Tonje Knutsen Sørdalen

• Tur, Anil Osman; Sørdalen, Tonje Knutsen; Halvorsen, Kim Aleksander Tallaksen & Beyan, Cigdem (2026). Parameter-efficient vision–language adaptation with continuous metadata conditioning for animal re-identification. Expert Systems With Applications. ISSN 0957-4174. 332(C). doi: 10.1016/j.eswa.2026.133618. Full text in Research Archive Show summary

  Long-term animal re-identification (ReID) must remain robust to gradual morphological evolution and seasonal appearance shifts. Although recent vision–language models provide strong pretrained visual representations, adapting them to longitudinal ecological settings remains challenging, particularly under identity and temporal distribution shifts. We present a parameter-efficient CLIP adaptation framework for animal ReID and introduce a continuous metadata-conditioning mechanism that incorporates numerical attributes directly into the prompt representation during training. While low-rank visual adaptation, prompt-based supervision, and cross-modal alignment provide the adaptation framework, the proposed metadata-conditioning strategy constitutes the primary methodological contribution. By preserving the continuous structure of numerical metadata rather than discretizing it into textual categories, the proposed approach enables smooth modulation of the embedding space during training while maintaining a purely visual inference pipeline. Experiments on a seven-year longitudinal fish dataset and multiple wildlife benchmarks demonstrate improved performance under closed-set, open-set, and time-aware evaluation protocols. The results demonstrate that continuous metadata conditioning improves robustness to longitudinal appearance variation and temporal distribution shifts, while parameter-efficient adaptation enables a purely visual inference pipeline without requiring metadata at test time.

• Sørdalen, Tonje Knutsen; Malde, Ketil; Sauvaitre, C.; Skiftesvik, Anne Berit; Beyan, C. & Larsen, Torkel [Show all 7 contributors for this article] (2026). A wild fish image dataset for individual re-identification and phenotyping. Scientific Data. 13. doi: 10.1038/s41597-026-07045-1. Full text in Research Archive Show summary

  Computer vision can transform wildlife monitoring by automating phenotyping and individual identification. Achieving this, however, depends on access to large, well-curated image datasets that capture natural variation across individuals and years. Here, we present Melops, a longitudinal dataset comprising 24 578 images of 9 861 individual corkwing wrasse, Symphodus melops, collected over seven years through a capture–mark–recapture survey. Each fish was PIT-tagged for re-identification and photographed from both sides against a standardized white background with a colour reference. Alongside the images, we provide metadata including body length, sex, and reproductive state. To support deep learning applications, the dataset includes both the original photographs and automatically cropped images focusing on the whole fish or specific body regions. Together, these resources provide a foundation for developing computer vision methods for individual re-identification, colour pattern analysis, sex classification and other visual phenotyping tasks. Beyond this species, Melops can serve as a model for similar datasets in other taxa. Because it contains thousands of individuals with repeated observations, it provides a rare opportunity to explore temporally aware re-identification and phenotypic change in wild fish.

• Ellis, Benjamin A.; Sørdalen, Tonje Knutsen; Briffa, Mark; Skiftesvik, Anne Berit; Wilson, Alexander D.M. & Halvorsen, Kim Aleksander Tallaksen (2026). Visual cues elicit differential aggression towards female and female mimics in the corkwing wrasse. Behavioral Ecology. ISSN 1045-2249. 37(3). doi: 10.1093/beheco/arag022. Full text in Research Archive Show summary

  Alternative reproductive tactics (ARTs) are consistent intrasexual differences in behavior that maximize reproductive success through different means. In corkwing wrasse (Symphodus melops), smaller males mimic female visual morphology (hereafter sneakers) to steal fertilizations from nest-building parental males (hereafter nesting males). We investigated whether this visual mimicry effectively deceived 47 nesting males across 4 sites by quantifying 2 agonistic responses (attack latency and attack frequency) to models depicting unfamiliar females, sneakers and rival nesting males. Contrary to expectations, nesting males exhibited faster agonistic responses towards female models, with lower attack latencies compared with both male types. Additionally, the relationship between model type and attack frequency varied between sites, with reduced attack frequencies at sites subject to increased levels of anthropogenic disturbance. These results provide evidence that nesting males can visually discriminate between females and sneakers even in the absence of behavioral cues, suggesting that visual mimicry alone may be insufficient for successful deception. This highlights the need to investigate alternative sensory modalities, such as behavioral mimicry, to explain the successful reproduction of sneaker males and the maintenance of ARTs in this species.

• Canovi, Noemi; Ellis, Benjamin A.; Sørdalen, Tonje Knutsen; Allken, Vaneeda; Halvorsen, Kim Aleksander Tallaksen & Malde, Ketil [Show all 7 contributors for this article] (2024). Trajectory-based fish event classification through pre-training with diffusion models. Ecological Informatics. ISSN 1574-9541. 82. doi: 10.1016/j.ecoinf.2024.102733. Full text in Research Archive Show summary

  This study contributes to advancing the field of automatic fish event recognition in natural underwater videos, addressing the current gap in studying fish interaction and competition, including predator-prey relationships and mating behaviors. We used the corkwing wrasse (Symphodus melops) as a model, a marine species of commercial importance that reproduces in sea-weed nests built and cared for by a single male. These nests attract a wide range of visitors and are the focal point for behavior such as spawning, chasing, and maintenance. We propose a deep learning methodology to analyze the movement trajectories of the nesting male and classify the associated events observed in their natural habitat. Our approach leverages unsupervised pre-training based on diffusion models, leading to improved feature learning. Additionally, we introduce a dataset comprising 16,937 trajectories across 12 event classes, making it the largest in terms of event class diversity. Our results demonstrate the superior performance of our method compared to several deep architectures. The code for the proposed method and the trajectories can be found at https://github.com/NoeCanovi/Fish_Behaviors_Generative_Models .

• Theberge, Kaitlyn; Sørdalen, Tonje Knutsen; Pugh, Tracy L. & Kindsvater, Holly K. (2024). Assessing the effects of female protections on size structure and spawning potential in two clawed lobster fisheries subject to varying exploitation levels. Fisheries Research. ISSN 0165-7836. 280. doi: 10.1016/j.fishres.2024.107155. Full text in Research Archive
• Olsen, Ørjan Langøy; Sørdalen, Tonje Knutsen; Goodwin, Morten; Malde, Ketil; Knausgård, Kristian Muri & Halvorsen, Kim Aleksander Tallaksen (2023). A contrastive learning approach for individual re-identification in a wild fish population. Proceedings of the Northern Lights Deep Learning Workshop. 4. doi: 10.7557/18.6824. Full text in Research Archive Show summary

  In both terrestrial and marine ecology, physical tagging is a frequently used method to study population dynamics and behavior. However, such tagging techniques are increasingly being replaced by individual re-identification using image analysis. This paper introduces a contrastive learning-based model for identifying individuals. The model uses the first parts of the Inception v3 network, supported by a projection head, and we use contrastive learning to find similar or dissimilar image pairs from a collection of uniform photographs. We apply this technique for corkwing wrasse, Symphodus melops, an ecologically and commercially important fish species. Photos are taken during repeated catches of the same individuals from a wild population, where the intervals between individual sightings might range from a few days to several years. Our model achieves a one-shot accuracy of 0.35, a 5-shot accuracy of 0.56, and a 100-shot accuracy of 0.88, on our dataset.

• Knutsen, Jan Atle; Kleiven, Alf Ring; Olsen, Esben Moland; Knutsen, Halvor; Espeland, Sigurd Heiberg & Sørdalen, Tonje Knutsen [Show all 15 contributors for this article] (2022). Lobster reserves as a management tool in coastal waters: Two decades of experience in Norway. Marine Policy. ISSN 0308-597X. 136. doi: 10.1016/j.marpol.2021.104908. Full text in Research Archive Show summary

  The positive effects of reduced fishing pressure in marine protected areas (MPAs) are now well documented globally. Yet, evidence of MPA benefits from long-term replicated before-after control-impact (BACI) studies and their usefulness in protecting target species are still rare, especially in northern temperate areas. Scientific rigor in the monitoring of MPAs is considered important for obtaining trust and compliance and can increase interest and enthusiasm for the benefits of marine conservation. Off the coast of southern Norway, a MPA imple- mentation process started up in 2002. Based on comprehensive consultations with local fishers and managers, four experimental lobster reserves were appointed in 2004. Two years later (2006), the reserves came into effect as the Norwegian Directorate of Fisheries implemented regulations as a by-law of the Saltwater Fisheries Act that effectively banned all fixed gear. Long-term monitoring of the MPAs and adjacent control areas has enabled a rigorous scientific evaluation of the effects of these MPAs on lobster populations, including effects on density, growth, demography, behavior, and phenotypic diversity. As protection effects started to manifest, the lobster reserves attracted high public attention and were soon considered a credible supplement to traditional fisheries management. In the period from 2002 to 2021, more than 50 lobster reserves have been implemented in Norway. Here, we review the experiences since the lobster reserves were designated, implemented, and embraced by local communities in Norway, and over two decades have become an important tool for fishery management. Thoughts on the future of MPAs along the coast of Norway are discussed.

• Sørdalen, Tonje Knutsen; Halvorsen, Kim Aleksander Tallaksen & Olsen, Esben Moland (2022). Protection from fishing improves body growth of an exploited species. Proceedings of the Royal Society of London. Biological Sciences. ISSN 0962-8452. 289(1987). doi: 10.1098/rspb.2022.1718. Full text in Research Archive Show summary

  Hunting and fishing are often size-selective, which favours slow body growth. In addition, fast growth rate has been shown to be positively correlated with behavioural traits that increase encounter rates and catchability in passive fishing gears such as baited traps. This harvest-induced selection should be effectively eliminated in no-take marine-protected areas (MPAs) unless strong density dependence results in reduced growth rates. We compared body growth of European lobster (Homarus gammarus) between three MPAs and three fished areas. After 14 years of protection from intensive, size-selective lobster fisheries, the densities in MPAs have increased considerably, and we demonstrate that females moult more frequently and grow more during each moult in the MPAs. A similar, but weaker pattern was evident for males. This study suggests that MPAs can shield a wild population from slow-growth selection, which can explain the rapid recovery of size structure following implementation. If slow-growth selection is a widespread phenomenon in fisheries, the effectiveness of MPAs as a management tool can be higher than currently anticipated.

• Goodwin, Morten; Halvorsen, Kim Aleksander Tallaksen; Jiao, Lei; Knausgård, Kristian Muri; Martin, Angela Helen & Moyano, Marta [Show all 10 contributors for this article] (2022). Unlocking the potential of deep learning for marine ecology: overview, applications, and outlook. ICES Journal of Marine Science. ISSN 1054-3139. 79(2), p. 319–336. doi: 10.1093/icesjms/fsab255. Full text in Research Archive Show summary

  The deep learning (DL) revolution is touching all scientific disciplines and corners of our lives as a means of harnessing the power of big data. Marine ecology is no exception. New methods provide analysis of data from sensors, cameras, and acoustic recorders, even in real time, in ways that are reproducible and rapid. Off-the-shelf algorithms find, count, and classify species from digital images or video and detect cryptic patterns in noisy data. These endeavours require collaboration across ecological and data science disciplines, which can be challenging to initiate. To promote the use of DL towards ecosystem-based management of the sea, this paper aims to bridge the gap between marine ecologists and computer scientists. We provide insight into popular DL approaches for ecological data analysis, focusing on supervised learning techniques with deep neural networks, and illustrate challenges and opportunities through established and emerging applications of DL to marine ecology. We present case studies on plankton, fish, marine mammals, pollution, and nutrient cycling that involve object detection, classification, tracking, and segmentation of visualized data. We conclude with a broad outlook of the field’s opportunities and challenges, including potential technological advances and issues with managing complex data sets.

• Deininger, Anne; Martin, Angela Helen; Pardo, Juan; Berg, Paul Ragnar; Bhardwaj, Jyotirmoy & Catarino, Diana Sofia de Oliveira [Show all 15 contributors for this article] (2021). Coastal Research Seen Through an Early Career Lens—A Perspective on Barriers to Interdisciplinarity in Norway. Frontiers in Marine Science. 8. doi: 10.3389/fmars.2021.634999. Full text in Research Archive Show summary

  The value of interdisciplinarity for solving complex coastal problems is widely recognized. Many early career researchers (ECRs) therefore actively seek this type of collaboration through choice or necessity, for professional development or project funding. However, establishing and conducting interdisciplinary research collaborations as an ECR has many challenges. Here, we identify these challenges through the lens of ECRs working in different disciplines on a common ecosystem, the Norwegian Skagerrak coast. The most densely populated coastline in Norway, the Skagerrak coast, is experiencing a multitude of anthropogenic stressors including fishing, aquaculture, eutrophication, climate change, land runoff, development, and invasive species. The Skagerrak coastline has also been the focus of environmental science research for decades, much of which aims to inform management of these stressors. The region provides a fantastic opportunity for interdisciplinary collaboration, both within and beyond the environmental sciences. This perspective article identifies the barriers ECRs in Norway face in establishing interdisciplinary and collaborative research to inform management of coastal ecosystems, along with their root causes. We believe our discussion will be of broad interest to all research institutions who employ or educate ECRs (in Norway and worldwide), and to those who develop funding mechanisms for ECRs and interdisciplinary research.

• Halvorsen, Kim Aleksander Tallaksen; Larsen, Torkel; Browman, Howard; Durif, Caroline; Aasen, Nicolai Ljøstad & Vøllestad, Leif Asbjørn [Show all 10 contributors for this article] (2021). Movement patterns of temperate wrasses (Labridae) within a small marine protected area. Journal of Fish Biology. ISSN 0022-1112. 99(4), p. 1513–1518. doi: 10.1111/jfb.14825. Full text in Research Archive Show summary

  The movement patterns of three commercially important wrasse (Labridae) species inside a small marine protected area (~ 0.15 km2) on the west coast of Norway were analysed over a period of 21 months. The mean distance between capture and recapture locations varied between 10 and 187 m, and was species and season specific. The extent of movement was not related to body size or sex. These results imply that a network of small strategically located marine protected areas can be used as management tools to protect wrasses from size- and sex-selective fishing mortality.

• Knausgård, Kristian Muri; Wiklund, Arne; Sørdalen, Tonje Knutsen; Halvorsen, Kim Tallaksen; Kleiven, Alf Ring & Lei, Jiao [Show all 7 contributors for this article] (2021). Temperate fish detection and classification: a deep learning based approach. Applied Intelligence - The International Journal of Research on Intelligent Systems for Real Life Complex Problems. ISSN 0924-669X. 52. doi: 10.1007/s10489-020-02154-9. Full text in Research Archive Show summary

  A wide range of applications in marine ecology extensively uses underwater cameras. Still, to efficiently process the vast amount of data generated, we need to develop tools that can automatically detect and recognize species captured on film. Classifying fish species from videos and images in natural environments can be challenging because of noise and variation in illumination and the surrounding habitat. In this paper, we propose a two-step deep learning approach for the detection and classification of temperate fishes without pre-filtering. The first step is to detect each single fish in an image, independent of species and sex. For this purpose, we employ the You Only Look Once (YOLO) object detection technique. In the second step, we adopt a Convolutional Neural Network (CNN) with the Squeeze-and-Excitation (SE) architecture for classifying each fish in the image without pre-filtering. We apply transfer learning to overcome the limited training samples of temperate fishes and to improve the accuracy of the classification. This is done by training the object detection model with ImageNet and the fish classifier via a public dataset (Fish4Knowledge), whereupon both the object detection and classifier are updated with temperate fishes of interest. The weights obtained from pre-training are applied to post-training as a priori. Our solution achieves the state-of-the-art accuracy of 99.27% using the pre-training model. The accuracies using the post-training model are also high; 83.68% and 87.74% with and without image augmentation, respectively. This strongly indicates that the solution is viable with a more extensive dataset.

• Moland, Even; Fernández-Chacón, Albert; Sørdalen, Tonje Knutsen; Villegas-Rios, David; Thorbjørnsen, Susanna Huneide & Halvorsen, Kim Aleksander Tallaksen [Show all 14 contributors for this article] (2021). Restoration of Abundance and Dynamics of Coastal Fish and Lobster Within Northern Marine Protected Areas Across Two Decades. Frontiers in Marine Science. 8(674756). doi: 10.3389/fmars.2021.674756. Full text in Research Archive Show summary

  This article reviews a suite of studies conducted in a network of coastal Marine Protected Areas (MPAs) in Skagerrak, Southeast Norway. In 2006, Norway’s first lobster reserves were implemented, with the aim of protecting European lobster (Homarus gammarus) through a ban on fixed gear. A before–after control-impact paired series (BACIPS) monitoring program was initiated to evaluate effects of protection on depleted lobster populations. Experimental trapping and capture-recapture techniques were combined to track demography of populations, also including movement of individuals within and beyond MPAs and adjacent control areas. Further, population genetics and parentage studies were applied, allowing for estimation of gene flow, and novel work on sexual selection in lobsters. Additional studies have evaluated MPA effects on coastal cod (Gadus morhua), and on commercially harvested labrids (Ctenolabrus rupestris and Symphodus melops) and anadromous brown trout (Salmo trutta). Together, these studies reported effects of protection pertaining to increased population density, survival, body size and phenotypic diversity, changes in emigration and interaction with surrounding fisheries, and alteration of selection pressure on morphological- and behavioral traits. Designation of MPAs in close collaboration with fishers and managers, long-term monitoring, inclusion of citizen science and evolving research protocols—also including fisheries data—have revealed novel effects of protection and harvesting on marine populations, thus providing substantial contributions to conservation science. Moreover, knowledge of MPA effects on coastal species has impacted harvest regulations showing the utility of MPAs as empirically documented management tools in Norway.

• Halvorsen, Kim Aleksander Tallaksen; Sørdalen, Tonje Knutsen; Larsen, Torkel; Browman, Howard; Rafoss, Trond & Albretsen, Jon [Show all 7 contributors for this article] (2020). Mind the depth: The vertical dimension of a small-scale coastal fishery shapes selection on species, size, and sex in wrasses. Marine and Coastal Fisheries. 12(6), p. 404–422. doi: 10.1002/mcf2.10131. Full text in Research Archive Show summary

  Small‐scale fisheries (SSFs) tend to target shallow waters, but the depth distributions of coastal fish can vary depending on species, size, and sex. This creates a scope for a form of fishing selectivity that has received limited attention but can have considerable implications for monitoring and management of these fisheries. We conducted a case study on the Norwegian wrasse fishery, a developing SSF in which multiple species are caught in shallow waters (mean depth = 4.5 m) to be used as cleaner fish in aquaculture. Several of these wrasses have life histories and behaviors that are sensitive to selective fishing mortality, such as sexual size dimorphism, paternal care, and sex change. An experimental fishery was undertaken over three sampling periods in 2018. Data on catch, length, and sex of wrasses across a depth gradient (0–18 m) were collected and analyzed. We found that depth distributions were species specific and the vertical overlap with the fishery was high for Corkwing Wrasse Symphodus melops and Ballan Wrasse Labrus bergylta, which were most abundant at depths less than 5 m. Three other wrasse species had invariant or increasing abundance with depth and were therefore less likely to be negatively impacted by this fishery. Body size was positively correlated with depth for these wrasses, and sex ratio became more male biased for the Corkwing Wrasse, the only species that could be visually sexed. This study demonstrates that depth can have strong effects on fishing selectivity at multiple scales and that such knowledge is necessary to develop management strategies that balance fishing mortality sustainably across species, sizes, and sexes. We recommend that management priorities be directed toward the Ballan and Corkwing wrasses—the species having the highest vertical overlap with the fishery. Furthermore, CPUE was strongly affected by seasonality and positively correlated with increasing wave exposure for one of the species. This underscores the general importance of standardizing catch data for several environmental covariates when monitoring species that are affected by SSFs.

• Kindsvater, Holly K.; Halvorsen, Kim Aleksander Tallaksen; Sørdalen, Tonje Knutsen & Alonzo, Suzanne H. (2020). The consequences of size-selective fishing mortality for larval production and sustainable yield in species with obligate male care. Fish and Fisheries. ISSN 1467-2960. 21(6), p. 1135–1149. doi: 10.1111/faf.12491. Full text in Research Archive
• Sørdalen, Tonje Knutsen; Halvorsen, Kim Aleksander Tallaksen; Vøllestad, Leif Asbjørn; Moland, Even & Olsen, Esben Moland (2020). Marine protected areas rescue a sexually selected trait in European lobster. Evolutionary Applications. 13(9), p. 2222–2233. doi: 10.1111/eva.12992. Full text in Research Archive Show summary

  Marine protected areas (MPAs) are increasingly implemented worldwide to maintain and restore depleted populations. However, despite our knowledge on the myriad of positive responses to protection, there are few empirical studies on the ability to conserve species’ mating patterns and secondary sexual traits. In male European lobsters (Homarus gammarus), the size of claws relative to body size correlates positively with male mating success and is presumably under sexual selection. At the same time, an intensive trap fishery exerts selection against large claws in males. MPAs could therefore be expected to resolve these conflicting selective pressures and preserve males with large claws. We explored this hypothesis by contrasting claw size of males and females in three pairs of MPAs and nearby fished areas in southern Norway. By finding that male lobsters have up to 8% larger claws inside MPAs compared to similarly sized males in fished areas, our study provides evidence that MPAs rescue a secondary sexual trait. Recovery from harvest selection acting on claws is the most likely explanation; however, the higher abundance of lobster inside MPAs does not rule out a plastic response on claw size due to increased competition. Regardless of the underlying cause, our study demonstrates (a) the value of protected areas as a management tool for mitigating fisheries‐induced evolution and (b) that MPAs help maintaining the scope for sexual selection in populations with vulnerable life histories and complex mating system.

• Olsvik, Erlend; Trinh, Christian M.D.; Knausgård, Kristian Muri; Wiklund, Arne; Sørdalen, Tonje Knutsen & Kleiven, Alf Ring [Show all 8 contributors for this article] (2019). Biometric fish classification of temperate species using convolutional neural network with squeeze-and-excitation. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. 11606 LNAI, p. 89–101. doi: 10.1007/978-3-030-22999-3_9. Full text in Research Archive
• Sørdalen, Tonje Knutsen; Halvorsen, Kim Aleksander Tallaksen; Harrison, Hugo B.; Ellis, Charlie D.; Vøllestad, Leif Asbjørn & Knutsen, Halvor [Show all 8 contributors for this article] (2018). Harvesting changes mating behaviour in European lobster. Evolutionary Applications. 11(6), p. 963–977. doi: 10.1111/eva.12611. Full text in Research Archive Show summary

  Removing individuals from a wild population can affect the availability of prospective mates and the outcome of competitive interactions, with subsequent effects on mating patterns and sexual selection. Consequently, the rate of harvest‐induced evolution is predicted to be strongly dependent on the strength and dynamics of sexual selection, yet there is limited empirical knowledge on the interplay between selective harvesting and the mating systems of exploited species. In this study, we used genetic parentage assignment to compare mating patterns of the highly valued and overexploited European lobster (Homarus gammarus) in a designated lobster reserve and nearby fished area in southern Norway. In the area open to fishing, the fishery is regulated by a closed season, a minimum legal size and a ban on the harvest of egg‐bearing females. Due to the differences in size and sex‐specific fishing mortality between the two areas, males and females are of approximately equal average size in the fished area, whereas males tend to be larger in the reserve. Our results show that females would mate with males larger than their own body size, but the relative size difference was significantly larger in the reserve. Sexual selection acted positively on both body size and claw size in males in the reserve, while it was nonsignificant in fished areas. This strongly suggests that size truncation of males by fishing reduces the variability of traits that sexual selection acts upon. If fisheries continue to target large individuals (particularly males) with higher relative reproductive success, the weakening of sexual selection will likely accelerate fisheries‐induced evolution towards smaller body size.

• Halvorsen, Kim Aleksander Tallaksen; Larsen, Torkel; Sørdalen, Tonje Knutsen; Vøllestad, Leif Asbjørn; Knutsen, Halvor & Olsen, Esben Moland (2017). Impact of harvesting cleaner fish for salmonid aquaculture assessed from replicated coastal marine protected areas. Marine Biology Research. ISSN 1745-1000. 13(4), p. 359–369. doi: 10.1080/17451000.2016.1262042. Full text in Research Archive Show summary

  Wrasse (Labridae) fisheries have increased markedly in Norway since 2010. Wrasse are being used as cleaner fish in salmonid aquaculture to control sea-lice infestations. However, fundamental knowledge on the demography and abundance of the targeted wrasse populations in Norwegian waters is lacking, and the consequences of harvesting at the current intensity have not been assessed. Here, we compared catch per unit effort (CPUE), size, age and sex ratio of goldsinny wrasse (Ctenolabrus rupestris) and corkwing wrasse (Symphodus melops) between marine protected areas (MPAs) and control areas open for fishing at four localities on the Skagerrak coast in Southern Norway. The CPUE of goldsinny larger than the minimum size limit was 33–65% higher within MPAs, while for corkwing three of four MPAs had higher CPUE with the relative difference between MPAs and control areas ranging from −16% to 92%. Moreover, corkwing, but not goldsinny, was significantly older and larger within MPAs than in control areas. Sex ratios did not differ between MPAs and control areas for either species. Our study suggests that despite its short history, the wrasse fisheries have considerable impacts on the target populations and, further, that small MPAs hold promise as a management tool for maintaining natural population sizes and size structure. Goldsinny, being a smaller-sized species, also seems to benefit from the traditional minimum size limit management tool, which applies outside MPAs. The final version of this research has been published in Marine Biology Research. © 2017 Taylor & Francis

• Halvorsen, Kim Aleksander Tallaksen; Sørdalen, Tonje Knutsen; Vøllestad, Leif Asbjørn; Skiftesvik, Anne Berit; Espeland, Sigurd Heiberg & Olsen, Esben Moland (2017). Sex-and size-selective harvesting of corkwing wrasse (Symphodus melops)-a cleaner fish used in salmonid aquaculture. ICES Journal of Marine Science. ISSN 1054-3139. 74(3), p. 660–669. doi: 10.1093/icesjms/fsw221. Full text in Research Archive Show summary

  Fishery-induced changes in sex ratios can have negative effects on reproductive rates and affect sexual selection and evolutionary trajectories. Here, we investigate sex- and size-selectivity of the fishery for corkwing wrasse (Symphodus melops) in Western Norway. The males that build and guard nests (nesting males) grow faster than females and sneaker males. Corkwing wrasse were tagged (n = 1057) during (May–June) and after (July) the spawning period in 2014 within a no-take site and in a nearby site open for fishing. We monitored the fishery within and nearby the tagging sites from June to October and sampled recaptures on all commercial fishing trips. Fishing mortality was higher for nesting males than for females (open site; tagged during spawning: 36% vs. 29%; after spawning: 49% vs. 36%) and was found to be caused by nesting males having higher capture probabilities than females, independently of body size. The fishing mortality of sneaker males did not differ from nesting males, but the sample of sneaker males in the study was small. The probability of being retained by fishers once captured depended on size, but not on sex. The small no-take site reduced fishing mortality, and the sex- and size-selectivity was similar to the open site. By demonstrating the capture process to be male-selective in a commercial fishery, our study highlights the importance of assessing sex-selectivity in similar fisheries. If detected, management measures protecting the sexes more equally should be implemented, such as slot-size limits and marine-protected areas large enough to account for sexual differences in spatial behaviour.

• Halvorsen, Kim Aleksander Tallaksen; Sørdalen, Tonje Knutsen; Durif, Caroline; Knutsen, Halvor; Olsen, Esben Moland & Skiftesvik, Anne Berit [Show all 9 contributors for this article] (2016). Male biased sexual size dimorphism in the nest building corkwing wrasse (Symphodus melops): implications for a size regulated fishery. ICES Journal of Marine Science. ISSN 1054-3139. 73(10), p. 2586–2594. doi: 10.1093/icesjms/fsw135. Full text in Research Archive
• Ellis, Charlie; Hodgson, DJ; Andre, Carl; Sørdalen, Tonje Knutsen; Knutsen, Halvor & Griffiths, Amber (2015). Notes on Hardy's uncertainty principle for the Wigner distribution and Schrödinger evolutions. PLOS ONE. 525(1). doi: 10.1016/j.jmaa.2023.127116. Full text in Research Archive

View all works in NVA

• Brøten, Gina; Sørdalen, Tonje Knutsen; Marcussen, Johanna Bjånes & Zimmermann, Fabian (2026). Data quality in small-scale fisheries: Bridging new methods and regulations with field realities. Full text in Research Archive
• Brøten, Gina; Zimmermann, Fabian; Marcussen, Johanna Bjånes & Sørdalen, Tonje Knutsen (2026). Crabs on camera: from underwater video to automated coastal monitoring using object detection. Full text in Research Archive
• Sørdalen, Tonje Knutsen; Larsen, Torkel; Skiftesvik, Anne Berit; Malde, Ketil & Halvorsen, Kim Aleksander Tallaksen (2025). How cameras and AI can replace traditional tags for individual identification of fish. Full text in Research Archive
• Ellis, Benjamin; Sørdalen, Tonje Knutsen; Briffa, Mark; Skiftesvik, Anne Berit; Wilson, Alexander & Halvorsen, Kim Aleksander Tallaksen (2025). Visual cues and familiar recognition inform agonistic responses towards females and female mimics in the corkwing wrasse (Symphodus melops). Full text in Research Archive
• III, James David Poling; Sørdalen, Tonje Knutsen; Halvorsen, Kim Aleksander Tallaksen; Thormar, Jonas; Perry, Diana & Larsen, Torkel (2025). FjordFish. doi: 10.5281/ZENODO.17496689. Full text in Research Archive Show summary

  FjordFish is an underwater imagery dataset (3002 images, 6350 annotated objects) across 27 classes (table 1) designed for training object detection models to detect and classify underwater marine fauna. The dataset focuses on North Atlantic coastal fish species inhabitating the Norwegian and Swedish coasts. Images were collected from videos taken during baited and unbaited remote underwater video surveys along the Swedish and Norwegian coasts. Images of coastal cod were also collected from a semi-natural mesocosm at the Institute of Marine Research Flødevigen research station. All images are fully framed to include diverse backgrounds and species assmeblages. Labels are provided in the YOLO format for YOLOv8 and newer, but can be converted to other formats. Data managers: David Poling - University of Agder, Norway - james.d.poling@uia.no Tonje Knutsen Sørdalen - Institute of Marine Research, Norway - tonjesordalen@gmail.com If using this dataset for research, please cite: Poling, J. D., Sørdalen, T. K., Halvorsen, K., Thormar, J., Perry, D., & Larsen, T. (2025). FjordFish [Data set]. Zenodo. https://doi.org/10.5281/zenodo.17496689

• Sørdalen, Tonje Knutsen; Malde, Ketil; Skiftesvik, Anne Berit; Sauvaitre, Canelle; Beyan, Cigdem & Larsen, Torkel [Show all 7 contributors for this article] (2025). MelopsReID: A wild fish image dataset for re-identification. doi: 10.5281/ZENODO.17099925. Full text in Research Archive
• Halvorsen, Kim Aleksander Tallaksen; Ellis, Benjamin; Kindsvater, Holly; Alonzo, Suzanne; Skiftesvik, Anne Berit & Browman, Howard [Show all 7 contributors for this article] (2025). The untapped potential of pit telemetry for studying behavioral ecology of coastal fish. Full text in Research Archive Show summary

  Acoustic telemetry is often the go-to method for studying behavior and movement ecology of marine fish. However, the cost and size of tags and arrays may limit its application on small bodied, sedentary species, and it can also be challenging to study interactions at fine spatial scales or to track a representative proportion of the population. Passive integrated transponder (PIT) tags are relatively cheap and small and can be detected remotely using antennas that allow for tracking of individual behaviors and social interactions over long time spans. Through a 6-year case study on commercially important wrasses in a Norwegian MPA, we demonstrate how PIT telemetry enables us to obtain new insights into key behavioral processes—such as sexual selection, social networks, predator-prey dynamics and homing behavior—that remain difficult to study with conventional approaches. We also highlight the benefit of combining pit-tagging and remote detection with other methods that can provide complementary knowledge, such as camera monitoring, artificial intelligence and genetics.

• Halvorsen, Kim Aleksander Tallaksen & Sørdalen, Tonje Knutsen (2024). Arven etter Dannevig: Autonom kystovervåkning. Full text in Research Archive Show summary

  45 deltagere (påmelding). 2 timers seminar. Velkommen til Havforskning og teknologi på Flødevigen Forskningsstasjon På Flødevigen havforskningsinstitutt I Arendal jobbes det med mange utfordringer. Kollaps i kysttorsk bestanden, tap av tareskog, økte havtemperaturer og etablering av fremmede arter viser et hav i endring. Men det finnes muligheter for å gjøre noe med utfordringene! Vi blir kjent med flere spennende forskningsprosjekter, for eksempel: Automatisert Kystovervåkning, prosjekt ARVEN: Kystovervåking med bruk av kunstig intelligens/maskinsyn og kamerateknologi. Utvikle bedre og mer effektiv økosystemovervåking: Artsbestemme, måle og telle fisk og skalldyr under vann Prosjekt Coastvision: Bruk av undervannskameraer med mønstergjenkjenning (Computervision) for å skille arter, individer og kjønn. Fisk, for eksempel torsk, kan telles, spores og identifiseres. Dette er viktig for å forstå hvilke faktorer som påvirker reproduksjon, som for eksempel andre arter og miljøendringer. Et mål er å utvikle gode modeller for å re-identifisere fisk. Prosjekt IdentiFisk: Digitale verktøy, maskinsyn, for sikker identifisering av rømt og vill laksefisk. Kan det bli til verktøy som sportsfiskere kan benytte?

• Sørdalen, Tonje Knutsen; Malde, Ketil & Halvorsen, Kim Aleksander Tallaksen (2024). Decoding female mimicry in a fish using deep learning. Full text in Research Archive
• Lien, Anne; Nygård, Malene Østreng; Laugen, Ane Timenes; Lie, Asbjørn; Slettan, Audun & Johansen, Beate Strøm [Show all 33 contributors for this article] (2023). Utbygging vil ødelegge for biologisk mangfold på campus Gimlemoen. NB! 34 forfattere totalt. p. 19–19. Full text in Research Archive
• Larsen, Jorunn; Nygaard, Malene; Halvorsen, Kirsti Vindal; Borg-Stoveland, Stian; Sørdalen, Tonje Knutsen & Fiskvatn, Kari [Show all 7 contributors for this article] (2023). Når natur forsvinner: Bærekraftig utvikling eller bare kraftig utvikling i Kristiansand kommune? Full text in Research Archive
• Allken, Vaneeda; Sørdalen, Tonje Knutsen; Beyan, Cigdem; Malde, Ketil; Browman, Howard & Halvorsen, Kim Aleksander Tallaksen (2022). A deep learning approach for individual re-identification (re-ID) of fish in the wild. Full text in Research Archive
• Halvorsen, Kim Aleksander Tallaksen & Sørdalen, Tonje Knutsen (2016). Trøbbel i havet. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2015). Stor hummer må fredes, minstemål er ikke nok. Full text in Research Archive Show summary

  hummerne kan være redningen for fremtidens hummerfiske. Men da må de beskyttes.

• Sørdalen, Tonje Knutsen (2015). Marinbiologi, med hjerte for havet. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2015). Jakten på svart hummer. argument. ISSN 1504-6087. Full text in Research Archive Show summary

  Vi høster hva vi sår, også under havoverflaten.

• Sørdalen, Tonje Knutsen (2015). Jakten på svart hummer. Vi høster hva vi sår, også under havoverflaten. Harvest : Mennesket & naturen. Full text in Research Archive
• Sørdalen, Tonje Knutsen & Halvorsen, Kim Aleksander Tallaksen (2015). Legal size limit implies strong fisheries selection on sexually selected traits in a temperate wrasse providing male-only parental care. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2015). Havforskerne. [TV]. TV agder. Full text in Research Archive Show summary

  Havforskerne fisker hummer i reservatene i Tvedestrand og Lindesnes. Hvem vinner konkurransen om å få flest hummer i teinene?

• Halvorsen, Kim Aleksander Tallaksen & Sørdalen, Tonje Knutsen (2015). Sex biased fishing mortality and movement patterns of the dimorphic corkwing wrasse (Symphodus melops). Full text in Research Archive
• Sørdalen, Tonje Knutsen (2014). Forsker Grand Prix 2013. [TV]. NRK. Full text in Research Archive
• Høgetveit, Olav & Sørdalen, Tonje Knutsen (2014). Undring og mangfald: Tonje Knutsen Sørdalen 2:5. [TV]. NRK2. Full text in Research Archive Show summary

  Samtaler med norske forskarar. Hummeren lever lenge, 70-100 år, men han er i dag ein truga art langs norskekysten. På 50-60 talet fiska vi meir enn 1000 tonn årleg her i landet. No er fangsten nede i 50-60 tonn. Korleis formerar denne arten seg og korleis sikre at hummeren framleis lever langs kysten vår slik at ein også i framtida kan eta denne delikatessen? (2:5)

• Sørdalen, Tonje Knutsen (2014). Fisheries effect on mating behavior and mate choice in wild European lobsters. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2014). Effekten av fiske på paringsadferd og makevalg hos viltlevende hummer. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2014). Hummerforskning; Et casestudie fra Flødevigen reservat og kontrollområde. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2014). Effekt av fiske på hummerens makevalg. Full text in Research Archive
• Sørdalen, Tonje Knutsen & Halvorsen, Kim Aleksander Tallaksen (2013). Forskningsdagene 2013. Full text in Research Archive
• Sørdalen, Tonje Knutsen & Halvorsen, Kim Aleksander Tallaksen (2013). Hvorfor er forskning innen marin biologi viktig? Eksempler fra arbeid med hummer og grønngylt. Full text in Research Archive
• Sørdalen, Tonje Knutsen & Halvorsen, Kim Aleksander Tallaksen (2013). "Transvesitter, store menn og løsnluppne damer". Agderposten. ISSN 0805-3723. Full text in Research Archive
• Halvorsen, Kim Aleksander Tallaksen & Sørdalen, Tonje Knutsen (2013). Transvestitter, store menn og løsslupne damer. Agderposten. ISSN 0805-3723. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2013). Ekko; et samfunns- og naturvitenskapelig program. [Radio]. NRK p2. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2013). "Er det kun størrelsen det kommer an på?". [Journal]. Kragerø blad Vestmar sommermagasin. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2013). "Frykter kardinalfeil". [Journal]. Dagens Næringsliv. Full text in Research Archive
• Sørdalen, Tonje Knutsen (2012). Multiple paternity assessment and paternity assignment in European lobster (Homarus gammarus). Full text in Research Archive
• Kleiven, Alf Ring; Marcussen, Johanna Bjånes; Sørdalen, Tonje Knutsen; Meeren, Terje van der; Agnalt, Ann-Lisbeth & Meeren, Gro Ingleid van der (2024). Ny målemetode for hummar. Havforskningsinstituttet. Full text in Research Archive Show summary

  Fiskeridirektoratet ber Havforskingsinstituttet om å greie ut om føremoner og ulemper med å nytte ryggskjoldet (carapax) som mål for å fastsetje minstemål og maksimalmål, og å gje tilråding på bakgrunn av dette. Totallengde har vore standard metode i Noreg. Ryggskjoldlengde er eit mer stabilt mål, ein betre indikator på vekst og modning og kan vere betre for dyrevelferda. Ryggskjoldlengde er også det som vanlegvis brukast i andre land i Europa.

• Halvorsen, Kim Aleksander Tallaksen; Sørdalen, Tonje Knutsen; Larsen, Torkel; Rafoss, Trond & Skiftesvik, Anne Berit (2020). Kunnskapsbasert innovasjon for optimal ressursutnyttelse i leppefiskeriet — Sluttrapport i prosjekt 272202 - Regionalt forskningsfond Agder. Havforskningsinstituttet. Full text in Research Archive Show summary

  Rensefisk benyttes for å kontrollere mengden lakselus i oppdrett av laksefisk i sjø i Norge. Behovet for rensefisk har økt markant siden 2010 på grunn av resistensutvikling mot medikamentelle behandlinger hos lakselus (Skiftesvik et al. 2014; Aaen et al. 2015; Rueness et al. 2019). På tross av at produksjonen av oppdrettet rensefisk, primært rognkjeks, har økt kraftig de siste årene, så har det ikke redusert behovet for villfanget leppefisk som rensefisk. Leppefiskeriet har blitt en svært viktig næring for kystfiskere i Agder. Da ålefisket ble forbudt i 2006, gikk mange over til å fiske leppefisk som kan utføres med samme redskap og fartøy. På Sørlandet har fiskerne leveringsavtaler med oppkjøpere som transporterer og videreselger leppefisken til oppdrettsanlegg hovedsakelig i Midt- og Nord-Norge. Fiskeriet på Skagerrakkysten er ikke jevnt fordelt, hoveddelen (ca 80 %) av fangsten tas i Agder (Figur 1). Fiskepresset på leppefisk anses å være lavt fra Telemark og til Svenskegrensen. Ytterligere informasjon om fangsttall, verdi og fartøy i de ulike regionene finnes på Fiskeridirektoratets nettsider: https://www.fiskeridir.no/Yrkesfiske/Tall-og-analyse/Fangst-ogkvoter/Fangst-av-leppefisk. Man har inntil nylig hatt lite kunnskap om hvor mye beskatning leppefiskbestandene tåler og hvilke økologiske konsekvenser et eventuelt overfiske kan medføre. På Skagerrakkysten fangstes det på tre ulike leppefiskarter, bergnebb (Ctenolabrus rupestris), grønngylte (Symphodus melops) og berggylte (Labrus bergylta). Grasgylt (Centrolabrus exoletus) og Rødnebb/Blåstål (Labrus mixtus) regnes som bifangst på Sørlandet siden de ikke egner seg for langdistansetransport. Det forekommer noe bruk av disse artene lokalt på Vestlandet. I tillegg tas det en del bifangst av andre fisk og skalldyr, som kan forårsake skade på leppefisken, samt av kyst-torsk og hummer, ikoniske arter som er truet av overfiske på Skagerrak-kysten (Fernández-Chacón et al. 2015; Sørdalen et al. 2018). Taskekrabbe brukes som agn i leppefisket, men må fiskes med egne krabbeteiner. Regelverket pålegger at all bifangst som fanges i Kunnskapsbasert innovasjon for optimal ressursutnyttelse i leppefiskeriet 1 - Bakgrunn for prosjektet 4/21 leppefiskredskap skal slippes ut på fangststedet. I Agder er det dokumentert høyere forekomster av leppefisk innad i bevaringsområder sammenliknet med nærliggende områder med åpent fiske, noe som tyder på at fiskeriet kan ha en bestandsregulerende effekt (Figur 2; Halvorsen et al. 2017). HIs råd om kvoter og reguleringer har vært basert på en føre-var-tilnærming for å stoppe en videre ekspansjon i fiskeriet før man har tilstrekkelig kunnskap om bestandenes tåleevne og svingninger. Fiskeridirektoratet forvalter bestandene ved bruk av flere ulike reguleringer, herunder gytetidsfredning, fluktåpninger, artsspesifikke minstemål, fartøykvoter og totalkvoter for ulike regioner. På Sørlandet er totalfangsten begrenset til fire millioner leppefisk årlig. For å gi råd om forsvarlig uttak av bestandene etablerte HI i 2011 et nettverk av referansefiskere som leverte fangstrapporter. Bestandsutviklingen ble representert ved regionale trender i fangst-per-enhet-innsats (Catch per unit effort; CPUE, heretter), men datagrunnlaget var for variabelt til å kunne konkludere med at endringer i CPUE reflekterte faktiske endringer i bestandsstørrelse. I dette prosjektet har vi fokusert på å ta i bruk ny teknologi og forbedre rapporteringsrutiner i fiskeriavhengig datainnsamling for å få kunnskap om leppefiskens romlige økologi. For fiskeriet på Skagerrak vil dette kunne bidra til mer kunnskap om ressursfordelingen og hvordan de kan utnyttes bedre og mer effektivt, samtidig som at dette muliggjør en bedre bestandsovervåking og grunnlag for å gi mer presise kvoteråd. I Prosjektet har vi også utnyttet tilgjengelige dataserier for å få en oversikt over bestandsutviklingen i Skagerrak. Merkeforsøk har blitt gjennomført for å få bedre kunnskap om romlig adferd og fangbarhet. Resultatene har blitt fortløpende inkludert i HIs kunnskapsstøtte til Fiskeridirektoratet, og det nasjonale referansefiskernettverket har tatt i bruk metodene for fangstrapportering som er utviklet i dette prosjektet. Det arbeides med å ferdigstille publikasjoner for fagfellevurderte tidsskrifter, og i denne rapporten vil det kun gis en kort oppsummering av disse studiene.

• Kleiven, Alf Ring; Moland, Even; Sørdalen, Tonje Knutsen; Espeland, Sigurd Heiberg & Meeren, Gro Ingleid van der (2017). Evaluering av effekten av forvaltningstiltak på hummer og forslag til tiltak. Havforskningsinstituttet. Full text in Research Archive

View all works in NVA