This blog post is provided by Tuomas Kankaanpää and tells the #StoryBehindThePaper for the article “Microclimate structures communities, predation and herbivory in the High Arctic“, which was recently published in Journal of Animal Ecology.
While we can predict the physical nature of climate change with high accuracy, our understanding of how it will affect complex biological systems is still unarguably poor. In my PhD thesis, I, Tuomas Kankaanpää, set out to continue our research group’s long-term efforts: to describe in detail what one of the world’s simplest ecosystems, High-Arctic tundra, looks like and how climate change might affect its functioning. The Spatial Foodweb Ecology Group (SFEG), led by Tomas Roslin and split between the University of Helsinki (Finland) and the Swedish University of Agricultural Sciences (Sweden), has been studying the ecological impacts of climate change in Zackenberg, Northeast Greenland for over a decade with a focus on quantifying trophic interactions between arctic flora and fauna.
Greenland, remote but perfect
The main challenge in studying the dynamics of species communities or interaction webs is the sheer number of observations required. Typical species communities are characterized by few common species and a long tail of rare ones. Of the latter, some may simply be species which are difficult to monitor. If we want to quantify the diet of each species, the sampling effort required increases exponentially with the number of species considered. Our solution to this is to select a simple model system: parasitoid food webs of the High Arctic. The benefits are many: Firstly, these communities are so species-poor that the species are counted in tens rather than thousands (Várkonyi & Roslin, 2013). Secondly, interactions between parasitoids and their hosts are at least somewhat specific and often restricted by physiology, since the parasitoid larva will spend its early life immersed in host tissue. Thirdly, the High Arctic is THE place to be for monitoring the effects of climate change, as the rate of warming here is more than double that of global averages. What is more, there is essentially no confounding disturbance from other human activity. Our main study area in Zackenberg valley is nested in the world’s largest, uninhabited terrestrial national park in Northeast Greenland.
The final perk of this distant study site is the solid foundations laid by the Danish and Greenlandic monitoring program, later joined by my advisor Tomas Roslin. At the time of writing, team SFEG has already worked at Zackenberg for a decade, and established e.g. what the meta-web of all the terrestrial animals looks like in the area (H. K. Wirta et al., 2015). We have also created a solid tool for many kinds of molecular ecology approaches: a nearly complete library of DNA-barcodes (H. Wirta et al., 2016) and more recently also of mitogenomes (Ji et al., 2020).
With a sense of adventure as an added bonus, I set out to sample insects in Greenland. In spite of serious weight shaving attempts, including trap poles made of carbon fibre, my luggage weighed 80kg, which I then had to manage through four flights on aircraft of diminishing size. The beginning of the growth season in Zackenberg can vary with up to a month between years and the flights are scheduled very early in the year. Thus, it is always a gamble in terms of the snow situation that you land in. When I arrived for the first time, I hit the aftermath of the snowiest winter in memory. As a result, the landscape was still fully covered in snow in the beginning of June. While not optimal for data collection, this gave me a chance to experience the total silence of a white world, where migratory birds had yet to arrive and the streams to start flowing. This also meant I could select my trapping locations as spring advanced.
The rest of the season more than made up for my involuntary skiing holiday and the days spent idle. Once the season picked up, my daily routine of emptying traps, counting flowers and flower damage, and then sorting all samples easily took 16 hours per day (not to brag, but to encourage better planning). On top of this, I spent the nights collecting moth larvae to screen for parasitoids. The larvae of the focal species Sympistis zetterstedtii are mainly nocturnal even in the land of the midnight sun. They get up in the cool night air to feed on Dryas buds and flowers and, thus, avoid the bulk of warmth loving parasitoids. This meant that I also had to head out again at 11 pm, to now crawl around the heathlands looking for the larvae. In doing so, I was hoping no vagabond polar bear would take me for a strange land seal. This kind of stupid working intensity is only made possible by the great research support at the Zackenberg research station and the cake always available ad libitum.
In order to catch a good number of parasitoids, I deployed three types of traps: Malaise traps, the staple choice for any hymenopterologist; emergence traps, to better estimate local phenology and abundance; and sticky traps, which are cheap and light with the added benefit that the insects caught will usually not be touching each other, and thereby not contaminate each other with DNA. This makes the catches from sticky traps more suitable for molecular studies than those from, for example, pitfall traps.
It turns out that as arctic parasitoids prefer not to fly even if disturbed, a single tiny sticky trap placed on the ground catches an order of magnitude more parasitoids than a malaise trap. Moreover, in many species a female caught on a trap will continue to attract males through pheromones. When snow-buntings discovered that our traps offered a tasty fly buffet, we swiftly added metal wire cages on top of our sticky traps. This had no effect on parasitoids but reduced unwanted bycatch such as blow-flies and butterflies – and of course removed any risk of catching vertebrates. Barring ethical considerations (slow death compared to other killing traps) and the problem of finding decent quality products as downsides, sticky traps appeared a superior choice for parasitoid collection, especially for molecular ecology studies.
Molecular methods have opened many new possibilities for food web ecology. The most interesting one is of course the scope for characterizing trophic interactions in most animal groups. Whether your interest concerns the liquified remains of spider prey (Eitzinger et al., 2019; Schmidt et al., 2018) or aerial catches of damselflies (Kaunisto, Roslin, Sääksjärvi, & Vesterinen, 2017), molecular methods allow you to arrive at a (semi-)quantitative description of what the animals have fed on. In some cases, the tiniest of droppings will be enough (Kaunisto et al., 2017). In entomological ecology, the mere identification of specimens to a species level using DNA-barcodes is of much use. Importantly, it also frees the expert taxonomists from having to identify a thousand specimens of the most common variety and allows them to focus on the truly interesting specimens and taxa.
SFEG has been using methods based on DNA and especially on DNA-barcodes for a good while. My colleagues had previously developed ways to screen for host associations from adult parasitoids, i.e. to detect the DNA-residues still present in the parasitoid’s gut or in folds of its exoskeleton (H. K. Wirta et al., 2014). My ambitious goal was to simply adapt these protocols, which were based on Sanger sequencing, to an NGS-platform, and to scale up the sample sizes to the quantities needed to build truly quantitative food webs. For this, I wanted to use a single set of DNA-extracts to identify both the parasitoids and their respective hosts. Yet, extracting and amplifying highly fragmented DNA can still be a bit finicky, and can become problematic when building massively multiplexed sequencing libraries. As a result, I ended up not achieving enough reliable trophic associations to really do quantitative analyses at the network level. Such is science sometimes (or actually often).
Luckily, the communities of parasitoids are informative on their own. The way in which we can analyse community data has also been rapidly advancing during the recent decades. The field of community ecology has matured from descriptive analyses to comprehensive statistical frameworks, which now allow the joint modelling of responses by all community members – while simultaneously estimating the effects of species traits, taxonomy as well as of the spatial and temporal structure of the study. We used Hierarchical Modelling of Species Communities (HMSC; Ovaskainen & Abrego, 2020; Ovaskainen et al., 2017; Tikhonov et al., 2019), a joint species distribution modelling framework developed on our campus at the University of Helsinki. Thus, I was able to answer many of my original study questions based on the mere composition of parasitoid communities, reconstructed through molecular identification of the component species.
Species traits can link responses and food web roles
The patterns we observed at the landscape level, as well as in a parallel study based on distributed, pan-arctic sampling (Kankaanpää et al., 2020), suggest that the parasitism strategy of a parasitoid species is a key factor in determining its responses to temperature. Many parasitoids can be classified as either idiobionts or koinobionts in how they use their hosts. Idiobionts are the ones that consume their host quite rapidly and often from the outside. Koinobionts on the other hand can lay dormant inside their host, waiting for it to grow and develop into a larger meal, before starting to eat it from the least important tissues and on. Many parasitoid experts have noted that there appear to be systematic differences between parasitoids adhering to a idiobiont or koinobiont lifestyle. Especially for the parasitoids specializing on the same host species, many differences seem to follow logically from how the respective groups utilise their hosts, with idiobiont pupal parasitoids being larger, longer-lived and less sensitive to host phenology. The fine study by Timms et al. (2016) was the first to make me realise that the specificities related to parasitism strategies may even explain global patterns. While not the focus of that paper (Timms et al. 2016), but rather something tucked away in the supplement, it showed that the abundance of idiobionts diminishes more steeply towards high latitudes. This pattern later proved apparent in our pan-arctic sampling and was visible across sites within a landscape, as now seen in our paper in Journal of Animal Ecology. Overall, the responses detected suggest that the warming climate might make arctic parasitoid communities more generalistic in nature. Such a change could pose a threat to arctic moths, by causing indirect competition from newcomers from the south.
In my thesis work, new methods in community and food web ecology allowed me to make sense of how the effects of environmental change ripple through the ecosystem. Although acquiring such data from the field and in the laboratory proved quite hard at times, a good step in the artic mud is necessary if we want to avoid blindly trusting what simulation models predict, and rather test them against empirical evidence.
Eitzinger, B., Abrego, N., Gravel, D., Huotari, T., Vesterinen, E. J., & Roslin, T. (2019). Assessing changes in arthropod predator–prey interactions through DNA-based gut content analysis—variable environment, stable diet. Molecular Ecology, 28(2), 266–280. doi: 10.1111/mec.14872
Ji, Y., Huotari, T., Roslin, T., Schmidt, N. M., Wang, J., Yu, D. W., & Ovaskainen, O. (2020). SPIKEPIPE: A metagenomic pipeline for the accurate quantification of eukaryotic species occurrences and intraspecific abundance change using DNA barcodes or mitogenomes. Molecular Ecology Resources, 20(1), 256–267. doi: https://doi.org/10.1111/1755-0998.13057
Kankaanpää, T., Vesterinen, E., Hardwick, B., Schmidt, N. M., Andersson, T., Aspholm, P. E., … Roslin, T. (2020). Parasitoids indicate major climate-induced shifts in arctic communities. Global Change Biology, 26(11), 6276–6295. doi: 10.1111/gcb.15297
Kaunisto, K. M., Roslin, T., Sääksjärvi, I. E., & Vesterinen, E. J. (2017). Pellets of proof: First glimpse of the dietary composition of adult odonates as revealed by metabarcoding of feces. Ecology and Evolution, 7(20), 8588–8598. doi: 10.1002/ece3.3404
Ovaskainen, O., & Abrego, N. (2020). Joint Species Distribution Modelling: With Application in R. doi: 10.1017/9781108591720
Ovaskainen, O., Tikhonov, G., Norberg, A., Guillaume Blanchet, F., Duan, L., Dunson, D., … Abrego, N. (2017). How to make more out of community data? A conceptual framework and its implementation as models and software. Ecology Letters, 20(5), 561–576. doi: 10.1111/ele.12757
Schmidt, N. M., Mosbacher, J. B., Eitzinger, B., Vesterinen, E. J., Roslin, T., & Schmidt, N. M. (2018). High resistance towards herbivore-induced habitat change in a high Arctic arthropod community. Biology Letters, 14, 20180054.
Tikhonov, G., Opedal, Ø. H., Abrego, N., Lehikoinen, A., de Jonge, M. M. J., Oksanen, J., & Ovaskainen, O. (2019). Joint species distribution modelling with the R‐package Hmsc. Methods in Ecology and Evolution, 2020(September 2019), 1–6. doi: 10.1111/2041-210x.13345
Timms, L. L., Schwarzfeld, M., & Sääksjärvi, I. E. (2016). Extending understanding of latitudinal patterns in parasitoid wasp diversity. Insect Conservation and Diversity, 9(1), 74–86. doi: 10.1111/icad.12144
Várkonyi, G., & Roslin, T. (2013). Freezing cold yet diverse: dissecting a high-Arctic parasitoid community associated with Lepidoptera hosts. The Canadian Entomologist, 145(02), 193–218. doi: 10.4039/tce.2013.9
Wirta, H. K., Hebert, P. D. N., Kaartinen, R., Prosser, S. W., Várkonyi, G., & Roslin, T. (2014). Complementary molecular information changes our perception of food web structure. Proceedings of the National Academy of Sciences, 111(5), 1885–1890. doi: 10.1073/pnas.1316990111
Wirta, H. K., Vesterinen, E. J., Hambäck, P. A., Weingartner, E., Rasmussen, C., Reneerkens, J., … Roslin, T. (2015). Exposing the structure of an Arctic food web. Ecology and Evolution, 5(17), 3842–3856. doi: 10.1002/ece3.1647
Wirta, H., Várkonyi, G., Rasmussen, C., Kaartinen, R., Schmidt, N. M., Hebert, P. D. N., … Roslin, T. (2016). Establishing a community-wide DNA barcode library as a new tool for arctic research. Molecular Ecology Resources, 16(3), 809–822. doi: 10.1111/1755-0998.12489