Clouds can reflect the visible radiation, here water clouds are more effective. But clouds also have a greenhouse effect. They trap long wave radiation: they reflect back a substantial amount of the outgoing heat radiation, which then leads to warming at the surface. Water clouds and ice clouds have very different properties in trapping the heat radiation. Also the specific composition of the ice particles has a very large effect on this radiation trapping. And this is one of the big questions—how to describe and model this in an accurate way. Already to get precise observations is difficult, since you have so many different possibilities in the atmosphere: different layers of clouds, overlapping clouds etc. Almost infinitely many cloud configurations can occur.
Is the cloud structure not something that one can look at with satellite measurements very precisely?
Satellite measurements are important, but not enough. Especially not, when you have layered clouds, since you only see the top layer. You can use radar measurements, but radar has a limited range and sometimes, it is not that easy to interpret what you see on a radar reflection measurement. A very interesting technique to investigate clouds is Lidar (light detection and ranging; similar to radar, but using light instead of microwaves). There are even satellites equipped with Lidar systems. This has really enhanced our understanding of the cloud dynamics in the past few years.
You work on weather prediction models. Weather prediction is not only about telling people whether it will rain tomorrow but also about sending out warnings for heavy weather events in time. I think of storm surges or windstorms, that can be potentially very dangerous and lead to huge casualties.
One of the main goals, we are pursuing at the European weather centre is to improve short and medium term predictions of what we call ''severe weather events''. Today we can give reliable predictions of such events on rather short timescales. One to two days, I would say. Our goal for the next 10 years is to improve the quality of these predictions on a timescale of two to four days. We are quite confident that we can do this if we get an increased funding for our computer resources and also for maintaining and developing the research we are doing. We need to continuously improve the models and tools—both with respect to forecast models but also to use the observations that we have in the best possible way. We have to be able to describe the current state of the weather system very accurately to make good quality predictions. This aspect takes more than 50% of our efforts. Because the weather system developments are highly initial state dependent, it is absolutely crucial to get the initial state right.
You have done research on Arctic warming, as well. I heard that warming in the Arctic region is almost twice as large as the global average. Why?
This is correct, if you talk about the warming at the surface. There are a number of different factors which can explain this. One factor is the sea ice. When you have warmer temperatures in general, then the sea ice melts. As the sea ice melts, you get less reflection of visible radiation during summer time. That means now, that more heat is stored in the Arctic during summer time. And this heat is released in the dark part of the year, which means that the winter temperature does not go as far down as would be the case otherwise. This is a feedback process, which enhances warming. Another factor is the higher temperature impact of greenhouse gas warming in the Arctic. As we usually have shallow inversion layers in the Arctic a certain increase in the greenhouse gas forcing gives a larger surface temperature response in the Arctic than it does in the tropics.
But this alone cannot explain all of the enhanced warming in the Arctic, there must be other factors. One specific factor which we have done research on is the possible feedback from both the atmospheric heat transport into the Arctic and the transport of water vapor into the Arctic. When you increase the CO2 concentration in the atmosphere, you can see from simulations that also the transport of heat and humidity changes. And this also seems to give a contribution to this increased warming in the Arctic. So there are different factors that contribute, but we still do not have a complete understanding why the warming in the Arctic is twice as large as the global average. I strongly suspect that the missing pieces of this puzzle are again connected to the clouds.
Has this enhanced warming in the Arctic region also some influence on global climate phenomena?
That might be the case, but it is hard to give a definite answer. The enhanced warming of the Arctic could have an effect on the so called thermohaline circulation of the ocean. But this is difficult to say at the moment. For the ocean system, there is still a lack of observations. We need more measurements of the ocean circulation and the ocean temperature distribution. This is necessary to investigate a number of mechanisms that are theoretically proposed. Until now, less resources have been put into deep ocean measurements than for example into atmospheric measurements. This is partially driven by the need for weather predictions but also by technology advances. Considerable resources have been put into the industry that build and develop satellite technology.
In the media, climate research or global warming are often discussed in a very simplified way. It seems, people are looking for short and simple answers, even if the subject is highly complex. How do you as a researcher experience this issue?
Well, since I work here at the European Weather Centre, I concentrate on weather predictions and do not take part in the climate debate. But from the time when I worked as a climate researcher in Sweden, I have some experience with the media and public discussions about climate. I think an important point to convey, is the message that climate research is based on fundamental physics. We have the laws of physics and use those to describe the dynamics of the climate system: atmosphere, ocean, ice and so on. It is not a purely statistical description, in the sense that we would use data from the past to make extrapolations into the future by statistical models. If you can convey that in 30 s during a television interview, then you are really good. But yes, sometimes you have to use simplified arguments to make a point clear.
Prof. Kallen, thank you for the interesting discussion.
Was this article helpful?