IMSC2016: Final day

The fifth and last day of the 13th International Meeting on Statistical Climatology (IMSC) has ended and with it a great week here in the Rocky mountains. It started today with the first homogenisation session and the talks covered a wide range. Among this the worldwide organisation of climate data generation, the proposal of a new homogenisation methodology and finally an overview on future challenges for homogenisation. As I had myself worked during my PhD on quality control of data this topic is of special interest for me and I was happy to see this variety of talks in this field.

Low clouds

It was followed with a session on nonlinear methods. As it was the final day, the talks within the sessions covered a wider area, which was good. Finally the day ended for me again with a homogenisation session and as before, the talks were of high quality.

As it was the last day I would like to take a look back on the week. The weather was fantastic, apart from the last day, when the clouds and rain got in. The conference and many talks were really interesting. The mixture of so many different topics gave a great overview on the many flavours of statistical application in climate science. Many scientists, with different backgrounds, on various levels within their career led to a great knowledge exchange and new views on the topics. It was really well organised and so it was easy to concentrate on the good things of a conference. Therefore, the meeting was really worth a visit so perhaps again in three years at the next IMSC.

IMSC2016: Day four

Day four of the IMSC in Canmore and once again many good talks to a wide range of topics. The day started with a downscaling session and covered emulators, handling of natural variability and perfect model frameworks. For me the imitation of complex results by simple models (emulators) have its interesting sides, but also frightens me a bit. I am a big fan of simple models and love to applicate them, but also learned that their results need a lot of statistical handling to deliver some acceptable results in their range of definition. Using them at the border of this definition or even out of it leads usually to inacceptable results. Sure, the simple models designed for the emulators are defined for this task, but it is still a very challenging topic and I am happy that some give it a try.

Conference hotel

The following session of nonlinear methods, with talks about coincidence and network analysis. This was followed by a talk of responsibility of climate science (and their consumers). Apart from some common misuse of statistical methods it also covered the call for communicating more completely the methods and assumptions used in studies. This very important challenge is rather complicate in modern science. Reviewers ask for reducing the method section only to the necessity part and so in many journals their readers just really see the results, without understanding the many assumptions went into it to create it. Statistics is a game of assumptions and so it is essential that they and the exact application of methodologies are added in detail to the papers. Nevertheless, it is the task of the reviewers to ask for it and for the authors to press for it to get it in. “Open methods, open data, open models” are required to replicate a scientific result and that should always be the aim of a publication (and yes, sometimes this is complicate, but trying is what counts here).

After lunch break I visited two more sessions, both covering extreme events. They included many interesting talks with a wide range of topics. Tomorrow will be the final day of the meeting and it will include homogenisation, a topic I really look forward to.

IMSC2016: Own Presentation

It is halftime here in Canmore and the IMSC got its highlights today. It started with a podium discussion on one of the WCRP grand challenges, the one on climate extremes. The aim is that in the next years scientists will collaboratively try to move forward on this and six other topics. Prediction/projections of extremes, may it be droughts or heavy precipitation, is complicate and so the agenda for this topic is long. In the discussion many topics were highlighted, which would help to bring the field forward. The most important of this, at least from my view-point, is the problem of data availability. Still many countries do not share their climate data and still many information can be found in archives, but is not yet digitised. But data is everything (ok, at least a lot) in climate science. Without it new developments and proper projections are not possible and without political incentives it is doubtful whether there will be in the next years a decisive move forward. Another topic of interest for me is the problem about uncertainties and the different understanding of it between the different fields. What is uncertainty? What is included? How it is exactly defined? Many questions around this topic highlight the proble of missing standardisation and an appropiate format in the community to share uncertainty information. Uncertainty is more than just a standard deviation value, much more.

Next were two talks on extreme value theory followed and touched more the theoretical side of the field. The lunch break was filled with an award ceremony.

The next session on my list was on climate model evaluation and covered a wide range of topics. Among this was my talk on the application of the EMD in highlighting differences between different initialisation procedures in decadal hindcasts. It is always nice to show own work, even when this contribution was originally planned just as a poster due to its early stage. A final session on downscalling ended the day for me.

IMSC2016: Poster day

The second day started with a detection and attribution session. The talks ranged from theoretical approaches to the plans for the future large projects in this field. This was followed by a talk on extremes before the next session focussed on teleconnections. In the latter session as expected the EOF analysis got some attention.

From the meeting hall

After the lunch break I enjoyed a session on the great challenges of climate extremes. Talks therein applied statistics to some interesting real applications, so showed that different methodologies can be really helpful for the community.

The final session of the day was reserved for the poster session. For me posters are always a very important part of a conference and it is good when it is placed at the right time into the schedule. It offers people to talk and get to know each other and helps therewith to make a conference a success. Today some interesting posters were presented on a wide range of topics. Tomorrow, at the halftime of the conference, I got my own talk and so I look forward to the remaining days here in Canmore.

IMSC2016: The start

Every three years a group of statisticians and geo-scientist for a meeting to discuss the newest developments of the statistical side of climatology. This meeting, the International Meeting on Statistical Climatology, is an interesting meeting point of many scientist to exchange new ideas. This year, its 13th edition takes place in Canmore, roughly an hours drive west of Calgary, in the Rocky Mountains (so yes, that is the area where the 1988 Winter Olympics took place).


Myself have hoped to visit the conference three years ago, but at that time it had not worked out. Now I got the chance to enjoy the conference and of course, a nice location like this is added extra value. Anyway, the focus is on the statistical part and especially for me to get new ideas and insights into current statistical methodologies. The format of the conference is a quite interesting one. The day starts with two sessions for everyone, where in each session just three longer talks are given. After lunch it switches to two parallel sessions, where six talks are given each. That format allows to get deeper into a topic and still see a lot of new and interesting stuff.

This day started with a nice overview on verification techniques and new methodologies and strategies in this field. It was followed by interesting talks on attribution of extreme events. After lunch I followed a session on climate variability and one on model evaluation. All these topics cover a wide range of statistical topics, which made it a great start for this conference.