An honest comment

The party is over, EGU 2019 is history. Over the days I have written some impressions I got from the conference, but sticked mostly to the positive points (Day 1, 2, 3, 4, 5). As always, the EGU itself, the opportunities it offers to talk to all kind of scientists and listen to their advancements was amazing. But compared to the last years the organizers have implemented several changes to the program, on which I would like to address a few honest comments. I am very aware that there are many people in the background who do a hell of a job to make this conference possible and they face the impossible task to make everybody happy and the challenge that every year more and more scientists want to attend the conference.

The topics I want to address in this post are the following

  • increase from 6 to 7 talks per session
  • limit to one contribution per scientists (instead of two)
  • poster sessions in parallel to talks
  • networking session instead of poster session in the evening
  • general comments

Most of these topics came up in many many conversations during the days with other scientists, people are complaining every year (usually about the drinks and the rooms for certain sessions), but this year there were constant discussions about all sorts of complaints.

Number of talks per session

In order to have enough time for the in parallel running poster sessions the number of talks was increased from 6 to 7 per session. A reason given was that more scientists want to have a talk rather than a poster and therefore there should be more space for talks. I personally have expected a worse impact by this change. Yes, surely, after six talks you are tired and do not listen anymore to the last one as you should, but as many do not follow the whole session anyway (see poster parallel to talks below), it seems to work. What I find more problematic is the notion that only because scientists prefer talks they should given more of them. First, this devalues posters and second, talks are up to now seen as something like an award for a good abstract and an interesting topic. Devaluing posters and talks at the same time is certainly not good, but as we have to see it as a business decision, it is acceptable. Especially as it has a reason behind it (more time for posters).

Limit for contributions

When I found out about this change I was honestly deeply worried. I have made the argument in the past that this change will heavily impact early career researchers.

The argument for this is two fold. On the one side the big shots will get a talk. Generally EGU is good in promoting young scientists to get talks at their conference. And with more talks available, it is hoped that the opportunities are not diminished for the ECR’s. I have no overview whether this has worked, but I give the benefit of the doubt. The other side is potentially much more damaging. In theory most senior scientists will now have only one abstract and most of them will end up in talks (I neglect at this points that most senior scientist are able to put fake first authors at the first place of the author list to have as many contributions as they want). That means, they will not have a poster where they are obliged to fulfill their attendance time. And it is the poster sessions where ECR’s have a chance to get in contact with the senior scientists, might it be to make their names known or make important steps for their career. By risking to loose these opportunities (see someone you are interested in a talk and then go to their poster to talk with them), ECR’s are in risk of loosing chances to progress. I understand that there are now too many scientists to give everybody a second contribution, but I am not really convinced by this argument. Because when you walked through the poster halls, many many lines in of poster boards where empty. So there is space for more posters. Might there be ways of change? Sure, but they might all be more complicate to implement (allow two contributions, with a first and a second, but the second is not guaranteed to be allowed to get through… but as many author lists are cheated it will be hard to implement as scientists will find ways around it).

Empty poster boards over the day

Poster sessions in parallel to talks

The biggest change was to put posters in parallel to talks. It is not uncommon to do this at conferences, AGU is famous for that. Nevertheless, I do not think that this is a good decision. There are massive consequences for the conference, not only for the posters, but for the talks as well. Compared to my last experiences at EGU I got the impression that more scientists leave the talks in between. This leads to disturbances, especially in the small rooms. As they are overfilled anyway, people leaving are ending in pushing and and asking for letting them through. Also posters are devalued by this. I was surprised that in the poster session I attended there was still proper crowd (and that at the end of the conferences), but in the end you have now to choose: see the headline science in a talk or much more of potentially non-headline stuff in a poster session. It might work when you are a disciplinary scientists (as I have written before, EGU is not really prepared for interdisciplinary visitors) as ideally there are just a few sessions you really care about and those are all coordinated that the poster sessions do not overlap the talk sessions. But honestly, this hardly ever is the case. As a consequence I have certainly missed out of many interesting discussions at the posters this year as well as many talks (usually ECR talks), which I had to skip because I still wanted to have the chance to see scientists at their posters.

The networking session

Networking is important. Anyway, the networking session was introduced to full the space for the poster session. And in my opinion, the introduction of the final hour in the evening has completely failed. At the first day you saw quite a lot of people in the poster halls talking and drinking, the typical catch up at the start of a conference. But the following day a huge amount of scientists used this hour to make an early end to the conference day and went to town for food. So basically the conference was shorten every day by an hour. I also missed the walk through the poster halls at the end of the day. Have a drink, go to sessions you potentially wouldn’t have had walked in on purpose and having nice chats with young and senior scientists. That is for myself (and many others I spoke) the most important part of a conference. This is gone at the EGU. Most scientists take down their poster either directly after their poster session in the morning/afternoon or at the beginning of the networking session. There is hardly anybody at their posters anymore and especially those you want to talk about are gone. In my opinion this has heavily damaged the usefulness of the conference. So at least make the network session another attendance time for the poster. Having one in the morning/afternoon and then another in the evening for everybody who preferred talks over the day would be proper way to minimize the impact.

General comments

Yes, complaints at EGU are the usual, but having this many heard over the days in close to all conversation I had is not a good sign. And we are now not talking about the little thinks like drinks (you can never make that right) or room choices (they are always too busy in the small rooms and to empty in the big ones). And yes, there are some complaints I can fully understand on the minor topics (offering springs to not to use so many plastic bottles, but giving out the drinks in the evening in plastic bottles? Really?!) The topics mentioned above are much more damaging for the EGU and the experience the scientists take from it. It leads to devaluing posters, disturbances in the talks sessions, ECR’s only confronted with people who care about their topic and hardly anymore a wider audience, even more the feeling you have missed important research and opportunities, and the impression that that what many people valued so highly at the EGU is gone. While I do not expect that there will be a roll back of the changes for next year, there will be a bigger topic coming up for the future of the EGU, which I heard in so many conversations: Is Vienna still the right place for the EGU? We all love Vienna, not too expensive, a nice conference centre, good opportunities to live, eat and socialize during the conference. Nevertheless, EGU has outgrown the conference centre. The queues at the small rooms get longer and longer, many do not get in and then look for a second choice. In some sessions the number of people in the room where double the number of seats available. Together with the changes made this year and the consequences that even more scientists want to switch between sessions and posters, I do not see how Vienna can be the long-term choice of the EGU anymore. The alternative would be to actively limit the number of scientists attending, e.g. by rejecting abstracts. Honestly, nobody wants that, but the changes implemented this year haven’t made the conference experience better.

EGU 2019: Final day

The final day at EGU 2019 is done. It was an exciting day here in Vienna, which ended with my poster at the final poster session of the conference. But let’s start in the morning. After hanging up my poster I joined the Energy Meteorology session for a couple of talks. Next I went to the data assimilation and prediction session, which is a stark contrast to the the one before. While in the Energy session data assimilation will be explained in some nice pictures, the latter session is always highly theoretical and filled with a lot of equations.

The second session of the day was for me happening at the dynamics of the atmospheric circulation session. It described well the influence of eddies on the atmospheric circulation and thermodynamics and showed the change of wind shear over time. In between I took a look at a talk about skewed ensembles, which gave me a lot to think about.

After lunch break, which saw a bit of sun, but was at the same time quite chilly outside, I visited a PICO session on paleo-databases. PICO are still not really my favorite type of presentation, but with this topic it was quite alright and let to interesting themes and discussions. Again I interrupted the session to see another talk on Gaussian process regression.

Final session was then the poster session. I showed a topic quite similar to last year. I was surprised that even for the late session I had some interesting discussions as usually these sessions are quite empty. With that the conference ended from my side. It was a successful one from my side of view as many discussions will hopefully open doors for future collaborations. Concerning the conference organisation today I will write another post in the upcoming days, as it will require an honest comment. Up to then, until next time in Vienna.

Attending PALSEA and QUIGS meeting 2018

Just after I have attended a conference on seasonal and decadal prediction at NCAR in Boulder/Colorado, my next stop was at the  PALSEA and QUIGS meeting  in Galloway/New Jersey. It was the third time that I attended a PALSEA and as the other two times it was a great opportunity for me. The main topic of the conference was the sea-level and climate change during the last and other interglacials.

Many talks surrounded the available data and their interpretation. In this context it is a very complex task to bring together on the one side the many different proxies and evidences of sea-level height during that period and on the other side to explain their causes by the changes of ice-sheets in the higher latitudes and the reaction on them by the Earth. As a consequence there were many interesting discussions surrounding this field and many different viewpoints were heard. Specific questions around when the last interglacial exactly started, how high the sea-level was during that period and how the exact evolution  of  sea-level happened during that time were often discussed during this week. Most discussions were evidence driven, trying to make sense of the sometimes contradicting results and their uncertainties.

Also the understanding of the consequences of these results played an important role, as sea-level change is not happening isolated. It requires the build up or melting of ice shoot and with it a change in climate. In a climate system, as we know from looking around us today, that is highly connected all the climate sub-components, especially atmosphere, cryosphere and ocean have to tell the same story.

Myself have presented a poster on the last interglacial sea-level evolution and had several interesting discussions on this topic. Also the field trip, which showed us the study fields of salt marches, which are used to investigate Holocene sea-level change at the East American coast was very informative. Seeing the data collection first hand always helps to understand the topic better and getting better results at the computer in your own office. All in all it was a great opportunity for me, which was made possible by the financial contribution by the organisers, which allowed me to attend.

International Conference on S2S2D

Last week NCAR in Boulder (Colorado) hosted the second edition of the International Conference on Subseasonal to Decadal prediction. It covered the climate prediction from a few weeks up to a few years and hosted with around 350 scientists a good representation of the community in this field. During most of the days the conferences was split into a subseasonal to seasonal (S2S) and a seasonal to decadal (S2D) session.

The International Conference on S2S2D poster

I personally visited only the S2D part, as my current work focuses on this topic. The first day looked into the mechanisms of predictability and the typical candidates, like ocean, soil moisture and stratosphere, were discussed. The second day shifted then more to the modelling of these phenomena. The weather services presented their new prediction systems and new approaches to modelling were discussed. As a third topic covered the handling of the predictions. It looked at calibration and other technique to make the prediction really useful. This lead to the fourth topic, which discussed the decision-making process basing on the prediction. Here, the applications were the main focus points and many different phenomena and their predictability were shown. Topic number five looked at the statistical verification. It presented new approaches to access the skill of the models. The final session of the S2D session looked at the frontiers of earth system prediction and therein especially at the handling  of carbon within the models. Afterwards in a combined session of both parts many different aspects on the future of research in this field were brought up. Among others the topics of temporal dependence of forecast skill and the so-called ‘signal to noise paradox’ lead to a lively discussion.

My personal contributions were threefold. I showed on a poster in the first session how the Summer NAO can be predicted using ensemble sub-sampling. In the second session I presented a poster on the view that sub-sampling can be viewed as a post processing procedure and can so explain why it works. The  talk in the fifth session then covered  the  2D categorical EMD score.

All in all it was a great conference, with  many interesting discussions and a great overview over this interesting field. Certainly many impulses will come from this and will give not only my own research a  new push.

EGU 2018: Poster day

Day three of the EGU 2018 in Vienna and today was the day for my poster. But beforehand an interesting day of presentation sessions was on the schedule. It started for me with a session on data assimilation in palaeo-climatology. As I come originally from meteorology, it is always interesting how the statistical methodology once developed for short term prediction applied onto completely different timescales. Next up was the GIA session, which included some sea-level talks.

In the afternoon the first session was the one on post-processing, in which I also had my poster. Various statistical methodologies and workflows where presented to generate more gain from a dynamical (weather) forecast. Final presentation session was then on corals and their ability to give us information mainly of the ENSO in the past.

The final session  of the day was then the poster session. I had nice discussions on my topic of statistical-dynamical prediction and my take on why it works. Tomorrow will be the day of my talk, where I will present an alternative to the common used ACC and RMSE.