By Physicist Dr. Ralph Alexander
Recent research has cast doubt on the influence of two watery entities – ocean currents and clouds – on future global warming. But, unlike many climate studies, the two research papers are grounded in empirical observations rather than theoretical models.
The first study examined so-called deep water formation in the Labrador Sea, located between Greenland and Canada in the North Atlantic Ocean, and its connection to the strength of the AMOC (Atlantic Meridional Overturning Circulation). The AMOC forms part of the ocean conveyor belt that redistributes seawater and heat around the globe. Despite recent evidence to the contrary, computer climate models have predicted that climate change may weaken the AMOC or even shut it down altogether.
Deep water formation, which occurs in a few localized areas across the world, refers to the sinking of cold, salty surface water to depths of several kilometers because it’s denser than warm, fresher water; winter winds in the Labrador Sea both cool the surface and increase salinity through evaporation. Most climate models link any decline in North Atlantic deep water formation, due to global warming, to decreases in the strength of the AMOC.
But the researchers found that winter convection in the Labrador Sea and the adjacent Irminger Sea (east of Greenland) had very little impact on deep ocean currents associated with the AMOC, over the period from 2014 to 2018. Their observational data came from a new array of seagoing instruments deployed in the North Atlantic, including moorings anchored on the sea floor, underwater gliders and submersible floats. The devices measure ocean current, temperature and salinity.
Results for the strength of the AMOC are illustrated in the figure below, in which “OSNAP West” includes the Labrador and Irminger Seas while the more variable “OSNAP East” is in the vicinity of Iceland. As can be seen, the AMOC in the Labrador Sea didn’t change on average during the whole period of observation. The study authors caution, however, that measurements over a longer time period are needed to confirm their conclusion that strong winter cooling in the Labrador Sea doesn’t contribute significantly to variability of the AMOC.
Understanding the behavior of the AMOC is important because its strength affects sea levels, as well as weather in Europe, North America and parts of Africa. Variability of the AMOC is thought to have caused multiple episodes of abrupt climate change, in a decade or less, during the last ice age.
The second study to question the effect of water on global warming involves clouds. As I described in an earlier post, the lack of detailed knowledge about clouds is one of the major limitations of computer climate models. One problem with the existing models is that they simulate too much rainfall from “warm” clouds and, therefore, underestimate their lifespan and cooling effect.
Warm clouds contain only liquid water, compared with “cool” clouds that consist of ice particles mixed with water droplets. Since the water droplets are usually smaller than the ice particles, they have a larger surface area to mass ratio which makes them reflect the sun’s radiation more readily. So warm clouds block more sunlight and produce more cooling than cool, icy clouds. At the same time, warm clouds survive longer because they don’t rain as much.
The research team used satellite data to ascertain how much precipitation from clouds occurs in our present climate. The results for warm clouds are illustrated in the following map showing the warm-rain fraction; red designates 100% warm rain, while various shades of blue indicate low fractions. As would be expected, most warm rain falls near the equator where temperatures are highest. It also falls predominantly over the oceans.
The researchers then employed the empirical satellite data for rainfall to modify the warm-rain processes in an existing CMIP6 climate model (the latest generation of models). The next figure, which shows the probability of rain from warm clouds at different latitudes, compares the satellite data (gray) to the model results before (maroon) and after (yellow) modification.
It’s seen that the modified climate model is in much better agreement with the satellite data, except for a latitude band just north of the equator, and is also a major improvement over the unmodified model. The scientists say that their correction to the model makes negative “cloud-lifetime feedback” – the process by which higher temperatures inhibit warm clouds from raining as much and increases their lifetime – almost three times larger than in the original model.
This larger cooling feedback is enough to account for the greater greenhouse warming predicted by CMIP6 models compared with earlier CMIP5 models. But, as the study tested only a single model, it needs to be extended to more models before that conclusion can be confirmed.