As a follow up to last week, below is an analysis of the Earth and Planetary Surface Processes section (EPSP) sessions at the AGU fall meeting — the number of sessions, number of posters, number of talks, and total presentations for the historic record. EPSP sessions started to appear in AGU starting in 2009, so this analysis is 2009-2015. All the data can be found on the the AGU Abstract Browser.

First, the session data — total sessions per year, Poster sessions per year, and talk sessions per year:

Total sessions have increased in number, driven by an increase in talk sessions (12 more sessions in 2015 compared to 2009). Poster presentations have not increased, staying stable at ~25 EPSP poster sessions.

Now the presentation data:

EPSP presentations at the fall meeting have increased by ~300 since 2009. This increase is mostly driven by poster presentations (200 new presentations) vs. the ~70 new talks that have been added. Note also the data on the ‘general geomorphology’ poster session, which draws a steady crowd of ~71 presentations. This seems to be the only consistent session from year to year aside from the Coastal Session (?).

Now let’s dive into the data for all of the poster sessions. EPSP has had a total of 190 poster sessions from 2009-2015. What is the typical session size? The distribution of posters per session can be seen below:

- The minimum is 2 posters in a session
- The maximum is 77 posters in a session
- The mean is 17 posters in a session

Have these numbers changed from 2009-2015? Below is a box plot showing poster session size distribution with range (whiskers), mean (red line) and Q1, Q3 breaks (blue) for each year from 2009-2016:

The mean seems to increase slightly, and the upper range definitely increases. This could definitely be a result of the big poster sessions (General Geomorphology and the Coastal Geomorphology session), which both draw lots of participants. Note that before 2013, the General Geomorphology session was broken into several smaller subsessions.

(this sort of analysis is not really necessary or interesting for talks, which are capped at 8 presentations per session)