The entire premise of the conclusion is that larger dust particles absorb light and warm the atmosphere, and yet the cited paper [16] I believe, does not appear to reach that conclusion at all. If anything, the contribution to cloud formation may offset the absorbed energy.
Models are little more than extrapolations made from assumptions, but I am having a hard time understanding the basis for the assumptions made here.
I think we are deep in 'it depends' territory here.
Albedo is about how much of the light is absorbed at all or bounced straight back into space. For light that is absorbed, altitude (the amount of greenhouse gases between them and vacuum) matters.
Dust or clouds in the upper atmosphere might absorb or reflect more light than the surface, but some of what they do absorb will re-radiate out into space. Dust clouds and real clouds are three dimensional, cloud density is limited by air temperature, and dust availability generally limited by rainfall. That's gonna be some pretty complex math, so I wouldn't be surprised if models disagreed.
> The entire premise of the conclusion is that larger dust particles absorb light and warm the atmosphere
Which is a questionable premise given that we already know large volcanic eruptions cool the planet (the latest was Mt. Pintanubo in 1992). And how do they do that? By putting lots of coarse dust into the atmosphere.
That’s part of the story. From Pinatubo, there is believed to also be a warming effect due to the stratospheric SO2 absorbing heat from below (related to the claims of the article), and in a secondary effect, potentially changing atmospheric circulation.
The argument, the way I understood it, is that the models already include effects you mention (the impact of the clouds is already correctly modeled) and even the dust impact is modeled, but that the parameters of the models represent the dust incorrectly, and they explicitly mention that such an impact to the clouds could also be reconsidered in the models:
"Accounting for this missing coarse dust increases the TOA coarse dust warming by about 0.15 W·m−2 (0.10 to 0.24 W·m−2), increases the fertilization of ocean ecosystems by dust deposition, and affects the distribution of global clouds and precipitation. Therefore, climate models must account for the missing coarse dust to accurately simulate its impact on clouds, biogeochemistry, and global climate."
But we already know that the models are already quite good, it seems it's just a possibility where some adjustments could be considered.
What most people who comment about the models miss is that the climate models don't predict the exact outcome (e.g. "what will be weather in the area X in N years", but are there to make a simulation of the outcomes that could be expected.
To compare with something now easier to relate to for everybody, it's like you can simulate the speed of the spread of the virus across a population where different people have different numbers of contacts, and you can have an acceptably good model of how many people have how many contacts (some contact a lot of people, some only a few) and how fast the infection will spread even if you can't predict which persons will actually be infected when. These kind of adjustments then often don't change significantly the outcome one expects to verify, even if the parameters inside are, observed there, different.
Given that they are modeling a chaotic system it is extremely important that they get even the tiniest details correct, or they are no better than a random guess.
Climate is much less chaotic than weather, as it is about long-term and long-range averages; the quality of climate models can be judged by feeding them historic inputs (such as the amount of anthropogenic greenhouse emissions) and comparing their outputs for the years until 2020 with the actual record. Even simple climate models are substantially better than random guesses.
Models are little more than extrapolations made from assumptions, but I am having a hard time understanding the basis for the assumptions made here.