Tick-borne diseases are an important burden on public health all over the globe, making accurate forecasts of tick population a key ingredient in a successful public health strategy. Over long time scales, tick populations can undergo complex dynamics, as they are sensitive to many non-linear effects due to the complex relationships between ticks and the relevant (numerical) features of their environment.

But luckily, capturing complex non-linear responses is a task that machine learning thrives on. In this contribution, Manley et al. (2023) explore the use of Gradient Boosted Trees to predict the distribution (presence/absence) and abundance of ticks across New York state.

This is an interesting modelling challenge in and of itself, as it looks at the same ecological question as an instance of a classification problem (presence/absence) or of a regression problem (abundance). In using the same family of algorithm for both, Manley et al. (2023) provide an interesting showcase of the versatility of these techniques. But their article goes one step further, by setting up a multi-class categorical model that estimates jointly the presence and abundance of a population. I found this part of the article particularly elegant, as it provides an intermediate modelling strategy, in between having two disconnected models for distribution and abundance, and having nested models where abundance is only predicted for the present class (see e.g. Boulangeat et al., 2012, for a great description of the later).

One thing that Manley et al. (2023) should be commended for is their focus on opening up the black box of machine learning techniques. I have never believed that ML models are more inherently opaque than other families of models, but the focus in this article on explainable machine learning shows how these models might, in fact, bring us closer to a phenomenological understanding of the mechanisms underpinning our observations.

There is also an interesting discussion in this article, on the rate of false negatives in the different models that are being benchmarked. Although model selection often comes down to optimizing the overall quality of the confusion matrix (for distribution models, anyway), depending on the type of information we seek to extract from the model, not all types of errors are created equal. If the purpose of the model is to guide actions to control vectors of human pathogens, a false negative (predicting that the vector is absent at a site where it is actually present) is a potentially more damaging outcome, as it can lead to the vector population (and therefore, potentially, transmission) increasing unchecked.

Boulangeat I, Gravel D, Thuiller W. Accounting for dispersal and biotic interactions to disentangle the drivers of species distributions and their abundances: The role of dispersal and biotic interactions in explaining species distributions and abundances. Ecol Lett. 2012;15: 584-593.
https://doi.org/10.1111/j.1461-0248.2012.01772.x

Manley W, Tran T, Prusinski M, Brisson D. (2023) Modeling tick populations: An ecological test case for gradient boosted trees. bioRxiv, 2023.03.13.532443, ver. 3 peer-reviewed and recommended by Peer Community in Ecology. https://doi.org/10.1101/2023.03.13.532443

The study of animal movement is essential for understanding their behaviour and how ecological or global changes impact their routines [1]. Recent technological advancements have improved the collection of movement data [2], but limited statistical tools have hindered the analysis of such data [3–5]. Animal movement is influenced not only by environmental factors but also by internal knowledge and memory, which are challenging to observe directly [6,7]. Routine movement behaviours and the incorporation of memory into models remain understudied.

Researchers have developed ‘MoveFormer’ [8], a deep learning-based model that predicts future movements based on past context, addressing these challenges and offering insights into the importance of different context lengths and information types. The model has been applied to a dataset of over 1,550 trajectories from various species, and the authors have made the MoveFormer source code available for further research.

Inspired by the step-selection framework and efforts to quantify uncertainty in movement predictions, MoveFormer leverages deep learning, specifically the Transformer architecture, to encode trajectories and understand how past movements influence current and future ones – a critical question in movement ecology. The results indicate that integrating information from a few days to two or three weeks before the movement enhances predictions. The model also accounts for environmental predictors and offers insights into the factors influencing animal movements.

Its potential impact extends to conservation, comparative analyses, and the generalisation of uncertainty-handling methods beyond ecology, with open-source code fostering collaboration and innovation in various scientific domains. Indeed, this method could be applied to analyse other kinds of movements, such as arm movements during tool use [9], pen movements, or eye movements during drawing [10], to better understand anticipation in actions and their intentionality.

References

1.           Méndez, V.; Campos, D.; Bartumeus, F. Stochastic Foundations in Movement Ecology: Anomalous Diffusion, Front Propagation and Random Searches; Springer Series in Synergetics; Springer: Berlin, Heidelberg, 2014; ISBN 978-3-642-39009-8.
https://doi.org/10.1007/978-3-642-39010-4
 
2.           Fehlmann, G.; King, A.J. Bio-Logging. Curr. Biol. 2016, 26, R830-R831.
https://doi.org/10.1016/j.cub.2016.05.033
 
3.           Jacoby, D.M.; Freeman, R. Emerging Network-Based Tools in Movement Ecology. Trends Ecol. Evol. 2016, 31, 301-314.
https://doi.org/10.1016/j.tree.2016.01.011
 
4.           Michelot, T.; Langrock, R.; Patterson, T.A. moveHMM: An R Package for the Statistical Modelling of Animal Movement Data Using Hidden Markov Models. Methods Ecol. Evol. 2016, 7, 1308-1315.
https://doi.org/10.1111/2041-210X.12578
 
5.           Wang, G. Machine Learning for Inferring Animal Behavior from Location and Movement Data. Ecol. Inform. 2019, 49, 69-76.
https://doi.org/10.1016/j.ecoinf.2018.12.002
 
6.           Noser, R.; Byrne, R.W. Change Point Analysis of Travel Routes Reveals Novel Insights into Foraging Strategies and Cognitive Maps of Wild Baboons. Am. J. Primatol. 2014, 76, 399-409.
https://doi.org/10.1002/ajp.22181
 
7.           Fagan, W.F.; Lewis, M.A.; Auger‐Méthé, M.; Avgar, T.; Benhamou, S.; Breed, G.; LaDage, L.; Schlägel, U.E.; Tang, W.; Papastamatiou, Y.P. Spatial Memory and Animal Movement. Ecol. Lett. 2013, 16, 1316-1329.
https://doi.org/10.1111/ele.12165
 
8.           Cífka, O.; Chamaillé-Jammes, S.; Liutkus, A. MoveFormer: A Transformer-Based Model for Step-Selection Animal Movement Modelling. bioRxiv 2023, ver. 4 peer-reviewed and recommended by Peer Community in Ecology.
https://doi.org/10.1101/2023.03.05.531080
 
9.           Ardoin, T.; Sueur, C. Automatic Identification of Stone-Handling Behaviour in Japanese Macaques Using LabGym Artificial Intelligence. 2023, https://doi.org/10.13140/RG.2.2.30465.02402
 
10.         Martinet, L.; Pelé, M. Drawing in Nonhuman Primates: What We Know and What Remains to Be Investigated. J. Comp. Psychol. Wash. DC 1983 2021, 135, 176-184, doi:10.1037/com0000251.
https://doi.org/10.1037/com0000251

Ecosystems shelter a huge number of species, continuously interacting. Each species interact in various ways, with trophic interactions, but also non-trophic interactions, not mentioning the abiotic and anthropogenic interactions. In particular, pollination, competition, facilitation, parasitism and many other interaction types are simultaneously present at the same place in terrestrial ecosystems [1-2]. For this reason, we need today to improve our understanding of such complex interaction networks to later anticipate their responses. This program is a huge challenge facing ecologists and they today join their forces among experimentalists, theoreticians and modelers. While some of us struggle in theoretical and modeling dimensions [3-4], some others perform brilliant works to observe and/or experiment on the same ecological objects [5-6].
In this nice study [6], Magrach et al. succeed in studying relatively large plant-pollinator interaction networks in the field, in Mediterranean ecosystems. For the first time to my knowledge, they study community-wide interactions instead of traditional and easier accessible pairwise interactions. On the basis of a statistically relevant survey, they focus on plant reproductive success and on the role of pollinator interactions in such a success. A more reductionist approach based on simpler pairwise interactions between plants and pollinators would not be able to highlight the interaction network structure (the topology) possibly impacting its responses [1,5], among which the reproductive success of some (plant) species. Yet, such a network analysis requires a fine control of probable biases, as those linked to size or autocorrelation between data of various sites. Here, Magrach et al. did a nice work in capturing rigorously the structures and trends behind this community-wide functioning.
To grasp possible relationships between plant and pollinator species is a first mandatory step, but the next critical step requires understanding processes hidden behind such relationships. Here, the authors succeed to reach this step too, by starting interpreting the processes at stake in their studied plant-pollinator networks [7]. In particular, the niche complementarity has been demonstrated to play a determinant role in the plant reproductive success, and has a positive impact on it [6].
When will we be able to detect a community-wise process? This is one of my team’s objectives, and we developed new kind of models with this aim. Also, authors focus here on plant-pollinator network, but the next step might be to gather every kind of interactions into a huge ecosystem network which we call the socio-ecosystemic graph [4]. Indeed, why to limit our view to certain interactions only? It will take time to grasp the whole interaction network an ecosystem is sheltering, but this should be our next challenge. And this paper of Magrach et al. [6] is a first fascinating step in this direction.

References

[1] Campbell, C., Yang, S., Albert, R., and Shea, K. (2011). A network model for plant–pollinator community assembly. Proceedings of the National Academy of Sciences, 108(1), 197-202. doi: 10.1073/pnas.1008204108
[2] Kéfi, S., Miele, V., Wieters, E. A., Navarrete, S. A., and Berlow, E. L. (2016). How structured is the entangled bank? The surprisingly simple organization of multiplex ecological networks leads to increased persistence and resilience. PLoS biology, 14(8), e1002527. doi: 10.1371/journal.pbio.1002527
[3] Gaucherel, C. (2019). The Languages of Nature. When nature writes to itself. Lulu editions, Paris, France.
[4] Gaucherel, C., and Pommereau, F. Using discrete systems to exhaustively characterize the dynamics of an integrated ecosystem. Methods in Ecology and Evolution, 10(9), 1615-1627. doi: 10.1111/2041-210X.13242
[5] Bennett, J. M. et al. (2018). A review of European studies on pollination networks and pollen limitation, and a case study designed to fill in a gap. AoB Plants, 10(6), ply068. doi: 10.1093/aobpla/ply068
[6] Magrach, A., Molina, F. P., and Bartomeus, I. (2020). Niche complementarity among pollinators increases community-level plant reproductive success. bioRxiv, 629931, ver. 7 peer-reviewed and recommended by PCI Ecology. doi: 10.1101/629931
[7] Bastolla, U., Fortuna, M. A., Pascual-García, A., Ferrera, A., Luque, B., and Bascompte, J. (2009). The architecture of mutualistic networks minimizes competition and increases biodiversity. Nature, 458(7241), 1018-1020. doi: 10.1038/nature07950

Polar and subpolar regions are fragile environments, where the introduction of alien species may completely change ecosystem dynamics if the alien species become keystone species (e.g. Croll, 2005). The increasing number of human visits, together with climate change, are favouring the introduction and settling of new invaders to these regions, particularly in Antarctica (Hughes et al. 2015). Within this context, the joint use of Species Distribution Models (SDM) –to assess the areas potentially suitable for the aliens– with other measures of the potential to become successful invaders can inform on the need for devoting specific efforts to eradicate these new species before they become naturalized (e.g. Pertierra et al. 2016).
Bazzichetto et al. (2020) use data from a detailed inventory, SDMs and trait data altogether to assess the drivers of invasion success of six alien plants on Possession Island, in the remote sub-Antarctic archipelago of Crozet. SDMs have inherent limitations to describe different aspects of species distributions, including the fundamental niche and, with it, the areas that could host viable populations (Hortal et al. 2012). Therefore, their utility to predict future biological invasions is limited (Jiménez-Valverde et al. 2011). However, they can be powerful tools to describe species range dynamics if they are thoughtfully used by adopting conscious decisions about the techniques and data used, and interpreting carefully the actual implications of their results.
This is what Bazzichetto et al. (2020) do, using General Linear Models (GLM) –a technique well rooted in the original niche-based SDM theory (e.g. Austin 1990)– that can provide a meaningful description of the realized niche within the limits of an adequately sampled region. Further, as alien species share and are similarly affected by several steps of the invasion process (Richardson et al. 2000), these authors model the realized distribution of the six species altogether. This can be done through the recently developed joint-SDM, a group of techniques where the co-occurrence of the modelled species is explicitly taken into account during modelling (e.g. Pollock et al. 2014). Here, the addition of species traits has been identified as a key step to understand the associations of species in space (see Dormann et al. 2018). Bazzichetto et al. (2020) combine their GLM-based SDM for each species with a so-called multi-SDM approach, where they assess together the consistency in the interactions between both species and topographically-driven climate variations, and several plant traits and two key anthropic factors –accessibility from human settlements and distance to hiking paths.
This work is a good example on how a theoretically meaningful SDM approach can provide useful –though perhaps not deep– insights on biological invasions for remote landscapes threatened by biotic homogenization. By combining climate and topographic variables as proxies for the spatial variations in the abiotic conditions regulating plant growth, measures of accessibility, and traits of the plant invaders, Bazzichetto et al. (2020) are able to identify the different effects that the interactions between the potential intensity of propagule dissemination by humans, and the ecological characteristics of the invaders themselves, may have on their invasion success.
The innovation of modelling together species responses is important because it allows dissecting the spatial dynamics of spread of the invaders, which indeed vary according to a handful of their traits. For example, their results show that no all old residents have profited from the larger time of residence in the island, as Poa pratensis is seemingly as dependent of a higher intensity of human activity as the newcomer invaders in general are. According to Bazzichetto et al. trait-based analyses, these differences are apparently related with plant height, as smaller plants disperse more easily. Further, being perennial also provides an advantage for the persistence in areas with less human influence. This puts name, shame and fame to the known influence of plant life history on their dispersal success (Beckman et al. 2018), at least for the particular case of plant invasions in Possession Island.
Of course this approach has limitations, as data on the texture, chemistry and temperature of the soil are not available, and thus were not considered in the analyses. These factors may be critical for both establishment and persistence of small plants in the harsh Antarctic environments, as Bazzichetto et al. (2020) recognize. But all in all, their results provide key insights on which traits may confer alien plants with a higher likelihood of becoming successful invaders in the fragile Antarctic and sub-Antarctic ecosystems. This opens a way for rapid assessments of invasibility, which will help identifying which species in the process of naturalizing may require active contention measures to prevent them from becoming ecological game changers and cause disastrous cascade effects that shift the dynamics of native ecosystems.

References

Austin, M. P., Nicholls, A. O., and Margules, C. R. (1990). Measurement of the realized qualitative niche: environmental niches of five Eucalyptus species. Ecological Monographs, 60(2), 161-177. doi: https://doi.org/10.2307/1943043
Bazzichetto, M., Massol, F., Carboni, M., Lenoir, J., Lembrechts, J. J. and Joly, R. (2020) Once upon a time in the far south: Influence of local drivers and functional traits on plant invasion in the harsh sub-Antarctic islands. bioRxiv, 2020.07.19.210880, ver. 3 peer-reviewed and recommended by PCI Ecology. doi: https://doi.org/10.1101/2020.07.19.210880
Beckman, N. G., Bullock, J. M., and Salguero-Gómez, R. (2018). High dispersal ability is related to fast life-history strategies. Journal of Ecology, 106(4), 1349-1362. doi: https://doi.org/10.1111/1365-2745.12989
Croll, D. A., Maron, J. L., Estes, J. A., Danner, E. M., and Byrd, G. V. (2005). Introduced predators transform subarctic islands from grassland to tundra. Science, 307(5717), 1959-1961. doi: https://doi.org/10.1126/science.1108485
Dormann, C. F., Bobrowski, M., Dehling, D. M., Harris, D. J., Hartig, F., Lischke, H., Moretti, M. D., Pagel, J., Pinkert, S., Schleuning, M., Schmidt, S. I., Sheppard, C. S., Steinbauer, M. J., Zeuss, D., and Kraan, C. (2018). Biotic interactions in species distribution modelling: 10 questions to guide interpretation and avoid false conclusions. Global Ecology and Biogeography, 27(9), 1004-1016. doi: https://doi.org/10.1111/geb.12759
Jiménez-Valverde, A., Peterson, A., Soberón, J., Overton, J., Aragón, P., and Lobo, J. (2011). Use of niche models in invasive species risk assessments. Biological Invasions, 13(12), 2785-2797. doi: https://doi.org/10.1007/s10530-011-9963-4
Hortal, J., Lobo, J. M., and Jiménez-Valverde, A. (2012). Basic questions in biogeography and the (lack of) simplicity of species distributions: Putting species distribution models in the right place. Natureza & Conservação – Brazilian Journal of Nature Conservation, 10(2), 108-118. doi: https://doi.org/10.4322/natcon.2012.029
Hughes, K. A., Pertierra, L. R., Molina-Montenegro, M. A., and Convey, P. (2015). Biological invasions in terrestrial Antarctica: what is the current status and can we respond? Biodiversity and Conservation, 24(5), 1031-1055. doi: https://doi.org/10.1007/s10531-015-0896-6
Pertierra, L. R., Baker, M., Howard, C., Vega, G. C., Olalla-Tarraga, M. A., and Scott, J. (2016). Assessing the invasive risk of two non-native Agrostis species on sub-Antarctic Macquarie Island. Polar Biology, 39(12), 2361-2371. doi: https://doi.org/10.1007/s00300-016-1912-3
Pollock, L. J., Tingley, R., Morris, W. K., Golding, N., O'Hara, R. B., Parris, K. M., Vesk, P. A., and McCarthy, M. A. (2014). Understanding co-occurrence by modelling species simultaneously with a Joint Species Distribution Model (JSDM). Methods in Ecology and Evolution, 5(5), 397-406. doi: https://doi.org/10.1111/2041-210X.12180
Richardson, D. M., Pyšek, P., Rejmánek, M., Barbour, M. G., Panetta, F. D., and West, C. J. (2000). Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions, 6(2), 93-107. doi: https://doi.org/10.1046/j.1472-4642.2000.00083.x

What kinds of studies are most needed to understand the effects of global change on nature? Two deficiencies stand out: lack of long-term studies [1] and lack of data on species interactions [2]. The paper by Mennerat and colleagues [3] is particularly valuable because it addresses both of these shortcomings. The first one is obvious. Our understanding of the impact of climate on biota improves with longer times series of observations. Mennerat et al. [3] analysed an impressive 18-year series from multiple sites to search for trends in parasitism rates across a range of temperatures. The second deficiency (lack of species interaction data) is perhaps not yet fully appreciated, despite studies pointing this out ten years ago [2,4]. The focus is often on species range limits and how taking species interactions into account changes species range predictions based on climate alone (climate envelope models; [5]). But range limits are not everything, as the function of a species (or community, network, etc.) ultimately depends on the strengths of species interactions and not only on the presence or absence of a given species [2,4]. Mennerat et al. [3] show that in the case of birds and their nest parasites, it is the strength of the interaction that has changed, while the species involved stayed the same. Mennerat et al. [3] found nest parasitism to increase with temperature at the nestling stage. They have also searched for trends of parasitism dynamics dependence on the host, but did not find any, probably because the nest parasites are generalists and attack other bird species within the study sites. This study thus draws attention to wider networks of interacting species, and we urgently need more data to predict how interaction networks will rewire with progressing environmental change [6,7].

References

[1] Lindenmayer, D.B., Likens, G.E., Andersen, A., Bowman, D., Bull, C.M., Burns, E., et al. (2012). Value of long-term ecological studies. Austral Ecology, 37(7), 745–57. doi: 10.1111/j.1442-9993.2011.02351.x
[2] Tylianakis, J.M., Didham, R.K., Bascompte, J. & Wardle, D.A. (2008). Global change and species interactions in terrestrial ecosystems. Ecology Letters, 11(12), 1351–63. doi: 10.1111/j.1461-0248.2008.01250.x
[3] Mennerat, A., Charmantier, A., Hurtrez-Bousses, S., Perret, P. & Lambrechts, M.M. (2019). Parasite intensity is driven by temperature in a wild bird. bioRxiv, 323311. Ver. 4 peer-reviewed and recommended by PCI Ecology. doi: 10.1101/323311
[4] Gilman, S.E., Urban, M.C., Tewksbury, J., Gilchrist, G.W. & Holt, R.D. (2010). A framework for community interactions under climate change. Trends in Ecology & Evolution, 25(6), 325–31. doi: 10.1016/j.tree.2010.03.002
[5] Louthan, A.M., Doak, D.F. & Angert, A.L. (2015). Where and when do species interactions set range limits? Trends in Ecology & Evolution, 30(12), 780–92. doi: 10.1016/j.tree.2015.09.011
[6] Bartley, T.J., McCann, K.S., Bieg, C., Cazelles, K., Granados, M., Guzzo, M.M., et al. (2019). Food web rewiring in a changing world. Nature Ecology & Evolution, 3(3), 345–54. doi: 10.1038/s41559-018-0772-3
[7] Staniczenko, P.P.A., Lewis, O.T., Jones, N.S. & Reed-Tsochas, F. (2010). Structural dynamics and robustness of food webs. Ecology Letters, 13(7), 891–9. doi: 10.1111/j.1461-0248.2010.01485.x