Head in a Cloud

troposphere and stratosphere meet blogosphere

Head in a Cloud random header image

Work in progress…. Please Comment!

July 10th, 2006 by andrewd · 7 Comments

Editor’s note: We would like to welcome Prof. Andrew Dessler of Texas A&M University as a guest blogger at Head in A Cloud! Please note this post was written by him and not Sean Davis.

The effects of convection on the summetime extratropical overworld

Much effort has been expended over the last decade in assessing the impact of deep convection on the extratropical lower stratosphere. It is now clear that summertime convection has an important impact on the chemical composition of the lowermost stratosphere (that part of the stratosphere with potential temperature θ < 380 K).

Recently, Dessler and Sherwood [2004, hereafter DS04] showed that summertime convection plays an important role in the Northern Hemisphere (NH) extratropical H2O budget at 380 K, the boundary between the lowermost stratosphere and the so-called overworld (the stratosphere with θ > 380 K). DS04 focused on the 380-K surface and did not look at altitudes above 380 K. I am now investigating how high summertime mid-latitude convection penetrates into the extratropical NH overworld.


To address this issue, I will use measurements of H2O made by the Halogen Occultation Experiment (HALOE), which was carried aboard the Upper Atmosphere Research Satellite [e.g., Dessler et al., 1998]. The measurements were obtained between 1994 and 2005, and are well validated [Harries et al., 1996; Park et al., 1996; Bruhl et al., 1996; SPARC, 2000], with systematic errors in the lower stratosphere of 20% for all three constituents. I use HALOE version 19 here, adjusted per instructions from the HALOE science team. The θ of each measurement is determined using daily temperature and pressure fields from the National Center for Environmental Prediction (NCEP) that are provided with the HALOE data.

Depth of penetration of convection into the overworld

For context, I plot in Figure 1 the average H2O at 380 K measured by the HALOE between June 15 and August 15 of 1994 through 2005; similar plots appeared in DS04 and Randel et al. [2001]. The data show maxima in H2O between 30°N and 40°N over North America and Asia. DS04 connected these 380-K maxima to deep convection, arguing that convection moistens the lower stratosphere in regions where the relative humidity is low.

figure 1

I now focus on the latitude range 30°N-40°N and calculate the summertime H2O anomaly as a function of θ and longitude. For each θ surface, HALOE data obtained between June 15 and August 15 of 1994 through 2005 and between 30°N and 40°N are interpolated to that θ surface. These data are then separated into 30 longitude bins and the average of each bin is calculated. The anomaly is calculated by subtracting the value of the lowest-average bin from all of the bins. This process is repeated at each θ surface and the resulting anomaly is plotted in Fig. 2.

figure 2
Figure 2 shows that there are two longitudes where large H2O anomalies exist: around 50° longitude, over the Asian monsoon, and around 270° longitude, over North America. This is consistent with Fig. 1, and is anyway not surprising since these regions are well known to be the site of vigorous deep convection. Over North America (270° longitude), the anomaly decreases monotonically with height, from ~0.65 ppmv at 380 K (15.5 km) to ~0.1 ppmv at 410 K (17.5 km).

Over Asia (50° longitude), the anomaly also decreases monotonically with height, from ~0.60 ppmv at 380 K (15.5 km) to ~0.1 ppmv at 460 K (19 km). Thus, the anomaly over North American and Asia are of similar magnitude at 380 K, but the convective anomaly over the Asian monsoon extends about 50 K (1.5 km) higher than the North American convection anomaly. This is not unexpected because, by most metrics, convection over the Asian monsoon is far stronger than over North America[e.g., Dunkerton, 1995].

I am arguing in this post that upward advection of high H2O air by convection is responsible for the H2O anomalies in Fig. 2. One might ask whether horizontal advection of high H2O from lower or higher latitudes might also be playing a role. Figure 1 shows that this is impossible at 380 K: H2O is lower both poleward and equatorward of the mid-latitude maxima, so meridional advection cannot explain the maxima. To show this is also true at other θ, I plot in Figure 3 a latitude-height cross-section through the Asian monsoon. The crosses indicate the latitude of maximum H2O on selected θ levels. This plot shows that between 380 and 450 K, H2O between 30°N-40°N is higher than air both poleward and equatorward. Thus, meridional transport cannot be responsible for the high H2O found there. In addition, H2O tends to decrease with increasing θ in this season and over the θ range of interest here, so downward transport also cannot explain the high values. I conclude that only upward transport can explain the H2O anomalies, and that the only possible explanation for this upward transport is convection.

figure 3

While convection reaching the overworld, even as high as 460 K, might seem surprising, previous studies provide some evidence to support this conclusion. Fromm et al.[2000] reported several occurrences of forest fire smoke in the overworld, while Fromm and Servranckx [2003] described observations of forest fire smoke as high as 460 K. Livesey et al. [2004] observed a biomass burning product, methyl cyanide, at 100–68 hPa (16–19 km or 400-450 K) in the summertime NH mid-latitudes. In situ data analyzed by Jost et al.[2004] showed biomass burning products at altitudes between 380 and 400 K. Wang et al. [2003] has modeled summertime continental convection and has simulated convection reaching θ > 400 K, well into the overworld.


Bruhl, C., et al. (1996), HALOE ozone channel validation, Journal of Geophysical Research, 101, 10,217-10,240.

Dessler, A.E., M.D. Burrage, J.-U. Grooss, J.R. Holton, J.L. Lean, S.T. Massie, M.R. Schoeberl, A.R. Douglass, and C.H. Jackman (1998), Selected science highlights from the first five years of the Upper Atmosphere Research Satellite (UARS) program, Reviews of Geophysics, 36, 183-210.

Dessler, A.E., and S.C. Sherwood (2004), The effect of convection on the summertime extratropical lower stratosphere, Journal of Geophysical Research, 109, D23301, DOI: 10.1029/2004JD005209.

Dunkerton, T.J. (1995), Evidence of meridional motion in the summer lower stratosphere adjacent to monsoon regions, Journal of Geophysical Research, 100, 16,675-16,688.

Fischer, H., et al. (2003), Deep convective injection of boundary layer air into the lowermost stratosphere at midlatitudes, Atmos. Chem. Phys., 3, 739-745.

Fromm, M., J. Alfred, K. Hoppel, J. Hornstein, R. Bevilacqua, E. Shettle, R. Servranckx, Z.Q. Li, and B. Stocks (2000), Observations of boreal forest fire smoke in the stratosphere by POAM III, SAGE II, and lidar in 1998, Geophysical Research Letters, 27, 1407-1410.

Fromm, M.D., and R. Servranckx (2003), Transport of forest fire smoke above the tropopause by supercell convection, Geophysical Research Letters, 30, 1542, DOI: 10.1029/2002GL016820.

Harries, J.E., et al. (1996), Validation of measurements of water vapor from the Halogen Occultation Experiment (HALOE), Journal of Geophysical Research, 101, 10,205-10,216.

Hegglin, M.I., et al. (2004), Tracing troposphere-to-stratosphere transport above a mid-latitude deep convective system, Atmospheric Chemistry and Physics, 4, 741-756.

Hess, P.G. (2005), A comparison of two paradigms: The relative global roles of moist convective versus nonconvective transport, Journal of Geophysical Research-Atmospheres, 110, -.

Jost, H.-J., et al. (2004), In-situ observations of mid-latitude forest fire plumes deep in the stratosphere, Geophysical Research Letters, 31, L11101, DOI: 10.1029/2003GL019253.

Livesey, N.J., M.D. Fromm, J.W. Waters, G.L. Manney, M.L. Santee, and W.G. Read (2004), Enhancements in lower stratospheric CH3CN observed by the Upper Atmosphere Research Satellite Microwave Limb Sounder following boreal forest fires, Journal of Geophysical Research, 109, D06308, DOI: 10.1029/2003JD004055.

Park, J.H., et al. (1996), Validation of HALOE CH4 measurements from the UARS, Journal of Geophysical Research, 101, 10,183-10,203.

Poulida, O., R.R. Dickerson, and A. Heymsfield (1996), Stratosphere-troposphere exchange in a midlatitude mesoscale convective complex, 1. Observations, Journal of Geophysical Research, 101, 6823-6836.

Randel, W.J., F. Wu, A. Gettelman, J.M. Russell, III, J.M. Zawodny, and S.J. Oltmans (2001), Seasonal variation of water vapor in the lower stratosphere observed in Halogen Occultation Experiment data, Journal of Geophysical Research, 106, 14,313-14,325.

Wang, P.K. (2003), Moisture plumes above thunderstorm anvils and their contributions to cross-tropopause transport of water vapor in midlatitudes, Journal of Geophysical Research, 108, 4194, DOI: 10.1029/2002JD002581.

Tags: stratosphere

7 responses so far ↓

  • 1 jfleck at inkstain » Blog Archive » A Promising New Climate Blog // Jul 11, 2006 at 12:20 pm

    [...] One of my frustrations with the climate blogging world is the relative abundance of climate wars-type discussion and the relative dearth of discussion of climate science. Sean Davis has started a new blog that looks promising in this regard: Head in a Cloud. Great early discussion, including a nice piece by Andrew Dessler on work he’s doing on deep convection in the lower stratosphere (convection is another one of those topics that came up in one of those maddeningly science-free discussions in the comments here a while back – Andrew adds some actual science). [...]

  • 2 Bloody Fingers » Head in a Cloud // Jul 11, 2006 at 1:16 pm

    [...] There are only a couple of posts right now, but one of them is from no other than Andrew Dessle, famous for his work on convective systems (well, he’s famous to me ). Definitely a new important stop in the blogosphere. [...]

  • 3 seand // Jul 11, 2006 at 2:45 pm

    John, Vincent, thanks for the flattering remarks and exposure for my blog. To me the success of this blog will hinge largely on how exciting and informative the discussions are.

    In that vein, I’d like to ask a couple of (perhaps simplistic) questions related to Andy’s post.

    1. First, a question about underworld vs. overworld stratosphere (I might want to include definitions of these in the Acronym soup page, even though they’re not acronyms). What physical basis is there for distinguishing between an underworld and overworld stratosphere? Does it reflect a difference in the chemistry or dynamics of the stratosphere, or both?

    2. Similar to the first question… What are the implications for a moistening of the overworld on ozone chemistry/column amount? As I recall, WV can play a large role in ozone destruction in the lower stratosphere, but I’m not as sure about the upper stratosphere.

    3. You addressed the issue of meridional transport by pointing to figure 1 and saying that it is drier at 380K both poleward and equatorward of the midlatitudes, and that therefore that is not the source of moistening. Is there any mechanism by which air could be transported poleward AND upward at subtropical latitudes, bringing moist air up in way that would still be consistent with figure 1?

    4. What do you think these figures would look like if it was possible to take a ‘snapshot’ during NH summer? Would you see such a smooth behavior, or would there be pockets of very large anomalies related to specific convective events?

    Please excuse my rudimentary knowledge of stratospheric dynamics/chemistry, if these questions seem too simplistic. Thanks! Sean

  • 4 seand // Jul 11, 2006 at 4:45 pm

    Ohh… and one more question. What about isotopes? Is there any way to confirm your ideas here using in-situ isotope data? It seems like there should be a way to do that. I forget if the Harvard isotope instrument is aboard the ER-2 (which could fly in the overworld) or WB-57 (which might not be able to?). Another option w.r.t isotopes would be the MIPAS satellite, which although not operational for very long, was attempting to get some isotope information as I recall (contact me offline for a contact on this one, if interested).

  • 5 Andrew Dessler // Jul 11, 2006 at 11:14 pm


    Here are a few thoughts:
    1) the concept of overworld vs. lowermost stratosphere was first put forth by Hoskins, B.J. (1991), Towards a PV-θ view of the general circulation, Tellus, 43AB, 27-35. I think a good and perhaps more accessible explanationg is in my 1995 paper on the lowermost stratosphere (available here). These air masses are somewhat different in both dynamics and chemistry.

    2) Water plays an important role in stratospheric chemistry and increasing water will likely reduce stratospheric ozone [Dvortsov and Solomon, 2001; Kirk-Davidoff et al., 1999]. In addition, water vapor in the lower stratosphere is itself an important greenhouse gas with significant radiative forcing of the troposphere [Forster and Shine, 1999; Smith et al., 2001].

    3) I can’t think of any mechanism like the one you describe that would be fast enough. Do you have anything in mind in particular?

    4) Good question. One might be able to do something like that with MLS, but I’m not sure there’s enough data to make it worthwhile. Clearly, the anomaly field would not be as smooth as the 12-year average. However, I suspect that you’d still see the same overall signature of monsoonal convection.

    One more question: This is a great question. In fact, I was working on an isotope paper when the idea for this paper occurred to me. So the short answer is yes, isotopes are a really interesting and (we think) confirming line of evidence. That paper has been submitted to GRL. I’ll write a post about it when it gets accepted.


  • 6 seand // Jul 12, 2006 at 8:53 am

    Thanks for your responses.

    RE #3 above, I didn’t have any specific mechanism in mind. My knowledge of stratospheric dynamics is severly limited though, which is why I asked.

    A few other suggestions as well… Since you are making the case for the effects of summertime convection on the overworld, perhaps an alternate (or additional) way of showing figure 1 would be to have the anomalies be for monsoon season months (JJA?), relative to the annual mean in each box, or something similar to that. One should see positive anomalies from the monsoon, unless there is a time lag. The time lag, if there is one, would be an interesting phenomena itself. Perhaps a hovmueller type plot for anomaly vs. &theta as a function of time for the grid box 30-40N/60-90E or 30-40N/270-300E? That might show the annual cycle effect and lag time…

  • 7 Andrew Dessler // Jul 12, 2006 at 10:41 am

    Concerning your other suggestions: your suggestion of considering annual means is a reasonable alternative methodology. as I see it, the problem w/ that approach is that the annual mean has a strong seasonal cycle associated with the so-called “tape recorder”. you would need to subtract off this annual cycle. I’m not sure how you’d do this, and I think it introduces some significant uncertainties. it’s an interesting idea, though …

Leave a Comment