American Geophysical Union (AGU) meeting is a geoscience conference hold each year around Christmas in San Francisco. It represents a great opportunity for PhD students like me to show off their work and enjoy what the west coast has to offer.

However, with nearly 24 000 attendees, AGU Fall Meeting is also the largest Earth and space science meeting in the world. As such, it represents an interesting data set to dive into the geoscience academic world. In this post, I explore the data set collected in the previous post.

Organization of the meeting

For a new attendee like me two years ago, the amount of information available is clearly overwhelming. For instance, nearly 3500 poster and 1000 oral presentations are scheduled every day of the conference. As each oral presenter is offered a 15 min time slot, this represents 250 hours of talks and, assuming a reasonable 15 slides by presentations, nearly 15000 slides of geoscience stuff, i.e. a lot to process for a single human being ;)

Nevertheless, the organization of the conference makes it easy to focus on specific field of interest. Indeed, the conference is organized by sessions, themselves organized into 27 different sections which cover almost every possible geoscience subject, from the fluid dynamics of the inner core in the Earth deep interior section to the formation of terrestrial planets if the planetary science one.

Sessions are proposed by conveners and are scheduled in the conference if they manage to gather a sufficient number of abstracts. If a session does so, it is granted with a 2 hours time slot for an oral session and a 4 hours slot for a poster presentation. The decision of whether each contribution goes in the oral or the poster slot is taken by the session conveners. In particular, an oral awarded contribution will get a 15 min time slot to convince other fellows of their good work.

The geoscience community is very large and certain field are much more represented that others. To address this question, I decided to look at the total number of contributors in each section as a proxy for its popularity. Note that, this measure can count the same person many times. For instance, if I participate in three abstracts in the same section, I’ll be counted three times. I could have used the number of contribution by section as a popularity measure instead but this measure failed to account for the collaborative aspect of research. Indeed, the average number of person by contribution is close to 5 and can be up to 7 for instance in planetary science.

According to this measure, atmospheric science is by far the most popular topics in the geoscience academic world for the past two years. It is then followed by hydrology and bio-geosciences. Together, these three thematics clearly dominates the rest of the field. It would be interesting to see if the research funding by field follows the same distribution.

Perhaps surprisingly, historical fields like volcanology, seismology or again mineral and rocks physics do not appear to be the most popular subjects during this week.

Also worth noticing, Global Environmental Change is also part of the top 5. Given the increasing importance of this thematic throughout our society, this is actually not surprising. Taking a closer look, we can see that an important part of the research investigations in this field focus of renewable every (4.7% of all the contributions), carbon sequestration-related problem (about 7%) and global climate-related problems.

Deeper look in the Planetary Science community

The planetary science community, whom I belong during my PhD, is driven by space mission. They bring the data to constrain our imagination and often, lower our expectations. 2015 has been a very exciting year in term of space mission success and this is again nicely reflected in the conference program; the three more popular sessions all deal with important planetary science mission

  • The MAVEN mission from NASA which investigates the martian atmosphere
  • The Rosetta mission from ESA which, after 10 years of inactivity, wakes up and manage its rendezvous with the comete Chury (Churyumov-Gerasimenko).
  • The Dawn mission from NASA which investigates two of the three known protoplanets of the asteroid belt, Vesta and Ceres.

Where do the contributors come from ?

As I discussed in the previous post, the institution for each contributor is also available on the website and therefore, the total number of contributors is also available by country.

The distribution is clearly not homogeneous and in particular, heavily bias towards US institutions. This may simply reflects the fact that the AGU is hold is the US in which case I might have oversold this data set as being representative of the whole geoscience community … On the other hand, it might also reflect the importance of the US research in the geoscience community. The European equivalent of AGU is hold each year in Vienna and it would be interesting to compare both distributions to answer this question.

A second group, which detaches from the others by their number of contributions, is composed by Germany, Japan, United Kingdom, France, China and, to a lesser extent, Canada. Then, the total numbers of contributors drops below 2000. I was really surprised by the fact that only 23 countries were not represented this year in San Francisco. We can further look at the popularity of each topic by country which I plot on the map below.

In general, the distribution well reflects the ressources in the country. For instance, Mexico hosts of many active volcanoes and volcanology naturally appear in first position; Brazil hosts the richest ecosystem in the world with the Amazonia and bio-geoscience appears to be fairly popular there; Greenland focus on the cryosphere …

Conclusion

In the previous post, I discuss how to gather meta data on the American Geophysical Union (AGU) meeting which is one of the largest geoscience conference worldwide.

In this post, we barely scratch the surface of the data set by looking at the various field in geoscience and their relative importance during the conference. We have shown that atmospheric science, hydrology and bio-geoscience clearly are the most popular field during AGU these past two years. Generalizing to the whole geoscience community is tempting but surely would need further analysis. We also have shown that contributors from USA institutions are by far the dominant contributors at AGU which might decrease the view of this data set as an insight into the whole geoscience community. However, it could constitute a very interesting basis to study the dominant trend in each field separately.

In the following (and last) post, we will look directly into the abstract content of the different contributions. This will allow us to build a vanilla collaborator recommendation system based on a higher embedding of the abstract corpus ; ).