Data cleaning – finding the problems with turfmapper

All large data sets have problems. The community data gathered in the Between the Fjords projects are no exception – there are inevitable spelling mistakes, synonyms, misidentifications (for example of sterile Carex). Fortunately, having community data from several years, we can compare the community over time to identify and then fix these sorts of problems.

The first step to fixing the problems is to find them and data visualisation is key to this. I’ve written a small R package called turfmapper which takes community data and makes a figure like this.

Turf map from a Landpress site.

A cell is coloured if the species in present in that subturf in that year. The intensity of the colour indicated the cover of the species. Species that suddenly appear or disappear might need checking. In this turf, there are problems with Festuca and some other taxa that need fixing.

The vignette in the package shows how to make these figures for all turfs in a project using Rmarkdown.

Decomposition in alpine grasslands

Our paper on decomposition processes in Norwegian alpine grasslands is just published online in Ecosystems :D. Go check it out!

Long-Term Climate Regime Modulates the Impact of Short-Term Climate Variability on decomposition in Alpine Grassland Soils

FunCaB has a logo!

A picture speaks a thousand words. They’re great ways to communicate and summarise a lot of information in a very compact format. They’re also great for making links and associations. This is why we use logos, right? I had a go building a logo for the funcab project, which (hopefully!) distills the main aims of the project into one small image.


LandPress on radio!

img_4346This spring we brought EKKO-journalist Ivar Grydeland to Lygra. The trip resulted in two reportages on the radio program EKKO on P2. Click on the links below to hear about prescribed burning and the old Norse sheep (in Norwegian).

dsc00648Prescribed burning on Lygra:

The old Norse sheep:

img_4349Vigdis is explaining the processes above and below ground after a prescribed fire

Singin’ in the rain

This week feels like proper Norwegian fieldwork again. Rain drips off your nose, water runs up your sleeves, and data sheets are hung up to dry like washing in the evenings. Nonetheless, we are over halfway now in terms of vegetation composition analysis (check out the map below!), and by the end of the week we will have a further three sites completed.img_1528 img_1529 img_1535seedclim_sites_w2-3

Back in the field

It’s been a beautifully warm and sunny week in western Norway for doing fieldwork which, of course, we took advantage of! I had just enough time back in Bergen after the conference last week to do my laundry and re-pack for fieldwork.img_1377


dsc_1288There’s still a bit of snow in the mountains


Happy plant ecologists after completing our first site of the field season!


Veronica officinalis in flower (left); a not uncommon site when counting seedlings under a dense vegetation canopy! (right)

Thus far we’ve completed the vegetation surveying for two of our twelve sites. The ‘bryofighters’, our trusty plant removal team, and the carbon flux analysers, are also doing their rounds.


Spreading the word: presenting at the IAVS conference

Fieldwork for plant ecologists in Norway normally begins in June, and the hard work we put into our site maintenance and data collection pays off when we are able to share our findings with vegetation scientists from around the world. From the 20th to the 24th of June 2017, a conference was held on the beautiful island of Sicily, and for those who could take time out of their busy field schedules this was a great opportunity to catch up with colleagues and present, discuss and disseminate our work.

This conference was organised by the IAVS, an international association for plant and vegetation scientists from around the world. This was the 60th annual meeting that the IAVS has organised and its theme this year was extremely relevant to the research we conduct in Bergen: “Vegetation patterns in natural and cultural landscapes”. The programme covered topics from invasive species and conservation planning, to diversity patterns in grasslands, and varied from strictly theoretical to more applied approaches. And what a location! Palermo is beautifully situated on the north coast of Sicilly, and the shorefront was meters from the entrance of the hotel. The posters were presented in the botanical garden in the open air, surrounded by citrus trees and cacti. This is the first outdoor poster session I’ve attended – if only Bergen had such predictably good weather!

There was a lot of good work presented, much of it relevant to our own experiments. It was highlighted several times how important land-use change is in addition to climate change, and that the scale and method used in experiments should influence the interpretation of findings from ecological studies in grasslands.

Our own study sites in Norway are part of a cultural landscape, owned and managed by landowners and farmers. It is always exciting to present our findings on an international platform, which is only possible through these collaborations.

The presentations and posters were complemented by a mid-symposium excursion, giving all participants a day to digest the overwhelming amount of information we’d heard, and a chance to network in a relaxed setting and see a bit of the island. Eight excursions took us in groups to various parts of Sicilly to look at the flora and cultural landscape.

Aside from conferencing, there was also time to explore the city, and go for a few coastal runs!img_1322img_134019457727_10209907040618046_2060785164_oimg_1353

The Ethical dilemmas of the ‘Project’ PhD

Since the 18th Century, the world has undergone an industrial revolution, upturned the consensus on medical care, and invented a whole new way of conducting agriculture. But somehow, the institution of education has stayed somewhere in the past. School children are still subjected to the same teaching style that was used two hundred years ago. Universities are still segregated into the same fundamental schools, maintaining imagined divides between disciplines. Why is it acceptable that we treat infections with antibiotics and plough our fields with tractors, but PhDs are allowed to evolve at a high rate without societal consent?

The PhD was originally what we would nowadays call a Masters – it was a test of your theoretical understanding of a particular scientific field. It was only much later that conducting research was included as part of the qualification, and now the research more or less takes centre stage. Even more recently, many PhDs are now being created within research projects. They are written into grant applications, and are thus one of the project ‘deliverables’. It is now very normal to begin a PhD that is already carefully structured to meet the requirements of the funder. At first glance, this seems like a sensible way to conduct research: feed the student with already outlined research questions and hypotheses, give them a little flexibility on experimental design, and report back on your grant objectives at the end of the project’s life. But we should step back and think about what the purpose of the PhD really is. Isn’t it about developing students into young researchers? Surely a fundamental aspect of their training to becoming a fully-fledged researcher is critically dependent on learning to develop your own research questions? Without pertinent questions, research stagnates.

Despite this, more and more project PhDs are being funded. Students are considered cheap labour in the academic environment, a problem which stems from a scarcity of available funding. This results in top-down pressure on PhD students to narrow their research to answer the specific questions set out in the grant applications. Students are given less responsibility, thus creating a generation of followers and not free thinkers.

There is an argument that if students were really to put a lot more effort into idea development and critical thinking, the PhD would take many more years. However, like all other degrees, PhDs have a lifespan. In the UK, students must submit within three years to avoid failing. Although not as strict, most other countries that offer PhDs will not fund the student once the agreed time has run out, but will still allow the student to defend their work unpaid. The emphasis here is clearly on quantity, not quality. Students who drag their studies out unnecessarily is also not ideal, but some compromise must be found between these two extremes. The maturation of ideas takes time, and if this is to be considered a key part of the PhD training, then it should be prioritised to a greater extent.

If a student is not given the freedom to develop ideas on their own, there is much less room for failure. When experiments don’t go to plan, or an idea doesn’t hold out under theoretical scrutiny, a student learns much more from the experience if the development of the idea began with them and not with a supervisor. It is common knowledge that we learn from our mistakes. The short lifespan of a PhD often does not allow for such learning curves though, in particular with so much pressure to produce a suite of peer-reviewed articles. It makes you question what the real purpose of a PhD is: once a student has completed their training, they (potentially) have an entire lifetime to produce scientific papers. After the PhD they do not, however, have that ‘safe’ environment in which to test their wings, so to speak. From discussions with fellow PhD students, it has been mentioned that this training stage should come much earlier, during the Masters. But here I strongly disagree. Yes, the Masters is another crucial step in the development of young scientists, but it is an opportunity to be given an idea which the student then implements, experimentally or theoretically. The PhD, on the other hand, should then have more emphasis on that first stage – the development of good scientific questions, and on the final stage of communication.

Of course, being able to communicate your findings to your peers in the scientific community is an essential part of the training for a career in science, because this is currently the way in which science progresses. Indeed, this aspect of the training is often underappreciated, resulting in subject-specific jargon that prevents collaboration between disciplines. With a more universal and accessible language and terminology, a huge number of research possibilities open up, potentially facilitating the answers to some of the biggest questions in science today. I often find I solve many of my own research problems when I explain my work to a non-specialist, and this principle could, and perhaps should, be applied throughout academia.

However, there are far more PhD students successfully defending their theses than there are permanent jobs in science, and this mismatch between student numbers and permanent positions in academia is so large that many must turn to other sectors to find employment. Thus, developing other forms of communication should also be included in the training. Greater successful communication, and implementation, of scientific findings in the public domain draws on other skills. There is a worrying gap between the knowledge produced in academia and its eventual use. I believe this is a niche where many jobs could be created, where findings are better packaged for use in policy and management. But this starts with the training of PhD students, who can go on to bridge this gap.

Currently this appears to be mostly a European phenomenon, but there is a risk of it spreading. Public funding of scientific research is extremely competitive around the world, which could result in the adoption of this blinkered PhD system being adopted elsewhere to the same extent at which it is used in Europe. Researchers are often forced to behave very competitively in academia, to win grant money for their own research. The deliverables written into the grant applications can encourage bad scientific and ethical practice. Students exposed to this kind of behaviour will then go on to exacerbate the problem when they themselves later take on supervisory roles and practice this way of conducting science. There have been incidents in the scientific community of plagiarism and fabrication for a long time already, but perhaps with the rapid increase in the speed at which studies are conducted, these cases may become more widespread. The scramble for scientific recognition has driven some scientists to alter their experiments, or tweak their data post-experiment. Fixation on CV development can lead to the neglecting of other important aspects of a PhD.

This ties into a much bigger issue: creativity is not encouraged in young scientists. The blinkered attitude seen in supervisors can have dangerous consequences, by stamping out creativity in their students. Prospective PhD candidates are a special societal group, often approaching PhDs with heads full of ideas and creativity. If this mindset is not cherished and allowed to develop, we risk losing their invaluable contribution to science and society. The dropout rate from PhD programmes has been associated with the nature of the student-supervisor relationship (Löfström and Pyhältö, 2017). There is a myriad of other ethical issues which arise when we take a look at the student – supervisor relationship, which I will not touch on here. I think it is enough to note that in most, and I would argue all, the supervisor plays an enormous part in determining the outcome of the research and the success of the student following the PhD (Löfström and Pyhältö, 2014). The training provided for the supervisors, if any such training is provided, could be better targeted to address these issues.

Perhaps the solution is not to attempt to solve all of these issues within the framework of the project PhD, but to create an alternative. If we agree that there are two cravings in science – one for novel research that develops theoretical understandings, and one for integrating research into the public sector – then we should probably establish two systems. These parallel systems should be streamlined such that there is overlap in course requirements (and I strongly believe some level of training in philosophy and ethics should be one such requirement) but some differentiation in terms of the degree to which the student can design the experiments and thesis structure, and some specialisation of communication training.

Could doctoral training centres be the solution? Already, the UK is moving to a more integrated approach for the PhD, with a year in industry often included (EPSRC, 2016). Collaborating centres across the country provide specialised training, between which student cohorts move before settling in their chosen field. This is currently offered to Engineering and Physical sciences, but could easily be extended to other fields. A system like this offers the chance for the student to be exposed to a variety of topics, thus providing a broader knowledge base and hopefully more grounded justifications in their eventual research findings. The downside to such a set-up is that the student is not fully integrated into a single research group, and learning to work within a close-knit department is also a part of the learning process within the PhD.

However, the doctoral training centre strategy depends on the current structure of the scientific community remaining intact. An alternative suggestion to resolving the ethical dilemmas of the project PhD is to revolutionise the way that science is conducted. Capping the number of students to be taken in each year, and providing more permanent positions in research, would help to balance out the growing gap in the job market. This, in turn, might help to slow down the frenzied speed at which studies are conducted, allowing more time for reflection on our findings and conducting more thorough checks of our data and interpretation. The way things currently stand, students are inclined to rush through analyses and interpretation in order to publish their findings, purely by following the example set by their supervisors. This does not set a good precedent for the quality of our findings, and it should make us question what the goal of our research really is. As a result of the system that we have all helped to construct, most data collection and paper writing is done by PhD students, and if the mentality within the PhD is to push results into journals as fast as possible in order to staple a few articles together and call it a thesis then maybe we have grounds to be worried. Sadly, the job market bays for more and more publications and citations, adding yet another step into the vicious cycle.

I attended a workshop in November 2016, where scientists, post-doctoral students, and PhD students attended to discuss the integration of ecology and climatology to answer ecological questions at greater scales, and to help validate climate models. During the introduction round, every PhD student but one stood up and introduced themselves not by their burning desire to find out more about a particular area of ecology, or where their research interests lie. Instead, everyone associated themselves with a project. What does this say about the inquisitiveness of the next wave of academics to shape the direction that science will take? If the current societal demands on science shift dramatically in the coming years, I expect there may be a chance to resolve the PhD dilemma by way of student intake control. And yes, unless there is a unanimous revolution within the scientific community, I doubt the situation will change in this direction without societal and political structures changing first. However, there is still hope that the existing PhD set-up might adapt itself, if these issues of funding and training goals are discussed. I think it is about time that education joined the development trend, and at least made an attempt to catch up with its neighbouring institutions.


“Centres for Doctoral Training – EPSRC Website.” Accessed March 15, 2017.

Löfström, Erika, and Kirsi Pyhältö. “Ethical Issues in Doctoral Supervision: The Perspectives of PhD Students in the Natural and Behavioral Sciences.” Ethics & Behavior 24, no. 3 (May 4, 2014): 195–214. doi:10.1080/10508422.2013.830574.

Löfström, Erika, and Kirsi Pyhältö. “Ethics in the Supervisory Relationship: Supervisors’ and Doctoral Students’ Dilemmas in the Natural and Behavioural Sciences.” Studies in Higher Education 42, no. 2 (February 1, 2017): 232–47. doi:10.1080/03075079.2015.1045475.