Global Meridional Circulation Profiles.. and more
Typically I examine global climate geostrophically. In that perspective, the planetary atmosphere is not unlike a river with the two poles as its banks. And I’m comfortable with that because hydrologists (and I am one) seldom peer deeply into the vertical profiles of any river. We typically find what we need from simple metrics such as total flow rates, as measured at various stream gages.
Certainly it’s strange to think of the poles of a spinning planet as the banks of an atmospheric river. But consider this example below for geostrophic wind velocities for July 1979, now that it has been unrolled from the global spherical shape onto a rectangular map.
I think now it should be more clear. It will help to know about the streamlines I added. The black lines originate from the left border (0E) and the few red lines originate from the right border (360E). Yellow lines originate from about 80N and 80S. The strongest lateral winds are always zonal (roughly parallel to any given latitude). The darker the shade, the slower the lateral flows. It’s encouraging to compare a solstice type (deep summer or winter) of map like this and see things that confirm what nominal mainstream meteorology teaches. For example, July is monsoon time in the northern hemisphere (NH) and so the Equatorial Trough (the slower westward moving air in the tropics) has shifted to the north from the equator.
Meanwhile in the Southern Hemisphere (SH), the eastward moving winds are very high. This again conforms to meteorological expectations. And that pattern is roughly mirrored for the other solstice (January) below. I use these maps quite a bit, for all kinds of explorations.
As you probably can see, there is no real beginning or end to the flows, because in truth, this is not a river. The flows of the atmosphere derive from a number of more three dimensional textbook defined features including the commonly cited Coriolis effect and the thermal gradients that are indicated to drive the widely recognized Hadley and Brewer-Dobson meridional circulations. Those circulations emerge from the surficial layers of our solar-furnace-heated equatorial atmosphere, and are described to lift to the upper elevations, whereupon they stream towards both poles. Upon the arrival of this now very cold and dehydrated air over the poles, the air masses subside back towards the surface. Think of patterns such as the Polar Vortex and you can imagine these cold dollops of air now sliding southwards* in radial patterns away from each pole.
There is a rich spectrum of variations within this broad construct. For my part, I often can develop a greater understanding of polar circulations by remapping back to spherical coordinates, as the next two images show. And of course, these are two polar views of the July 1979 map above.
One can really get a sense of the degree of downwelling of chilled polar air from images such as these. When the streamlines that start at 80N for example, radiate far to the south, then my observational/evaluational experience has been that, depending on the month, we’ll see another Polar Vortex (solsticial times like January) or a larger Arctic Ice Cap (equinoctial times like September).
And by contrast, if the downwelling is not so great then, since Nature and the Geostrophic atmosphere abhor a vacuum, the eastward flowing mid latitude winds expand to the north. I’m definitely speaking in some simplified terms. The handful of researchers who live and breath this topic would prefer to speak in terms of those Hadley and Brewer-Dobson circulations, along with Rossby Waves, angular momentums and more.
Here is a handy Hadley and Brewer-Dobson (BD) illustration from , which itself was borrowed from . The BD circulation is the highest white arrow path. The Hadley circulations are no doubt covered by the lower white arrrow outlines. I appreciate the achievements of the many scientists starting roughly in the 1600s to now, who have helped to develop these simplified profiles. By the way, this practice of representing the entire global atmosphere by averaging meridional values across all longitudes and then simply displaying as a function of latitude, IS the convention.
These schematics help in a rough way, not unlike my geostrophic maps above. But they are not directly based on observations. They are drawings. And as hard as one might look, actual observation – derived profiles like these are almost nowhere to be found. Moreover, the scientists average across all longitudes as I noted, so much is really left to the imagination.
That’s why I’ve been putting in some time to try to reconcile my rich geostrophic flavored maps with their convention. For my new longitudinally averaged meridional profiles, I naturally work from the ECMWF resource. That includes fully 3D atmospheric coverages of many variables including zonal, meridional, and vertical winds. Here are the associated profiles then, with a few closeups, for the same two months of January and July of 1979, including again the featured image at the top of this post. These follow the conventions of the other researchers to some extent, and my conventions as well. The vertical axis is in pressure levels hPa, which as a metric is not too far off from elevation units of km. To be clear, I’m not contouring pressures, as many do. I’m using pressure as a proxy for altitude, and that is a convention that many others adopt as well for these types of graphics. Note that I always plot color fields of the vertical wind strength, and the mapping is such that dark blue ALWAYS indicates downward flows. Any other lighter shade of blue, and any subsequent color indicates upward circulations. The arrow fields are complementary.
These first two are two enlargements of the same January map, in order to bring out some of the fine detail. The next two are similar enlargements for the July map.
When one also (very loosely) considers the geostrophic expressions, these profiles seem complementary as well. For example, and fair eye-squinting warning, for either month, try comparing the latitude bands of greatest vertical flux (either up or down) to the longitude of lowest zonal flux (either eastward or westward). This isn’t perfect at all, but in general you should find that when air is busy moving robustly up or down, it isn’t moving so fast laterally. This recognition was key to my own discoveries about Solar Forcing of our climate .
COUNT ME IN ON THE BD CIRCULATION
In other words, count me in as someone who nominally accepts the BD, the Hadley and the geostrophic as seamlessly integrated. Yes, the conventional representations of the BD and the Hadley are so schematic and simplified that they don’t do justice to the concepts. I’ll perhaps advance further in some other looks at these profiles, perhaps for longer time spans and different parameter maps, including hopefully latent heat and moisture.
BUT COUNT ME OUT ON THE REST
Yes, the circulation of air from equator to poles and back, through multiple degrees of directional freedom, whether meridional or otherwise, and likely through ever shifting combinations, seems most plausible. Yet many items in  seem to gloss over outstanding contradictions with more popular descriptions of climate and our atmosphere.
In popular climate literature, the BD is limited to circulation above the tropopause (see Wikipedia example below). But any who review the schematic from  above can see that the tropopause is patchy, not continuous. See those three horizontal slabs with the hatch lines? That is the tropopause.
The actual BD literature, including the observationally-based Figure 5 from , using the same ECMWF data that I’ve been using, reinforces that BD flows emerge from the Earth’s surface, not its raggedy tropopause.
The tropopause may only be observable when the atmospheric column is moist. Water condensing at the tropopause height makes the tropopause. The temperature inflection there maybe primarily due to the release of latent heat from the condensing vapor.
In this case the author of  reaches into the unknown a little bit further, mentioning the word “ozone” over 80 times, but the word “solar” is never mentioned. Notably the paper also indicates that no one has effectively quantified the BD flows. The climate models predict a speeding up of (upper atmosphere) meridional flows from equator to poles, but this has never yet been observed.
I’ve made some final plates since it was convenient. These cover a coarser pattern so I could include vectors more clearly. I’ve also taken the meridional profiles and made them averages for the full set of 480 longitude columns. And then I rotated the meridional maps to align to some plan view maps, for basically the same lateral winds, but close to the surface. Nothing will ever be perfect, but this gives a somewhat more integrated sense of meridional, vertical, and zonal circulation, if you are happy interpret various seasonal flows to according to these month, parameter, layer and spatial averaging limits.
For both cases, the y axis is only the reduced grid value (due to coarsening from the original). The maps on the left are of lateral near-surface wind magnitudes with an overlay of the very same as vectors. There is a different take through the right images. There the highest altitudes are to the now left and the land surface is the right boundary. The South Pole makes the bottom boundary and the North Pole is the top boundary. For all, the color scales are simply auto-assigned, so they are different for each map even within the same plate. But they are always mapped with red highest positive and blue the largest negative values, or zero. With this in mind you can confirm flow magntitudes and directions. Finally, on the right panels, the blue shades indicate relative westward flow and red shades indicate relative eastward flow.
Between all of the sets, one can get a sense of flow directions and scales. I can also already even if prematurely speculate about the high vertical flows that ring Greenland and Antarctica. This is perhaps from Brewer Dobson vertical downwelling “jets” over each pole, which lead to a “spray” of air, like a splash if you will, which surrounds the jet. Fortunately it is somewhat easy to at least start to confirm through these convenient images. First, a plan view of Greenland’s winds at the surface over the month of February 1979. The color contours represent the vertical wind magnitudes.
Next, I developed a profile of these winds along 75N. I’ve checked somewhat for consistency and this seems correct so far.
Typically the Brewer Dobson – Hadley circulation papers I am familiar with don’t explore this resolution of detail. Rather they try to capture a bigger picture. I understand and am still learning in any case. But no matter how I slice up the vertical satellite atmospheric wind coverages, the data is stunningly beautiful, and that’s at least one good reason to feature some of the slices in a blog post. On a more scientific mode, getting a sense of trends of winds over time is a key goal to me.
With many of these aspects in mind, and more, I wrote a Solar post about wind trends and the so-called Arctic Amplification. It is an epic post I think. I’ve placed it in private mode for the time being while I collaborate on a paper for submission to peer review.
In this example, I’ve moderately corroborated, with real data, the essential claims of . But there is more to explore so another post will follow to look at the temperature and moisture profiles for the same months.
*Like molasses, as friend and mentor Bob Endlich at CAF once taught me.
 Butchart, N. 2014. The Brewer-Dobson circulation Reviews of Geophysics 52|2 pp. 157-184
 Plumb, R.A. and Mahlman, J.D. 1986. The Zonally Averaged Transport Characteristics of the GFDL General Circulation/Transport Model. Journal of the Atmospheric Sciences. 44|2 298-327
 Wallace, M.G., 2019, Application of lagged correlations between solar cycles and hydrosphere components towards sub-decadal forecasts of streamflows in the Western US. Hydrological Sciences Journal Volume 64 Issue 2. doi: 10.1080/02626667.2019.
3848total visits,2visits today