Where have all the insects gone?

Recently, we commissioned one of Journal of Animal Ecology‘s most experienced Associate Editors, Simon Leather, to compile a Virtual Issue on his great passion – insects. The journal has published many classic insect ecology papers over the years and Simon does a great job of highlighting some of these as well as many new papers that we hope will go on to become classics themselves. In his preface to the VI, Simon bemoans the fact that back in the 1970s, when he first began subscribing to the journal, there were many more papers on insects than there are now and that the journal has perhaps become vertebrate-centric in recent years.

This got me thinking – is this really true? And if it is, then why do we publish fewer entomological papers now than back then? Are we alone in this trend or is it common across other general ecological journals? And, either way, should we be worried about the taxonomic distribution of our papers?

The data

As a scientist, I’m not a great fan of anecdotes so I wanted some quantitative data to answer these questions. I went to Web of Science and used a key word search to establish how many papers Journal of Animal Ecology published over the last 40 years on insects and other taxonomic groups. For comparison, I did the same for the ESA journal, Ecology. To account for year to year variation in the number of papers published, I simply calculated the percentage of papers in each category. So, what did I find?

The results

Annoyingly, Simon was correct – as usual (Figure 1). Over the last 4 decades, the number of papers JAE published that included ‘insect’ as a key word has roughly halved: in the 1970s, more than 40% of our papers included insects, whereas in the 2010s it averages at around 20%. Other invertebrate taxa (crustaceans, annelids and non-insect arthropods such as spiders) have either remained constant over time or have also declined slightly.

By contrast, the number of vertebrate papers has increased over the same period. For example, the proportion of papers that include mammals has nearly doubled in the last 40 years, from about 14% to 26%; bird papers increased from about 15% to 30% between the 1970s and 1990s, and has since either stabilised or declined slightly; and the number of fish papers has increased from about 7% to 12%. Of course, many of the papers we publish will include multiple taxa, perhaps even whole communities, so we need to interpret these findings with some caution.


Post 1_Fig 1

Figure 1. Trends in the number of citations per taxon in Journal of Animal Ecology.

This lead to me to wonder whether this trend is shared by other ecological journals and can therefore be explained by genuine changes among the research community. Perhaps the priorities of funding agencies have changed, such that researchers are increasingly working on charismatic mega-faunae to the detriment of more numerous (and some might say, more interesting!) smaller organisms like insects? Or maybe researchers simply can study vertebrates more easily these days with the increased availability of techniques and tools such as remote sensing and bio-logging, for example.

A comparison with our US counterpart, Ecology, suggests that these taxonomic shifts are not universal (Figure 2), and similar trends are seen for Oikos (data not shown). For none of the 8 taxa I searched for was there a major temporal change in its frequency of occurrence in this journal, though as an aside it is noteworthy that whereas JAE now publishes around twice as many papers on mammals as fish, for Ecology this trend is reversed, so each journal has its own taxonomic biases!


Post 1_Fig 2

Figure 2. Trends in the number of citations per taxon in Ecology (data for the period 1978-1990 are excluded due to poor data quality).

Potential explanations for the trends

So, what has brought about the demise in invertebrate studies published in JAE and the upsurge in vertebrate studies?

In his preface to the VI, Simon alludes to potential biases in the senior editors of the journal. Back in the 1970s, the journal editors were often entomologists, whereas latterly they have often been dominated by vertebrate ecologists. However, there has been a fairly frequent turnover of senior editors over this period, and until very recently, the 4 senior editors were evenly split between those that work mostly invertebrates (Mike Boots & Ken Wilson) and those that study vertebrates (Tim Coulson & Graeme Hays).

Based on my own personal experiences as a senior editor, I certainly do not believe there has been any great conspiracy to exclude insects or to promote mammals. Every paper reviewed by JAE is sent to experts in their respective fields and I do not feel that entomologists are any harder on their fellow ecologists than are mammalogists.  Of course, we cannot exclude the possibility that senior editors (or associate editors or reviewers), subconsciously favour taxa close to their hearts at the expense of other taxa.

I suspect, however, that the change is more a reflection of the subject areas that these taxa are the subject of. We probably publish more papers now in the fields of demography, evolutionary ecology, spatial ecology and disease ecology, and these are fields that tend to be dominated by vertebrate systems. These may also appear to be more novel when vertebrate systems are involved and the authors use fancy technological, modelling or statistical tools. So, the challenge for entomologists is to ensure that their studies have sufficient novelty to compete in a highly competitive environment (JAE currently rejects around 85% of the manuscripts we receive). Acceptance rates for insect papers are similar to those for papers on birds and mammals, so the number of entomological papers we publish reflects the number we receive – send us more good papers and we will likely publish more.

Another possibility is that the shortfall of entomology papers in JAE is balanced by an increase in the number of insect papers published in our sister journals, Journal of Applied Ecology – which began in the early 1960s, three decades after JAE was launched – and Functional Ecology – which began in the late 1980s. But, no, a quick look at these two journals suggests that they also publish fewer insect papers now – a decline from ~36% to ~20% over the last 30 years of Applied, and a decline from ~45% to ~30% over the last 20 years in Functional. In Applied, their place has been taken by birds and mammals, and in Functional by birds and fish.

So perhaps this tells us something about British ecologists (or at least ecologists that publish in the British Ecological Society journals)? Arguably, there are relatively fewer insect ecologists in the UK now than there once were, so perhaps this is the cause of entomology’s decline. Or perhaps entomologists are just choosing to publish in more specialist entomological journals where they may better reach their target audience – certainly there has been an increase in the number of specialist entomological journals over this period.

What should we do?

Whatever the reason for the gradual decline in the number of invertebrate studies we publish, should we be worried? And do we need to do something about it?

My feeling is that the apparent slump in insect ecology (in BES journals, at least) is not terminal and there is no need to panic!

For a start, nearly a quarter of all of JAE papers feature insects – comparable to the proportion of papers on both birds and mammals, and a similar proportion to that seen in Ecology. And what about those of us working on earthworms or frogs, I hear you cry, let’s have more annelid and amphibian papers – there are many more of them than birds and mammals! Well, if the number of papers we published on each taxon reflected the number of species on the planet, then for every 1000 insect papers we publish, we should publish just 31 papers on fish, 13 on reptiles & amphibians, 10 on birds, and a miserly 5 papers on mammals! Clearly, this would be ridiculous.

Besides, these things often go in phases – as an example, notice the recent glut of high-profile papers on pollination ecology driven by global concerns over the decline in bee numbers. New advances in molecular ecology, computational biology and remote sensing will likely translate into a fresh wave of exciting invertebrate studies, and we encourage authors to consider Journal of Animal Ecology as a home for these, as well as their (our) more traditional hypothesis-driven insect ecology studies – as always, the key criteria for publication are genuine novelty and broad ecological appeal.

A more important point, perhaps, is that Journal of Animal Ecology focuses on broad ecological themes, principles and concepts, and most of the papers we publish use specific species, taxa or communities as model systems for understanding these. So it shouldn’t really matter how the taxonomic balance of the journal changes as long as the quality of the papers remains high and they help to improve our understanding of general ecological principles. For taxon-specific papers, there are plenty of excellent specialist journals, including one or two edited by our good friend Simon.

The challenge

So, reader, what do you think?

Ken Wilson (Executive Editor, Journal of Animal Ecology)

Posted in Uncategorized | 3 Comments

Name: UK government. Animal ecology test score: 0

© Andrew Byrne

© Andrew Byrne

Every now and again animal ecology findings make it into the news. Press coverage often focuses on cases where a species is on the edge of extinction, has erupted to plague proportions, or exhibits some quirky behaviour. One of the positive things about such coverage is that the public appreciates that animal ecology is a mature field of study that uses high-tech methods of data collection, cutting-edge statistical methods and mathematically elegant models. But all too often animal ecology stories are little more than a curiosity, chosen to fill the ‘And finally…’ slot. Occasionally animal ecology research influences government policy – something that has happened with the control of tuberculosis (TB) in cattle. However, this particular case is not a good news story – sound animal ecology advice is being ignored by the current UK government. The reason? A cynic might speculate that it is because following best animal ecology practice might lead to conclusions at odds with what the government seems unjustifiably determined to do.

I believe that policy should always be guided by the best possible evidence available. If I am offered policy based on science, or policy based on conjecture, anecdote and innuendo, I will go with the science-based view as long as it is ethical and humane. I suspect that such a position is considered rather extremist by the current, and recent, British administrations, but I consider it defensible.

Everyone I have spoken to on the issue of TB in cattle wants it eradicated. I have not spoken to the Secretary of State for Environment, Food and Rural Affairs about it. This is perhaps a little surprising. I was a member of the Independent Expert Panel (IEP) appointed to assess the efficacy, humaneness and safety of the badger cull by the Secretary of State’s department – Defra. In retrospect, I think the IEP should have had access to the Secretary of State so we could present our findings and discuss them directly with him. He might have found it useful. I won’t agree to serve on such a government committee again without agreed access to the appropriate minister.

So what were the culls about? Previous research in England had convinced the government that a reduction in badger numbers of at least 70% would be sufficient to eradicate TB in cattle. But how best to achieve this? Gassing, trapping or shooting at night? Gassing was not an option since it was banned for being inhumane decades ago. So that left shooting and trapping. If badger numbers could be humanely reduced by 70% by controlled shooting then a workable solution to TB in England would have been found. So two areas were identified – one in Gloucestershire and one in Somerset – and planning for the multi-year pilot culls commenced.

The first job of the IEP was to devise methods to assess efficacy and humaneness. The methods needed to be robust to fraud by anti-cull protesters making the cull look less effective than it was, and by contractors returning badger carcasses shot elsewhere to make the cull look more effective. The IEP came up with the following method to assess effectiveness: hair traps were used to sample the badger population in the pilot areas, with individuals uniquely identified through genotyping. Hair samples were also taken from culled animals and individually identified with the same genotyping methods. The proportion of the original sample among culled animals gives an estimate of the effectiveness of the cull. Robust estimates of population size can also be obtained using our approach. The method does make assumptions, and we devised a suite of statistical analyses to check for biases and to estimate uncertainties. Once the cull was over, and all analyses were conducted, we were able to say with 95% confidence that the culls failed to deliver anywhere near the 70% target. The probability of either cull having achieved the requisite 70% or more reduction in badger numbers are similar to me – a middle-aged, overweight, unfit Brit – being selected to captain the Brazilian football team in the World Cup. Zero. The culls were not effective, and we can say that with strong statistical support based on the analysis of high quality data.

The assessment of humaneness is a little less certain, but was based on survival analysis with censoring of animals that were shot at. There is greater uncertainty around our conclusions of this analysis. However, we were able to conclude that it was highly improbable that the culls met Defra’s humaneness target of no more than 5% of badgers taking more than 5 minutes to die.

The IEP also made several recommendations on improvements to the way the cull is delivered that the government accepted. For example, we made recommendations on the way that contractors are trained.

So that was year 1 of the pilot culls. Year 2 is approaching. Given the success of the animal ecology methods used, presumably the government would continue to use these tried and tested methods? Methods that are hard to cheat. Methods based on mark-recapture analysis, which is arguably the most innovative statistical development in animal ecology in the last 25 years. Surprisingly, not, despite the IEP recommending it. The government has not announced exactly what they are going to do, but they will not use methods that allow the effectiveness of the continuing pilots to be assessed in year 2 in the same way they were assessed in year 1. Any results they do achieve will be incomparable. If one of my undergraduate students made such an elementary mistake in an exam essay they would be heavily marked down. A change of protocol half way through an experiment reveals such a limited understanding of the scientific method that I am tempted to speculate that the government no longer wants to know whether the pilots are effective or humane. They just want to cull badgers, regardless of whether the population or humaneness consequences can be assessed.

In addition to changing the protocols, there is to be no more independent oversight of the ongoing culls. So who will oversee the analysis of data and the interpretation of results? The same folk that have decided to change the protocols half way through the experiment? I’m not a dyed-in-the-wool Bayesian, but this is a case where I think I might be justified in working with a well-informed prior that the conclusions will be unlikely to stand up to scrutiny.

Government agencies are stuffed full of very competent scientists. Presumably the concerns that they must have raised are being wilfully ignored by government. I wonder why? I wonder if the government no longer wants to know the answer to whether their ongoing pilot culls will deliver the required outcome. I wonder if conducting the pilot culls is the easiest way for the government to look as if it is tackling the awful issue of bovine TB, even though a large body of animal ecology has concluded it is unlikely to be the solution in England? I fear we may hear that the second year is a success once it is over. But such a statement would be hollow.

Not all government policy can be based on science. Often ministers need to work out how to carve up funds. There may be no right or wrong answer on how to do this, and the decision may be based on who shouts loudest, or what seems ‘right’ given the minister’s philosophy. But when animal ecology – and more generally science – can inform a policy debate, scientific approaches must be used and scientific conclusions should not be ignored. The government’s decision to ignore best scientific practice has not been justified by the Secretary of State. I’d be surprised if he changes his mind. U-turns are seen as a sign of weakness. But what is incredibly sad about the whole sorry affair is we are missing an opportunity to assess whether the pilot culls that the government implemented can solve the dreadful scourge of bovine TB. The existing evidence strongly suggests that culling is not the solution in England, and that the ongoing culls were on course to add more evidence in support of this view. The government’s recent actions rob us of this evidence. And this means we will be delayed in solving the TB problem, that farmers will continue to carry the cost of this dreadful disease for years to come and that badgers will be culled without justification. The issue is not the badgers moving the goalposts as the Secretary of State famously claimed. It is the government. But why they have moved them to make it so easy to score an own goal in the fight against TB is beyond me.

You can download the IEP report here. If you want animal ecology to be relevant to policy, and not just a curiosity used by the media for a bit of light relief, speak up for it! Writing to your MP about it and being vocal on social media is an easy way to make an impact. If enough ecologists speak up for their field, future governments perform better in the use of animal ecology evidence.

Tim Coulson
Editor, Journal of Animal Ecology
(twitter: @tncoulson)

Posted in Uncategorized | 2 Comments