- 1Met Office, Atmospheric Processes and Parameterizations, United Kingdom of Great Britain – England, Scotland, Wales (joseph.cartonkelly@metoffice.gov.uk)
- 2Met Office, Atmospheric Processes and Parameterizations, United Kingdom of Great Britain – England, Scotland, Wales (anthony.jones@metoffice.gov.uk)
- 3Met Office, Atmospheric Processes and Parameterizations, United Kingdom of Great Britain – England, Scotland, Wales (paul.field@metoffice.gov.uk)
We have developed a double moment aerosol model appropriate for numerical weather prediction (NWP) within the Met Office’s Unified Model framework. Our aerosol model (Sol-Insol) is fully coupled to a double moment cloud microphysics model (CASIM) permitting aerosol-cloud interactions to be resolved at high-resolution. As a test we investigate how our setup performs against traditional ACI parametrizations used for NWP by comparing model results to observations from the Café-Brazil field campaign over the Amazon. Our simulations contribute to the CleanCloud model intercomparison project (MIP) aimed at improving aerosol-cloud interactions across timescales. As a case study, we use 14th January 2023, a date selected due to HALO aircraft observations being recorded alongside a pronounced mesoscale squall line.
The MIP protocol sets out 3 regional domains surrounding the ATTO site - a 6.6km resolution grid over the Northern part of South America into Central America, a 3.3km resolution grid inside this and then a 1.6km resolution grid covering the Amazon. Three model configurations were selected to isolate the impacts of aerosol on cloud and precipitation: a control run with no aerosol scheme, fixed cloud CDNC and a Cooper ICNC vs T relation (CONTROL); a coupled aerosol-cloud run with arbitrary initialised aerosols (ARB-AER); and a coupled aerosol-cloud run with aerosol initialised to CAMS reanalysis data (CAM-AER).
We constrain the simulated aerosol, cloud and radiation properties using a range of co-located satellite, surface (ATTO) and air (HALO) observations. Our results show that initialising the aerosol to CAMS reanalysis concentrations leads to slightly improved results relative to CONTROL while ARB-AER performs significantly worse when looking at observed top of the atmosphere radiation, highlighting the high sensitivity of cloud to ambient aerosol. Further experiments comparing ground and flight observations to model metrics help to support the conclusion that CAM-AER is performing better than ARB-AER and highlights potential improvements to NWP hydro-forecasts from using realistic aerosol properties.
How to cite: Carton-Kelly, J., Jones, A., and Field, P.: Investigating a double moment, fully coupled aerosol-cloud model over the Amazon, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3245, https://doi.org/10.5194/egusphere-egu25-3245, 2025.
