A species that is locally common can be globally rare and vice versa. But why? Turns out that tolerance of climatic conditions drives plant species commonness towards global spatial scales, while at finer local scales, competitive ability is relatively more decisive. Accounting for this scale dependence in species occupancy is important when anticipating the effects of climate change or invasive species at local vs. broader scales.

Above: Arctic vegetation, like on the slopes of this mount Saana in North-Finland, is threatened by climate change. Species that are specialized to cold environments are directly affected by warming climate, while those not being strong in competition with other species are threatened also locally by the spread of boreal species northwards. Photo by Miska Luoto.

Why are some species common while others are rare? Trying to answer this question has a long history in biogeography, but despite the decades of studies and suggested and supported reasons, there is no ultimate answer. The quest for the answer is even more topical now, when climate change and invasive species together with other human related actions alter the environment. Indeed, maybe one should rather ask why and which species will become more common or rarer in the future?

While reading through examples of studies investigating the reasons behind species commonness vs. rarity – a feature called ‘occupancy’ in ecology – I stumbled on a study by Heino and Tolonen (2018). They made a short note that the used spatial scale might have affected the outcome of their study, where they found that habitat availability was the most important driver of occupancy while species’ traits or taxonomy played only minor role.

Editors’ choice / Cover article: (Free to read online for two years.)
Mod, H. K., Rissanen, T., Niittynen, P., Soininen, J., & Luoto, M. (2023). The relationships of plant species occupancy to niches and traits vary with spatial scale. Journal of Biogeography, 00, 1– 13. https://doi.org/10.1111/jbi.14608 

Having studied spatial scale and its effect on decisiveness of different drivers behind ecological phenomena, our immediate thought then was that habitat availability, as representing species preferences of environmental conditions, could be more decisive at broader spatial scales where environmental conditions are thought to determine species growth and survival. Instead, at more local scales, where organisms are close enough to each other to compete, biotic interactions would dictate which species get along in a specific location. As species’ ability to compete can be deduced from some of its traits (such as size, effectiveness of resource usage, reproductive capacity), what the species is like, in comparison to other species, would appear important for occupancy only at very fine spatial scales. Species preference and tolerance of environmental conditions are called abiotic ‘niche marginality’ and ‘niche breadth’, respectively, describing how specialist or generalist a species is in terms of environmental conditions, while from species traits one can derive a measure of how varying the species itself can be, i.e., ‘intraspecific trait variability’ (ITV), and a measure of how much its traits deviates from the traits of other species (‘trait distinctiveness’). High niche breadth and low niche marginality should thus lead to high occupancy through availability of potential habitats, while high ITV and trait distinctiveness would lead to high occupancy due to being assets in competition with other species.

In our study we thus wanted to investigate whether spatial scale affects the role of these niche and trait measures in driving species occupancy. For this we needed information on how often a species is encountered at study areas of different scales and information of niches and traits of these species. For this we chose four arctic study areas that varied in size from a few square kilometers to all of terrestrial non-glaciated area north of Arctic Circle and in studied plot sizes from 0.04 m2 to 4 km2. The analyses were done for 106 plant species occurring in the study areas with varying occupancies and niche and trait metrics.

The four study areas of different scales are located north of the Arctic Circle. From each study area we mapped in the field or derived from open databases species occurrences to calculate occupancy, i.e., how often species are encountered, per spatial scale. The location of the study area at the finest scale and the extent of the study area at the next finest scale are marked with yellow and orange squares, respectively. The study area at the second coarsest scale covers mainland Finland, Sweden, and Norway north of Arctic Circle (in purple) and the study area at the coarsest scale covers all terrestrial non-glaciated area north of Arctic Circle (in blue).

The results supported our hypothesis. At the two largest study areas species occupancy was most related to their niche breadth: the species that tolerate range of climatic conditions are more common than those that can only stand certain type of conditions. In contrast, at the more local and finer scales, species that are strong competitors, such as those that can adjust their resource use effectiveness, were found to be more common than those that have traits indicating lower competitive ability.

So, coming back to the burning question: ‘why and which species will become more common or rarer in the future?’, the answer according to our results is that it depends on the spatial scale. Towards the global scale, the species that tolerate varying environmental conditions are likely to remain as common as they are now under a changing climate, while those specializing to certain environmental conditions are at risk of becoming even rarer. On a local scale, in turn, tolerating varying climatic conditions is not much of an asset in being or becoming common, whereas a good ability to compete with other species drives commonness and those that don’t have the properties to compete with other species are at risk to become even locally extent. This can mean that the Arctic is in double jeopardy: cold-adapted arctic vegetation as a whole is under the threat of warming climate while the spread of competitive boreal species northwards threatens the arctic plants also locally.

However, our study do not provide the ultimate answers to the question of underlying mechanisms of species occupancy nor to the question of the future of the Arctic. It still remains to be tested if our findings hold for other species groups than plants and for other niche and trait metrics than those we used. Also, there are additional factors influencing the future of Arctic environment and vegetation than those included in our study. Thus, the quest continues!

Written by:
Heidi Mod, university lecturer, Department of Geosciences and Geography, University of Helsinki, Finland

Tuuli Rissanen, PhD candidate, Department of Geosciences and Geography, University of Helsinki, Finland

Pekka Niittynen, Post doc, Department of Geosciences and Geography, University of Helsinki, Finland & Department of Biological and Environmental Science, University of Jyväskylä, Jyväskylä, Finland

Janne Soininen, professor, Department of Geosciences and Geography, University of Helsinki, Finland Miska Luoto, professor, Department of Geosciences and Geography, University of Helsinki, Finland

Additional information:
https://www.helsinki.fi/en/researchgroups/biogeoclimate-modelling-lab

The geographic ranges of mammals in Africa are limited in size by the variation in habitats across space (habitat heterogeneity), but surprisingly not by the variation in elevations (topographic heterogeneity). Mammalian ranges will be sensitive to future habitat destruction and alteration, as climate change and human impacts continue to intensify.

Above: The cheetah (Acinonyx jubatus) is one of many African mammals whose range size may be constrained by the variation in habitats across landscapes, i.e., habitat heterogeneity (picture taken by Michael S. Lauer).

In the summer of 2019, I attended the Ecological Society of America’s annual conference in Louisville, Kentucky. I remember walking around a massive room with hundreds of posters, and as I made my rounds, I couldn’t help but notice that many of the posters had the word “heterogeneity” in their title. Fast forward a few months to the fall of 2019, when my co-author and aca-brother (Dr. Benjamin Shipley) introduced me to some of the aspects of mammalian biogeography that were still unknown. Fast forward again to the spring of 2020, when I took a course titled “Biodiversity on a Changing Planet” taught by my co-author and graduate advisor Dr. Jenny McGuire. Why am I listing three seemingly unrelated experiences? As it turns out, all three together served as the foundation of our recent paper. Had I not gone to the conference, I would not have internalized how important landscape heterogeneity (i.e., the variation in environmental conditions across space) is to ecology and biogeography. Had I not conversed with Ben, I would not have been initially enticed to study mammalian geographic ranges and the associated knowledge gaps. And had I not taken the course, I would have not been assigned the project that morphed into the initial draft of this paper. All of this is to say that sometimes the most exciting ideas come from a “heterogeneity” of experiences and perspectives.

Cover article: (Free to read online for two years.)
Lauer, D. A., Shipley, B. R., & McGuire, J. L. (2023). Habitat and not topographic heterogeneity constrains the range sizes of African mammals. Journal of Biogeography, 50, 846 – 857. https://doi.org/10.1111/jbi.14576  

Ultimately, it was the combined perspectives of myself, Ben, and Jenny that led us to ask the following question: are mammalian geographic range sizes in Africa more constrained by the variation in habitats (habitat heterogeneity) or the variation in physical elevations (topographic heterogeneity) across space? We knew from prior research that species ranges are constrained in heterogeneous landscapes because to co-exist, each species adapts to the specific environmental conditions of specific areas. But we wanted to take this idea a step further. We were motivated to compare the effects of habitat and topographic heterogeneity specifically because they represent two fundamentally different things about landscapes. Habitat heterogeneity is fluid, as habitats come and go rapidly depending on how climates change and how humans behave. Consider, for example, a landscape that is heterogeneous because it possesses small patches of intertwined forest and grassland habitats. Such a landscape could become rapidly homogeneous if its trees are cut down and it transforms into an extensive grassland. Contrast that to topographic heterogeneity, which is much more set in stone. A landscape that is heterogeneous because it is hilly is likely to remain hilly for a long time. Hills and mountains don’t disappear overnight.

The landscape on the left has low habitat but high topographic heterogeneity, as it exhibits only forest habitat but is hilly. The landscape on the left is the opposite, as it exhibits forest, grassland, and aquatic habitats but is flat (pictures taken by Daniel A. Lauer).

At first, we were thinking that both habitat and topographic heterogeneity would at least have some constraining effect on species range sizes. But in a fashion that often makes science so exciting, we were surprised to arrive at an unexpected result. While we found evidence that habitat heterogeneity had a strong limiting effect on range size, our analyses suggested that topographic heterogeneity had no effect at all! This contrast became particularly interesting and important to us as we thought about its implications. Essentially the contrast suggests that when landscape heterogeneity limits mammalian ranges, it does so via its fluid, subject-to-change habitats, and not via its rigid physical structure. Consequently, the persistence and geographic occurrences of mammals may be highly susceptible to habitat change, destruction, and alteration by climate change and human land use. We channeled these ideas to think about how our results could be meaningful for both ecological theory and environmental conservation. Regarding the former – we were able to add a new layer of nuance to the theory of how landscape heterogeneity constrains ranges. And regarding the latter – we would suggest that conservation efforts should focus on the small-ranged mammals that occur in regions of high habitat heterogeneity, particularly considering that a small range can often mean a higher extinction risk.

Along our collaborative journey I learned many things, and together we opened some avenues for future researchers to explore. I internalized the power that many experiences, as well as ideas from other scientists, can have in forming a meaningful scientific question. I learned how cool it can be to answer such a question by collecting data from many sources and combining that data into a single analytical pipeline. And I learned that the most exciting science can emerge from the results that were wildly unexpected. Our work calls for an enhanced understanding of the specific species that live in regions of high habitat heterogeneity in Africa. What are these species? What are their functional traits and evolutionary histories? How do they compare to species that occur in more homogeneous environments? What would be the most effective conservation strategies to prevent their extinction? Time will tell, but for now we can say that the sky (or the heterogeneity of the landscape?) is the limit.

Written by:

Daniel A. Lauer (Ph.D. Candidate)^1,2, Benjamin R. Shipley (Ph.D.)², Jenny L. McGuire (Assistant Professor)^1-3

¹Interdisciplinary Graduate Program in Quantitative Biosciences, Georgia Institute of Technology, Atlanta, GA, USA

²School of Biological Sciences, Georgia Institute of Technology, Atlanta, GA, USA³School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA, USA

Additional information:
LinkedIn (Daniel Lauer): https://www.linkedin.com/in/daniel-lauer/
ResearchGate (Daniel Lauer): https://www.researchgate.net/profile/Daniel-Lauer-2
Twitter (Daniel Lauer): @DannyLauer

ResearchGate (Benjamin Shipley): https://www.researchgate.net/profile/Benjamin-Shipley

Instagram (Jenny McGuire): mapsnbones
Twitter (Jenny McGuire): @JennyMcGPhD
Webpage (Jenny McGuire): https://www.mcguire.gatech.edu

Plants can reproduce clonally or by seeds. What are the circumstances that clonal reproduction is favoured, and which type of species are most likely to be clonal?

Above: Plants reproduce asexually via rhizomes.

Sexual reproduction from seeds is common in the plant kingdom. However, many plants reproduce through vegetative propagation or clonal growth such as sprouting from rhizomes, thus they are called clonal plants. Back in 2018, we studied the Australian flora and found that the proportion of clonal species increases with latitudes across the whole continent. Thereafter, we asked why — we would like to know the underlying mechanisms for this geographic pattern. Despite having many studies on the prevalence of sexual versus clonal reproduction and its associations with abiotic conditions at smaller scales, we still have much to learn about the circumstances under which sexual versus clonal reproduction are favoured, and which type of species are most likely to be clonal or sexual at the continental scale.

Editors’ choice / Cover article: (Open access)
Zhang, H., Chen, S.-C., Bonser, S. P., Hitchcock, T., & Moles, A. T. (2023). Factors that shape large-scale gradients in clonality. Journal of Biogeography, 50, https://doi.org/10.1111/jbi.14577. 

To better understand the ecological and evolutionary significance of reproductive mode, we studied 4116 seed species and their 914456 occurrence records in the Australia flora, together with a series of factors that might influence plant clonality. In this way, we could directly compare the effects of the four plant characteristics and sixteen environmental variables on determining the likelihood of a species to exhibit clonal reproduction or not.

Plants reproduce by seeds (left panel; sexual reproduction organ of the walking stick palm Linospadix monostachya; Photo: Si-Chong Chen) and by rhizomes (right panel; clonal reproduction organ of Hydrocotyle sibthorpioides; Photo: Hongxiang Zhang).

We found that plant characteristics explained more than two times the variation in the probability of species having clonal reproduction than did environmental variables. Our findings suggest that we may need to consider species’ traits as a coordinated suite that respond to environmental conditions, rather than studying them one at a time in the future. For example, one potential direction for future research is whether big plants (e.g. trees and tall plants) tend to sexual reproduction and have greater seed dispersal ability and distances to counterbalance the longer time to first reproduction, while small plants (e.g. herbs and shorter plants) tend to be clonal reproduction and have higher seed dormancy and seed persistence in the soil. A hot topic about relationships between environmental conditions and sexual vs clonal reproduction is whether stress conditions favour sexual or clonal reproduction. Sexual reproduction is found to be predominant in stress conditions such as drought and low fertility, while clonal reproduction is common in harsh conditions, e.g. low temperature and high altitudes. Our results showed that clonality tended to be favoured when environmental resource availability was high, like high water availability and high soil nutrient conditions. Therefore, we suggest clonal reproduction may be a ‘weapon’ of plants for population expansion in resource-abundant sites, rather than as a reproductive assurance under environmental stress. This would be an interesting and important direction for future research at multiple geographical regions and scales

Our study is the first continental-scale cross-species analysis to date of plant characteristics and environmental factors in shaping plant clonality. The findings advance understanding of broad patterns in reproductive strategies and help improve understanding of species’ capacity to adapt and migrate in response to future climate change.

Written by:
Hongxiang Zhang, Full Professor, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences.
Si-Chong Chen, Full Professor, Wuhan Botanic Garden, Chinese Academy of Sciences.

Additional information:
https://scholar.google.com/citations?hl=en&user=lWQ3WBEAAAAJ&view_op=list_works&sortby=pubdate
https://scholar.google.com/citations?user=BcKbEYsAAAAJ&hl=en

Twitter: @SichongChen, @AngelaMoles

Area, environmental heterogeneity, scale and the conservation of alpine diversity.

Above: Phyteuma hemisphericum from the Sierra de Villabandín, Cantabrian Mountains, Spain;
photograph by Borja Jiménez-Alfaro

This project was an extension of our work on the relative importance of geographic distance and environmental difference to the beta diversity of alpine plant communities (Malanson et al. 2022). We developed a simulation of all 23 of the mountain ranges in southern and central Europe based on data from 18,000 plot surveys. However, we decided that we could learn more by simulating each range independently. Further, we recognized that additional quantification of environmental heterogeneity within ranges could complement our earlier work (Jimenez-Alfaro et al. 2021). This concurrence led us to the key problem description by Udy et al. (2021) and the simulation model by Ben-Hur (2020) that was similar to what we had developed. Together, these reinvigorated our longstanding interest in the current relevance of island theory to alpine biogeography and the combination of field data and simulation results allowed us to assess the types of environments in which area per se becomes important in the maintenance of species diversity.

Areas of alpine vegetation are found in 23 mountain ranges in southern and central Europe from the Baetic System in the southwest, in an arc through the central Alps, shown here, to the Hellenides in the southeast (photograph courtesy of Harald Pauli).

Cover article: (Open Access)
Malanson, G. P., Testolin, R., Pansing, E. R., & Jiménez-Alfaro, B. (2023). Area, environmental heterogeneity, scale and the conservation of alpine diversity. Journal of Biogeography, 50, 743– 754. https://doi.org/10.1111/jbi.14573

Beyond the interest to reread MacArthur and Wilson and realize how blithely they dealt with heterogeneity, thinking about the relevance of area per se was motivating. The degree to which the complex topography, and thus heterogeneity, of mountains will provide microrefugia for current alpine species depends on that effect. While our empirical analysis of the 23 ranges is too coarse to address microrefugia, our virtual microcosms show that area per se will, indeed, be important – and this can be seen intuitively in the basic species-area curve upon which MacArthur and Wilson built their theory. The conservation of heterogeneity in microrefugia is no panacea for the impacts of climate change on alpine species diversity.

In local areas, extents can be quite small, as here in the Sierra de Villabandín, Cantabrian Mountains Spain. Fragmentation within the 23 ranges was included in the simulations (photo: Borja Jiménez-Alfaro).

In the process of developing the simulation, the question of the Heterogeneity – Effective-Area Tradeoff (HEAT) emerged. Its significance was driven home in conversation with Kostas Triantis at the International Biogeography Society meeting in Vancouver. Our simulation did not produce this pattern in our basic runs, but additional modifications allowed HEAT to evolve more often when our simulation grid had greater discrete steps of environment in space. While this indicated that HEAT could be induced as a model artifact, it also tells that real landscapes, which often are not continuous, could generate HEAT.

Teaming with others in AlpVeg (https://www.alpveg.com/), next we will pursue questions of microrefugia using finer scale data. These studies may extend globally and would use the GLOBALP database (https://www.givd.info/index.xhtml).

Extension to finer scale analyses will rely in part on legacy data from worldwide plots (here, Glacier National Park, USA) in the GLOBALP database (photo: Calypso Ecological LLC, used by contract).

Written by:
George P Malanson
Coleman-Miller Professor Emeritus, The University of Iowa

Additional information:
https://clas.uiowa.edu/geography/people/george-malanson

References:
Ben-Hur, E. & Kadmon, R. 2020. Heterogeneity–diversity relationships in sessile organisms: A unified framework. Ecology Letters 23, 193–207
Jiménez-Alfaro, B., Abdulhak S, Attorre, Bergamini A, Carranza ML, Chiarucci, A., Ćušterevska R, Dullinger S, Gavilán RG, Giusso del Galdo G, Kuzmanov N, Laiolo P, Loidi J, Malanson GP, Marcenó C, Milanović D, Pansing ER, Roces-Díaz JV, Ruprecht E, Šibik J, Stanisci A, Testolin R, Theurillat J-P, Vassilev K, Willner W, Winkler M. 2021. Postglacial determinants of regional species pools in alpine grasslands. Global Ecology & Biogeography 30, 1101-1115.
Malanson, G. P., Pansing, E. R., Testolin, R., Abdulhak, S., Bergamini, A., Ćušterevska, R., Marcenò, C., Kuzmanović, N., Milanović, Đ., Ruprecht, E., Šibík, J., Vassilev, K., Willner, W., & Jiménez-Alfaro, B. 2022. Explanation of beta-diversity in European alpine grasslands changes with scale. Ecosphere 13, e4066
Udy, K., Fritsch, M., Meyer, K. M., Grass, I., Hanß, S., Hartig, F., Kneib, T., Kreft, H., Kukunda, C. B., Pe’er, G., & Reininghaus, H. 2021. Environmental heterogeneity predicts global species richness patterns better than area. Global Ecology & Biogeography 30, 842–851

Rivers sometimes change their way, caused by geological events. During such river rearrangement, what happened to the inhabitants? We investigated the genetic traces remaining in genomes of the descendants and look for a way to find unidentified geological events.

Above: An upper reach of mountain stream on Honshu Island, Japan.

Honshu Island, the main island of the Japan archipelago, has a mountainous landscape with a complex array of strike-slip and thrust faults. Numerous streams rise from the mountain ranges called the Central Divide of Japan, where two major watersheds are divided (The Sea of Japan and the Pacific watersheds). However, the drainage divide is not always apparent, and headwaters of different drainages are close to each other in some locations. As a regional angler, I was wondering if these rivers were connected or disconnected easily by heavy rainfalls, general climate change or geomorphological movement. If so, what happened in aquatic organisms, particularly to their genetic structures and distributions, when their habitats were drastically changed? Accordingly, thinking that stream capture, in which an upper stream of a river is captured by and displaced to another adjacent river, should have a prominent impact on the gene flow of stream fishes, then the history of the landscape should have synchronized the ecology of aquatic organisms. Thus, population genetics should be a good candidate as a witness of the geological events.

White-spotted charr is a stenothermal cold-water adopted salmonid distribute in North-eastern Pacific, including the Japan archipelago. This species is generally landlocked in the upper streams in Honshu Island, the edge of their distribution. These characteristics would meet requirements bearing witness to geological events (unless there had been human mediated translocation). First, we investigated the genetic structure of white-spotted charr from three major watersheds, the Sea of Japan, the Pacific and Lake Biwa. Samples were collected from the upper reaches close to the divides to avoid the bias of artificial stockings. As a result, fish from three major watersheds were clearly classified into three distinctive genetic structures (Fig. 2).

Editors’ choice article: (Free to read online for two years.)
Masuda, T., Shimono, Y., Kishi, D., & Koizumi, I. (2023). Systematic headwater sampling of white-spotted charr reveals stream capture events across dynamic topography. Journal of Biogeography, 50, 453– 466. https://doi.org/10.1111/jbi.14553

However, we found several exceptions in which genetic structure of charr was inconsistent with the current watershed group. One such exception was seen at a site known to have experienced stream capture (Fig. 3). Although site J1 belongs to the Sea of Japan watershed now, fish from the site were perfectly classified to the Lake Biwa genetic cluster. This result is completely consistent with the history of the river capture, in which site J1 used to be upstream of site B1, and subsequently, the catchment including J1 was captured by the river flowing north to the Sea of Japan, resulting in isolation from B1. In accordance with the geological history, the genetic structure of J1 charr was completely assigned to the Lake Biwa genetic group almost without contamination with genes from the Sea of Japan group. Thus, the population at J1 is a subdivision of the Lake Biwa genetic group, and designated as an ‘exclave case’. It would have been impossible for downstream fish (the Sea of Japan group) to move up into J1 territory because of some intervening gorges and steep waterfalls. The exclave case would be a typical consequence of river capture events. We found another example of an exclave in which the Pacific and Lake Biwa genetic group were related, indicating the unidentified broad area drainage rearrangement.

Typical individuals of white-spotted charr from the Sea of Japan (upper), the Pacific (middle) and Lake Biwa (bottom).

We also observed other genetic and regional inconsistencies that produced genetic admixture in some locations near the divides. This pattern could be a trace of the transient river connections by natural phenomena, although some of such cases might be attributed to human mediated translocation.

(a) Population genetic structure of white-spotted charr from the sample sites. (b) Genetic structure of some populations from the three major watersheds.(c) An example of a population illustrating the “exclave” pattern.

Thus, we proposed the possibility to discover the geological history of drainage rearrangement from the genetic aspects of aquatic organisms. Verification with other aquatic species including invertebrates would further support the hypotheses. We are investigating the genetic structure of Rhynchocypris oxycephalus and Cottus pollux from the locations with probable stream capture history in addition to sites of the current research.

Written by:
Taro Masuda¹, Yoshiko Shimono², Daisuke Kishi³ and Itsuro Koizumi^4
1Laboratory of Marine Biology, Faculty of Agriculture, Setsunan University, 45-1 Nagaotoge-cho, Hirakata, Osaka, 573-0101, Japan
² Laboratory of Weed Science, Division of Agronomy and Horticultural Science, Graduate School of Agriculture, Kyoto University, Kitashirakawa-Oiwake Cho, Kyoto
³ Gero Branch, Gifu Prefectural Research Institute for Fisheries and Aquatic Environments, 2605-1 Hane, Hagiwara, Gero, Gifu, 509-2592, Japan
⁴ Faculty of Environmental Earth Science, Hokkaido University, N10W5 Sapporo, Hokkaido 060-0810, Japan

Global body size distributions in dragonflies and damselflies are shaped by temperature and predators

Above: A model replica of a fossil dragonfly (Urogomphus giganteus) in Museum für Naturkunde (Berlin) that lived about 140 million years ago. Dragonflies and damselflies have an unusually rich fossil record, compared to other insect groups. Photo: Erik Svensson.

Dragonflies and damselflies (Odonata) is an old and fascinating insect order, comprising about 6400 species globally. Although not one of the most diverse insect orders, these insects are wellknown to layperson and increasingly popular among amateur naturalists and photographers, due to their interesting behaviours and rich diversity in size, shape and colouration. It is also an insect group with a dramatic macroevolutionary history and a rich fossil record. The ancestors  of Odonata that existed about 300 million years before present, included the now extinct genus Meganeura, where some individuals had a wing span of about 70 cm, the largest flying insects that ever have existed on Earth (Clapham and Karr 2012; Waller and Svensson 2017). The large sizes of these now extinct insects have been explained as being a result of the higher atmospheric oxygen levels during the Carboniferous period, which were over 30 %, compared to the situation today (about 20 %), as flying insects are dependent on high oxygen levels as they breathe through trachea, small openings in the cuticle (Grimaldi and Engel 2005).

Cover article: (open access)
Svensson, E. I., Gómez-Llano, M., & Waller, J. T. (2023). Out of the tropics: Macroevolutionary size trends in an old insect order are shaped by temperature and predators. Journal of Biogeography, 50, 489–502. https://doi.org/10.1111/jbi.14544 

However, even among today’s surviving species of dragonflies and damselflies there is considerable variation in size that remains to be explained. Consider, for instance, the large  Helicopter Damselflies (Megaloprepus caerulatus) in Central and South America with a wing span of about  190 mm vs. the smallest Wisp damselflies of the genus Agriocnemis with wing spans  less than 20 mm. How can we explain this large variation in size today across the globe and which ecological and environmental factors were important in shaping today’s geographic variation in size in Odonata?

A Helicopter Damselfly (Megaloprepus caerulatus), photographed during a night walk in the rainforest reserve La Selva in Costa Rica in February 2020. This is the world’s largest damselfly, where some males can have wing spans of 190 mm. Photo: Erik Svensson.

To answer this question, we compiled size data of dragonflies and damselflies from field guides, scientific articles and internet sources with the aim to create a global database of phenotypic traits of this enigmatic insect order. The first result of this work – to which we owe a great debt to the many students who helped us out – was published in the journal Scientific Data (Waller et al. 2019) and as an open trait database: The Odonate Phenotypic Database (http://www.odonatephenotypicdatabase.org/shiny/odonates/?).

Armed with this database and size information for 775 species of Odonata ranging from the tropics to the temperate zone, we first investigated if there was any latitudinal size gradient. In other words; do dragonflies and damselflies become consistently smaller or larger in size as we moved away from the species-rich tropics where most lineages have their evolutionary origin? It turned out that there was indeed a systematic pattern: at higher latitudes in the temperate zone, dragonflies (suborder Anisoptera), but notably not damselflies (suborder Zygoptera), are larger than in the tropics. This geographic pattern, which is known as the classical “Bergmann’s Rule”, is wellknown among endothermic animals like birds and mammals, but has also been documented among some ectothermic animal groups, including dragonflies and damselflies, and some other insect groups. How can we explain this latitudinal size gradient, then?

The most obvious explanation is temperature. Insects are known to develop faster and reach a smaller size when ambient temperatures are high, like in the tropics. Indeed, we found a significant effect of temperature when analyzing global size variation in Odonata, taking in to account phylogenetic relatedness and by using modern comparative methods. However, temperature is not the sole answer. Two intriguing pieces of data reveal a more complex explanation for these latitudinal size gradients.

First, when taking in to account other environmental factors, such as bird species diversity (a proxy of predation risk on Odonata, as we know that birds are important predators on these large insects), we found that the effect of bird diversity was three times stronger in explaining size variation than the effect of temperature. Moreover, the effect of bird diversity was unlikely to solely be a general diversity effect, unrelated to predation, as mammal diversity (which served as an independent control variable) did not show such a strong effect. This strongly suggests that the diversity of avian predators has a negative effect on the average size of dragonflies and damselflies across the globe.

An additional piece of evidence in this puzzle comes from how the latitudinal size gradient in Odonata has changed over macroevolutionary time, during the last 210 million years.  To investigate this, we used data on extinct species of Odonata from the Paleobiology Database (https://paleobiodb.org/#/). By combining size and age information and information about the paleo-latitude of different-sized fossils, we found that the latitudinal size gradient has changed over macroevolutionary time. The recent latitudinal size gradient where the largest species are found at the highest latitudes is actually of recent origin. In the past the latitudinal size gradient had a different sign, and the largest species were found at low latitudes in the tropics. These findings from the fossil record decisively show that a simple explanation based on temperature cannot fully explain why the largest species are found at the highest latitudes as there is no reason to think that they should respond differently to temperature now than they did in past geological times. Instead, we interpret these changing latitudinal size gradients as a result of the evolutionary radiation of birds, that emerged on the geological scene about 150 million years ago. Birds then diversified rapidly after the most recent mass extinction at the end of the Cretaceous Period, 65 million years ago.

In the tropics, bird predation on Odonata is high, as illustrated by this Rufous-tailed Jacamar (Galbula ruficauda) that has caught a Buenos Aires Darner (Aeshnidae: Rhionaeschna bonariensis) in REGUA wetland reserve in the Atlantic Forest of Brazil in January 2018. Photograph by E.I. Svensson.

As birds are major predators on Odonata, we suggest that as they started to diversify at lower latitudes, their increased presence selected against larger sized species that could, due to their higher dispersal capacity, subsequently escape predation by invading the temperate region with lower predation pressure. Thus, increased pressure from birds in combination with higher dispersal capacity of large-bodied species of Odonata partly shaped the current latitudinal size gradient, alongside with temperature.

Our study therefore illustrate that both temperature and birds were responsible for creating the current latitudinal size gradient and a single explanatory factor is thus insufficient to fully explain today’s global size distribution. We hope that our work will stimulate research on other organismal groups and that researchers will consider both abiotic factors like temperature but also biotic factors like predator pressure when seeking to explain latitudinal size gradients. However, research on both Odonata and other insects is limited by lack of body size and ecological information, particularly for tropical taxa. In the present study, for instance, we only had access to size data for 775 out of the total of 6400 species of Odonata that are known, or about 12 %. This mainly reflects the lack of field guides and data from the tropics. Clearly, our study has only laid the foundation for future work in this area and there is an urgent need for more data from the tropics for this and other insect groups.

References:
Clapham, M. E., and J. A. Karr. 2012. Environmental and biotic controls on the evolutionary history of insect body size. Proceedings of the National Academy of Sciences of the United States of America 109:10927–10930.
Grimaldi, D., and M. S. Engel. 2005. Evolution of the insects. Cambridge University Press, New York.
Waller, J. T., and E. I. Svensson. 2017. Body size evolution in an old insect order: No evidence for Cope’s Rule in spite of fitness benefits of large size. Evolution 71:2178–2193.
Waller, J. T., B. Willink, M. Tschol, and E. I. Svensson. 2019. The odonate phenotypic database, a new open data resource for comparative studies of an old insect order. Sci. Data 6:1–6.

Written by:
Erik Svensson
Professor, Department of Biology, Lund University, SE-223 62 Lund, SWEDEN

Additional information:
Website: https://portal.research.lu.se/en/persons/erik-svensson
Twitter: https://twitter.com/EvolOdonata
Mastodon: https://ecoevo.social/@EvolOdonata