The anthropogenic contribution to cloud condensation nuclei is known to be large due to pollutant emissions. However, the natural processes that regulate cloud condensation nuclei over many regions are less well understood. The CLAW hypothesis (Charlson et al., 1987) has provided the intriguing prospect of oceanic and atmospheric systems exhibiting in a coupled way to react to the changing climate in a manner that opposes the global warming. It was proposed that the rapid oxidation of dimethylsulfide (DMS) in the atmosphere leads to the formation of a non-sea-salt aerosol (NSS-SO₄^2-), which, when oxidized, constitutes nuclei required for the condensation of water vapor, and thus to form cloud over the oceans. Since clouds reflect a part of the solar radiation that reaches the earth, they contribute to cool the planetary surface. Therefore, DMS has been pointed out as a negative greenhouse gas, which could counterbalance the heating effects of greenhouse gases such as CO₂ and CH₄. Over the last three decades, CLAW hypothesis has been stimulating a great deal of research globally. More complete understanding of the interactions between the marine ecosystem and climate is a real challenge, having been described as one of the "Hilbert problems" for the geosciences in the 21^st century (Ghil, 2001). Understanding the marine biota-climate links can only be achieved through a combination of models, field/laboratory experiments, and satellite observations.

The sea-to-air flux of the DMS is thought to constitute an important radiative impact on climate, especially in the remote polar regions (Gabric et al., 2001). The sea-to-air flux of sulfur due to DMS release constitutes a significant proportion of the total atmospheric sulfur burden. Numerous studies have examined the DMS-aerosol-climate hypothesis. A global database of DMS seawater concentrations has been compiled (Lana et al., 2011). An empirical algorithm relating DMS sea-water concentration to the mixed layer depth (MLD) and surface chlorophyll concentration has been derived (Simó and Dachs, 2002). Climate models predicted a strong spatial variation to the global warming with large sea-surface temperature and salinity changes in the polar oceans (Hirst, 1999). It is relevant to note that Simó and Dachs (2002) derived an inverse relation between MLD and DMS concentration, suggesting that DMS levels would likely increase with increased stratification. Gabric et al. (2004) indicated that the greatest perturbation to DMS flux took place at high latitudes in both the hemispheres, with little change predicted in the tropics and sub-tropics. The largest change is simulated in the Southern Hemisphere between 50°S and 60°S (Gabric et al., 2005). The authors recently used CMIP5 (Coupled Model Intercomparison Project Phase 5) model output to simulate changes in relevant model forcings under 1×CO₂ (contemporary) and 4×CO₂ (which could be reached by 2100 under the worst case emissions scenarios); the simulation results suggest that the annual mean DMS sea-to-air flux would increase by 323% in the Barents Sea (Qu et al., 2020), and would increase more than triple in Greenland Sea (Qu et al., 2018).

Qu et al. (2021) use a global climate atmospheric model (AOGCM) (Rotstayn and Lohmann, 2002) to assess the global radiative impact of the perturbed DMS sea-to-air flux on contemporary climate to the one that has been projected to the year of 2090 (at least) for the polar region. In order to estimate the climate response to a prescribed meridionally-variable change in zonal DMS sea-to-air flux, they performed several simulations in which the AOGCM was run with incorporated sulfur cycle, coupled to a mixedlayer ('q-flux') ocean model. Their results indicated that the DMS flux would increase significantly in polar regions during the latter part of the 21^st century. The DMS flux perturbations are likely to cause a cooling in surface temperature in the Antarctic and Arctic Oceans due to their relatively unpolluted environment.

An advanced global climate atmospheric model was used in that paper (Qu et al., 2021) to incorporate the authors' several-year research on both the control run (baseline B00) with contemporary input forcings (late 20^th century) and the modified run (indicated as B01) by applying perturbed DMS flux (by 21^st century) in the polar regions (10° latitude bands). After comparing both Arctic and Antarctic Oceans, they found that much higher simulated increase of DMS flux led to higher increase of biogenic sulfur emitted into the Antarctic Ocean (especially in the austral summer) as compared with that into the Arctic Ocean. However, with more anthropogenic sources of continental aerosols, higher biogenic sulfur concentrations appear mostly in Arctic Ocean, although the regional increase rate is rather lower. They finally concluded that the DMS sea-to-air flux perturbations lead to the surface temperature cooling 1 K in Antarctic Ocean and 0.8 K in Arctic Ocean. These again proved that there would be more cooling caused by the perturbation in the polar regions. As comparing to their early research of the DMS flux perturbation for both southern hemisphere and northern hemisphere, the surface temperature cooling was 0.8 K for southern hemisphere and 0.4 K for northern hemisphere (Gabric et al., 2013).

However, uncertainty still remains on the overall oxidation process and the importance of particular mechanisms in determining the balance of the stable intermediates and the final end products of DMS oxidation (Von Glasow and Crutzen, 2004). In the recent three decades, the CLAW hypothesis has not been tested adequately due to its complexities. There appeared positive and negative feedback due to increased biogenic sulfur emissions. The role of DMS oxidation products in the formation of CCN (cloud condensation nuclei) in the marine boundary layer (MBL) has been debated and remains controversial (Ayers and Cainey, 2007; Quinn and Bates, 2011). Ayers and Cainey (2007) pointed out that DMS is not the only species responsible for the climate change. Sulfur dioxide will preferentially react on the surface of aerosol particles, especially sea salt, rather than nucleate to form new particles.

Due to its remoteness, key information of sulfur cycling is lacking in the polar region. Understanding how global warming and the ongoing reduction of seasonal sea-ice cover will affect the strength of these different sulfur emissions is challenging (Levasseur, 2013). More comprehensive sampling in different regions and a better understanding of the key processes governing DMS production in and around the sea ice, and the sensitivity of these processes to projected changes in the polar environment are equally important and full of challenge.