I am giving/gave a talk this morning at a Festschrift celebrating the 25th anniversary of the Graduate Ecology program at the Universidade Federal de Minas Gerais (UFMG), the large state university in one of the larger states/cities in Brazil. So first congratulations to the program and many thinks to the organizers (especially Marco Mello, Adriano Paglia and Geraldo Fernandes) for inviting and hosting me.
I was invited to give the talk based on my blogging, which is sort of a new trendy thing in ecology. So I foolishly offered to give a perspective on the past 25 years of ecology and what the next 25 years of ecology will contain, because I like to think about such things. But as I prepared my slides I increasingly got nervous because these are topics no one person should claim expertise on!
However, I did come up with a couple of data-driven graphics that I thought readers might find interesting.
Publication trends
First I did some statistics on rates of publishing by country (using Web of Science so biased to English journals). I picked out the US, several European Countries and Brazil and China. What would you guess the trends are? First, the total # of papers published per decade is increasing at a phenomenal rate, so everybody is publishing more. But as a percent of published papers, most European countries are holding steady (although some countries like Germany started to publish in English later than other countries like Sweden so they show a big increase in the 1980s or 1990s), the US is slowly declining and China and Brazil are increasing rapidly.
According to Web of Science which is English journal-biased. RoW is rest of world.
Research topic trends
Secondly, and more interesting to me, I did a Wordle on the titles of the top 200 cited papers in 1989 and the top 200 cited papers in 2012 (yes it is 2014 but I found I had to go back to 2012 to get citations that had settled down to papers that were truly the top instead of just the ones published in January).
The two Wordles are for 1989:
Word cloud for titles of top 200 cited papers in 1989 (click for a full size image)
And 2012:
Top 200 for 2012 (click for full size image)
There are some obvious differences. But before I comment, I am curious to see what you all see (that is the point of a word cloud after all) I hope you all will share your thoughts on what has or has not changed in 25 years (OK 23). I’ll try and add my thoughts in the comments after others have had a chance at analysis.
PS – if you’re curious you can download my slides for the talk from figshare. The first 1/3 matches what you read above. The last 2/3 mostly matches themes I’ve hit on before here in my posts on DE. Although students might enjoy the next to last slide on advice to students.
Dynamic Ecology 
The obvious thing that jumps out is that the term “ecology” is used less frequently in titles. Is that a sign of a field becoming more established and/or sure of itself? How many physics papers use the term “physics” in the title, for example? Ditto chemistry?
Within these broad trends there are some interesting smaller patterns – if you’ve not seen it, this old post of mine is relevant here and might be of interest: http://jeffollerton.wordpress.com/2013/06/17/the-cliff/
Marco and Adriana were great hosts when I visited UFMG last year, please pass on my regards if you’re still there.
Hi Jeff – yeah visually the “Ecology” is the single biggest change. I took it to mean we’re publishing less papers with titles like “The ecology and distribution of organism X”. But I like your interpretation. I do have the source data (200) titles. I will have to go look. And thanks for the link to your post.
So looking at title’s there are a lot of grand proclamations on the state of the field (e.g. Wien’s “Spatial scaling in ecology”) and a very large fraction of the ecology usages are “landscape ecology”.
So to be honest other than the peak of landscape ecology in 1989 which is notable in other words too (like landscape), I’m not sure what to make of the frequency of the use of the word ecology (maybe 1989 was a year of summarizing and pronouncing on the state of the field).
Hi Jeff! You’re welcome! Next time we’ll take you to the rocky meadows too.
Thanks Marco, I look forward to that 🙂
My speculations: I see the rise of “Species” and “Distribution” as a consequence of the stronger focus on biodiversity, with “traits” as a tag-along when trying to explain species differences. From the 1989 cloud I like Jeff’s explanation for “Ecology”, but also suspect that might be due to a stronger relative influence of systems/ecosystem ecology(?), which might fit with more prominent “landscape”, “use”, “water”, “organic” and “effect”. That a stronger methodological focus has led to more “model” feels obvious, as well as larger data sets/spatial scales has led to more “global” and “distribution” (along with “climate” “change”). I’m surprised that gene/genetic hasn’t made more of an impact in 2012. It would also be interesting to see how these patterns/differences are influenced by the fact that they are based on the top-200 papers from each year, as compared to all papers from each respective year. The 1989 patterns should be more influenced by what has actually paid off (as in leading to many citations over the long term), while the 2012 pattern should be more influenced by direct impact/hot topics (papers quickly gathering citations over 2 years).
I’m with Jeff and Brian on the decline of “The ecology of…” papers since 1989. (Instead we seem to write papers on “The effect of…”, thus explaining the rise of “effect”). We’re much more quantitative these days, hence the rise of “model”. I’m with Tobias on the rise of “species” indicating the rise of biodiversity (and the associated “trait” paradigm shift/bandwagon depending on your point of view). That’s probably also why “community” is a big riser–if you’re studying diversity, you’re almost necessarily studying community ecology. The rise of “distribution” I think indicates the rise of biogeography, global climate change ecology (“global”, “climate”, and “change” are also big risers), and species distribution modeling (aka “niche” modeling, thus partially explaining the rise of “niche” and “model”).
“Microbial” ecology is on the up, thanks to gene sequencing technology.
Interesting that “landscape”, “scale”, “cover”, “use”, and “history” are among the big decliners. Did landscape ecology and studies of land use change and land use history peak in the late 80s? At least in terms of highly cited papers?
I thought the term “invasive” would have been much larger.
Keep in mind that this is only the top 200 papers bycitation, so it is biased to papers that are of interest to many fields of ecologists (one of the reasons that methodological papers feature prominently if you look at the actual list of titles). I wonder how different the results would be if it were done on all papers or the top 10,000. I do think invasive would be much bigger
So some of the things I notice in the word clouds:
1) Landscape ecology, spatial ecology, as well as scale/grain/extent all peak around 1989 and then decrease to 2012
2) Evolution, life history and related topics were bigger in 1989 than 2012, I consider this decline unfortunate.
3) As others noted, diversity, biodiversity and species have all risen rapidly
4) Global change and climate change have increased a lot like one would expect
4) Niche had a revival (I’m sure it would have been big in the 1970s too but its very small 1989
5) Many new areas like microbial ecology, phylogenetics appear
6) Traits increased way more than I thought
7) Marine and corals also increased – maybe that artificial barrier is finally breaking down a bit?
Increase in corals and marine maybe another signal of increasing focus on climate change and biodiversity?
And what do people think of the 4x increase in # of papers published in 25 years. Everybody knows papers are increasingly rapidly, but that amount is mind-boggling to me. I am quite certain we don’t have 4x as much good/useful science being done as 25 years ago. The Red Queen publish-or-perish effect is out of control!
Would be interesting to break that down into increases in the size of established journals vs. founding of new journals. And to associate with some measure of the fragmentation of the field. Because personally, it doesn’t *feel* to me like there’s 3-4x as much stuff coming out that I need up with, compared to when I started grad school in 1995. Presumably because I don’t feel any particular need to keep up with the venues or subfields where a good chunk of that 4x growth has occurred.
Isn’t some of the increase an artefact of WoS starting to index more journals over the years that they have not included as back issues? I recall reading somewhere that this needs to be taken into account when assessing trends (though I don’t remember where).
That’s going to be minor compared to the real increase though. For some subfields the increase has been 10-fold even within a fixed set of journals, as I showed for pollination ecology in the link I posted above.
Brian, I liked your slides,
Glad to see “Marine” rose into the 2014 word cloud.
But I have to ask, what is “publication fact #1”?
Hi Andrew – yes I’m sure somethings were lost without the speaking to go with them #1 is the fact that I can take a draft manuscript and make it look like a typeset professional (or at least almost professional) document complete formulas and pictures in 5 minutes (more time would obviously be more professional). Fact #1 is we do not need publishers to typeset for us. For a long time this was a real value added but no more.
Ha! I was staring at the journal page, trying to figure out the message and did not realize it was homemade in 5 minutes. Point made.
I’m glad someone asked this question. I was in the audience, but I couldn’t understand this portion of the talk. I was never the best student in English classes.
Just a little observation:, the correct names of the professors are Andriano Paglia and Geraldo Fernandes.
And thanks for the great talk. It was very interesting and inspiring.
Adriano*
Thank you so much for coming to our symposium, Brian! Your talk was very inspiring and surely gave us a lot to think about. Things are changing at a fast pace in Ecology and science in general, and we want to stay fine-tuned with the zeitgeist. Soon we’ll post some ideas in the symposium’s blog: http://simposioecmvs2014.wordpress.com.
Interesting stuff, Brian! A couple of observations:
1) All “Fig. 1: this temporal keyword search in WoS shows how important my topic is!” introductions in manuscripts must be divided by the values in your “Ecology Papers – whole world” to have any meaning. I’m amazed the raw data are still allowed to be published anywhere (but they are).
2) By “England” do you mean “UK” in the 2nd figure? Most research in the UK is funded by RCUK – as far as I know, there’s no “England only” equivalent. Even after September the 18th, Wales and Northern Ireland will still contribute to and receive money from RCUK.
It’d be nice to see this info on a per-capita basis. IIRC Nordic countries do very well (but also tend to invest properly) in such comparisons. However, they also do well in terms of research output per unit of money (€, $, £) invested.
Finally, I’d love to see a word cloud from the 1970s to have even more historical context! Does “Model” cycle in and out of fashion?
Fully agree on #1 – basic fact of academic life # of papers is growing exponentially – exponential growth is so strong that basically means # of papers iis growing by any break out you can conceive of – by country, by university, by topic, etc are growing, so including an appropriate denominator is essential.
Re #2 – WoS doesn’t use UK as a country in its country field, it uses England (and presumably Scotland and Wales and N Ireland but I didn’t check). This came to my attention when I found zero papers when searching for # of ecology papers in UK, United Kingdom and variations. To get a UK trend you would have to add these back together. I was lazy. Presumably adding Scotland in particular would increase the share attributable to the UK. Presumably the trends would be the same though.
It would indeed be interesting to see some more word cloud data points. Maybe there is a future blog post for me..
Pingback: What are the big ecological innovations of the last century? #ESA100 | EcoTone: news and views on ecological science
Pingback: Changes in research interests in amphibian ecology and evolution over the last 20 years | Germán Orizaola
Pingback: Guest post: Doing ecology on a rollercoaster in Brazil | Dynamic Ecology