Most of the latest models from a popular climate modeling platform underestimate solar absorption by biomass burning aerosols emitted from fires in Central and Southern Africa and carried by smoke plumes over the southeast Atlantic Ocean, according to a new study.
Every year during peak dry season, between June and September, fires spread across the African savanna, including fires used to clear agricultural waste from fields. The burning vegetation emits aerosols composed of organic matter, black carbon, and a mix of inorganic species, which then travel in smoke plumes over the southeast Atlantic. Unlike other particles that tend to dominate aerosol concentrations over the ocean, like sulfate and sea salt, these biomass-burning aerosols strongly absorb radiation from the sun.
The findings in the Oct. 8 issue of Science Advances explain why about 75% of climate models from the sixth phase of the Coupled Model Intercomparison Project (CMIP6) that predict general circulation in the planet's atmosphere do not fully capture the intense warming that is occurring at the top of the atmosphere over tropical Africa.
"We had our suspicions," said Marc Mallet, an atmospheric scientist at the University of Toulouse in France and the first author of the study. "What we really didn't know was how well the CMIP6 models represent the aerosol optical properties and especially the strong solar-absorption that have been highlighted recently by state-of-the-art on-site measurements obtained at the surface and measurements made by heavily instrumented aircraft."
"We have used satellite and surface remote-sensing observations to show that some models correctly represent these optical properties while others underestimate the absorption," he added.
A number of recent international investigations, including the French-led AEROCLO-SA project, the U.S.-led ORACLES and LASIC projects and the U.K.-led CLARIFY-2017 project have demonstrated that these aerosols create a warming effect at the top of the atmosphere over the southeast Atlantic. Some recent research has suggested that the solar absorption of biomass burning aerosols appears to be higher than previously thought, raising the question of whether the climate effects of these aerosols are adequately represented in the latest CMIP6 configurations.
"Constraining aerosol absorption in models is a difficult task and there has been considerable debate on how well the current global climate models represent absorption properties and their role in radiative forcing and climate warming," said Mallet. "Moreover, few studies have looked specifically at how well global climate models represent the combination of low-level clouds and smoke aerosols in this specific region."
To investigate how well (or how poorly) CMIP6 models simulate low clouds, aerosols from burning vegetation in Central and Southern Africa, and the solar absorption of these aerosols over the southeast Atlantic, Mallet and colleagues compared model predictions for aerosol reflection of solar radiation for July, August, and September of 2003 to 2014 against measurements obtained from the PARASOL satellite. They also compared model predictions with observations retrieved from AERONET (an international ground-based aerosol remote sensing network) at three research stations, as well as the 2017 version of the Max Planck Institute Aerosol Climatology dataset and recent aircraft observations.
Unlike the results of a similar assessment based on comparing eight global models with aircraft observations, Mallet and colleagues found that the CMIP6 models do not fully capture the strong solar absorption of biomass-burning aerosols over the southeast Atlantic.
"What we show is that biomass burning aerosols lead to a positive radiative forcing (warming) over this particular region while many of the CMIP6 models exert a negative radiative forcing and hence a cooling," said Mallet. "This radiative forcing is accompanied by a heating in the biomass burning layer that can impact low-level clouds but also atmospheric dynamics and precipitation over tropical Africa."
The scientists point to several possible reasons why the CMIP6 models may underestimate the solar absorption of these aerosols. The models may underestimate the concentration of the smoke itself, perhaps due to challenges with satellites detecting small fires, inadequately represent the chemical composition of the aerosols, fail to account for the correct transport altitude of the smoke plumes, or neglect to account for how the particles' optical properties change as the smoke aerosol ages.
Mallet noted that it is still difficult to determine in which other parts of the world CMIP6 models might not fully account for biomass burning aerosols, or to know whether the models generally underestimate aerosol absorption at a global scale.
"We do find this kind of situation in other areas such as off the coast of California where intense fires in summer transport smoke over the Pacific Ocean," said Mallet. "It would be really interesting, for example, to study how the CMIP6 model behaves over this region, too."