Cutting Edge > Topics > Hurricanes-Climate Change Connection > Workshop 08 > Discussions > Questions for Jenni Evans

« Workshop Discussion Threads

Questions for Jenni Evans  

Please post your questions and comments here following Jenni Evans' presentation.


Share edittextuser=24 post_id=2834 initial_post_id=0 thread_id=974

By what physical mechanism is the ocean mixed layer expected to cool with Global Warming? What confidence do we have in that?


Share edittextuser=595 post_id=2838 initial_post_id=0 thread_id=974

Hi, Jenni: thanks very much for this very informative presentation! I have a question about the impact of hurricanes on climate. You mentioned the possibility of cooler SSTs due to mixing of the oceans - would this be a potential long-term effect (several years or longer) or one that would persist for a short time following a hurricane?


Share edittextuser=1902 post_id=2839 initial_post_id=0 thread_id=974

For the questions relating to ocean cooling with hurricanes, we observe cooling of the ocean surface from the passage of individual hurricanes now due to mixing of cooler water from below the mixed layer. This persists for a short time following each hurricane.

The first question is whether the net effect of all hurricanes has a cooling effect (albeit slight) on the tropical oceans in the present climate. If it does, then depending on how things change, this signal might also change.

In terms of current simulations of a warmer climate, if the sub-surface ocean cools and the surface warms, then the cooling at the surface due to this mixing from below will be larger (a bigger change in temperature).

What is my confidence? Low at this point, because we have to answer the question of whether this is a season-long net feedback and my inferences about the changes in SST and sub-surface temperatures from GCM would have to be right.

So why did I pose this idea? Because it has been discussed and debated quite widely and because I felt that thinking about hurricane impacts back *onto* the climate system is important. If hurricanes contribute to the net climate that we observe - and I think that they must - then we need to think about their interactions in both directions.


Share edittextuser=2144 post_id=2842 initial_post_id=0 thread_id=974

The GCMs are set up to run under modern boundary conditions. With global warming we are likely to leave those boundary conditions. For example if we shut off North Atlantic Deep Water production, we will significantly change heat transport from the subtropical North Atlantic to the high latitudes in the North Atlantic. What effect would this have on hurricanes.


Share edittextuser=2124 post_id=2844 initial_post_id=0 thread_id=974

Jenni, I got the impression from your presentation that (coupled atmosphere/ocean) GCMs predict cooler subsurface temperatures (and warmer surface temperatures) in the tropics. This influences the magnitude of the hurricane-climate feedback via enhanced mixing, but why would subsurface temperatures cool in a warmer world? I can see how a warmer surface would increase the stability of the pycnocline and discourage mixing generally, for example, but that's not the same as cooling the subsurface.


Share edittextuser=436 post_id=2845 initial_post_id=0 thread_id=974

Jenni: Nice summary of the issues, and thank you for that link to the WMO report, I wasn't aware of that work and that seems very helpful in connection with the IPCC work. I've got two questions that I'm thinking about relative to your talk and potential activities for students. Do you think that the issues in the observational record would make it difficult to ask students to work with those data and draw any conclusions? And when talking to a general audience, in your opinion what is/are the most important aspect(s) of the linkage between hurricanes and climate change. Or phrased another way - is there a part of this that you think would be more useful for students to focus on in a lab/problem set?


Share edittextuser=2142 post_id=2846 initial_post_id=0 thread_id=974

Dave: I was thinking that strong hurricane winds are generally strong enough to mix the top 100-200 m of the ocean regardless of the strength of the pycnocline. I could see this as a potential limiting factor on the growth of strong hurricanes (upwelling water that is much colder) in the future. Perhaps this would be a negative feedback on hurricane intensity?


Share edittextuser=2142 post_id=2847 initial_post_id=0 thread_id=974

Todd: Yes, mixing of the ocean surface by hurricanes is an immediate negative feedback for the development not only of a particular hurricane (very short time scales), but also of any hurricanes that might develop in the area not long after (short time scales), and, if there are enough hurricanes in an ocean basin, maybe even for that region on the seasonal scale or longer, if I understand Jenni right. The think I'm not clear about is why the GCMs predict cooler subsurface temperatures (in and below the pycnocline) in the first place. If they do, then that would tend to limit hurricane intensity, as you note.


Share edittextuser=436 post_id=2849 initial_post_id=0 thread_id=974

What effect will reduction of the equator-to-pole temperature gradient (more warming at high latitudes) have on hurricane formation?


Share edittextuser=2126 post_id=2850 initial_post_id=0 thread_id=974

The "Hadley cell" is a planetary-scale feature in the global circulation that transports heat out of the tropics to higher latitudes. From there, the midlatitude storms keep transporting this heat towards the poles. By doing this, the atmosphere is trying to get rid of the temperature gradients created by the Sun heating the different areas of the globe by differing amounts (in the same way water boils when heated from below, moving around so that it all comes nearly to the same temperature).

Theories about the Hadley cell and the midlatitude storm track tie their strength to the magnitude of the difference between the temperatures in the tropics and those near the poles.

If the poles warm more than the tropics these circulations could reduce in strength, since less heat has to be transported out of the tropics to make the temperature difference smaller.

One signature of the Hadley cell in the tropics is in the amount of deep convection (the thunderstorms we need for developing hurricanes). The stronger the Hadley cell, the more thunderstorms we need to help with the heat transport. So, a weaker Hadley cell likely means less convection in the tropics - or less efficient convection (more smaller clouds and fewer really intense ones).

This is the basic idea for why reducing the equator to pole temperature gradient could cause a reduction in some aspects of hurricanes (most likely their frequency or the average lifetime - number of days - for an individual storm).


Share edittextuser=2144 post_id=2860 initial_post_id=0 thread_id=974

GCM have certainly been calibrated using modern boundary conditions, but some have been run for palaeoclimates and compared to available proxy records as well. These simulations provide some confidence that the models can provide realistic simulations beyond the narrow range of the current climate.

Past simulations have tried to address the problem of a large climate shift by taking into account the different patterns of ocean currents resulting from different continental configurations. This won't get you the whole way with hurricanes but, coupled with information about the other factors we talked about (deep thunderstorm activity, monsoons, wind changes with height) it will give you a sense of what might happen.

However, if the climate crosses over a bifurcation point such as you describe, I think that all bets are off...


Share edittextuser=2144 post_id=2861 initial_post_id=0 thread_id=974


I went back to check myself on the ocean sub-surface cooling. I realized that it had come from observations from the recent climate - see chapter 5 of the IPCC report. I'm attaching a couple of graphics taken out of that report.

Fig 5.2 shows the spatial distribution of ocean heat content (per unit area), showing decreasing trends in the tropical Indian Ocean and western tropical Pacific.

I will post an accompanying IPCC Fig 5.3 in a following message - I can't figure out how to attach two files now :(

Fig 5.3 shows T(lat,z) for each ocean basin. From this figure, the decrease in heat content comes from a sub-surface decrease in zonally averaged temperature. (I attached a copy cut out of the report since the IPCC copy on their website had a white out region not in the report copy).

The explanation given for this sub-surface cooling is a shallowing of the thermocline due to more active ENSO in the past 20 years or so. If we take the current model projections of a more ENSO-like mean climate, this cooling *may* persist. This projection is based on a shallower thermocline resulting from different tropical surface wind stresses due to the change in the Walker cell in this scenario.

Does this make sense? Certainly this cooling is not a confident prediction, but it is consistent with recent observations and current explanations for those observations.




Share edittextuser=2144 post_id=2866 initial_post_id=0 thread_id=974

I'm attaching IPCC WG1 Fig 5.3 here - in pdf format because of display issues I discuss in the previous post.

See the previous post for discussion of this figure and how I see it fitting into projections of sub-surface cooling.


IPCC WG1 Fig 5.3 (Acrobat (PDF) 84kB Oct21 08)


Share edittextuser=2144 post_id=2867 initial_post_id=0 thread_id=974

Jenni, From your explanation it sounds as if the sub-surface cooling is in the western side of the equatorial Pacific, where the mixed layer becomes shallower as the warm surface water shifts westward under El Nino conditions and the thermocline consequently comes closer to the surface. (Presumably the opposite would be occuring in the eastern Pacific.) So this would be a "local" (fixed location, at a depth within the thermocline) in a region, not a basin-wide effect averaged over the depth of the ocean below the mixed layer, which makes sense in a world where the oceans are otherwise on the average gaining heat. Thanks for tracking that down!


Share edittextuser=436 post_id=2874 initial_post_id=0 thread_id=974

Join the Discussion

Log in to reply

« Workshop Discussion Threads

« 20th Century GCM reconstructions vs. observations       Questions for Tom Knutson »