In the field of ecology, there is a growing interest in machine (including deep) learning for processing and automatizing repetitive analyses on large amounts of images collected from camera traps, drones and smartphones, among others. These analyses include species or individual recognition and classification, counting or tracking individuals, detecting and classifying behavior. By saving countless times of manual work and tapping into massive amounts of data that keep accumulating with technological advances, machine learning is becoming an essential tool for ecologists. We refer to recent papers for more details on machine learning for ecology and evolution (Besson et al. 2022, Borowiec et al. 2022, Christin et al. 2019, Goodwin et al. 2022, Lamba et al. 2019, Nazir & Kaleem 2021, Perry et al. 2022, Picher & Hartig 2023, Tuia et al. 2022, Wäldchen & Mäder 2018).

In their paper, Nakagawa et al. (2023) conducted a systematic review of the literature on machine learning for wildlife imagery. Interestingly, the authors used a method unfamiliar to ecologists but well-established in medicine called rapid review, which has the advantage of being quickly completed compared to a fully comprehensive systematic review while being representative (Lagisz et al., 2022). Through a rigorous examination of more than 200 articles, the authors identified trends and gaps, and provided suggestions for future work. Listing all their findings would be counterproductive (you’d better read the paper), and I will focus on a few results that I have found striking, fully assuming a biased reading of the paper. First, Nakagawa et al. (2023) found that most articles used neural networks to analyze images, in general through collaboration with computer scientists. A challenge here is probably to think of teaching computer vision to the generations of ecologists to come (Cole et al. 2023). Second, the images were dominantly collected from camera traps, with an increase in the use of aerial images from drones/aircrafts that raise specific challenges. Third, the species concerned were mostly mammals and birds, suggesting that future applications should aim to mitigate this taxonomic bias, by including, e.g., invertebrate species. Fourth, most papers were written by authors affiliated with three countries (Australia, China, and the USA) while India and African countries provided lots of images, likely an example of scientific colonialism which should be tackled by e.g., capacity building and the involvement of local collaborators. Last, few studies shared their code and data, which obviously impedes reproducibility. Hopefully, with the journals’ policy of mandatory sharing of codes and data, this trend will be reversed. 

REFERENCES

Besson M, Alison J, Bjerge K, Gorochowski TE, Høye TT, Jucker T, Mann HMR, Clements CF (2022) Towards the fully automated monitoring of ecological communities. Ecology Letters, 25, 2753–2775. https://doi.org/10.1111/ele.14123

Borowiec ML, Dikow RB, Frandsen PB, McKeeken A, Valentini G, White AE (2022) Deep learning as a tool for ecology and evolution. Methods in Ecology and Evolution, 13, 1640–1660. https://doi.org/10.1111/2041-210X.13901

Christin S, Hervet É, Lecomte N (2019) Applications for deep learning in ecology. Methods in Ecology and Evolution, 10, 1632–1644. https://doi.org/10.1111/2041-210X.13256

Cole E, Stathatos S, Lütjens B, Sharma T, Kay J, Parham J, Kellenberger B, Beery S (2023) Teaching Computer Vision for Ecology. https://doi.org/10.48550/arXiv.2301.02211

Goodwin M, Halvorsen KT, Jiao L, Knausgård KM, Martin AH, Moyano M, Oomen RA, Rasmussen JH, Sørdalen TK, Thorbjørnsen SH (2022) Unlocking the potential of deep learning for marine ecology: overview, applications, and outlook†. ICES Journal of Marine Science, 79, 319–336. https://doi.org/10.1093/icesjms/fsab255

Lagisz M, Vasilakopoulou K, Bridge C, Santamouris M, Nakagawa S (2022) Rapid systematic reviews for synthesizing research on built environment. Environmental Development, 43, 100730. https://doi.org/10.1016/j.envdev.2022.100730

Lamba A, Cassey P, Segaran RR, Koh LP (2019) Deep learning for environmental conservation. Current Biology, 29, R977–R982. https://doi.org/10.1016/j.cub.2019.08.016

Nakagawa S, Lagisz M, Francis R, Tam J, Li X, Elphinstone A, Jordan N, O’Brien J, Pitcher B, Sluys MV, Sowmya A, Kingsford R (2023) Rapid literature mapping on the recent use of machine learning for wildlife imagery. EcoEvoRxiv, ver. 4 peer-reviewed and recommended by Peer Community in Ecology.  https://doi.org/10.32942/X2H59D

Nazir S, Kaleem M (2021) Advances in image acquisition and processing technologies transforming animal ecological studies. Ecological Informatics, 61, 101212. https://doi.org/10.1016/j.ecoinf.2021.101212

Perry GLW, Seidl R, Bellvé AM, Rammer W (2022) An Outlook for Deep Learning in Ecosystem Science. Ecosystems, 25, 1700–1718. https://doi.org/10.1007/s10021-022-00789-y

Pichler M, Hartig F Machine learning and deep learning—A review for ecologists. Methods in Ecology and Evolution, n/a. https://doi.org/10.1111/2041-210X.14061

Tuia D, Kellenberger B, Beery S, Costelloe BR, Zuffi S, Risse B, Mathis A, Mathis MW, van Langevelde F, Burghardt T, Kays R, Klinck H, Wikelski M, Couzin ID, van Horn G, Crofoot MC, Stewart CV, Berger-Wolf T (2022) Perspectives in machine learning for wildlife conservation. Nature Communications, 13, 792. https://doi.org/10.1038/s41467-022-27980-y

Wäldchen J, Mäder P (2018) Machine learning for image-based species identification. Methods in Ecology and Evolution, 9, 2216–2225. https://doi.org/10.1111/2041-210X.13075

Two fields of research have grown considerably over the past twenty years: the investigation of human-wildlife conflicts (e.g. see Treves & Santiago-Ávila 2020), and multispecies occupancy modelling (Devarajan et al. 2020). In their recent study, Lauret et al. (2023) combined both in an elegant methodological framework, applied to the study of the co-occurrence of fishing activities and bottlenose dolphins in the French Mediterranean.

A common issue with human-wildlife conflicts (and, in particular, fishery by-catch) is that data is often only available from those conflicts or interactions, limiting the validity of the predictions (Kuiper et al. 2022). Lauret et al. use independent data sources informing the occurrence of fishing vessels and dolphins, combined in a Bayesian multispecies occupancy model where vessels are "the other species". I particularly enjoyed that approach, as integration of human activities in ecological models can be extremely complex, but can also translate in phenomena that can be captured as one would of individuals of a species, as long as the assumptions are made clearly. Here, the model is made more interesting by accounting for environmental factors (seabed depth) borrowing an approach from Generalized Additive Models in the Bayesian framework. While not pretending to provide (yet) practical recommendations to help conserve bottlenose dolphins (and other wildlife conflicts), this study and the associated code are a promising step in that direction.

REFERENCES

Devarajan, K., Morelli, T.L. & Tenan, S. (2020), Multi-species occupancy models: review, roadmap, and recommendations. Ecography, 43: 1612-1624. https://doi.org/10.1111/ecog.04957

Kuiper, T., Loveridge, A.J. and Macdonald, D.W. (2022), Robust mapping of human–wildlife conflict: controlling for livestock distribution in carnivore depredation models. Anim. Conserv., 25: 195-207. https://doi.org/10.1111/acv.12730

Lauret V, Labach H, David L, Authier M, & Gimenez O (2023) Using integrated multispecies occupancy models to map co-occurrence between bottlenose dolphins and fisheries in the Gulf of Lion, French Mediterranean Sea. Ecoevoarxiv, ver. 2 peer-reviewed and recommended by PCI Ecology. https://doi.org/10.32942/osf.io/npd6u

Treves, A. & Santiago-Ávila, F.J. (2020). Myths and assumptions about human-wildlife conflict and coexistence. Conserv. Biol. 34, 811–818.  https://doi.org/10.1111/cobi.13472

Are you an ecologist who uses a computer or know someone that does? Even if your research doesn’t rely heavily on advanced computational techniques, it likely hasn’t escaped your attention that computers are increasingly being used to analyse field data and make predictions about the consequences of environmental change. So before artificial intelligence and robots take over from scientists, now is great time to read about how experts think computers could make your life easier and lead to innovations in ecological research. In “Data-based, synthesis-driven: setting the agenda for computational ecology”, Poisot and colleagues [1] provide a brief history of computational ecology and offer their thoughts on how computational thinking can help to bridge different types of ecological knowledge. In this wide-ranging article, the authors share practical strategies for realising three main goals: (i) tighter integration of data and models to make predictions that motivate action by practitioners and policy-makers; (ii) closer interaction between data-collectors and data-users; and (iii) enthusiasm and aptitude for computational techniques in future generations of ecologists. The key, Poisot and colleagues argue, is for ecologists to “engage in meaningful dialogue across disciplines, and recognize the currencies of their collaborations.” Yes, this is easier said than done. However, the journey is much easier with a guide and when everyone involved serves to benefit not only from the eventual outcome, but also the process.

References

[1] Poisot, T., Labrie, R., Larson, E., & Rahlin, A. (2018). Data-based, synthesis-driven: setting the agenda for computational ecology. BioRxiv, 150128, ver. 4 recommended and peer-reviewed by PCI Ecology. doi: 10.1101/150128

Host-parasite interactions are ubiquitous on Earth. They are present in almost every conceivable ecosystem and often result from a long history of antagonist coevolution [1,2]. Recent studies on climate change have revealed, however, that modification of abiotic variables are often accompanied by shifts in the distributional range of parasites to habitats far beyond their original geographical distribution, creating new interactions in novel habitats with unpredictable consequences for host community structure and organization [3,4]. This situation may be especially critical for endangered host species having small population abundance and restricted distribution range. The infestation of bird species with larvae of the muscid fly genus Philornis is a case in point. At least 250 bird species inhabiting mostly Central and South America are infected by Philornis flies [5,6]. Fly larval development occurs in bird faeces, nesting material, or inside nestlings, affecting the development and nestling survival.
Recent reports indicate significant reduction of bird numbers associated with recent Philornis infection, the most conspicuous being Galapagos finches [7,8]. One way to prevent this potential effect consists in to examine the expected geographical shift of Philornis fly species under future climate change scenarios so that anticipatory conservation practices become implemented for endangered bird species. In this regard, Ecological Niche Modeling (ENM) techniques have been increasingly used as a useful tool to predict disease transmission as well as the species becoming infected under different climate change scenarios [9-11]. The paper of Cuervo et al. [12] is an important advance in this regard. By identifying for the first time the macro-environmental variables influencing the abiotic niche of species of the Philornis torquans complex in southern South America, the authors perform a geographical projection model that permits identification of the areas susceptible to be colonized by Philornis species in Argentina, Brazil, and Chile, including habitats where the parasitic fly is still largely absent at present. Their results are promissory for conservation studies and contribute to the still underdeveloped issue of the way climate change impacts on antagonistic ecological relationships.

References

[1] Thompson JN (1994) The Coevolutionary Process. University of Chicago Press.
[2] Poulin R (2007) Evolutionary Ecology of Parasites: (Second Edition). Princeton University Press. doi: 10.2307/j.ctt7sn0x
[3] Pickles RSA, Thornton D, Feldman R, Marques A, Murray DL (2013) Predicting shifts in parasite distribution with climate change: a multitrophic level approach. Global Change Biology, 19, 2645–2654. doi: 10.1111/gcb.12255
[4] Marcogliese DJ (2016) The distribution and abundance of parasites in aquatic ecosystems in a changing climate: More than just temperature. Integrative and Comparative Biology, 56, 611–619. doi: 10.1093/icb/icw036
[5] Dudaniec RY, Kleindorfer S (2006) Effects of the parasitic flies of the genus Philornis (Diptera: Muscidae) on birds. Emu - Austral Ornithology, 106, 13–20. doi: 10.1071/MU04040
[6] Antoniazzi LR, Manzoli DE, Rohrmann D, Saravia MJ, Silvestri L, Beldomenico PM (2011) Climate variability affects the impact of parasitic flies on Argentinean forest birds. Journal of Zoology, 283, 126–134. doi: 10.1111/j.1469-7998.2010.00753.x
[7] Fessl B, Sinclair BJ, Kleindorfer S (2006) The life-cycle of Philornis downsi (Diptera: Muscidae) parasitizing Darwin’s finches and its impacts on nestling survival. Parasitology, 133, 739–747. doi: 10.1017/S0031182006001089
[8] Kleindorfer S, Peters KJ, Custance G, Dudaniec RY, O’Connor JA (2014) Changes in Philornis infestation behavior threaten Darwin’s finch survival. Current Zoology, 60, 542–550. doi: 10.1093/czoolo/60.4.542
[9] Johnson EE, Escobar LE, Zambrana-Torrelio C (2019) An ecological framework for modeling the geography of disease transmission. Trends in Ecology and Evolution, 34, 655–668. doi: 10.1016/j.tree.2019.03.004
[10] Carvalho BM, Rangel EF, Ready PD, Vale MM (2015) Ecological niche modelling predicts southward expansion of Lutzomyia (Nyssomyia) flaviscutellata (Diptera: Psychodidae: Phlebotominae), vector of Leishmania (Leishmania) amazonensis in South America, under climate change. PLOS ONE, 10, e0143282. doi: 10.1371/journal.pone.0143282
[11] Garrido R, Bacigalupo A, Peña-Gómez F, Bustamante RO, Cattan PE, Gorla DE, Botto-Mahan C (2019) Potential impact of climate change on the geographical distribution of two wild vectors of Chagas disease in Chile: Mepraia spinolai and Mepraia gajardoi. Parasites and Vectors, 12, 478. doi: 10.1186/s13071-019-3744-9
[12] Cuervo PF, Percara A, Monje L, Beldomenico PM, Quiroga MA (2020) Environmental variables determining the distribution of an avian parasite: the case of the Philornis torquans complex (Diptera: Muscidae) in South America. bioRxiv, 839589, ver. 5 peer-reviewed and recommended by PCI Ecology. doi: 10.1101/839589

Hyperaccumulator plants can concentrate heavy metals present on the soil in their tissues, avoiding their toxic effects and potentially discouraging herbivores (Martens & Boyd, 1994). But not all herbivores are necessarily discouraged, and access to locally abundant resources with low interspecific competition from other herbivores, can affect feeding choices. Godinho et al. performed a series of controlled laboratorial trials to evaluate if herbivores (spider mites) avoid tomato plants with high concentrations of Cadmium under alternative scenarios, namely: the presence/absence of conspecifics, the presence/absence of a competitor species (a congeneric mite), and the relative abundance of contaminated plants.

They found that when looking for plants to lay their eggs, individual spider-mites (females) do not seem to discriminate between plants with or without cadmium, despite a significantly lower performance on the former. However, they consistently chose plants without Cadmium in set-ups where 200 mites are faced with this decision together. This preference was consistent and independent from the relative abundance of cadmium-free plants, but only when mites do this decision collectively. In addition, this preference was stronger than that for plants where interspecific competition was lower, with mites preferring to face high competition from congeneric herbivores than laying their eggs on Cadmium contaminated plants. 

Taken together these experiments suggest that aggregation is a key mechanism by which spider mites can avoid metal contaminated plants. As good research often does, these experiments open several important questions that will need to be addressed in the future. In particular, it will be important to clarify what are the sensorial and behavioural mechanisms that allow this decision/outcome when spider mites make this choice collectively but lead to a different outcome (no choice) when they face this decision alone. Additionally, it will be interesting to explore the potentially adaptive (or non-adaptive) consequences of this behaviour in terms of individual and inclusive fitness. One thing seems certain: both the abiotic and the biotic context can affect spider mite choices, and both need to be considered to advance our understanding about the trade-offs between plant defence mechanisms and associated herbivore decisions and fitness. 

References

Martens, S. N., & Boyd, R. S. (1994). The ecological significance of nickel hyperaccumulation: a plant chemical defense. Oecologia, 98(3–4), 379–384. https://doi.org/10.1007/BF00324227

Godinho, D. P., I. Fragata, M. C. de la Masseliere, S. Magalhaes 2024 Spider mites collectively avoid plants with cadmium irrespective of their frequency or the presence of competitors. bioRxiv, ver. 4, peer-reviewed and recommended by PCI Ecology 2023.08.17.553707. https://doi.org/10.1101/2023.08.17.553707