Friday, December 19, 2014

Looking at Clouds with Ulrike Lohmann at AGU14

This year's Charney Lecture at the AGU Fall Meeting was given by Professor Ulrike Lohmann from ETH Zurich. Her topic was:

A21N-01 Grand Challenges in Understanding Clouds: From Ice Crystal Formation to Their Influence on Climate (Invited)
Ulrike Lohmann, ETH Swiss Federal Institute of Technology Zurich, Zurich, Switzerland; ETH Zurich, Zurich, Switzerland

As she said, you can just enjoy clouds by looking at them, "looking at the beauty that is out there". Although you can tell she loves clouds from an aesthetic viewpoint, she is also curious about them from a scientific perspective. Her talk outline was:
  • Importance of clouds for radiative forcing - which means there is an external perturbation to the system;
  • Ice crystal formation in the laboratory and in the field - ice crystals are understood much less than liquid clouds
  • Importance of clouds for climate change
  • Climate engineering with cirrus clouds

So you can see that she covered a very wide range. I won't try to summarise the whole talk. You can see it yourself via AGU Virtual Options (see here for details). The easiest way is probably to sort the videos alphabetically, which sorts by the number (A21N).

Professor Lohmann started by saying that because CO2 is a well-mixed greenhouse gas, you don't need to look at its distribution over the world. Instead she put up a map of aerosols showing different aerosols in different colours - dust as orange/red, sea salt as blue, organic/black carbon as green and sulfates as white. I found the map she used on the NASA website - as always, click to enlarge it:

This portrait of global aerosols was produced by a GEOS-5 simulation at a 10-kilometer resolution. Dust (red) is lifted from the surface, sea salt (blue) swirls inside cyclones, smoke (green) rises from fires, and sulfate particles (white) stream from volcanoes and fossil fuel emissions.
Source: NASA Image credit: William Putman, NASA/Goddard

Fascinating! As you can tell from the caption, this image was produced from a GEOS-5 simulation. It's not actual observations at a point in time. But it's probably real enough. The NASA page has a bit more information, describing it as:
High-resolution global atmospheric modeling run on the Discover supercomputer at the NASA Center for Climate Simulation at Goddard Space Flight Center, Greenbelt, Md., provides a unique tool to study the role of weather in Earth's climate system. The Goddard Earth Observing System Model, Version 5 (GEOS-5) is capable of simulating worldwide weather at resolutions of 10 to 3.5 kilometers (km).

Brandon R. Gates in the comments found an animated version of this on YouTube, and he's right - it's stunning. Click in the bottom right to view it full screen (or on YouTube):

An aside: If you look at Australia over summer time, you can see the effect of the massive bushfires - which burnt all around where I live. It's the green swirls.

Professor Lohmann explained that the dust mainly comes from the Sahara, plus you can see the dust over Australia, too. The green colour is the organic matter and black carbon from wildfires and burning biomass, as well as from trees warming in the summer (see this NASA article). She also pointed to the sulfate aerosols, some of which come from natural sources others from industry as shown by the white blotches over China.

The patterns of aerosols happen because they only last in the atmosphere for a short time - from days to weeks. They don't have time to mix well, unlike a long-lived gas like CO2. That makes them "harder to understand because they are spatially variable". It's hard to pin them down to study the effects in vivo!

Scientists do know this much. It's complicated :)
  • Some aerosols scatter solar radiation, which leads to cooling locally and when spread out leads to a more general cooling.
  • Some, like black carbon or soot, absorb solar radiation, which leads locally to a cooling because for a bit it stops the solar radiation from reaching the surface, but then "if you wait some time this turns into a warming".
  • Aerosol particles also influence cloud formation.

Prof Lohmann put up some nice diagrams showing the above, and also showing "clean clouds" and "dirty clouds", the latter being contaminated by aerosol pollutants. In fact, as she said, without aerosols there wouldn't be clouds forming as they do. Water vapour would need to be a lot more concentrated to get clouds. Without aerosol particles, relative humidity would need to be several hundred per cent to form a cloud.

A polluted cloud has more but smaller water droplets. This increases the overall surface of the cloud and reflects more radiation back to space. She also said that it's possible that having more and more smaller water droplets may make it harder for them to join together for precipitation - though that is not known with any certainty.

To look at the effects of aerosols compared to clouds with no aerosols, Prof Lehmann went to the Arctic in summer, where the aerosol concentration was more than 100 times less than that on the continents. She showed this video from YouTube. Watch it. It's a fun demonstration comparing clean and polluted air (using pollution from a cigarette lighter compared to no aerosols):

She then referred to the radiative forcing chart from the IPCC AR5 WG1 report, which you've probably seen before.

Figure TS.6 Radiative Forcing (RF) and Effective Radiative Forcing (ERF) of climate change during the industrial era.  Forcing by concentration change between 1750 and 2011 with associated uncertainty range (solid bars are ERF, hatched bars are RF, green diamonds and associated uncertainties are for RF assessed in AR4). Source: IPCC AR5 WG1

The chart distinguishes between the forcings in regard to confidence levels, showing that the aerosol-cloud interactions have high uncertainty (the bar) and low confidence (far right). Ulrike explained that as said before, one of the reasons for the uncertainty is because aerosols are short-lived and move about a lot. They are spatially inhomogeneous Also because clouds have very small-scale structures.

She then talked about how to find clouds formed by aerosols and put up a MODIS image showing the clouds from ship tracks, which were formed by aerosols in the Marine Boundary Layer. (The Marine Boundary Layer is the part of the atmosphere that is in direct contact with the ocean and therefore influenced by the ocean.) I found a YouTube video showing this, from a NASA website. The NASA webpage describes how the clouds are seeded by particles in the exhaust emitted by ships.

Professor Ulmann showed a chart demonstrating how aerosol particles in a ship track were observed with aerosol concentrations at ten times what it is outside the ship track, plus more. Apparently ship tracks don't form very often. Particular conditions are needed in the Marine Boundary Layer for them to form.

Now this is already quite a long article and I haven't even got to the second point, which is about how aerosols influence ice clouds. This is her area of special interest so I shouldn't stop writing at this point, but I will - and just relax and watch the rest of the video. If it's anything like what I've seen so far, I'm going to learn a whole lot of new stuff.

I might come back and add some to this article (but no promises), in which case I'll add it as an update and put a link to the update up the top of this article.

As I wrote earlier, you can see the video for yourself on the AGU virtual options page. Look for the Charney Lecture. Here's the code and title again: A21N Atmospheric Sciences Charney Lecture. And here's the link explaining how to get to the AGU14 Virtual Options (to see the videos and poster sessions).

About Professor Ulrike Lohmann

A short bio from ETH Zurich:

Ulrike Lohmann is Full Professor for Experimental Atmospheric Physics in the Institute for Atmospheric and Climate Science since October 2004.
She was born in 1966 in Berlin (Germany) and studied from 1988 to 1993 Meteorology at the Universities of Mainz and Hamburg. In 1996, she obtained her PhD in climate modelling from the Max Planck Institute for Meteorology. Prior to her current appointment, she was a post-doctoral fellow at the Canadian Centre for Climate Modelling and Analysis in Victoria and an Assistant and Associate Professor at Dalhousie University in Halifax (Canada). She was awarded a Canada Research Chair in 2002 and was elected as a fellow of the American Geophysical Union in 2008.
Her research focuses on the role of aerosol particles and clouds in the climate system. Of specific interest are the formation of cloud droplets and ice crystals and the influence of aerosol particles on the radiation balance and on the hydrological cycle in the present, past and future climate. She combines laboratory work, field measurements, satellite data and different numerical models.
Ulrike Lohmann has published more than 180 peer-reviewed articles. She was a lead author for the Fourth and Fifth Assessment Reports of the Intergovernmental Panel for Climate Change (IPCC). She is the coordinator of the EU FP7 project BACCHUS. At ETH, she is the head of the Institute for Atmospheric and Climate Science since 2006.


  1. Anyone interested in clouds could do worse than read 'The Cloudspotter's Guide' by Gavin Prector-Pinney (2006). Brilliant layperson introduction to both the aesthetics and science of clouds.

    You might even consider joining the 'Cloud Appreciation Society' (Gavin is the founder). One of their manifesto pledges is "to fight 'blue-sky thinking'". Who could argue with that?

  2. Sou,

    Gosh that GEOS-5 aerosol image is fascinating. I need a poster of it, stat. Interesting to see climate engineering with cirrus clouds featured. Do you have any sense of how that portion of her lecture was received?

    Skipping down to the AR5 radiative forcing chart, I think my brain almost understands the distinction between RF and ERF but I'm not sure I grok it completely. How I would say it is that RF is the straight up direct effect, ERF is like a net effect after interactions with other things in the system. Is that about right?

    1. PS, not to be missed, GEOS-5 aerosols in motion. I knew there had to be one:

    2. Thanks, Brandon. I've added it to the article. You can see lots in the video. I noticed the bushfires in Victoria (SE Australia) show up really strongly.

    3. You bet. The hurricane type eddies tracking across the Atlantic and Pacific were quite cool. Various places where I'd expect wild fires and/or agricultural burning were represented well. Sumatra for instance. I couldn't quite pick out Java in that plume, but I know from experience there is always something on fire in Jakarta. About midway through is a colossal explosion of white stuff in sub-Saharan Africa I can't fathom. An intentional what if, perhaps.

    4. Here is the definition from Chapter 8 of AR5:

      The two most commonly used measures of radiative forcing in this chapter are the radiative forcing (RF) and the effective radiative forcing (ERF). RF is defined, as it was in AR4, as the change in net downward radiative flux at the tropopause after allowing for stratospheric temperatures to readjust to radiative equilibrium, while holding surface and tropospheric temperatures and state variables such as water vapor and cloud cover fixed at the unperturbed values.

      ERF is the change in net top-of-the-atmosphere downward radiative flux after allowing for atmospheric temperatures, water vapour, and clouds to adjust, but with surface temperature or a portion of surface conditions unchanged. While there are multiple methods to calculate ERF, we take ERF to mean the method in which sea surface temperatures and sea ice cover are fixed at climatological values unless otherwise specified. Land-surface properties (temperature, snow and ice cover and vegetation) are allowed to adjust in this method. Hence ERF includes both the effects of the forcing agent itself and the rapid adjustments to that agent (as does RF, though stratospheric temperature is the only adjustment for the latter). In the case of aerosols, the rapid adjustments of clouds encompass effects that have been referred to as indirect or semidirect forcings (see Chapter 7, Figure 7.3 and Section 7.5), with some of these same cloud responses also taking place for other forcing agents (see Chapter 7, Section 7.2). Calculation of ERF requires longer simulations with more complex models than calculation of RF, but the inclusion of the additional rapid adjustments makes ERF a better indicator of the eventual global mean temperature response, especially for aerosols. When forcing is attributed to emissions or used for calculation of emission metrics, additional responses including atmospheric chemistry and the carbon cycle are also included in both RF and ERF (see Section 8.1.2). The general term forcing is used to refer to both RF and ERF.

    5. Ayee, thanks for the legwork on that one, I feel like I should have been able to find that myself. So the distinction mainly hinges on the time lag between the relatively immediate stratospheric readjustment to radiative equilibrium and the slower surface and trophosphere responses. I think it finally begins to gel for me.

  3. Ulrike Lohman is one of my research heroes from the time I was working on cloud structure. She has such an enormous wide range of knowledge, from cloud microphysics (all the different cloud particles and how they form) to cloud dynamics, climate modelling and the assessment of the influence of clouds on climate (radiative forcing). Because clouds are so complicated and so variable on all space and time scales, they are hard to study, but she always finds beautiful creative ways to "isolate" single problems to be able to study and understand them in detail.

    She also said that it's possible that having more and more smaller water droplets may make it harder for them to join together for precipitation - though that is not known with any certainty.

    I think that the fact that it is harder to make precipitation with smaller droplets is not that uncertain. Although the quantification surely is difficult. What she probably wanted to say is that it is uncertain how that changes the life time of cloud and what that does to the radiative balance of the Earth.

    If you have a typical cloud droplet of 10 micrometre, 0.01 millimetre, and a typical rain droplet of 1 millimetre, you see that the difference in size is about a factor 100. That means that the difference in volume is 100*100*100 = 1 000 000. Thus you have to combine 1 million droplets to get one rain droplet. If they are smaller, even more. A further complication is that small droplets have almost no vertical velocity and thus also almost no differences in vertical velocity, which they need to bump into each other.

    One of the problems which are still not fully understood is how it is possible for a cumulus cloud (a shower) to start producing rain so fast after its first appearance. There are claims that the rain can start within 15 minutes and that is something our cloud models cannot reproduce (without "fudging", for example assuming that there were a few large droplet or dirt particles present already in the beginning, which might be true, but might just as well not be true).


Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL. Details here.

Click here to read the HotWhopper comment policy.