Saturday, December 31, 2011

Hadley Circulation in the Pliocene

I have been continuing my attempts to understand why trends and predictions for global drying under global warming are in the opposite direction from the Pliocene climate.  I have had a little email back and forth with Aiguo Dai (author of many of the PDSI papers I have been referring to in this series of blog posts) and he reminds me of the importance of the Hadley cell in driving the desert areas of the globe in this map from yesterday:

The Hadley cell is a name for the main climatic mechanism operating between the tropics and about 30o north and south of the equator:
What happens here is that at the equator air is warmed by the strong sun, as well as made moist by the high sea surface temperature (SST) there so that warm moist air rises (dropping lots of rain on tropical rainforests in the process) and then spreads out polewards at a height of 10-15km before descending back to earth at around 30o north and south. Having lost much of its moisture in the ascent, the air is dry when it descends and this gives rise to the various deserts north and south of the equator.  The actual pattern of ascent and descent in the modern climate is well illustrated in this figure from the Wiki:

Here the blue is ascending air velocity and the pink is descending.

Now, since the Pliocene climate apparently involved much less desert:

it seems that it must have had a weaker Hadley circulation.  Why so?  The best story I've been able to find is in Brierley et al Greatly Expanded Tropical Warm Pool and Weakened Hadley Circulation in the Early Pliocene.  This 2009 Science paper argues based on evidence from fossil fauna that the SST distribution in the Pliocene involved warm surface temperatures 15o further north and south than at present:

The main point here is the difference between the soft grey line (postulated Pliocene) and modern (thin solid black line).  Note the huge shift - we are talking about 10o-15o of latitude so in the ballpark of a thousand miles in both directions.  Obviously this chart relies on a rather small number of data points and the need to translate from the Atlantic speaks volumes - so it's probably indicative rather than dispositive.

Anyway - when they put this much larger tropical warm pool into an atmosphere-only climate simulation the Hadley circulation decreases substantially:
The Northern Hemisphere branch of the Hadley circulation weakens on average by roughly 30%. During boreal winter, the reduction reaches nearly 40%. The center of the Hadley cell moves northward by ~7°, whereas the cell’s latitudinal extent (26) increases by 3° to 4°. These changes of the Northern Hemisphere circulation are a robust response to changes in meridional SST gradient as evinced by additional sensitivity calculations (table S2). The volume transport of the circulation’s southern  branch nearly halves, making the southern Hadley cell even weaker than the northern cell. Whether this is a genuine feature of the early Pliocene climate remains to be seen, because there are only a few SST data points currently available in the Southern Hemisphere
This makes sense as much warmer water at the bottom (descending) leg of the Hadley cell is going to reduce the temperature difference between northern and southern ends of the cell that drives the circulation.  And in their simulations, this indeed shows up as fewer subtropical dry zones:
You want to look at the shrinkage of the pale/violet zones in the Pliocene map at bottom.  So the picture that emerges is that the Pliocene climate involved more widespread (and gentler) rainfall than the pre-industrial modern climate but that this was predicated on a much warmer ocean that extended tropical conditions much further north and south than the present ocean temperature distribution.

So again this points me in the direction of thinking about the difference in response times here - the ocean takes around a thousand years to completely turn over so presumably it's going to take around that amount of time before it has fully responded to the changes humans are making in the atmosphere - the present deep ocean water knows nothing about what we've done and it will presumably continue to well up into (and thus cool) the surface ocean for hundreds of years to come.  Look at this last map which comes from NASA GISS and shows the estimated temperature change between 1880-1900 and 2000-2010:

Clearly the oceans have warmed much less than the land - and a lot of the eastern Pacific has barely warmed at all (presumably because this is the site of a lot of upwelling of deep old ocean water).

Thus even if we've created a Pliocene concentration of carbon dioxide in the atmosphere, it will be a long time before we have a Pliocene ocean and thus a Pliocene weaker Hadley cell and gentler distribution of wet and dry (assuming this provisional picture of the Pliocene climate turns out to be right).


William M. Connolley said...

Since I drew the pic of vertical velocity that you used from wiki, back in the days when I had access to the data...

Thinking about the paleo changes, it is hard to see a world without a Hadley cell, which is what some of the no-Sahara stuff might seem to imply. The solution is likely to be / possibly in the seasonal migration: the ITCZ somewhat follows the solar equator, but with variations, so altering things a bit might get you from a permanently dry area into one with rain half the year at least, say.

You can play with the data from ECMWF here: if you like; you want the vertical velocity on the 500 hPa surface.

Stuart Staniford said...

Belette - but that would just move the desert, not shrink it?

Also - thx for the loan of your pic!