Status: this preprint is currently under review for the journal ESSD.
Beyond 'Trawler' in the Adriatic Sea: Reconstructing Eight Years of Fishing Pressure by Distinct Trawling Strategies
Carmen Ferrà,Enrico Nicola Armelloni,Giuseppe Scarcella,Pamela Lattanzi,and Anna Nora Tassetti
Abstract. Trawling fisheries in the Adriatic Sea are among the most spatially intensive fishing activities in the Mediterranean, with significant implications for marine ecosystems and resource management. Understanding their spatiotemporal dynamics is essential for informing sustainable fisheries management and mitigating ecosystem impacts. We present an 8-year (2015–2022) high-resolution (0.01° × 0.01°) dataset of monthly trawling effort indicators reconstructed from Automatic Identification System (AIS) data across Geographical Sub-Areas (GSAs) 17 and 18. A key novelty of this dataset is the explicit classification of the three main trawling strategies – bottom, pelagic, and beam trawling. Reconstructed from 858 fishing vessels identified as performing trawling during the study period, the dataset includes detailed information on fishing hours, fishing days, vessel counts, gear type, and departure/destination GSAs, harmonised and disaggregated by country and trawling method. This harmonised and disaggregated dataset enables high-resolution assessment of fleet dynamics, fishing intensity, and spatial footprint. The novel reconstructed trawling-effort dataset is publicly available at: https://doi.org/10.17882/114473 and represents a valuable tool for researchers, fisheries managers, and policymakers. It directly supports Ecosystem-Based Fisheries Management (EBFM) and the revision of existing management plans in line with General Fisheries Commission for the Mediterranean (GFCM) recommendations.
Received: 16 Jun 2026 – Discussion started: 30 Jun 2026
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Review of Ferrà et al., Beyond 'Trawler' in the Adriatic Sea: Reconstructing Eight Years of Fishing Pressure by Distinct Trawling Strategies
General comments
This is a useful and timely AIS-derived dataset in the Adriatic Sea which improves upon more general methods used for example by Global Fishing Watch, in particular by providing gear specific information. I recommend it for publication as a data paper but I would suggest to address certain general criticisms of AIS, with a particular inclusion of Hintzen et al., (2025) which brings up concern on how AIS derived data (from GFW in particular) can overestimate the fishing effort by sometimes specifying certain activities, like echosounder use, as fishing activities. I know of Couce et al., (2020) as well which provides similar trawling hours information for the North Sea and it would be nice to see some comparison of methods from that and potentially other similar datasets/methods. I think you can use that information to provide more compelling arguments to how the current methods paper may improve upon other methods while remaining transparent with potential limitations. Overall, I think it’s a nice contribution and while the framing could be improved to connect it to other studies, I think it can still be very useful for scientists, managers and policy makers.
Detailed comments
Abstract
L22 Can you please state explicitly how this supports EBFM? I think you mean because it gives you some good indications/proxies of bottom disturbance (and improvements on other AIS derived information like from GFW) but less informed readers might not pick this up right away. Same for supporting the revision of management plans as it’s not so clear (even though it sounds nice).
Introduction
L32 Try and harmonize the use of EBM and EBFM which is in the abstract.
L64-67 Here I suggest also including critiques of how GFW uses AIS data. I’m really just thinking about the study from Hintzen et al., (2025) which compared GFW fishing events with independently recorded haul data and found that GFW derived estimates could substantially overestimate the actual trawling being deployed. I would just briefly mention here that study or that critiques exist and go back to this topic in the discussion. Like would your study and improvements on AIS derived fishing effort also be less likely to fall under these criticisms from Hintzen et al., which point out, for instance, that vessels searching for fish using sonar equipment may be mistaken for fishing activities (by GFW derived AIS analysis at least)? Also, it the same critiques are still valid, then this paper should at least mention that as potential limitations. This could potentially strengthen your arguments that this product is a valid improvement compared to prior GFW methods (at least in the Adriatic).
Methods
L99 Not a huge deal but “together constitute” probably can be phrased simpler (less wordy) with “represent”, “encompass” or something else. I just think it could flow better.
L99-100 It is interesting information to know that these GSA’s are the most highly collaborative fisheries management zone in the Adriatic but it comes out sounding less relevant here compared to the rest of the paper. I would still try to fit it in but suggest breaking up that sentence. Also, isn’t GSA 17 and 18 the total Adriatic? The sentence makes it seem like it is just part of the Adriatic (though I admit I’m less familiar with that sea).
L103-104 again, see if this is enough to evade criticism from Hintzen et al., or like minded scientists (not new text needed here but potentially in the discussion).
L118 so does manually validated mean that local experts took a look and said it was okay? Shouldn’t it then say “visually” validated or something like that. Consider clarifying.
Results
L175-176 the threshold application for excluding analysis (because they’re probably not trawling below that value), shouldn’t that be introduced in methods not results (I mean I see that it flows kinda nice here but still)? Also for others to reproduce this work, it would be useful in the methods.
L179 ok I’ll be the annoying reviewer that suggests more content here. You have some nice qualitative descriptions of the gear specific effort data in figures 1 and 2. Is it possible to also add a small summary table showing more quantitative descriptions like total fishing hours by gear per GSA or other spatial information?
L192 I hate to be a stickler but you do use the term significant here in the results to show statistical significance so perhaps just use ‘marked’, or ‘substantial’ instead of significant here.
L215-217 the reasons for the Albanian increase may be valid though I wonder if that information belongs in the results and not the discussion (possible improved reporting and fleet expansion).
L242 Just to be pragmatic and clear to everyone, can you please specify that these percentages are coming from the full 2 GSA areas (combined) to form the denominator used for the percentages (The Adriatic Sea). I think it can be specified once briefly early in the paragraph (kind of like on line 386 in the discussion).
Discussion
L404 in this paragraph you can add the example of the fleet expansion and better reporting from Albania potentially leading to the results.
I don’t think you should go overboard on the limitations of the method but a few things come to mind for me. While you might be able to make some assumptions based on gear, I don’t see any swept area or swept area ratios for bottom impacting gears. This may be a bit difficult for otter trawling (trawl doors make variable swept areas) but it’s relevant because combining gear and gear width information will let you know more about the actual seabed disturbance much more than fishing hours. Also swept area ratio is directly used for analyses like relative benthic status (RBS; Pitcher et al., 2017) and I don’t think the study gives indications of gear penetration depth (though they can be inferred/assumed I guess).
Perhaps most important, I think that a discussion integrating the criticisms of GFW methods from Hintzen et al., 2025 is appropriate because they likely also apply in some ways to the current study. I do think this work improves upon the more general GFW methods for that particular area but nonetheless there may be general criticisms which may need to be addressed.
Also the inclusion of Couce et al., 2020 (North Sea) and other similar works are missing from the discussion. Try and add a paragraph which compares and contrasts other work which similarly tries to spatially quantify bottom trawl effort.
References
Couce, E., Schratzberger, M., & Engelhard, G. H. (2020). Reconstructing three decades of total international trawling effort in the North Sea. Earth System Science Data, 12(1), 373–386. https://doi.org/10.5194/essd-12-373-2020
Hintzen, N. T., Brigden, K., Kaastra, H.-J., Mackinson, S., Pastoors, M. A., & van de Pol, L. (2025). Bias in Global Fishing Watch AIS data analyses results in overestimate of Northeast Atlantic pelagic fishing impact. ICES Journal of Marine Science, 82(3), fsaf033. https://doi.org/10.1093/icesjms/fsaf033
Pitcher, C. R., Ellis, N., Jennings, S., Hiddink, J. G., Mazor, T., Kaiser, M. J., Kangas, M. I., McConnaughey, R. A., Parma, A. M., Rijnsdorp, A. D., Suuronen, P., Collie, J. S., Amoroso, R., Hughes, K. M., & Hilborn, R. (2017). Estimating the sustainability of towed fishing-gear impacts on seabed habitats: A simple quantitative risk assessment method applicable to data-limited fisheries. Methods in Ecology and Evolution, 8(4), 472–480. https://doi.org/10.1111/2041-210X.12705
Carmen Ferrà,Enrico Nicola Armelloni,Giuseppe Scarcella,Pamela Lattanzi,and Anna Nora Tassetti
Data sets
AIS-derived monthly fishing effort and aggregated pressure indicators over eight years (2015–2022)Carmen Ferrà, Enrico Nicola Armelloni, Giuseppe Scarcella, Pamela Lattanzi, and Anna Nora Tassetti https://www.seanoe.org/data/01033/114473/
Model code and software
R4AIS: an R workflow to process AIS data for fisheryAlessandro Galdelli, Enrico Nicola Armelloni, Carmen Ferrà, Jacopo Pulcinella, Giuseppe Scarcella, and Anna Nora Tassetti https://zenodo.org/records/4761890
Carmen Ferrà,Enrico Nicola Armelloni,Giuseppe Scarcella,Pamela Lattanzi,and Anna Nora Tassetti
Viewed
Total article views: 25 (including HTML, PDF, and XML)
HTML
PDF
XML
Total
BibTeX
EndNote
22
1
2
25
0
2
HTML: 22
PDF: 1
XML: 2
Total: 25
BibTeX: 0
EndNote: 2
Views and downloads (calculated since 30 Jun 2026)
Cumulative views and downloads
(calculated since 30 Jun 2026)
Viewed (geographical distribution)
Total article views: 13 (including HTML, PDF, and XML)
Thereof 13 with geography defined
and 0 with unknown origin.
Trawling fisheries in the Adriatic Sea are among the most spatially intensive fishing activities in the Mediterranean, with important implications for marine ecosystems. Understanding their spatiotemporal dynamics is essential to support sustainable management. The dataset derived from vessel tracking data provides monthly information on fishing effort and distinguishes between different trawling methods, offering a valuable tool for researchers and fisheries managers.
Trawling fisheries in the Adriatic Sea are among the most spatially intensive fishing activities...
Review of Ferrà et al., Beyond 'Trawler' in the Adriatic Sea: Reconstructing Eight Years of Fishing Pressure by Distinct Trawling Strategies
General comments
This is a useful and timely AIS-derived dataset in the Adriatic Sea which improves upon more general methods used for example by Global Fishing Watch, in particular by providing gear specific information. I recommend it for publication as a data paper but I would suggest to address certain general criticisms of AIS, with a particular inclusion of Hintzen et al., (2025) which brings up concern on how AIS derived data (from GFW in particular) can overestimate the fishing effort by sometimes specifying certain activities, like echosounder use, as fishing activities. I know of Couce et al., (2020) as well which provides similar trawling hours information for the North Sea and it would be nice to see some comparison of methods from that and potentially other similar datasets/methods. I think you can use that information to provide more compelling arguments to how the current methods paper may improve upon other methods while remaining transparent with potential limitations. Overall, I think it’s a nice contribution and while the framing could be improved to connect it to other studies, I think it can still be very useful for scientists, managers and policy makers.
Detailed comments
Abstract
L22 Can you please state explicitly how this supports EBFM? I think you mean because it gives you some good indications/proxies of bottom disturbance (and improvements on other AIS derived information like from GFW) but less informed readers might not pick this up right away. Same for supporting the revision of management plans as it’s not so clear (even though it sounds nice).
Introduction
L32 Try and harmonize the use of EBM and EBFM which is in the abstract.
L64-67 Here I suggest also including critiques of how GFW uses AIS data. I’m really just thinking about the study from Hintzen et al., (2025) which compared GFW fishing events with independently recorded haul data and found that GFW derived estimates could substantially overestimate the actual trawling being deployed. I would just briefly mention here that study or that critiques exist and go back to this topic in the discussion. Like would your study and improvements on AIS derived fishing effort also be less likely to fall under these criticisms from Hintzen et al., which point out, for instance, that vessels searching for fish using sonar equipment may be mistaken for fishing activities (by GFW derived AIS analysis at least)? Also, it the same critiques are still valid, then this paper should at least mention that as potential limitations. This could potentially strengthen your arguments that this product is a valid improvement compared to prior GFW methods (at least in the Adriatic).
Methods
L99 Not a huge deal but “together constitute” probably can be phrased simpler (less wordy) with “represent”, “encompass” or something else. I just think it could flow better.
L99-100 It is interesting information to know that these GSA’s are the most highly collaborative fisheries management zone in the Adriatic but it comes out sounding less relevant here compared to the rest of the paper. I would still try to fit it in but suggest breaking up that sentence. Also, isn’t GSA 17 and 18 the total Adriatic? The sentence makes it seem like it is just part of the Adriatic (though I admit I’m less familiar with that sea).
L103-104 again, see if this is enough to evade criticism from Hintzen et al., or like minded scientists (not new text needed here but potentially in the discussion).
L118 so does manually validated mean that local experts took a look and said it was okay? Shouldn’t it then say “visually” validated or something like that. Consider clarifying.
Results
L175-176 the threshold application for excluding analysis (because they’re probably not trawling below that value), shouldn’t that be introduced in methods not results (I mean I see that it flows kinda nice here but still)? Also for others to reproduce this work, it would be useful in the methods.
L179 ok I’ll be the annoying reviewer that suggests more content here. You have some nice qualitative descriptions of the gear specific effort data in figures 1 and 2. Is it possible to also add a small summary table showing more quantitative descriptions like total fishing hours by gear per GSA or other spatial information?
L192 I hate to be a stickler but you do use the term significant here in the results to show statistical significance so perhaps just use ‘marked’, or ‘substantial’ instead of significant here.
L215-217 the reasons for the Albanian increase may be valid though I wonder if that information belongs in the results and not the discussion (possible improved reporting and fleet expansion).
L242 Just to be pragmatic and clear to everyone, can you please specify that these percentages are coming from the full 2 GSA areas (combined) to form the denominator used for the percentages (The Adriatic Sea). I think it can be specified once briefly early in the paragraph (kind of like on line 386 in the discussion).
Discussion
L404 in this paragraph you can add the example of the fleet expansion and better reporting from Albania potentially leading to the results.
I don’t think you should go overboard on the limitations of the method but a few things come to mind for me. While you might be able to make some assumptions based on gear, I don’t see any swept area or swept area ratios for bottom impacting gears. This may be a bit difficult for otter trawling (trawl doors make variable swept areas) but it’s relevant because combining gear and gear width information will let you know more about the actual seabed disturbance much more than fishing hours. Also swept area ratio is directly used for analyses like relative benthic status (RBS; Pitcher et al., 2017) and I don’t think the study gives indications of gear penetration depth (though they can be inferred/assumed I guess).
Perhaps most important, I think that a discussion integrating the criticisms of GFW methods from Hintzen et al., 2025 is appropriate because they likely also apply in some ways to the current study. I do think this work improves upon the more general GFW methods for that particular area but nonetheless there may be general criticisms which may need to be addressed.
Also the inclusion of Couce et al., 2020 (North Sea) and other similar works are missing from the discussion. Try and add a paragraph which compares and contrasts other work which similarly tries to spatially quantify bottom trawl effort.
References
Couce, E., Schratzberger, M., & Engelhard, G. H. (2020). Reconstructing three decades of total international trawling effort in the North Sea. Earth System Science Data, 12(1), 373–386. https://doi.org/10.5194/essd-12-373-2020
Hintzen, N. T., Brigden, K., Kaastra, H.-J., Mackinson, S., Pastoors, M. A., & van de Pol, L. (2025). Bias in Global Fishing Watch AIS data analyses results in overestimate of Northeast Atlantic pelagic fishing impact. ICES Journal of Marine Science, 82(3), fsaf033. https://doi.org/10.1093/icesjms/fsaf033
Pitcher, C. R., Ellis, N., Jennings, S., Hiddink, J. G., Mazor, T., Kaiser, M. J., Kangas, M. I., McConnaughey, R. A., Parma, A. M., Rijnsdorp, A. D., Suuronen, P., Collie, J. S., Amoroso, R., Hughes, K. M., & Hilborn, R. (2017). Estimating the sustainability of towed fishing-gear impacts on seabed habitats: A simple quantitative risk assessment method applicable to data-limited fisheries. Methods in Ecology and Evolution, 8(4), 472–480. https://doi.org/10.1111/2041-210X.12705