Smoke, science, and sharks

By Ross Herbert

In the August of 2017 the Cloud-Aerosol-Radiation Interactions and Forcing – Year 2017 (CLARIFY) measurement campaign took place on a tiny island in the middle of the southeast Atlantic Ocean where we were surrounded by whales, sharks, and most importantly, stratocumulus clouds. During August and September dense layers of strongly absorbing smoke from biomass burning in central Africa were transported over these semi-permanent clouds. These events provided us with a unique opportunity to understand the interactions between the cloud, smoke layers, and radiation; processes which remain key uncertainties within our understanding of global aerosol radiative forcing, cloud feedbacks, and, ultimately, climate change. In this blog I will discuss my experience of participating in the measurement campaign and also outline the high-resolution cloud modelling work that I am currently doing.

The island (with Bear Grylls)

The Ascension Island is situated at 8°S 14°W in the middle of the Atlantic Ocean, 1600km from the coast of Africa and 2250km from Brazil. The island, with a diameter ~ 10km, is a volcanic island natively populated by nesting turtles, colourful land crabs, and (once) huge bird colonies, and non-natively by rats (that appeared with the boats), cats (that were let loose on the island to control the rats but decided to eat birds instead and are now outlawed), and donkeys (that are tolerated). In the mid-nineteenth century, following encouragement from Charles Darwin, the royal navy began bold plans to increase precipitation on the island by ‘greening’ the upland reaches of the largest mountain with anything that would grow. The result: bamboo, bananas, wild ginger, and guava, and the birth of the aptly named ‘Green Mountain’; however, most of the island remains volcanic and devoid of vegetation.

Figure 1:On top of Sister’s peak looking towards Green Mountain

The measurement campaign

My role on the month-long campaign was a mission scientist. Along with several others we worked as a team alongside the pilots to plan the sorties and then join the crew onboard the FAAM BAe-146 research aircraft. For the sorties we would use forecasts and satellite observations to plan flights depending on what our objectives were; this might include incoming plumes of smoke, co-ordinated satellite overpasses, inter-comparison flights, and interesting cloud or convective features.

Figure 2: The FAAM BAe-146 (right) and NASA P3 (left) on Wideawake Airfield

The pilots required detailed flight plans of bearings, altitudes, distances, and times, so the planning would often take several hours. On flight days we would be up at 6am to check the most recent observations and make any last-minute changes to the sortie plan before heading to the airfield. During flights, the mission scientists used real-time data and the expertise of the instrument operators to fine-tune the sortie so that we were focusing on the correct feature. Smoke layers and stratocumulus-to-cumulus transitions are very poorly forecast so we had to make rapid decisions based on what we observed during flight. This information would be relayed to the mission scientist who sat in the cockpit and would ultimately make the final decision before informing the pilots.

This slideshow requires JavaScript.

The campaign was a huge success; we flew 28 sorties totalling 99 hours and collected data in every cloud-aerosol-radiation regime we hoped for and more. As well as the standard atmospheric state measurements (temperature, pressure, relative humidity etc..) we made detailed measurements of the clouds, the particles, and the gases within the atmosphere and smoke layers. This included number, size, and composition, and also radiative properties such as ability to scatter and absorb radiation. The measurements will help towards improving our understanding of how the smoke layers are formed, transported, evolve, and also how they affect radiative fluxes and interact with the stratocumulus clouds. We can also use the measurements to improve our representation of smoke layers and related processes in the computer models that help us predict the climate and forecast the weather. FYI the aircraft data will be available on the Centre for Environmental Data Analysis (CEDA)archive (ask if you want more specifics).

End of campaign. Time to go back to the office

So enough of the campaign, what am I actually doing on the project? I am working with Nicolas Bellouin and Ellie Highwood to investigate what happens to the cloud radiative properties (i.e., how much energy from the sun and the earth’s surface passes through and out of the cloud) when the layer of smoke is elevated above the cloud; this is a scenario that we encountered numerous times during the campaign and that we observe in satellite observations. Previous studies suggest that the smoke may act to thicken the clouds, which makes them brighter and more reflective to incoming solar radiation (sunlight), thus acting to cool the climate. The stratocumulus clouds and smoke layers cover vast areas of the ocean, therefore small changes to the radiative properties of the cloud may have important impacts. My work will help us understand this effect in more detail and understand how sensitive it is to the properties of the smoke layer and cloud.

Smoke above cloud + Sunlight = Heat = Cloud response = Semi-direct radiative effect = Cooling? or Warming?

In my work I am using the MetOffice Large Eddy Model (the LEM) to simulate the evolution of a stratocumulus cloud deck with an elevated layer of absorbing smoke. The LEM has very high spatial resolution (imagine dividing the atmosphere into boxes – high resolution means lots and lots of little boxes) that allows us to simulate the main turbulent motions that drive the movement of energy, winds, and moisture in the atmosphere (typical climate-scale models have spatial resolutions ~ 200 times greater). It is also coupled to a radiation scheme that allows us to represent upward and downward fluxes of radiation through the atmosphere, smoke layer, and cloud. From this we can represent the additional heating that is caused by the smoke and any subsequent changes to the cloud field and radiative properties; we can then finally determine whether the smoke is acting to cool or warm the climate. This specific radiative effect is commonly referred to as the semi-direct effect and in the context of the IPCC AR5 report can be seen as a rapid adjustment to the instantaneous radiative effect of aerosols.

These simulations are giving us very interesting results that are currently being written up for publication (sneak preview: it’s not as simple as we previously thought). Watch this space!

A longer version of this blog can be found at

This entry was posted in Africa, Atlantic, Atmospheric chemistry, Atmospheric circulation, Atmospheric optics, Climate, Climate change, Climate modelling, Clouds, Data collection, earth observation, Energy budget, Environmental hazards, Greenhouse gases, Measurements and instrumentation, Microphysics, Numerical modelling, Solar radiation, Weather forecasting, Wind. Bookmark the permalink.