Science, People & Politics, issue 2 (March-April), volume ii (2010), VII.
Tillmann Mohr, a special advisor to the secretary general of the World Meteorological Organisation, talks to Science, People & Politics.
Interview by Helen Gavaghan
Thank you for agreeing to this interview with Science, People & Politics. As a special advisor to the secretary general of the World Meteorological Organisation you have one of the most important jobs globally today in the field of weather forecasting and climate change.
I have no doubt. Human activity is causing global warming. It is incontrovertible.
The French National Weather Service reports on its website that the global temperature rise will be between 1.1 and 6.4 degrees centigrade by the end of the century. What do you think of this prediction?
Put simply it depends on the scenarios that were used in the simulations. The French website is reporting predictions for 2100 that depend on carbon dioxide concentrations in the atmosphere. There are models for carbon dioxide from 390 to 1100 parts per million (ppm). Associated with those concentrations there are predicted increases in global average temperature. Many other factors are also at play because carbon dioxide is not the only gas contributing to global warming as a result of human activity. Nor are atmospheric gases the only variable impacting global change. Aerosol are also very important.
Is the decrease in temperature over some parts of the Oceans the reason why climate change modellers are so interested in the way the oceans and atmosphere interact and want to see models developed that "couple" ocean and atmosphere?
Are you saying that if, globally, the increase were 3 per cent by 2100 then the way the temperature increases will vary regionally, and thus the way the climate changes will also vary regionally? So climate change might be very different on the Gold Coast where ocean impacts the region, compared with Chad in the centre of Africa. And is it that the average temperature increase globally will depend on human activity from farming to how many cars are on the road to what fuel we use to make electricity and thus how many chemicals we put into the atmosphere that trap heat on Earth and contribute to a global warming?
How damaging to mitigation strategies, sometimes for decades into the future, are the controversies that have surrounded the quality of data and data sets related to climate change? What questions ought government departments and local authorites (Federal and Cantons, for example) to be asking about the data on which they are making short, medium and long term planning decisions.
The criticisms are unfair. There are mistakes in the IPCC reports but there are hundreds of pages of report. And the errors do not undermine the overall conclusion that human activity has caused global warming.
Unlike Europe and other developing*(I think the previous word probably ought to have been developed, rather than developing. Insert by editor 22.10.11.) countries most developing countries do not have a well advanced climate sector within their national meteorological services. Other than the UK and France, which started as weather prediction services, most of Europe's meteorological services started out as climate services. Africa and south America have the fewest and least developed climate services. The WMO wants to see that situation change. The five categories of work within the global framework on climate services are: observation and monitoring; research and modelling; climate services; climate mitigation; and capacity building. By the way, though the climate services vary considerably all the WMO's member states have access to the same quality-controlled weather data.
How would you define weather and climate? For example when I think of weather I think of Thomas Jefferson recording temperatures and pressures. Whilst when I think of climate I think of it as being the typical atmospheric conditions for a particular geographical location. For example, the UK is damp and the Sahara is dry. Would you mind taking a shot at a more erudite and sophisticated explanation of weather and climate?
Weather is nothing other than physics and chemistry of the atmosphere. Climate is the average of the weather over a given locations over a 30-year timespan. This is a definition that has existed for 100 years.
I understand that all these satellite data - millions and millions of pieces of data twice a day - fit into computer simulations of the weather. So if instrumentation on an ocean buoy records an atmospheric temperature of 10 degrees c at 12.00 GMT on 10th March 2010 in the Bay of Biscay then all over the world there are computers containing grids of co-ordinates - like three dimensional graphs - that this data will be sent to. Some of the grids cover a three dimensional box above the whole northern hemisphere or Europe or just the atmosphere above France, Spain, Portugal and Britain, say. These computers will already have gone through a pretend scenario -- a best guess of what the weather will be like tomorrow, the day after the magazine publishes this interview. So the temperature measured by the buoy tomorrow can be compared with the temperature the computer will have guessed will be tomorrow's temperature. Is that right? Is that how numerical simulation works, guessing from an understanding of weather patterns how the weather will develop and then correcting the simulation so the the simulation starts its next future guess from a more accurate picture of reality?
Not quite. Let me clarify. Take a weather forecast model for two weeks rather than a climate model. At some stage in the beginning real data of pressure, temperature, humidity, wind direction and speed were analysed and with this analysis the forecast run started over a particular time period, say 6 hours. The numerical forecast model runs a set of equations, which describe the atmosphere and the surfaces the atmosphere is interacting with. After, say, 10 minutes of computer time, which might be a prediction of the weather 6 hours ahead, one can ask the computer what its predictions are. These results are checked against reality based on new observations. And a new forecast run is starting. This is repeated until the desired forecast period, say two weeks, is achieved.
What are the most important parameters to measure for weather forecasting and at what scales of time, temperature and radiation frequency and what are the most important parameters for monitoring climate? How has understanding this past 20 years deepened of which parameters are the most significant for weather forecasting and, separately, for climate monitoring?
The physical measurements needed are the same for climate and weather. But for climate one needs to be much more accurate (0.1 rather than one degree centigrade) because of the length of time over which one is averaging. The first World Climate Conference called in 1979 by the WMO established GCOS and the observational requirements were identified as Essential Climate Variables (ECVs). These were the variables that needed to be taken globally to assess the nature of climate and climate change. Within the set of ECVs there are 26 ECVs which can be derived from satellite measurements. These variables set by the GCOS might be calculated from data measured by R&D or operational satellites. Atmospheric ECVs include parameters such as the surface wind speed over the oceans, recordable via active scatterometers detecting the reflected radiation from the dynamic surface of oceans. Other ECVs (and these all need to be cross checked in many different ways) include vertical soundings for temperature, pressure and water vapour. This adds up to four or five million vertical soundings taken over the day. Only now do we have the computers to digest so much data.
Satellites have reduced considerably the errors in numerical weather prediction. Their importance for the quality of weather forecasts is today far beyond those of the conventional data like the measurement of the upper air soundings by ballons. Three satellite instruments at present in orbit called active microwave sounding units AMSU (see links below to a description of the instrument) have contributed alone to an 18 per cent reduction in the errors in numerical weather predictions.
I was a member of the German delegation. German foreign policy in the mid 80s was to increase the number of European international organisations in Germany. The number in Germany of such bodies was low compared with Switzerland, France, Austria and the UK. For example the WMO is based in Geneva and the European Centre for Medium Range Weather Forecasting is in England. In addition to Federal policy the town of Darmstadt in Southern Germany and the Rhein region also had an economic interest in attracting another European intergovernmental organisation to Germany, as did the German Weather Service. The European Space Agency invited those countries which had expressed an interest in forming EUMETSAT to participate in an intergovernmental conference in June 1986. Our delegation was headed by Dr. Heinz Reiser, the director of the German Weather Service. We prepared papers looking at the importance for the economy, foreign relationships and the stimulus EUMETSAT could provide for high tech. industry in the local economy. There were only a handful of people in the organisation when EUMETSAT was formed first in 1986. Now there are 500 people on staff or as contractors.
Both numerical weather prediction and satellites have become enormously important. When I was doing my Ph.D satellites were new. I studied satellite imagery for Typhoon Dinah in 1967 and my thesis in 1970 proposed a mechanism for the way that tropical storms become extra tropical cyclones. At the time of my Ph.D three mechanism were thought to be possible. I concluded from study of satellite imagery that a weakening of the vertical convection plus strengthening of horizontal winds disrupts the tropical storm structure but that when Typhoon Dinah reached the tropical latitudes to the North, where the Coriolis forces are stronger, the storm reformed as a cyclone. The general conditions to be met, I concluded, needed to be a surface pressure low and a fast moving low travelling horizontally in the upper atmosphere. The wind vector fields combine to shape a cyclone. My work had predictive value for cyclone strengths and paths starting with satellite imagery for a few years and then was superseded by numerical weather prediction.
How did you become a meteorologist, what attracted you to the field? Did you realise at the time what a thoroughly international field of study it is? One which simply cannot be undertaken effectively unless one views the Earth as a whole and without artificial boundaries such as State lines.
It was very simple. I went to a lecture by a professor who I thought was interesting. I had no thought of meteorology at the time, but he made both me and my friend laugh -- at him a little bit -- and then he interested us in what he was saying.
The French National Weather service.
Tillmann Mohr advises the secretary general of the World Meteorological Organisation about satellite matters and a global framework for climate services. Dr Mohr headed the German Weather Service (1992-1995) and helped shape the European Organisation for the Exploitation of Meteorological Satellites (formed 1986), a body tasked with relaying weather data from satellites to weather services from the Atlantic to the North Sea, Baltic, Black Sea, Mediterranean and all points in between. EUMETSAT also collects climate data. He was EUMETSAT's director general from 1995 until July 2004. He is a member of the European Union's Space Advisory Group.
Thanks to Martin Redfern Senior Producer with the BBC World Service, for comments about this piece. Martin is also an editorial advisor to Science, People & Politics. Typos, simple errors and production errors corrected within 24 hours of publication and url saved to publisher's record. Final correction 22.26 gmt 9.3.10.
Science, People & Politics©. All rights reserved.