ICES WGNEPS REPORT 2013 SCICOM STEERING GROUP ON ECOSYSTEM SURVEYS SCIENCE AND TECHNOLOGY ICES CM 2013/SSGESST:21 Report of the Working Group on Nephrops Surveys (WGNEPS) 5–8 November 2013 Barcelona, Spain
ICES WGNEPS REPORT 2013 SCICOM STEERING GROUP ON ECOSYSTEM SURVEYS SCIENCE AND TECHNOLOGY
ICES CM 2013/SSGESST:21
Report of the Working Group on Nephrops Surveys (WGNEPS)
5–8 November 2013
Barcelona, Spain
International Council for the Exploration of the Sea Conseil International pour l’Exploration de la Mer
H. C. Andersens Boulevard 44–46 DK-1553 Copenhagen V Denmark Telephone (+45) 33 38 67 00 Telefax (+45) 33 93 42 15 www.ices.dk [email protected]
Recommended format for purposes of citation:
ICES.2013. Report of the Working Group on Nephrops Surveys (WGNEPS), 5–8 November 2013, Bar-celona, Spain. ICES CM 2013/SSGESST:21. 27 pp.
For permission to reproduce material from this publication, please apply to the General Secretary.
The document is a report of an Expert Group under the auspices of the International Council for the Exploration of the Sea and does not necessarily represent the views of the Council.
© 2013 International Council for the Exploration of the Sea
ICES WGNEPS REPORT 2013 | i
Contents
Executive summary ................................................................................................................ 1
1 Introduction .................................................................................................................... 2
2 Changes to UWTV survey design, coverage and intensity .................................... 2
3 The Series of ICES Survey Protocols (SIPS) ............................................................. 9
4 Changes to trawl survey design, coverage and intensity ..................................... 11
5 Technological developments ..................................................................................... 14
6 Bio-rythmitic behaviour ............................................................................................. 16
Annex 1: List of participants............................................................................................... 21
Annex 2: Agenda ................................................................................................................... 23
Logistical Information ......................................................................................................... 23
Annex 3: WGNEPS terms of reference for the next meeting ........................................ 25
Annex 4: Recommendations ............................................................................................... 27
ICES WGNEPS REPORT 2013 | 1
Executive summary
The first meeting of the Working Group on Nephrops Surveys (WGNEPS), formerly known as the Study Group on Nephrops Surveys (SGNEPS), took place from the 5–8 November 2013 in Barcelona in Spain. The meeting was chaired by Dr Colm Lordan of the Marine Institute in Ire-land and included three participants from Ireland and Spain, two from Denmark and England and one participant from Northern Ireland, Scotland, Sweden and Portugal. WGNEPS is the co-ordination expert group for Nephrops UWTV and Trawls surveys within the ICES area. Various updates on survey design were presented, discussed and concluded on by the EG at the meeting. Coordination between surveys across countries and laboratories was also carried out at the meet-ing. WGNEPS adopted multi-annual ToRs in 2013 and this first meeting focused on compiling content for a Cooperative Research Report (CRR), which will document the UWTV methods, the assessment and advice approach currently used within ICES in one single document. This doc-ument remains in a draft from but should be completed for review at the 2014 meeting. Work also began on a SIPS document is due to be drafted at the 2014 meeting. WGNEPS had updates on two research areas; technological developments in automatic burrow recognition and bio-rythmitic behaviour of Nephrops. These areas an extremely important research topics for the EG.
2 | ICES WGNEPS REPORT 2013
1 Introduction
The first meeting of WGNEPS was hosted by Dr Jacopo Aguzzi and Dr Francisco Sardà at the ICM-CSIC (www.cmima.csic.es) in Barcelona. This group has evolved from the former SGNEPS and has adopted a multi-annual term of reference for ICES Science experts groups. WGNEPS is the coordination group for Nephrops UWTV and trawl sur-veys within ICES and has a quality assurance and development role. The main objec-tives of the 2013 meeting were to review changes in survey design, coverage, procedures, etc. (ToRs a and e) and to commence work on a Co-operative Research Report (CRR) which outlines the UWTV approach in an integrated way from survey to assessment and advice. Following a recommendation made to the survey groups at the WKESST, the group will develop a SIPS document for UWTV surveys at next year’s meeting.
2 Changes to UWTV survey design, coverage and intensity
SGNEPS 2012 made some recommendation that each country should assess the scope for adjusting survey intensity and coverage provided that the final survey achieved a RSE of <20%. WGNEPS 2013 reviewed any changes that were implemented.
Ireland
In 2012 and 2013, Ireland modified sampling intensity and increased survey coverage based on the recommendations of SGNEPS 2012. The total numbers of stations for 2012 and 2013 remains broadly similar ~330 to previous years (Figure 2.1). The numbers of stations in FU15, FU17 and FU22 were reduced in 2012 and 2013. New surveys were developed for FU16 and FU20–21 and improved coverage was achieved for FU19. In addition, the survey design was change from a randomized square grid to a random-ized isometric gird where all stations were equidistant. The CVs for surveys where sampling intensity was reduced either had no or minor decreases in relative precision. The individual survey reports and further details of the survey design and numbers of stations are available from the Marine Institute Open Access Repository at http://oar.marine.ie/handle/10793/59.
ICES WGNEPS REPORT 2013 | 3
Figure 2.1. Time-series of the total number of UWTV stations carried out by Ireland by Functional Unit (Stations in FU15 are carried out in collaboration with AFBI in UK-NI).
UK-Northern Ireland
From 2003 to 2011 a randomized fixed square grid for the western Irish Sea (FU15) Nephrops ground has been used. The grid spacing was 3.5 nautical miles (6.5km) and an adaptive approach is taken whereby stations are continued past the known perim-eter of the ground until the burrow densities are zero or very close to zero. The initial ground perimeter has been established using a combination of integrated logbook-VMS data (using the methods described in Gerritsen and Lordan, 2011), British Geo-logical Survey (BGS), and other sediment maps and previously collected UWTV data. The same ground boundaries have been used throughout the time-series. Following a review (Doyle et al., 2013) the grid design was changed from a 3.5 nautical mile square grid prior to 2012 to 4.5 nautical mile isometric grid in 2012. In 2013, the grid spacing was increased further to 5.0 nautical mile isometric grid. The main motivation to do this was to achieve full spatial coverage of FU15 while giving the option to reallocate ship time to increase coverage in other Functional Units (FU16, FU20–21 and FU19); also in line with SGNEPS recommendations (ICES, 2012b). Reducing the number of stations is not expected to have significantly affected the accuracy of the survey esti-mate given the apparent spatial autocorrelation in density across the area. Increasing the station spacing to 7 nautical miles did not affect the accuracy of the result in the removal analysis carried out by Doyle et al. (2013). The precision do not appear to have been significantly reduced either with as CV of 3%, which was in line with previous estimates well below the SGNEPS 2012 recommendation of 20%.
Conclusion: Despite the reduction in survey density, the precision of the surveys re-mains high and the high correlations and regular density surface indicates that the sur-veys are likely to be giving an accurate abundance estimate. The improvements in spatial coverage now mean that the vast majority of fished Nephrops in VII is covered with some UWTV survey. The ship time saved was used to develop new surveys in FU16 and FU20–21 which particularly challenging areas to achieve full coverage each year. WGNEPS endorsed the changes in design and coverage made by the Ireland and UK-Northern Ireland and recommended that these changes be maintained in future. The new and old surveys remain a suitable basis for the UWTV based assessment and advice.
0
50
100
150
200
250
300
350
400
450
Num
ber o
f sta
tions 2021
22
19
17
16
15
4 | ICES WGNEPS REPORT 2013
Scotland
Marine Scotland Science made no changes to any of the UWTV hardware in 2013 dur-ing either of the two DCF funded cruises. However two aspects of the survey design were implemented on the May/June offshore North Sea and West coast survey aboard MRV Scotia, with reduced implications for the MRV Alba-na-Mara cruise in August.
The first change was in relation to defining the boundary of the survey area in the North Minch. Historically the survey design in this area was based on sediment infor-mation from the BGS and stations were randomly generated within the different ap-propriate sediment strata. Analysis of VMS data showed that the Nephrops fleet in the North Minch often fished in areas not covered by the survey, and conversely the sur-vey occasionally focused on areas where the fleet avoided – even when considering the absence of data from the <12m section of the fleet. This approach underestimated the full extent of the fishing grounds in the North Minch. At WKNEPH 2013 it was agreed that following analysis of all available data that the boundary of the North Minch UWTV survey area should be defined using the union of yearly polygons as generated from the VMS data. This increased the survey area from 1775km2 to 2908km2.
The second change was partially due to the increased survey area in the NM but also in an attempt to reduce the relative standard error between survey area and number of stations in those areas, as discussed at SGNEPS 2012. As a result, during the MRV Scotia cruise, the number of planned survey stations were increased in the North Minch, South Minch and Devils Hole in an attempt to address this observation. To accommodate this increase in effort without any increase of the ship’s time, the sched-uled number of stations in the Sound of Jura and the Clyde (where the CV was rela-tively low) were reduced. However due to poor weather and the abundance of creels in the South Minch, and an incident where the TV cable was severed and three days were spent trying to retrieve the sledge from the seabed in the North Minch, the actual number of stations surveyed in these two areas was similar to previous years. Due to the sledge recovery activities no time remained to survey Devils Hole in 2013. During the MRV Alba-na-Mara UWTV cruise in August the number of planned stations in the Moray Firth were increased at the expense of stations in the Firth of Forth in an attempt to reduce the high CV in FU 9.
Conclusion: Both approaches attempted in 2013 (applying the re-defined North Minch boundary and varying the number of survey stations in each area to reduce the CVs) will be implemented again in both MSS UWTV surveys in 2014.
England
FU5
In autumn 2010 for the first time a UWTV survey was conducted on the Botney Gut / Silver Pit ground providing a starting point on using UWTV surveys to assess the Nephrops populations in this area, although, the following surveys were not very suc-cessful due bad weather and poor visibility. This revealed to be a particular difficult ground to survey, located in the middle of the North Sea the Botney Gut / Silver Pit area is completely exposed to bad weather and between 2010 and 2012 only 2 out of 4 attempts to survey this ground were successful. Also burrow identification in this area was notoriously difficult due to the high burrow densities of small burrows (mainly in the eastern side of the ground) and poor visibility. Due to funding constraints, Cefas is no longer covering this ground, although the possibility of having a collaborative sur-vey in future is being discussed with the Netherlands. More information regarding
ICES WGNEPS REPORT 2013 | 5
future options to reinstate this survey will be discussed in the near future and this in-formation will be passed on to the WGNEPS.
FU6
No gear or design changes were made to the survey in the Farn Deeps ground since the last SGNEPS update.
Survey time changes
In 2013 the UWTV survey of the Farn Deeps was carried out in June. Traditionally the survey was held in October but poor weather and low visibility caused by Nephrops trawling activity had caused problems in previous years. Due mainly to a very intense trawling activity in October many stations were repeated at least once, until a reason-able good visibility was achieved. This has been a recurring problem in previous sur-veys where extra time (up to 1 day) at sea was used to revisit stations with poor quality footage; it was recommended that the survey should be moved back, before the start of the fishing season and where the chances of having better weather conditions are higher. The 2013 survey was very successful, excellent clarity was achieved in all sta-tions and no time was lost to repeat stations, this proved to be a more cost efficient survey.
This move in survey time was discussed in the Benchmark meeting in February 2013, where the following topics were addressed:
• There is practically no targeted Nephrops fishing between May and Septem-ber; therefore there is minimal scope for fishery induced changes in stock abundance between the new survey time and the previous October timing.
• There are no migrations of animals to consider, Nephrops are on (or rather in) the ground all year-round. What affects their availability to the fishery is their emergence behaviour, which does not affect the ability to count the burrows.
• The only factor which may affect the burrow density between June and Sep-tember is any seasonality in the creation of new burrow complexes by juve-niles. There are no data regarding the timing of burrow creation with which to make an informed judgement as to whether this is likely to be an im-portant effect.
The group concluded therefore that the FU 6 TV series should continue to be regarded as a single series and that where possible work should be undertaken to investigate if the survey timing change was likely to have affected results.
Outlook for reduction of the station density
At over 110 stations covering an area of 3000km2, the FU6 survey provides one of the highest resolution coverage of the ground with a low CV (typically 2%). The annual CV is determined by bootstrapping the survey (with replacement) and the re-running the geostatistical model.
Investigations were commenced in order to evaluate the impact upon survey variance in a reduction of the station density and which station configuration would be optimal. The objective of the exercise was to see if there were permutations of existing positions but at lower density (70, 80, 90, 100 stations) which would provide robust estimates of the population abundance.
6 | ICES WGNEPS REPORT 2013
The number of possible permutations is very high and impractical to test all of them; therefore a random selection of stations at each of the densities was created. In order to ensure complete spatial coverage of the ground, only those permutations, which covered 90% of the latitudinal and longitudinal range of the full dataset, were consid-ered valid. Three hundred valid permutations were to be constructed for each density.
There remain one or two technical issues with this approach that require investigation before results can be produced.
An alternative approach would be to sequentially identify the stations with minimal influence on the abundance estimate (looked at over several years of survey results), however this has yet to be coded and identified as it is anticipated that the calculation time would be orders of magnitude greater than the first approach.
FU14
No gear or design changes were made to the survey in the Farn Deeps ground since the last SGNEPS update.
Due to the construction of the Walney Offshore wind farm in the south part of the ground, in 2010 and 2011 some stations were abandoned. VMS data indicated vessels were avoiding that part of the ground while the wind farm was in construction and fishing was not allowed around the construction side. In 2012 most of the bottom sta-tions were surveyed and in 2013 three stations were moved out the restriction area (highlighted in Figure 2.2: 14-AH, 14-AL, 14-AK).
Figure 2.2. Map of the main FU14 grid showing the change in the station positions for 14-AH, 14-AL, 14-AK.
ICES WGNEPS REPORT 2013 | 7
Conclusion: Cefas is no longer covering the Botney Gut/Silver Pit ground although in-ternational collaborations are currently being considered to carry out a future survey in this area. There is scope for reducing station intensity in the Farn Deeps. There was a change in timing from October to June due to poor weather and low visibility caused by Nephrops trawling activity. June proved to be a better time to achieve good coverage in the Farn deeps and it is more cost efficient as no time was lost to redo stations.
Denmark/Sweden
After some exploratory work in 2008 and 2009 and quite limited area coverage in 2010 the survey in the Skagerrak and Kattegat (FU 3 and 4) was conducted in collaboration between Denmark and Sweden with wider area coverage since 2011 (Figure 2.3). How-ever, important parts of the assumed distributional range of Nephrops were still not covered in 2013 for which the data analysis has yet not been finished. The survey is still developing and improving spatial coverage is expected for 2014.
Figure 2.3. Sampling locations and Nephrops burrow density in the UWTV survey in the Skagerrak and Kattegat (FU 3 and 4) in 2008 (26 stations, Denmark only), 2009 (47 stations, Denmark only), 2010 (72 stations, Denmark only), 2011 (146 stations) and 2012 (166 stations).
9° 10° 11° 12° 13°
0.00
0.50
1.00
100m
50m
25m
G
43
44
45
46
42
41
2008n/m2
9° 10° 11° 12° 13°
0.00
0.50
1.00
100m
50m
25m
G
43
44
45
46
42
41
2009n/m2
9° 10° 11° 12° 13°
0.00
0.50
1.00
100m
50m
25m
G
43
44
45
46
42
41
2010n/m2
SA 1
SA 2
9° 10° 11° 12° 13°
0.00
0.50
1.00
100m
50m
25m
G
43
44
45
46
42
41
2011n/m2
SA 1
SA 2
SA 3
SA 4
SA 5
SA 6
9° 10° 11° 12° 13°
0.00
0.50
1.00
100m
50m
25m
G
43
44
45
46
42
41
2012n/m2
SA 1
SA 2
SA 3
SA 4
SA 5
SA 6
8 | ICES WGNEPS REPORT 2013
Conclusion: Further effort is needed to ensure better spatial coverage. In defining the boundaries, it is important to integrate data from all available sources; VMS linked logbooks, interpolated habitat maps, multibeam or other direct habitat maps, trawl surveys, observer data. It is important to consider the biases and accuracy associated with these data.
Recommendations FU 3 and 4
• Revision of subarea boundary definitions based on update of VMS and other information (sediment map, logbook for non VMS trawlers and creel fishery)
• Consider extension of survey coverage towards the western Skagerrak and Swedish coastal waters (2 potentially new subareas)
• Revision of subarea station allocations based on their contribution to the overall relative standard error
• Build up a set of reference footage for FU3 and 4
The Netherlands
A representative from the Netherlands attended WGNEPS for the first time. They are currently scoping out the possibilities for survey development in collaboration with other countries.
Conclusion: WGNEPS recommended that stocks with annual landings >500–1000 should have dedicated UWTV surveys. The design and protocols for new surveys should follow best practice as outlined by WGNEPS and implemented on other sur-veys developed to date.
ICES WGNEPS REPORT 2013 | 9
3 The Series of ICES Survey Protocols (SIPS)
The main task of the WGNEPS 2014 meeting will be to prepare an UWTV SIPS docu-ment. The ground work for this was discussed at the WGNEPS 2013 meeting and the following recommendations were made:
Training and Reference Material
1 ) All institutes must produce training and reference material for areas cov-ered by UWTV. As a minimum this is to include annotated 1 minute video footage covering range of density and visibility and other burrowing species encountered and also a photograph guide of signature features of Nephrops burrows.
2 ) A standard operating procedure for training in Nephrops Burrow identifica-tion should be produced and implement on commencement of a survey.
3 ) A standard operating procedure should be produced for burrow counting. This is to include details of how many minutes are to be counted, warm-up session details, where to count on the screen and removal of minute counts where footage quality deteriorates.
4 ) Staff and protocol exchange between institutes where possible and espe-cially on collaborative surveys such as FU15, FU14 and other FUs
5 ) Warm up counts should be made for the first minute of every station, if a counter resumes counting after a break of more than 8 hours then counter needs to blind count a full station of 10 minutes.
Reference Footage Sets
1 ) Each institute should produce reference set to comprise of 10 runs of 5 minutes where the footage selected for these reference counts covers the range of visibility, Nephrops density and species complexes likely to be en-countered in each area. Each institute collated video footage from their ar-chives and burnt them onto DVD with each run comprising a separate chapter.
2 ) Agreed consensus counts must be made on these reference counts either us-ing independent national experts are with international exchanges.
3 ) If reference footage set outdated it should be updated regularly 5–10 years. It would be better to randomly present reference footage interspersed with annual footage.
4 ) There should be sufficient reference sets per area to ensure that they cannot be learned.
Quality Control of Survey Counts
1 ) Datasheets should be separate for each independent verification count. 2 ) The first counter should record all ancillary information to avoid duplica-
tion of work. 3 ) Institutes must to use Linns CCC on station basis to check counter con-
sistency. However, the threshold to be uses will be discussed and recom-mended by WGNEPS 2014.
10 | ICES WGNEPS REPORT 2013
4 ) Edge station of low density in sandy ground record 7 minutes and count 7 minutes.
5 ) Rocky ground record footage if possible for 1 minute and included as zero station in the workup since failing to do so introduces bias.
Quality Control of Navigation Data
1 ) When navigation data for both sled and ship track is collected that institute check the sled track should be used where it is of sufficient quality.
ICES WGNEPS REPORT 2013 | 11
4 Changes to trawl survey design, coverage and intensity
Portugal (Cristina)
FU 28 and 29 (Southwest and South Portugal)
No gear or design changes were made to the survey in the FU 28 and 29 grounds since the last SGNEPS update.
A regular grid composed by 22 rectangles in FU 28 and 59 rectangles in FU 29 is used, with one station within each rectangle. Each rectangle has 6.6 minutes of latitude x 5.5 minutes of longitude for the SW coast and vice-versa for the south coast, corresponding approx. to 33 nm2. The grid was designed for a trawl survey to cover the main crusta-cean fishing grounds within the range of 200–750 m (Figure 4.1). The areas deeper than 750 m, where the giant scarlet prawn occurs, are not covered.
The hauls are carried out during daytime with a speed of 3 knots and have duration of 30 minutes. Although the crustacean species are the target (Norway lobster, rose shrimp and red and blue shrimp), data from all other taxa and species are also col-lected, as well as marine litter.
Figure 4.1. Survey grid in FUs 28 and 29 overlaying the crustacean fishing grounds represented by VMS records.
In the year 2010, due to some technical problems of RV “Noruega” in covering areas deeper than 600m, the survey plan had to be adjusted. In 2011, the survey did not cover the whole area due to engine failure. No survey was carried out in 2012 and in 2013 the whole area was covered with 81 stations. The complete coverage in this year can only be compared with the 2009 survey (Figure 4.2).
12 | ICES WGNEPS REPORT 2013
Spain (Yolanda)
Figure 4.2. Norway lobster biomass indices distribution from crustacean surveys in the period 2005–2013.
FU 31- FU 25 (Cantrabrian Sea) and FU 26 (West of Galicia)
The SpGFS-WIBTS-Q4 covers the northern Spanish shelf comprised in ICES Division VIIIc and the northern part of IXa, including the Cantabrian Sea and off Galicia waters (Figure 4.3). However, this survey is not targeting to estimate Nephrops abundance. It is a bottom-trawl survey with a random stratified by depth strata sampling design. The number of hauls in each depth stratum is proportional to trawlable surface and the haul duration is 30 min. The aims are to collect data on the distribution, relative abun-dance and biology of commercial fish species such as hake, monkfish and white an-glerfish, megrim, four-spot megrim, blue whiting and horse mackerel. Abundance indices are estimated by length and in some cases by age, with indices also estimated for Nephrops, and data collected for other demersal fish and invertebrates. This survey
ICES WGNEPS REPORT 2013 | 13
usually starts at the end of the 3rd quarter (September) and finishes in the 4th quarter and extends from 30 to 800 m depths. Nephrops survey abundance index is not used in the assessment of FUs 25, 31 and 26.
Figure 4.3. Catches in biomass of Nephrops on the Northern Spanish Shelf bottom-trawl surveys during the last decade: 2003–2012.
FU 30 (Gulf of Cadiz)
No survey specially designed to evaluate abundance indices of Nephrops is carry out in FU 30. The bottom-trawl surveys SpSGFS-cspr-WIBTS-Q1 and SpGFS-caut-WIBTS-Q4 occur in the Southern part of ICES Division IXa, the Gulf of Cádiz, which corresponds to the FU 30. The area covers 7 224 Km2 and extends from 15 to 800 m depths. Sampling design follows a random stratified scheme with 5 depth strata (15–30 m, 31–100 m, 101–200 m, 201–500 m and 501–800 m; Figure 4.4). The number of hauls in each depth stra-tum is proportional to trawlable surface and the haul duration is 60 min. All hauls are carried out during daylight hours using Baka 40/60 trawl fear with a 43.6 m footrope and 60.1 m headline. An inner 20 mm mesh codend liner is used to prevent the escape of small individuals. This survey collects data on the distribution, relative abundance and biology of commercial fish such as hake, wedge sole, sea breams, mackerel and horse mackerel. Abundance indices are also estimated for crustaceans (mainly rose
Nephrops norvegicus42
4344
2 kg1 kg0.5 kg
2003
2 kg1 kg0.5 kg
2004
4243
44
2 kg1 kg0.5 kg
2005
2 kg1 kg0.5 kg
2006
4243
44
2 kg1 kg0.5 kg
2007
2 kg1 kg0.5 kg
2008
4243
44
2 kg1 kg0.5 kg
2009
2 kg1 kg0.5 kg
2010
4243
44
10 9 8 7 6 5 4 3 2
2 kg1 kg0.5 kg
2011
10 9 8 7 6 5 4 3 2
2 kg1 kg0.5 kg
2012
14 | ICES WGNEPS REPORT 2013
shrimp and Nephrops), cephalopods species and other invertebrates. The Nephrops abundance index from spring survey series is used in the assessment of FU 30. How-ever, this survey is not specifically directed to Nephrops and it is not carried out during the main Nephrops fishing season.
Figure 4.4. Grid and location of hauls from, SpSGFS-cspr-WIBTS-Q1 series (1993–2012) (a) and Nephrops distribution in yield in number individuals/h (b).
5 Technological developments
To aid in the current Nephrops population assessment procedure, collaborative research is currently being done between the Marine Institute Galway and Trinity College Dub-lin. A presentation on this work was given at the WGNEPS meeting. This research in-volves the exploration of computer vision techniques to:
i ) improve visibility ii ) analyse the content of the recorded survey videos automatically to identify
items of interest to scientists.
To improve the visibility, two techniques have been developed. The first technique improves visibility in the video directly by correcting the uneven lighting in these scenes. This correction is accomplished by modelling the light distribution on the sea-bed with a Gaussian-like degradation function, the parameters of which are estimated with corresponding points across consecutive frames. Sample results obtained from this method are shown in Figure 5.1. The second technique developed improves visi-bility by using the corrected video sequences obtained from the first technique to gen-erate a wide area view or mosaic of the surveyed seabed area. Figure 5.2 shows an example of a mosaic generated from 65 frames of survey typical survey video, using this technique.
200 m
500 m
100 m
30 m
ICES WGNEPS REPORT 2013 | 15
Figure 5.1. (Left) Original and Corrected (Right).
Figure 5.2. (Left) Original Video Frame, (Middle) Mosiac generated from 65 frames, and (Right) detected burrows (blue) and Nephrops (red).
For the content analysis part of this research three items of scientific interest are tar-geted in these videos for automatic identification: i) burrows, ii) Nephrops, and iii) Nephrops clusters. The identification of burrows and Nephrops involves two main stages of object detection and classification. For the object detection stage, novel image seg-mentation algorithms are developed that target candidate burrow and Nephrop re-gions, which are mainly dark in contrast and bright orange in colour respectively. These candidate regions are then classified using supervised learning frameworks (Support Vector Machines and K-Nearest Neighbour), based on their colour, size, shape and texture features. Based on a test set of 10 mosaics containing 6,530 burrows, 21 Nephrops and 37,168 other items, these purposed systems obtained average recall and precision values of 95% and 92% for burrows and 88% and 88% for Nephrops. Ex-amples of detected burrows and Nephrops obtained from a section of one of these test mosaics are given in right image in Figure 5.2.
The identification of the last item of interest, the Nephrops clusters is still ongoing. Cur-rently three of the four main clustering features used by scientists: i) Crescent shaped burrows, ii) claw marks, and iii) burrows containing Nephrops, can be identified with approximately 70% accuracy.
16 | ICES WGNEPS REPORT 2013
WGNEPS concluded that significant progress has been made and the approach has some important advantages:
1 ) The image enhancement and video mosaicking produces a high quality im-age that could be counted in parallel with the live footage and used to vali-date counts between readers,
2 ) The automatic detection show very promising results and could be used to reduce work load but more importantly reduce subjectivity in the process.
3 ) Having a mosaic of the video footage allows for various new statistics to be generated on number of holes, size of Nephops and burrows.
To assist in this research participants were asked to carry out verification recounts on the mosaics available on the WGNEPS SharePoint site by mid-December.
6 Bio-rythmitic behaviour
Burrow emergence rhythms deeply influence the commercial fishery of the Norway lobster (reviewed by Bell et al., 2006; Aguzzi and Sardà, 2008). Animals show a noctur-nal emergence in shallow water areas, phased at night-time, especially under full moon conditions, shifting the event of that timing as the depth increases (Chapman et al., 1975). Recently, crepuscular and fully diurnal catch patterns have been found in deeper water areas of the shelf and the slope, respectively (Aguzzi et al., 2003). Animals ap-parently relay on an “optimum light intensity” to carry out their seabed surface behav-ioural activities especially in relation to feeding (Oaklwey, 1979). Animals chiefly emerge for territorial control and for food retrieval (Katho et al., 2013). On the shelf animals carry out a first event of emergence at sunset remaining engaged in that activ-ity depending form their initial success (Aguzzi et al., 2004b). Those succeeding in food collection may re-enter their burrows, while other keep searching under the protection of darkness from visual predators. On the deeper slope a similar dynamic occurs but this time, animals use the faint midday light. Females apparently reduce their emer-gence duration when berried, carrying out what can be defined as “door-keeping” be-haviour (Aguzzi et al., 2007). Animals wait at the entrance of their burrows patrolling the surrounding seabed parcel. A sexual dimorphism in that behaviour may occur, with males more prone to emergence than females (Aguzi et al., 2008b).
Nephrops distribution as a wide bi-dimensional character, encompassing shelves and slopes as well as the North Atlantic and the Mediterranean (Aguzzi et al., 2004c). In this geographic range populations can be exposed to internal tides or inertial (i.e. atmos-pheric-driven) currents when tides are negligible (as in the case of the Mediterranean; Aguzzi et al., 2008a). Therefore, the modulation of Nephrops behavioural rhythms has to be very plastic and adaptable to local conditions (Aguzzi and Company, 2010). An-imals rely on optimum monochromatic blue (i.e. 480 nm) light intensity to set their behavioural activity rhythms upon the day-night cycles at different depths (Aguzzi et al., 2009a). In fact, this radiation is the sole invariantly present all over the photic and the disphotic oceanic depth range until the complete extinction of all-solar light in the deep-sea (Aguzzi et al., 2011a). A reduction in its intensity showed how laboratory tested animals’ shift their activity from nocturnal tom diurnal, as it can be observed when performing a temporally scheduled trawling in shelf and slope populations (Aguzzi et al., 2009b). This phenomenon, defined as a “temporal niche switching” oc-curs for a still unknown mechanism, likely involving the hormonal control (Chiesa et al., 2010).
ICES WGNEPS REPORT 2013 | 17
Such an activity shift could be mediated by melatonin, which is rhythmic only at sim-ulated day-night cycles of brighter intensity (as proxy of the shelf photic environment). Melatonin is the putative hormone regulating the activity rhythms of vertebrates and invertebrates, as well as the growth and reproduction (reviewed by Vivien-Roels and Pévet, 1993, Fukuhara et al., 2005; Aguzzi et al., 2006). The behavioural reaction of ani-mals to currents is currently unknown and it is under laboratory testing (Sbragaglia et al., 2013a), but apparently internal tides modulate emergence duration and hence com-mercial catchability (Bell et al., 2008). Melatonin may be involved in this control since rhythmically expressed in deep-sea species where light cycles are absent but internal tides act as modulator of behaviour (Wagner et al., 2007).
Currently, the true proportion of animals carrying out burrow emergence behaviour on a 24-h base, is still unknown (Sbragaglia et al., 2013b). Day-night laboratory tests always reveal a proportion of arrhythmic animals that may represent a demographic component of unknown behaviour in the field (Atkinson and Naylor, 1976; Aguzzi et al., 2004a). Moreover, some animal may inhibit the emergence of subordinate neigh-bours based on territorial aggressive displays (Menesatti et al., 2009, Aguzzi et al., 2011b). Dominant individuals may suppress the emergence of weaker conspecific, pos-sibly representing a natural mechanism preserving Nephrops populations despite a con-sistent increase in the fishing effort in the past decades (Sardà and Aguzzi 2012).
WGNEPS noted that emergence behaviour of Nephrops in the wild could be investi-gated using existing UWTV datasets because most laboratories collected density, num-bers out and numbers door keeping for every station. Most countries operate throughout the night. An example for two Irish survey areas is given below (Figure 6.1.). Data should be sent to Jacopo Aguzzi for further analysis.
18 | ICES WGNEPS REPORT 2013
Figure 6.1. The cumulative numbers of Nephrops out of burrows by hour for UWTV to date on the Aran Grounds (top) and the Smalls (bottom). Approximate sunrise times is shown as a blue bar and sunset as a red bar.
Bibliography Aguzzi, J., Bullock, N. M., Tosini, G. 2006. Spontaneous internal desynchronization of locomotor
activity and body temperature rhythms from plasma melatonin rhythm in rats exposed to constant dim light. Journal of Circadian Rhythms 4: 6.
Aguzzi, J., Company, J. B. 2010. Chronobiology of deep-water decapod crustaceans on continen-tal margins. Advances in Marine Biology 58: 155–225.
Aguzzi, J., Company, J. B., Abello, P. 2004a. Locomotor activity of continental slope Nephrops norvegicus (Decapoda: Nephropidae). Journal of Crustacean Biology 24(2): 282–290.
Aguzzi, J., Company, J. B., Costa, C., Menesatti, P., Garcia, J. A., Bahamon, N., Puig, P., Sardà, F. 2011a. Activity rhythms in the deep-sea: a chronobiological approach. Frontiers in Biosci-ence-Landmark 16: 131–150.
Aguzzi, J., Company, J. B., Sardà F. 2004b. Feeding activity rhythm of Nephrops norvegicus (L.) of the Western Mediterranean shelf and slope grounds. Marine Biology 144: 463–472.
0
50
100
150
200
250
300
350
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Hour of the day in BST (GMT+1)
Aran Grounds Nephrops Out BST
020406080
100120140160
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Hour of the day in BST (GMT+1)
Smalls Nephrops Out
ICES WGNEPS REPORT 2013 | 19
Aguzzi, J., Costa, C., Menesatti, P., García, J. A., Sardá, F. 2009a. Monochromatic blue light en-trains diel activity cycles in the Norway lobster, Nephrops norvegicus (L.) as measured by automated video-image analysis. Scientia Marina 73(4): 773–783.
Aguzzi, J., Maynou, F. J., Company, J. B., Rottlant, G., Sardà F. 2004c. The dimensional units of Nephrops norvegicus (L.) distribution: from burrows to populations. Proceedings of the Ma-rine Biogeography and Mediterranean Sea: patterns and dynamic of biodiversity. Italian Society of Biogeography (SIB); Biogeographia 25: 81–92 (ISSN: 1594–7610).
Aguzzi, J., Puig. P., Company J. B. 2008a. Hydrodynamic, non-photic modulation of biorhythms in the Norway lobster, Nephrops norvegicus (L.). Deep-Sea Research I 56: 366–373.
Aguzzi, J., Sbragaglia, V., Sarriá, D., García, J. A., Costa, C., del Río, J., Mànuel, A., Menestti, P., Sardà, F. 2011b. A new laboratory radio frequency identification (rfid) system for behav-ioural tracking of marine organisms. Sensors-Basel 11: 9532–9548
Aguzzi, J., Sanchez, J., Garcia, J.A., Sardà, F. 2009b. Day-night and depth differences in haemo-lymph melatonin of the Norway lobster, Nephrops norvegicus (L.). Deep-Sea Research I 56: 1894–1905.
Aguzzi, J., Sardà, F. 2008. A history of recent advancements on Nephrops norvegicus behavioural and physiological rhythms. Reviews in Fish Biology and Fisheries 18: 235–248.
Aguzzi, J., Sardà, F., Abelló, P., Company, J. B., Rottlant, G. 2003. Diel patterns of Nephrops norvegicus catchability (Decapoda: Nephropidae) in the western Mediterranean: a perspec-tive through depth and season. Marine Ecology Progress Series 258: 201–211.
Aguzzi, J., Sarriá, D., García, J. A., Del Rio, J., Sardà, F., Manuel, A. 2008b. A new tracking system for the measurement of diel locomotor rhythms in the Norway lobster, Nephrops norvegicus (L.). Journal of Neuroscience Methods 173: 215–224.
Atkinson, R. J. A., Naylor, E. 1976. An endogenous activity rhythm and the rhythmicity of catches of Nephrops norvegicus (L.). J. Exp. Mar. Biol. Ecol. 25:95–108.
Bell, M. C., Redant, F., Tuck, I. 2006. Nephrops species. In: Phillips BF (ed) Lobsters: biology, man-agement, aquaculture and fisheries. Blackwell Publishing, Oxford, pp. 412–461.
Bell, M. C., Elson, J. M., Addison, J. T., Revill, A. S., Bevan, D. 2008. Trawl catch composition in relation to Norway lobster (Nephrops norvegicus L.) abundance on the Farn Deeps grounds, NE England. Fisheries Research 90: 128–137.
Chapman, C. J., Johnstone, A. D. F., Rice, A. L. 1975. The behaviour and ecology of the Norway lobster, Nephrops norvegicus (L.). Proc 9th Europ. Mar. Biol. Symp.: 59–74.
Chiesa, J. J., Aguzzi, J., García, J. A., Sardà, F., de la Iglesia, H. 2010. Light intensity determines temporal niche switching of behavioral activity in deep water Nephrops norvegicus (Crusta-cea: Decapoda). Journal of Biological Rhythms 25: 277–287.
Katoh, E., Sbragaglia, V., Aguzzi, J., Breithaupt, T. 2013. Sensory biology and behaviour of Nephrops norvegicus. Advances in Marine Biology 64: 66–106.
Menesatti, P., Aguzzi, J.,Costa, C., García, J. A., Sardà, F. 2009. A new morphometric imple-mented video-image analysis protocol for the study of social modulation in activity rhythms of marine organisms. Journal of Neuroscience Methods 184: 161–168.
Oakley, S. G. 1979. Diurnal and seasonal changes in the timing of peak catches of Nephrops norvegicus reflecting changes in behaviour. In: Naylor, E., Hartnoll, R. G. (eds) Cyclical phe-nomena in marine plants and animals. Oxford, Pergamon Press, pp 367–373.
Sardà, F., Aguzzi, J. 2012. A review of burrow counting as an alternative to other typical methods of assessment of Norway lobster populations. Reviews in Fish Biology and Fisheries 22: 409–422.
20 | ICES WGNEPS REPORT 2013
Sbragaglia, V., Aguzzi, J., García, J. A., Chiesa, J. J., Angelini, C., Sardà, F. 2013a. Dusk but not dawn burrow emergence rhythms of Nephrops norvegicus (Crustacea: Decapoda). Scientia Marina. In Press (doi: 10.3989/scimar.03902.28C).
Sbragaglia, V., Aguzzi, J., García J. A., Sarriá, D., Gommariz, S., Costa, C., Menesatti, P., Vilaró, M., Manuel, A., Sardá, F. 2013b. An automated multi-flume actograph for the study of be-havioural rhythms of burrowing organisms. Journal of Experimental Marine Biology Ecol-ogy 446 177–185.
Vivien-Roels, B., Pévet, P. 1993. Melatonin: presence and formation in invertebrates. Experientia 49, 642–647.
Wagner, H. J., Kemp, K., Mattheus, U., Priede, I. G. 2007. Rhythms at the bottom of the deep-sea: cyclic current flow changes and melatonin patterns in two species of demersal fish. Deep-Sea Research I 54, 1944–1956.
ICES WGNEPS REPORT 2013 | 21
Annex 1: List of participants
Name Address Phone/Fax E-mail
Jacopo Aguzzi
Marine Science Institute (ICM-CSIC). Paseo Maritimo de la Barceloneta 37–49, 08003 Barcelona Spain.
0034 93 230 95 00
93 230 95 55 (fax)
Dick de Haan Wageningen IMARES (Institute for Marine Resources and Ecosystem Studies) POBox 68 1970 AB IJmuiden
Jennifer Doyle
Marine Institute Rinville, Oranmore Co. Galway Ireland
Phone 353 91 387200 Fax 353 91387201
Jordan P. Feekings
Technical University of Denmark National Institute of Aquatic Resources Nordsøen Forskerpark, Postboks 101 9850 Hirtshals
Direct +45 35883292
Colm Lordan (Chair)
Marine Institute Rinville, Oranmore Co. Galway Ireland
Phone +353 91 387200 Fax +353 91 387201
Ana Leocadio Centre for Environment, Fisheries and Aquaculture Science Lowestoft Laboratory Pakefield Road NR33 0HT Lowestoft Suffolk UK
Tel: 44 (0) 1502524280 Fax: 44 (0) 1502 513865
Alessandro Ligas
Agri-Food and Biosciences Institute (AFBI) 18a Newforge Lane, BT9 5PX Belfast, Northern Ireland (UK) www.afbini.gov.uk
Phone: +44 2890255013
Francisco Sardà
Marine Science Institute (ICM-CSIC). Paseo Maritimo de la Barceloneta 37–49, 08003 Barcelona Spain.
Tel. (+34) 93 230 95 00
Ken Sooknanan
Department of Electronic and Electrical Engineering, Trinity College Dublin, Dublin, Ireland
22 | ICES WGNEPS REPORT 2013
Name Address Phone/Fax E-mail
Yolanda Vila Instituto Español de Oceanografía IEO. Centro Oceanográfico de Cádiz Puerto Pesquero, Muelle de Levante s/n E-11006, Cádiz Spain
Phone +34 956294189 Fax +34 956294232
Cristina Silva IPMA Avenida de Brasilia PT-1449–006 Lisbon Portugal
Phone +351 213027096 Fax +351 213015948
Mats Ulmestrand
Swedish University of Agricultural Sciences Department of Aquatic Resources Institute of Marine Research Turistgatan 5 SE- 453 30 Lysekil Sweden
Phone +46 104784048
Adrian Weetman
Marine Scotland Science Marine Laboratory PO Box 101 AB11 9DB Aberdeen Torry UK
Phone +44 1224 295354
Kai Wieland Technical University of Denmark National Institute of Aquatic Resources Nordsøen Forskerpark, Postboks 101 9850 Hirtshals
Phone +45 35883276
ICES WGNEPS REPORT 2013 | 23
Annex 2: Agenda
Logistical Information
• Meeting times 10:00–18:00 Day 1 (9–18 thereafter) Finishing 13:00 on the 8th • Coffee breaks 11:00–11:15 and 15:45–16:00 • Lunch 13:00–14:00
New ToRs
*** Taking into account that a proportion of Nephrops discards is generally assumed to survive, agree on an appropriate implementation (i.e. calculation) to give catch advice for Nephrops under a discard ban or landing obligation. ***
Presentations given
Nephrops UWTV surveys in the Skagerrak and Kattegat (FU 3–4)
Kai Wieland 1, Mats Ulmestrand 2, Jordan Feekings 1 and Sven Koppetsch 2
1: DTU Aqua, Hirtshals, Denmark
2: SLU IMR, Lysekil, Sweden
Nephrops area definitions in the Skagerrak and Kattegat (FU 3–4)
Jordan Feekings1, Patrik Jonsson2, Kai Wieland1, Mats Ulmestrand2, and Jo-han Lövgren2
1 DTU Aqua, Hirtshals, Denmark
2 SLU IMR, Lysekil, Sweden
The results of Nephrops Trawl surveys in Portugal FU 28–29
Cristina Silva, IPMA, Lisbon
Developments on MSS UWTV surveys in 2013
Adrian Weetman, Marine Scotland Science, Aberdeen.
New technological applications to track the behaviour of animals in the laboratory and in the field.
Jacopo Aguzzi
Changes to coverage and sampling intensity for Irish UWTV surveys.
Jennifer Doyle and Colm Lordan, Marine Institute, Ireland.
Content Analysis of Underwater Survey Videos of Nephrops Habitats
Ken Sooknanan, Naomi Harte, David Corrigan, Anil Kokaram, Jim Wilson. Sigmedia, Trinity College Dublin, Ireland.
Long-term Management Plan; Bbuffer definition
Ewen Bell, Cefas, UK (via Skype)
Separable Cohort Analysis update
Ewen Bell, Cefas, UK (via Skype)
24 | ICES WGNEPS REPORT 2013
ICES Working Group on Nephrops Surveys (WGNEPS); an open lecture to Marine Sci-ence Institute (ICM-CSIC) staff.
Colm Lordan, Marine Institute, Ireland.
ICES WGNEPS REPORT 2013 | 25
Annex 3: WGNEPS terms of reference for the next meeting
2012/MA2/SSGESST15 The Study Group on Nephrops Surveys (SGNEPS) will be renamed the Working Group on Nephrops Surveys (WGNEPS), chaired by Colm Lordan, Ire-land, will meet in Barcelona, Spain, 5–8 November 2013 to work on ToRs and generate deliverables as listed in the Table below.
The second interim meeting of the Working Group on the Nephrops Surveys (WGNEPS), chaired by Colm Lordan, Ireland will meet in Lisbon, Portugal 4–6 No-vember 2014.
WGNEPS will report on the activities of 2013 (the first year) by 6 December 2013 and (second year) by 6 December 2014 to SSGIEOM.
ToR descriptors
TOR DESCRIPTION
BACKGROUND
SCIENCE PLAN
TOPICS
ADDRESSED DURATION
EXPECTED
DELIVERABLES
a To review any changes to design, coverage and equip-ment for the various Nephrops UWTV sur-veys.
To ensure surveys used by WKNEPH, WGCSE, WGNSSK are fit for purpose.
Recurrent annual up-date
Survey summary including and de-scription of alter-ations to the plan, to relevant as-sessment-WGs (WKNEPH, WGCSE, WGNSSK,) and SCICOM. Plan-ning of the up-coming surveys for the survey co-ordinators and cruise leaders.
b To ensure common approaches to proto-cols, quality control quality and assurance of UWTV data and make recommenda-tions in relation to standard operating procedures as neces-sary.
There is a need for survey protocols to be documented in the Series of ICES Survey Protocols (SISP).
2 years Manual for Nephrops UWTV Surveys
c To draft methods paper on the use of Nephrops UWTV surveys for the provision of managment advice.
UWTV surveys have become the main bais of managment advice for Nephrops stocks in ICES. There is a need to formally document the approach.
Year 1 Paper in primary literature or ICES CRR
d To review the WGISUR reports and discuss the utility of Nephrops UWTV surveys as platforms
Nephrops UWTV surveys have a role in relation to benthic habbitat monitoring and the collection of
Year 3 To update the SISP based on conslusions
26 | ICES WGNEPS REPORT 2013
for the collection of data for OSPAR and MFSD indicators.
other environmental and ecosystem variables.
e To review video enhancement, video mosaicking, automatic burrow detection and other new technological developments
WGNEPS should periodically review emerging technologies that might improve survey methodologies
Year 3 Technical paper on new technologies and update SISP as appropriate
g To review the design, coverage, results and uses of Nephrops trawl surveys in consultation with WGISDAA
There are trawl surveys for Nephrops in some area and trawling activity also takes place with UWTV surveys. These activities need rewiew and coordination.
Recurrent annual up-date
Survey summary including and description of alterations to the plan, to relevant assessment-WGs (WKNEPH, WGCSE, WGNSSK, WGHMM,) and SCICOM. Planning of the upcoming surveys for the survey coordinators and cruise leaders.
Summary of the Work Plan
Year 1 The main task will be to draft a CRR or review paper in the use of UWTV surveys as the primary basis of managment advice for Nephrops stocks. ToR A and G will be also addresed annually and plans for ToR B will be made.
Year 2 ToR B will be the main output, work for ToR D and E will be planned.
Year 3 Work will forcus on ToRs D and E as well as reviweing any changes in ToR A.
“Supporting information
Priority Nephrops are a valuable species whose stocks are potentially sucseptible to local depletion. UWTV surveys are an integral part of the stock assessment and management managment advice provided by ICES. WGNEPS is the international coordination group for Nephrops surveys focusing on planning, coloboration, quality control and survey development issues. This work is considered high priority.
Resource requirements
The research programmes which provide the main input to this group are already underway, and resources are already committed. The additional resource required to undertake additional activities in the framework of this group is negligible.
Participants The Group is normally attended by some 15–20 members and guests.
Secretariat facilities None.
Financial No financial implications.
Linkages to ACOM and groups under ACOM
This group will feed into the assessment working groups and subsequently on to ACOM
ICES WGNEPS REPORT 2013 | 27
Linkages to other committees or groups
There is a very close working relationship with all the groups of WKNEPH. It is also very relevant to stock assessment experts groups that used the survey results i.e. WGCSE and WGNSSK. Links with WGISUR to address ToR D and WGISDAA to address Tor G..
Linkages to other organizations
None
Annex 4: Recommendations
Recommendation Adressed to
1. WGNEPS propose that an UWTV Database to should be developed make the data avialable.
WGNEPS and ICES Data Centre
2. To explore the use of UWTV sleds and laboratory to investigate the effects of pulse fishing on Nephrops.
WGELECTA
3. A dedicated research project on Nephrops UWTV assumptions and uncertainties should be proposed for funding to the European Commission.
Regional Coordination Meeting of the DCF
4. SGNEPS recommends that survey coverage be expanded to other important fisheries not currently assessed e.g. Horns Reef FU33, Bottney Gut FU5 and Gulf of Cadiz FU30. Some of the IBTS surveys may be a suitable platform to do UWTV stations at night or with additional days.
Regional Coordination Meeting of the DCF, IBTS