The value of long-term monitoring of field experiments: a case from the sand grasslands of Central Hungary

The post provided by Ildikó Orbán and György Kröel-Dulay

Two subtypes of our studied vegetation type, showing the two species that are competing for dominance – a) Stipa borysthenica, and b) Festuca vaginata (Photo credit: György Kröel-Dulay)

This post refers to the article The role of drought, disturbance, and seed dispersal in dominance shifts in a temperate grassland by Orbán et al. published in the Journal of Vegetation Science (https://doi.org/10.1111/jvs.13199)

Gyuri (György Kröel-Dulay, senior author, who designed and conducted the fieldwork): I started this study in 1999, and I can recall two things that motivated this study back then: the hypothesized importance of seedling establishment in grassland dynamics, and the almost complete lack of experimental manipulations in Hungarian grasslands.

As part of my PhD, I spent one year in the United States in 1996/97 (under the joint supervision of Edit Kovács-Láng from Hungary and Debra Peters from the US), and part of my work was the application of an individual-based gap dynamics model (STEPPE), which was originally developed for North American short-grass steppe, to the sand grasslands of Hungary. In this model, small-scale disturbances regularly create gaps, where individuals of characteristic plant species (lifeforms) of the particular vegetation type establish, grow, compete for soil moisture, and die, which results in equilibrium vegetation, when averaged over long-term and large spatial scale.

After a long process of model parameterisation and reaching a reasonable model outcome, I conducted a sensitivity analysis, testing how changing each of the input variables (one by one) affects final model outputs. This sensitivity analysis highlighted the overarching importance of the „seedling establishment probability” parameter. This made me think about a field study where the seedling establishment is the focus – especially since very little knowledge was available on this from Hungarian sand grasslands.

A more general motivation for the particular study design was that while studying in the United States, I realised that experimental manipulation, as a research approach, was widely used in US grasslands. In Hungary, however, most grassland studies were observational at that time. Standardised and controlled small-scale experimental disturbances with special attention on seedling establishment of the dominant species seemed to be the simplest way to (a) gather information on seedling establishment and (b) gain a general understanding of grassland dynamics in controlled experiments.

Within the first few years of the experiment (1999-2002), we learned much about grassland recovery following disturbances and the huge between-year variability of seedling establishment in sand grasslands. However, one additional factor also came in: natural drought events in 2000, and especially in 2003 had dramatic effects on these grasslands and the recovery pathways after our experimental disturbances. While these unplanned extreme events almost overrode the effects of our previous experimental manipulations, they finally added an interesting twist to the original study design and made it possible to study how drought events and disturbances may interact.

When the effects of original experimental disturbances mostly disappeared in the early 2010s, we almost abandoned the study plots. Still, the repeated drought events in 2012 and 2013 reinforced the importance of drought in these grasslands. Also, by this time, we started to see the true value of long-term data sets: although it requires only two days of fieldwork per year, in the long run, it offers a unique insight into grassland dynamics.

Getting my PhD in 2002, I started new projects (experiments!). Some have been abandoned by now, and some are ongoing, but I always found time to resample these old plots set up in 1999. Since I could not get to the point of writing up the findings from this long-term study on my own, I was happy when Iló joined our team and was interested in the data set and the story there.

The take-home message? Continue monitoring long-term studies and include young colleagues to push the hard work forward.

A final note: in 2022, an unprecedented drought hit the region, and we have 20+ years of continuous reference data sets from the past. What a value!

Iló (Ildikó Orbán, first-author and previous PhD student of Gyuri): As an undergraduate student, I was involved in palaeoecological studies where one can observe vegetation dynamics over decadal, centennial and even millennial timescales! I found it a fascinating topic, but for my PhD, I wanted to get closer to the present and I wanted to be able to directly address questions about global anthropogenic stressors and modern climate change. That is how I found Gyuri’s research group and got to know Gábor (G. Ónodi, also a co-author on our paper) and many other colleagues studying changes in grasslands using (mostly, but not only) manipulation experiments.

I was really enjoying getting to know this new field, doing fieldwork on our experimental site – which to me, as a novice, felt like a strange cross between a foil greenhouse and a biological laboratory. We planned new, small-scale manipulation experiments involving drought treatments on native and invasive species for my PhD project. I worked hard, but eventually, I faced the usual issues that most PhD students face: 1) some experiments fail, and 2) there is never enough time. How will I show vegetation dynamics in my PhD dissertation?! It is generally very challenging to study environmental change that takes place on a longer time scale when one has to think in the timeframe of research projects that are, at best, 3 to 5 years long. Doing it within the timeframe of your doctoral studies comes with similar time constraints.

So, when Gyuri mentioned that he had a long-term dataset that covered (at that point) almost 20 years, I got very excited about the opportunity to look at longer-term processes. Looking into multiple, naturally occurring drought events instead of a singular, chronic drought treatment also allowed me to learn more about our studied ecosystem’s regeneration capacity and see dynamics that otherwise would have remained invisible to us using only data from manipulation experiments. His data collection on the early phases of establishment allowed a closer focus on the mechanistic understanding of processes.

I felt very lucky to be able to see the sites and observe and aid the sampling for a year or two, rather than just working with historical data. I also felt grateful to be able to build on my colleagues’ past work – and at the same time, it is funny to think that I was starting the 4th grade in primary school when it was first set up!

Overall, I think this cooperation was a win-win-win situation for us: Gyuri’s long-term efforts finally came to fruition, and all his results are now reaching a wider audience. I was grateful to have a different subproject during my PhD and learnt a lot through working with the dataset. Our research group’s results from manipulation experiments will also have an interesting comparison with this long-term observation record.

Brief personal summaries:

Ildikó Orbán is a vegetation ecologist interested in the effects of global change drivers on vegetation. The current research was part of her PhD project that tackled drought and disturbance effects on native dominant grasses and alien invasive plant species in the semi-arid grasslands of Central Hungary. She is currently a postdoctoral researcher at the University of Potsdam, Germany, looking into tipping points in dryland vegetation.

György Kröel-Dulay is a vegetation ecologist at the HUN-Ren Centre for Ecological Research at Vácrátót, Hungary and the leader of the Experimental Vegetation Ecology Research Group. His major research areas include the individual and combined effects of climate change, extreme events, plant invasion, land-use change on vegetation composition and vegetation dynamics. He mostly works in temperate grasslands and conducts long-term observational studies and complex manipulation experiments.