DLR, UNIVLEEDS) will be focussed on planning the measurement strategy and the use of the multiple airborne and ground based platforms. The detailed experimental strategy will be finalised in this period, including the completion of the flight plans and the detailed coordination between airborne platforms and ground based observation facilities. This will be achieved in conjuncture with WP3. Detailed aircraft measurements of the concentration, size distribution, composition and characteristics of CCN as well as number and size distributions of cloud droplets and rain drops will be made and this task will ensure that the instrumentation is properly prepared, calibrated and compared in advance of the IOP. A detailed timetable for this work will be developed between the partners. Ground based sondes and UAVs will enable an assessment of the cloud depth and top height of the low-level nocturnal stratus, whilst aircraft will be able to sample the later stages of the stratus as cloud top grows and cloud depths increase through the morning and will be used to probe the transition to cumulus and congestus clouds in the afternoon. In this way the full diurnal cloud cycle can be observed. Mapping of the cloud deck will be carried out in the sub-cloud layer to determine the aerosol characteristics, within the cloud at multiple levels and above the cloud layer. Vertical profiles will be used to derive the thermodynamics structure and the cloud microphysics. Repeated flights over the same flight tracks on different days will allow statistics to be constructed against which models can be tested.

    Task 4.2: Evaluation of existing model datasets (M21-36; UNIVLEEDS, KIT, MO, ETHZ, ECMWF, UPMC)
    Coord: KIT, B.Vogel
    Deliverable: M36 (November 2016)
    Work at UPMC: 4 PM
    An assessment of pre-existing model datasets will be conducted jointly by UNIVLEEDS, KIT, MO, ETHZ, ECMWF and UPMC to determine aerosol-cloud properties across the region and to explore model diversity and the main systematic biases. The regional extent, depth and cloud top height of the nocturnal stratus cloud will be analysed and statistics will be provided to guide the analysis of the field data and for further modelling work. Models will be used to probe the diurnal transition from full stratus cover, through break up and the development of fair weather cumulus during the afternoon and will include the development of congestus clouds. The onset of these transitions and their frequency across the region will be determined. Differences in CCN behaviour between the different models will be analysed. Results will subsequently be used in Tasks 4.4. and 4.5 where analysis of the pre-existing data will be compared with the new model runs to assess improvements in model performance using the latest model versions of the UM, COSMO-ART/M7, Chimere and ECHAM6-HAMMOZ. Models will be used to probe the diurnal transition from full stratus cover, through break up and the development of fair weather cumulus during the afternoon and will include the development of congestus clouds. The onset of these transitions and their frequency across the region will be determined. Differences in CCN behaviour between the different models will be analysed. Results will subsequently be used in Tasks 4.4. and 4.5 where analysis of the pre-existing data will be compared with the new model runs to assess improvements in model performance using the latest model versions of the UM, COSMO-ART/M7, Chimere and ECHAM6-HAMMOZ.

    Task 4.4: Process model studies of cloud-aerosol interactions (UNIVMAN, UBP, UPMC, DLR, UNIVLEEDS, MO, KIT)
    Coord: UNIVMAN, Coe.
    Deliverable: M54 (May 2018)
    Work at UPMC: 5 PM
    The effect of natural and anthropogenic aerosol on the formation and the persistence of low-level stratus clouds developing during nighttime, their transition through break up and formation of cumulus clouds to congestus formation will be investigated in this task. Convection-permitting UM (UKCA) simulations with two-way cloud-aerosol interactions will be evaluated against observations on a case-by-case and statistical basis by UNIVLEEDS and MO. Case studies using bulk microphysics at LES to climate grid-spacings will be compared with bin-LES simulations (UNIVLEEDS, UNIVMAN and MO). KIT will use COSMO-ART/M7 and UPMC will use Chimere in a series of detailed sensitivity simulations with different spatial resolutions and different treatments of the key processes, as for example aerosol dynamics. These simulations will allow quantification of the key processes and an evaluation how both resolution and the bulk multi-moment representations of aerosols and clouds in the UM and COSMO-ART/M7 impact the representation of processes and control simulated aerosol impacts. UNIVMAN, UBP and DLR will investigate detailed aerosol-CCN-cloud processes through analysis of the aircraft data and by comparison with box and LES models containing extended chemical schemes. If congestus clouds are sampled during the IOP that later turn into cumulonimbus, interactions between aerosols and cloud microphysics including the ice phase and structure will be investigated using cloud resolving models and the UM (UNIVMAN and UNIVLEEDS).

    Task 4.5: Regional modelling of clouds and aerosol interactions across the SWA (UNIVLEEDS, KIT, MO, ETHZ, ECMWF, UPMC, UNIVMAN)
    Coord: UNIVLEEDS, Marsham
    Deliverable: M54 (May 2018)
    Work at UPMC: 4 PM
    This task focuses on an assessment of the regional aerosol budget and its impact on clouds and precipitation and will be conducted jointly by UNIVLEEDS, MO, KIT, ETHZ, ECMWF and UPMC. The measurement database developed in Task 4.3 will be used to challenge the model runs (UNIVMAN) and the process understanding in Task 4.4 will be used to aid diagnosis of the performance of the models in capturing regional aerosol and cloud properties. The models will be set key challenges in terms of aerosol production, transport and daily cycle of different types of clouds. Model simulations will be used to quantify the impact of various, in particular anthropogenic, emissions on clouds during the IOP and hence on radiation and the monsoon itself (UNIVLEEDS). Using model simulations performed at various resolutions but with consistent aerosol scheme, results in terms of aerosol-cloud interactions will be upscaled from the regional to the global scale (with support of ECMWF). The contribution of deep convective clouds and the role of aerosols on influencing these may also be investigated in this Task. Tasks 4.4 and 4.5 will inform the sensitivity studies within WP7, which considers longer time and greater spatial scales and addresses the integral effects of the whole monsoon system.