Fisheries science JSFS , tập 78, số 1, 2012 1

Fish Sci (2012) 78:1–14 DOI 10.1007/s12562-011-0413-0 ORIGINAL ARTICLE Fisheries Recall bias in recreational summer flounder party boat trips and angler preferences to new approaches to bag and size limits Eleanor A Bochenek • Eric N Powell John DePersenaire • Received: 13 December 2010 / Accepted: 12 July 2011 / Published online: 27 October 2011 Ó The Japanese Society of Fisheries Science 2011 Abstract Three innovative approaches to bag and size limits were evaluated in the recreational summer flounder Paralichthys dentatus fishery Each approach was designed to reduce discard mortality while increasing angler satisfaction, yet still limiting recreational take within management goals Each was compared to the 2006 legal bag and size limits on party boat trips from New Jersey and New York Angler-specific catch data were collected during the trips, and anglers completed a questionnaire while sailing back to port Comparison of questionnaires to observer records revealed that anglers could not accurately recall the number of fish kept or released Anglers overestimated both kept and discarded fish by a factor of about two Neither fishing scenario, age, sex, nor years fished significantly influenced the accuracy of survey reports of kept fish Anglers on three of five boats over-reported landings Reported landings were nearly accurate on two boats Survey accuracy for reported discards was influenced by bag-and-size-limit scenario and differed among boats, sexes, and fishing experience, but no predictable pattern was evident In particular, bias in reporting was unrelated to angler sex, age, experience, and performance on observed trips or any other criterion measured in this study Anglers E A Bochenek (&) Á E N Powell Haskin Shellfish Research Laboratory, Institute of Marine and Coastal Sciences, Rutgers, The State University of New Jersey, 6959 Miller Ave., Port Norris, NJ 08349-3167, USA e-mail: bochenek@hsrl.rutgers.edu E N Powell e-mail: eric@hsrl.rutgers.edu J DePersenaire Recreational Fishing Alliance, PO Box 3080, New Gretna, NJ 08224, USA e-mail: jdepersenaire@joinrfa.org preferred the slot limit most and the 2006 legal bag and size limit least High grading and transfer of fish among anglers were rare occurrences Our study demonstrated that the summer flounder fishery is a consumptive fishery Keywords Angler recall Á Bag and size limits Á For-hire Á Summer flounder Introduction Summer flounder Paralichthys dentatus supports important commercial and recreational fisheries along the northeast coast of the United States This species is readily accessible to recreational anglers In 2006, summer flounder ranked fourth in total number of recreationally caught fish in the Atlantic and Gulf region [1] Of the top recreational species encountered in 2006, summer flounder ranked third in number of fish released but was not one of the top five harvested species [1]; thus the discard-to-landings ratio was extremely high By the late 1980s, summer flounder was severely overfished [2, 3] Consequently a stock-rebuilding program was initiated in the early 1990s Spawning stock biomass returned to near historically high levels by 2004 [3] The recreational fishery is managed through an annual harvest limit computed as landings plus 10% of discards based on the estimated discard mortality rate [2, 3] As the summer flounder stock rebuilt, more older, heavier fish became available to anglers Responsive harvest controls included shortened fishing seasons, increased minimum legal sizes, and reduced bag limits These efforts to constrain landings during successful stock rebuilding resulted in landings of larger, predominately female fish and increased regulatory discards because of the greater availability of fish smaller than the legal size limit in the rebuilding stock with a 123 corresponding increase in discard mortality The steady rise in discards coupled with the increasing minimum size limits effectively reduced the realized number of fish anglers could keep because total allowable catch includes both landings and dead discards Three alternative approaches to bag and size limits in the recreational summer flounder party boat fishery were evaluated in 2006 These were designed to reduce discard mortality while increasing angler satisfaction within constraints that would retain catch within management goals Bochenek et al [4] described the impact of each alternate bag-and-size-limit scenario on discards, landings, and the discard-to-catch ratio Powell et al [5] evaluated the health of discarded fish relative to the presumed 10% mortality rate [3] Inaccuracy in angler recall and nonresponses to angler surveys can bias recreational catch and effort data [6–11] Chase and Harada [12] noted that reducing the time between the event and the reporting of the event could possibly reduce the impact of recall bias Little information on recall bias is available for the summer flounder fishery The objectives of this study were to (1) evaluate angler Fish Sci (2012) 78:1–14 recall for the number of kept and discarded summer flounder on directed summer flounder party boat trips and (2) conduct an angler survey to determine their preferences for bag-and-size limit alternatives, reasons for discarding summer flounder, and whether they high grade Materials and methods Experimental design Five party boats fishing for summer flounder during the 2006 fishing season were selected for this study These boats encompassed a range of vessel sizes and areas fished along the coasts of New York (southwest coast) and New Jersey, with three vessels with homeports in New Jersey (NJ) and two vessels with homeports in New York (NY) Vessels ranged in size from 50 to 90 ft; angler capacity ranged from 50 to 131 anglers Locations fished included near-coast state waters, offshore federal waters, and bays and estuaries (Fig 1) Detailed descriptions of the party Fig Fishing trips by party boat Dots indicate the locations of each observed drift for the five party boats (A–E) 123 Fish Sci (2012) 78:1–14 boats, areas fished, and number of trips are provided in Bochenek et al [4] Four bag-and-size-limit scenarios regulating catch were compared on these trips: (1) the 2006 state-specified regulations (control scenario) that had a bag limit of eight fish with a minimum length of 16.500 (41.91 cm) for New Jersey and four fish with a legal size limit of C1800 (45.72 cm) for New York; (2) a reduced-minimum-size limit with a 1400 (35.56 cm) minimum length and the 2006 state-specified bag limit (NJ eight fish, NY four fish; (3) a slot limit in which anglers were allowed to keep two fish between 1400 and the state-specified minimum size limit with the remaining kept fish being greater than or equal to the statespecified minimum size limit, with the state-specified bag limit enforced; and (4) a cumulative size limit with kept fish C1400 set by conflating the state-specified size and bag limit to produce a cumulative number of inches that could be harvested, determined for, e.g., New Jersey, by conflating the 2006 legal bag limit and minimum size as fish 16.500 = 13200 Field sampling occurred only on weekdays (Monday– Thursday) during June through September with the fishing season split into an early (June to mid-July), mid (end of July to mid-August), and late (end of August to mid-September) season Each boat was sampled once over a consecutive 4-day period in each season The bag-and-sizelimit scenario was randomly selected without replacement for each 4-day period such that one scenario was sampled per day and such that each scenario was fished once by each boat in each season Most anglers were unaware of the change in regulation until they boarded the boat, although a few may have gleaned information from the boat owner/ operator prior to the trip Each angler was given a numbered tag and a one-page flier describing the current day’s fishing scenario One or two observers were present on each trip and collected angler-specific data on kept and discarded summer flounder as the fish were caught [4] At the end of the trip, observers gave each angler C14 years of age a one-page questionnaire and asked them to complete and return the survey to a marked box prior to leaving the boat Some younger anglers also completed the survey with assistance from their parents Fishermen remained anonymous, except for a numbered tag given to each angler at the inception of the trip used to relate observer data to the respective questionnaire Anglers were requested to record the unique tag number; number of summer flounder kept and discarded on the current trip; their gender, age, and fishing experience (years fished); and information pertaining to the angler’s preference for bag and size limit fished, reasons for releasing fish, and whether they high graded, that is whether they released otherwise legal fish that were deemed to be of lower quality Statistical analysis The difference in summer flounder kept or discarded between that observed during the trip (observed kept/discarded) and that reported on the angler survey (reported kept/discarded) was calculated as: D kept ¼ observed kept À reported kept and D discard ¼ observed discard À reported discard To standardize the degree of over- or under-reporting of kept or discarded summer flounder, the reporting accuracy (RA) was compared to the observed value For the fish kept or discarded: RA(kept) ¼ D kept=observed kept RA(discarded) ¼ D discard=observed discard: For perfect accuracy, RA = If reports were exactly twice the observed value, RA = -1.0 Thus the factor difference between the observed and reported values can be obtained as FDiff ¼ abs(RA À 1Þ: Reporting accuracy (RA) was not normally distributed as indicated by significant Kolmogarov-Smirnov one-sample tests (a B 0.05) Accordingly, nonparametric statistics were used to assess the degree of recall bias in angler reports of fish kept or discarded Wilcoxon signed-rank tests were used to test this expectation for kept and discarded fish across the entire study and, as well, by boat, bag-and-size-limit scenario, angler sex, angler age, and angler experience measured as years fished Angler age was assigned to a series of age groups for analysis identified subsequently as = anglers aged C70 years, = anglers aged 60–69, = anglers aged 50–59, = anglers aged 40–49, = anglers aged 30–39, = anglers aged 20–29, = anglers aged 10–19, and = anglers aged \10 Angler experience was binned into a series of fishing experience groups identified subsequently as 12y = years fished C50, 11y = years fished C45 to \50, 10y = years fished C40 to \45, 9y = years fished C35 to \40, 8y = years fished C30 to \35, 7y = years fished C25 to \30, 6y = years fished C20 to\25, 5y = years fished C15 to \20, 4y = years fished C10 to \15, 3y = years fished C5 to \10, 2y = years fished C2 to \5, and 1y = years fished \2 The tendency for anglers to over-report or under-report was assessed using a sign test This test merely compares the number of times anglers over-reported or underreported fish regardless of the degree of error Nonstandardized (D kept, D discarded) and standardized [RA(kept), 123 RA(discarded)] reporting accuracies were compared between boats, fishing scenarios, angler age groups, fishing experience levels, and angler sexes using Kruskal-Wallis tests Correspondence analysis [13] was used to visualize the interrelationship of boat, bag-and-size-limit scenario, angler age group, fishing experience group, and angler sex with angler survey responses Correspondence analysis is a data-reduction technique that permits evaluation of relationships within categorical datasets [14, 15] and is analogous to principal components analysis for continuous or meristic data [16] Inputs to the correspondence analysis included the following: survey responses that pertain to the angler’s preference for keeping and releasing legal summer flounder, survey responses that pertained to why anglers released some summer flounder that they could have kept legally, and survey responses that pertain to their preference for the three experimental bag and size limits and the control Additional variables included were angler age group, fishing experience group, and sex Supplementary variables positioned on the axes were boat and bag-andsize-limit scenario Factor loads for each variable for the first 10 dimensions resolved by correspondence analysis were used as variables describing each angler category and survey response in cluster analysis Response and angler category variables were clustered using an unweighted pair-group algorithm with Euclidean distance as the similarity index [17] To investigate possible reasons for angler bias in the reporting of fish kept and discarded, we added to the list of supplementary variables in correspondence analysis a series of variables describing the degree of angler bias These included whether the angler over-reported or under-reported landings and/or discards and whether the bias in reporting fell in the upper or lower 25% of all respondents in each regard We also distinguished anglers that landed fish from those that did not catch any legal-size fish Angler bias measured as the raw difference between observed and reported landings and discards and standardized to total angler catch was included Results Survey statistics In 2006, 76 summer flounder fishing trips (full day, morning half-day, afternoon half-day) were sampled on five party boats A total of 1,860 anglers fished on these boats and 1,090 anglers completed the survey (58.6%) Of these, 49 were discarded because of inaccurate trip or angler designations Thus, our study relies on 1,051 angler surveys 123 Fish Sci (2012) 78:1–14 The number of completed angler surveys was unevenly distributed among the three bag-and-size-limit scenarios and the control (Table 1) The fewest number of surveys was completed for control trips (19.5%) and the greatest number for reduced-minimum-size trips (30.6%) (Table 1) More surveys were completed on morning trips (N = 508) than on afternoon (N = 367) and full-day trips (N = 215) To a large extent, these differentials reflected the differential in the number of anglers participating in morning and afternoon trips and the tendency for observed vessels to carry out half-day rather than full-day trips [4] Anglers that completed surveys were predominately males (83.6%) and angler mean age was 47.6 years and ranged from to 85 We had asked that participants C14 years of age complete the survey However, 4.7% of respondents were younger than 14 and some parents assisted these younger anglers in completing the survey Other young anglers filled out their own questionnaire These data were included in the analysis Anglers had fished on average 27.7 years and ranged from new anglers with years of experience to anglers with 75 years of experience encompassing a broad range of angler experience from novices to the very experienced (Table 1) A total of 156 anglers reported that they did not catch a summer flounder Anglers reported keeping 2,108 summer flounder, an average of 2.0 fish per angler (Table 2) and discarding 3,676 summer flounder, an average of 3.5 fish discarded per angler Fishermen also disclosed that they discarded 3,297 fish below the minimum size, a per-angler average of 3.2 fish In extremum, one angler reported 50 discarded fish, all below the legal minimum size, and that all discards were dead Observers on this boat did not see one dead fish discarded nor any single angler catching 50 summer flounder When we asked anglers if they ever high graded, 169 anglers (15.5%) answered affirmatively, but only 3.4% reported high grading during the previous year (the 2005 fishing season) Fishermen were also questioned as to whether they gave fish away on the boat instead of high grading: 146 fishermen answered affirmatively, but only 2.4% did so during the previous year (the 2005 fishing season) (Table 2) Anglers were asked how many fishing trips they made on party boats, private boats, and from the shore, bank, or jetty in the previous fishing year (2005) Fishermen participating in this study averaged approximately eight trips on party boats, four trips on private boats, and three outings from the shore, bank, or jetty (Table 1) Angler preferences/descriptive statistics Anglers were asked to identify the bag-and-size-limit scenario under which they fished for that trip Only nine Boat A = 215 Boat B = 347 Boat C = 265 Boat D = 153 Slot limit = 231 Reduced size = 333 Cumulative size = 313 Control = 213 Full day = 215 PM = 367 AM = 508 No of surveys by time-of-day Unknown = Female = 178 Male = 911 Angler sex Range 6–85 Mean (SD) = 47.6 (17.66) Angler age (years) Range 0–75 Mean (SD) = 27.7 (19.27) Experience (years saltwater fished) Range 0–200 Mean (SD) = 8.1 (13.98) Trips on party boat in 2005 Range 0–150 Mean (SD) = 3.6 (9.24) Trips on private boat in 2005 Range 0–300 Mean (SD) = 3.4 (14.38) Trips on land in 2005 Mean (SD) = 3.4 (14.39) Range 0–100 Yes = 169 No = 897 No = 921 Yes = 146 Gave fish away ever Range 0–100 Mean (SD) = 2.4 (11.99) Gave fish away 2005 (% trips) N = 2,108 fish Range 0–15 Mean (SD) = 2.0 (2.09) Number kept NA Not answered N = 3,676 fish Range 0–50 Mean (SD) = 3.5 (5.59) Number discarded Angler estimates, this study only Anglers were also asked to record the number of summer flounder kept, discarded, and discarded below the minimum size NA = 24 High grade in 2005 (% trips) High grade ever N = 3,297 fish Range 0–50 Mean (SD) = 3.2 (5.35) Number discarded below size Table Angler responses to survey questions concerning whether they ever high graded while fishing for summer flounder, the percentage of trips on which the angler high graded while fishing for summer flounder during the 2005 fishing season, whether the angler ever gave some summer flounder away while fishing for summer flounder instead of high grading, and the percentage of trips the angler gave summer flounder away to another angler during the 2005 fishing season Boat E = 110 No of surveys by boat No of surveys by scenario Table Angler responses to survey questions pertaining to their current summer flounder fishing trip and the number of fishing trips taken in 2005 (the year prior to our study) on party boats, private boats, and from the shore, bank, or jetty (land) Fish Sci (2012) 78:1–14 123 Fish Sci (2012) 78:1–14 cumulative-size scenario, but more anglers indicated dissatisfaction with this scenario (*26%) than for the slotlimit and reduced-minimum-size scenarios (Table 4) Increased dissatisfaction by anglers for the cumulative-size scenario may come from the greater difficulty for anglers to keep track of their total inches caught More anglers expressed dissatisfaction with the control scenario than any of the alternatives Fishermen were asked about their predilections for keeping and releasing summer flounder that they catch Six hundred and fifteen anglers preferred to keep all legally allowed summer flounder, 246 anglers preferred to keep most of the summer flounder legally allowed, 49 anglers preferred to release most of the legally allowed summer flounder, and 17 anglers would release all legally allowed summer flounder These preferences support the generally held view that the summer flounder fishery is predominately a consumptive fishery, since a majority (66.3%) of the anglers would rather keep all legal summer flounder whereas only 1.8% of the anglers practice true conservation by releasing all summer flounder that could have been legally kept Forty-two percent responded that the question does not apply to them Anglers were queried as to why they released some of the summer flounder that could have been kept legally (Table 5) Only 16.9% (N = 126) responded that they release legally allowed summer flounder because they not consume them, whereas 63.6% (N = 473) of anglers anglers (0.8%) answered incorrectly Anglers were also asked to rate their satisfaction with the bag-and-size-limit scenario under which they fished that day on a five-point scale Most anglers rated the slot-limit scenario as highly preferred (56.7%), and 4.6% of the anglers ranked this scenario as not preferred (Table 3) For the cumulative-size scenario, 46.5% of the anglers ranked this scenario as highly preferred, and only 5.1% did not prefer this scenario The rankings for the reduced-minimum-size scenario were similar to the cumulative-size scenario In contrast, only 28.9% of the anglers highly preferred the control scenario, and 22.9% of the anglers ranked this fishing scenario ‘‘not preferred’’ (Table 3) The mean rank for the control, slot-limit, reduced-minimum-size, and cumulativesize scenarios was 3.2, 4.2, 4.1, and 4.0, respectively, with a indicating ‘‘highly preferred.’’ Fishermen were queried as to their relative preferences for the three alternative bag-and-size-limit scenarios and the control (Table 4) In this case, anglers were asked to respond concerning the desirability of each alternative as a future management scenario, even though they had participated in only one of the four alternatives For the 2006 state-specified bag and size limits (control scenario), about 32% of the anglers were satisfied and 32% not satisfied with this scenario In contrast, approximately 43% of the anglers favored the reduced-minimum-size scenario and about 47% of the anglers favored the slot-limit scenario Approximately, 33% of the anglers preferred the Table Angler responses to their preference for or satisfaction with the bag-and-size-limit scenario fished during their trip Scenario Slot limit Rank 1: ‘‘do not prefer’’ Rank Rank Rank Rank 5: ‘‘highly prefer’’ 25 41 110 Cumulative size 13 64 52 119 Reduced size 20 12 48 52 159 Control 38 11 47 22 48 Preference was ranked from to with equal to ‘‘do not prefer’’ and equal to ‘‘highly prefer’’ No explicit alternative was provided Table Angler responses for each question pertaining to the management of summer flounder and to the bag-and-size-limit options they would prefer Questions Agree Somewhat agree Somewhat disagree Disagree Don’t know Current legal size and bag limit for summer flounder in your state: NY fish at 1800 , NJ fish at 16.500 (control scenario) 280 165 122 288 33 Retain summer flounder C1400 inches and keep your state’s current bag limit: NY fish, NJ fish (reduced-minimum-size scenario) 379 210 102 145 38 A slot limit for summer flounder where fish can be kept between 1400 and your state’s current legal size limit and bag limit: NY fish, NJ fish (slot-limit scenario) 412 207 82 118 57 Fish for summer flounder under a cumulative total size limit (e.g., NJ: total size limit of 13200 , NY: total size limit of 7200 and with a minimum size limit of 1400 ) (cumulativesize scenario) 288 183 93 222 81 Does not include no-response answers 123 Fish Sci (2012) 78:1–14 Table Angler responses for questions pertaining to reasons why they release some of their summer flounder that they could have kept legally Questions Agree You don’t eat summer flounder 126 Somewhat agree Somewhat disagree 40 61 Disagree Don’t know 473 44 Some summer flounder were too small to keep 527 108 39 66 37 You caught what you want to eat 423 116 58 143 34 Your conservation ethic 340 145 57 124 75 You like to catch and release summer flounder 152 158 110 299 42 Does not include no-response answers disagreed with this reason About 68% of the anglers responded that they release these fish because some of the summer flounder were too small to keep and about 55% released fish because they caught what they wanted to eat Approximately 46% of the anglers responded that their conservation ethic was the reason for release of some fish However, in response to the direct question concerning their desire to carry out a catch-and-release fishing experience, only about 20% agreed or somewhat agreed and about 39% disagreed These responses also support the dominance of consumptive fishing as the prime motivating force for angler participation in this fishery Observed versus reported summer flounder catch The mean difference between summer flounder observed kept and reported kept (D kept) was -0.370, indicating that anglers reported landing more fish than they actually landed per trip The mean standardized accuracy of reports [RA(kept)] was -0.800 indicating that the reported number of kept fish exceeded the observed kept by a factor of 1.800 For discarded summer flounder, the mean difference (D discard) was -1.600 indicating that anglers also overestimated their discards The mean standardized accuracy was -1.423 Anglers tended to overestimate their summer flounder discards by about a factor of 2.423 The accuracy of angler reports for kept and discarded fish differed significantly among boats, but only among bag-and-size-limit scenario for kept fish (Table 6, KruskalWallis test) Angler age did not influence reporting accuracy overall, and angler sex influenced only the accuracy of reported discards with males being more biased with a mean D discard = -1.81, whereas for females the mean D discard = -0.56 Fishing experience influenced only the accuracy of reported landings (Table 6) Significant results not imply inaccurate reporting, however; they merely indicate that reporting accuracy varied between main-effect categories To further investigate the degree of accuracy within main-effect groups, we evaluated each category separately with tests designed to identify bias The nonstandardized differences in observed Table Results of Kruskal-Wallis tests comparing the differences in the observed and angler-reported kept and discarded summer flounder by boat, bag-and-size-limit scenario, age group, sex, and angler fishing experience D Kept D Discard Bag-and-size-limit scenario 0.0001 NS Boat 0.0311 \0.0001 Age NS NS Sex NS 0.0001 Fishing experience 0.0114 NS NS Not significant at a = 0.05 and reported discards were significantly different from zero for all bag-and-size-limit scenarios, boats, angler age groups except for the very youngest (age 1), sexes, and fishing experience groups (Table 7) That is, in no case was reporting accuracy unbiased relative to the true (observed) value For the three alternative bag-and-size-limit scenarios, discards were over-reported with the degree of overestimate ranging from a factor of 1.41 fish for the cumulative-size scenario to 1.61 fish for the reduced-minimum-size scenario In contrast, discards were overestimated by 2.04 fish for the control scenario Male anglers overestimated discards by a factor of 1.81 fish; female anglers were somewhat more accurate, overestimating discards by 0.56 fish (Table 8) Reporting was consistently better for kept fish than for discarded fish For kept fish, the difference between observed and reported fish was significantly different for three of five boats, all bag-and-size-limit scenarios but the slot limit, both sexes, five of eight angler age groups, and of 12 fishing experience groups (Table 7) Less experienced anglers tended to over-report landings more than anglers with greater experience Younger fishermen tended to report more accurately than older fishermen In order to standardize the level of bias in the reporting of kept and discarded summer flounder, we computed the standardized reporting accuracy (RA) (Table 8) The highest over-reporting of kept fish occurred for the control scenario at 3.922 Lesser reporting bias occurred for kept 123 Fish Sci (2012) 78:1–14 Table Results of sign tests (D kept, D discard) and Wilcoxon signed-rank tests [RA(kept), RA(discard)] to evaluate the differences in the observed and reported kept and discarded summer flounder by boat, bag-and-size-limit scenario, age, sex, and fishing experience Nonstandardized Standardized D Kept RA(kept) D Discard Table The factor difference in reporting accuracy for those categories in which the significance of reporting accuracy was evaluated in Table Median Mean Kept Discard Kept Discard Control -0.262 -0.333 -2.922 -1.318 Reduced min-size -0.190 -0.664 -0.449 -2.090 RA(discard) Scenarios Control (0.0066) (\0.0001) NS (0.0371) Cumulative size -0.293 -0.653 -0.403 -0.801 Reduced size (\0.0001) (\0.0001) (0.0014) (0.0001) Slot limit -0.426 -0.667 -0.426 -1.319 Cumulative size (\0.0001) Boat A -0.111 -0.688 -0.011 -1.031 Boat B -0.450 -0.629 -2.082 -1.941 Boat C -0.195 -1.296 -0.178 -1.289 Boat D -0.118 -0.257 -0.438 -1.899 Boat E -0.516 Slot limit NS (\0.0001) (\0.0001) (\0.0001) (0.0232) (\0.0001) (0.0002) Vessels Boat A Boat B NS (\0.0001) (\0.0001) (\0.0001) NS (\0.0001) (0.0039) (\0.0001) Boat C (\0.0001) (\0.0001) (0.0049) (0.0005) Boat D NS (\0.0001) NS (0.0289) Boat E (0.0013) (\0.0001) (0.0098) (0.0059) Age groups Age NS NS NS NS Age (\0.0001) (\0.0001) (\0.0001) (0.0035) Age Age Age Age Age Age NS NS (0.0145) (0.0080) (0.0003) (0.0070) (\0.0001) (\0.0001) (\0.0001) (\0.0001) (\0.0001) (0.0009) NS NS NS 0.0405 NS (0.0176) (0.0005) (0.0002) (\0.0001) (\0.0001) (\0.0001) (0.0057) Sexes Female (0.0003) (0.0016) (0.0173) (\0.0238) Male (\0.0001) (\0.0001) (\0.0001) (\0.0001) Experience groups 12y 11y 10y (0.0002) NS (0.0076) (\0.0001) (0.0002) (0.0058) NS (\0.0001) (0.0049) (\0.0001) NS (\0.0001) 9y NS (\0.0001) NS (0.0016) 8y (0.0034) (\0.0001) (0.0369) (\0.0001) 7y NS (0.0241) NS (0.0305) 6y NS (\0.0001) NS (0.0039) 5y 4y NS (0.0005) (0.0026) (0.0029) NS 0.0105 (0.0098) (0.0521) 3y (0.0009) (\0.0001) 0.0058 (\0.0001) 2y (0.0005) (\0.0001) NS (0.0077) 1y NS (0.0114) NS (0.0195) (0.0001) (0.0001) (0.0001) (0.0001) Overall statistics NS Not significant at a = 0.05 Abbreviations for angler age and years fished are presented in the ‘‘Materials and methods’’ section Parentheses indicate negative values fish for the alternative scenarios that ranged from the cumulative-size scenario for which anglers reported 1.403 more kept fish than observed to the reduced-minimum-size 123 -0.174 -0.600 -0.323 Age 0.000 -0.667 -0.607 -0.524 Age -1.000 -0.333 -0.957 -1.262 Age 0.000 -1.000 -0.511 -1.110 Age 0.000 -0.208 -0.228 -1.441 Age Age 0.000 0.000 -0.500 -0.625 -0.231 -0.325 -1.475 -1.945 Age 0.000 -0.547 -0.139 -1.005 Age 0.000 -0.500 -0.216 -1.235 Female 0.000 0.000 -0.210 -0.410 -0.167 -0.643 -0.861 -1.666 12y 0.000 -0.500 -0.212 -1.369 11y -0.333 -1.000 -0.154 -1.527 10y 0.000 -0.462 -0.183 -0.752 9y 0.000 -0.500 0.022 -0.746 8y 0.000 -0.414 -0.487 -1.610 Male 7y 0.000 -0.417 -0.017 -0.838 6y 0.000 -0.333 0.033 -1.213 5y 0.000 0.000 -0.056 -0.658 4y -0.111 0.000 -0.406 -0.478 3y 2y -0.250 0.000 -0.750 -0.667 -0.811 -0.918 -1.752 -0.539 0.000 -0.100 -0.560 -1.170 -0.174 -0.614 -0.800 -1.423 1y Overall values Reported are median and mean values for RA(kept) and RA(discard) Perfect accuracy occurs when RA() = The factor difference between the observed and reported value of kept or discarded fish can be obtained as |(RA - 1)| scenario for which anglers reported 1.45 more kept fish then observed Under-reporting occurred for some anglers, but only in two cases (fishing experience) was the mean RA(kept) value positive Discards were over-reported by anglers fishing under the reduced-minimum-size scenario by a factor of 3.090 In contrast, least bias occurred in the cumulative-size scenario in which anglers over-reported discards by a factor of 1.801 The value for the control (2006 legal) scenario was 2.318 Under-reporting occurred for some anglers, but in no case was the mean RA(discard) Fish Sci (2012) 78:1–14 value positive For boats, RA values for kept fish ranged from 1.011 to 3.082 for boats A and B, respectively For discards, RA ranged from 1.516 (boat E) to 2.941 (boat B) Female anglers tended to over-report kept and discarded fish less than male anglers (Table 8) RA fell between about and for kept fish for all angler age groups; the range for discards was 1.524–2.945 Regardless of the level of fishing experience, all anglers tended to over-report kept and discarded fish except for kept fish for anglers that had fished C20 to \25 years (6y) and C35 to \40 years (9y) who under-reported their catch RA for kept fish ranged from about to excluding the anglers that under-reported their kept fish; the range for discarded fish was about 1.5–2.8 (Table 8) Correspondence/cluster analysis Angler characteristics and survey responses were used to define axes in correspondence analysis and the degree to which any vessel or bag-and-size-limit scenario fell preferentially on one of the assessed characteristics Survey responses were little influenced by vessel or bag-and-sizelimit scenario overall nor did biases exist in the distribution of anglers by age, fishing experience, or sex between vessels and trips None of the vessels or bag-and-size-limit scenarios received a factor loading score [ |0.22| on any of the first 10 dimensions of the correspondence analysis Thus, most angler reports and angler descriptors (e.g., age, years fished) were relatively randomly distributed among boats and bag-and-size-limit scenarios To examine the interaction of survey responses and angler descriptive variables, the various responses and descriptive variables were clustered using the factor loading scores for the first 10 dimensions of the correspondence analysis to define between-variable and/or between-response similarity This analysis revealed the following interactions (Table 9) (1) Young anglers \20 years old and anglers of limited fishing experience (2–15 years) provided indecisive answers as to the reason for discarding summer flounder Most ‘‘somewhat disagreed’’ with the range of options provided for why they release some summer flounder from the survey (Table 9) Separately clustered were anglers with 15–20 years of experience These anglers were associated with ambivalent feelings about the bag-and-size-limit alternatives, recording the opinion ‘‘somewhat disagree’’ with the control, reduced-minimum-size, and cumulativesize scenarios (Table 5) In total, anglers with relatively little fishing experience tended to be more indecisive about management scenarios and reasons for discarding than older and more experienced anglers (2) Females clustered with anglers 20–30 years of age and anglers with less than years of experience The latter two criteria described most female anglers These anglers typically provided no response to the question concerning their preference for keeping or releasing legal-size fish, although they responded to other questions on the survey (3) A series of anglers responded that they did not know why they discarded fish (Table 4) These anglers were not dominantly male or female, nor were they dominantly in one age or fishing experience group (4) A series of anglers indicated that they did not have an opinion as to their preference for the four studied bag-and-size-limit scenarios (Table 5) These anglers also, interestingly, had no opinion when asked about their preference for landing or releasing legal-size fish Such anglers might be expected to express no opinion as to bag-and-size-limit scenario These anglers were not dominantly male or female, nor were they dominantly in one age or fishing experience group (5) A series of anglers did not respond to survey questions pertaining to why they released some summer flounder that they could have kept and to the management measures they prefer for summer flounder (Tables 4, 5) Failure to respond fell into two distinctive clusters differentiating the two questions, indicating that such responses came from two different groups of anglers These anglers also were not dominantly male or female, nor were they dominantly in one age or fishing experience group (6) Two clusters defined anglers that either fished primarily for consumption or did not The first cluster disagreed with the following reasons as to why they release some summer flounder that they could have kept legally: some summer flounder were too small to keep, they had caught what they wanted to eat, and their conservation ethic (Table 4) The second cluster of anglers agreed with the catch-and-release philosophy, agreeing that they not consume summer flounder and like to catch and release (Table 4) This latter group also expressed a preference for releasing most or all summer flounder caught Separation into two clusters is indicated by some anglers choosing to express the catch-and-release approach in two distinctive ways on the survey form Interestingly, in neither case were these anglers associated with a preference for any bag-andsize-limit scenario, nor were they associated dominantly with either sex, any angler age group, or any angler fishing experience group (7) The oldest anglers, with[50 years of fishing experience and [60 years old, were not associated with any discrete set of survey responses Such anglers responded in a multitude of ways to the survey questionnaire (8) Anglers with 40–50 years of fishing experience indicated discomfort with the three alternative bag-and-size-limit scenarios by checking ‘‘disagree’’ (Table 5) This angler group was unique in this respect and, significantly, was not associated with any predominant response to preference for the control scenario (Table 5) They were simply more likely to accept the status quo without stipulating a clear preference for it (9) An important group of anglers expressed a preference to keep most, but not all, legal-size fish These anglers were ambivalent as to why they discarded fish (Table 4) and 123 192 Table Correlation matrix of eight physicochemical properties and 15 sensory attributes Fish Sci (2012) 78:187–195 Moisture Salt content TTA TVB FAN Viscosity pH aw Sulphury meaty 0.87* -0.94* -0.92* 0.10 -0.81* -0.97* 0.12 0.35 Sweaty 0.85* -0.96* -0.89* 0.19 -0.86* -0.96* 0.13 0.40 Overripe cheese 0.82* -0.98* -0.82* 0.20 -0.87* -0.87* 0.08 0.41 Acid 0.95* -0.75 -0.95* -0.34 -0.62 -0.93* 0.14 0.04 Burnt 0.36 -0.39 -0.34 -0.33 -0.62 -0.26 0.90* 0.84* Fishy -0.17 0.29 0.07 -0.21 0.13 0.89* -0.94* -0.85* -0.09 -0.79 Salty -0.96* 0.94* 0.96* 0.12 Umami -0.86* 0.98* 0.90* -0.17 Sour -0.80 0.80 0.84 Ammonia 0.06 0.45 0.27 -0.87* 0.14 0.32 0.83* 0.98* -0.24 -0.35 0.94* 0.94* -0.29 -0.55 -0.20 0.68 0.89* 0.05 -0.11 -0.25 Bitter -0.97* 0.87* 0.98 0.18 0.77 0.99* -0.23 * P \ 0.05 Greasy mouthfeel -0.98* 0.91* 0.96 0.26 0.80 0.95* -0.29 -0.33 TTA Total titratable activity, TVB total volatile bases, FAN free amino nitrogen, aw water activity Grainy mouthfeel -0.96* 0.84* 0.98 0.25 0.78 0.97* -0.37 -0.34 Fishy aftertaste -0.55 0.22 0.61 0.59 0.35 0.52 -0.68 -0.31 Bitter aftertaste -0.95* 0.95* 0.96 0.04 0.85 0.99* -0.24 -0.38 and grainy mouthfeel, and bitter aftertaste had a positive correlation with salt content, TTA, FAN, and viscosity and a significant negative correlation with moisture content The result showed that the intensity of sulphury meaty, sweaty, overripe cheese, acid/vinegar, and ammonia was highly influenced by the moisture content of products Meanwhile, the intensity of salty, umami, sour, bitter, and bitter aftertaste correlated well with salt content, TTA, FAN, and viscosity PLS regression was performed to quantify the correlation between physicochemical parameters and sensory attributes Several regression methods are available in Unscrambler, such as PLS1, PLS2, PCR, MLR, and 3-WAY PLS In this research we used PLS2, which relates a group of X-variables to several Y-variables Physicochemical parameters were used as X matrix variables and sensory attributes as Y matrix variables Two PLS2 regression models were run in this research (Fig 2) PLS regression models were conducted based on important sensory attributes from the perspective of commercial trade: sensory taste and sensory odor Physicochemical parameters were included in the model based on hypotheses about the effects of the fermentation process, and the result of fermentation end products as well as data from the Pearson correlation tests (Table 5) Even though TTA and viscosity showed an excellent correlation with most of the sensory attributes mentioned above, they were excluded from PLS-R analysis to avoid misleading results, due to their standard deviation values as can be seen in Table The PLS2 model developed for sensory taste is presented in Fig 2a The loading plot PC1 versus PC2 explained 97% of X matrix and 96% by Y matrix R value between the predicted versus the measured data of sensory taste was 0.98, and the RM-SEC was 0.19 units of 123 measure These values also supported the goodness of taste model In the case of sensory odor, the loading plot PC1 versus PC2 explained 69% of X matrix and 93% of Y matrix The R value of sensory odor was 0.93, and the RM-SEC was 0.28 units of measure (Fig 2b) Note that in the figure, the variables in the Y matrix that correlated well with PCs that had been established previously are indicated In Fig 2a ‘‘Salty, 2’’ indicates that salty correlated well with PC2 In Fig 2b ‘‘Sulphury, 1’’ indicates that sulphury meaty odor correlated well with PC1 Discussion The sensory odor characteristics of LAP and MAP shown herein indicate the existence of volatile compounds due to the degradation of protein and lipids by autolytic and bacterial enzymes during fermentation Various volatiles, including acids, carbonyls, nitrogen-containing compounds, and sulfur-containing compounds are formed during fermentation and believed to be responsible for the distinct odor of fish sauce [3, 19–21] The intensity differed mainly with regard to the amount of salt that was added during the fermentation process Less salt was added to the MAP (17.5% w/w) than LAP (25% w/w) It is known that salt can inhibit microbial activity, thereby affecting the presence of volatile compounds In this study, the product with less salt added (MAP) showed more intensity in sulphury meaty, sweaty, overripe cheese, acid/vinegar, and ammonia odor than LAP (Fig 1a), although the TVB-N contents were high and similar for both samples (Table 3) It can be assumed that the amount of volatile compounds in LAP (higher salt added) was lower than in MAP (lower salt added) This is in agreement with previous studies of Fish Sci (2012) 78:187–195 193 Fig PLS-R plots of predicted versus measured data using full cross-validation a Salty, umami, bitter and bitter aftertaste corresponded with moisture, water activity (aw), salt content, and free amino nitrogen b Odors of sulphury meaty, overripe cheese, sweaty, and ammonia corresponded to moisture, aw, salt content, and total volatile bases MAP Most accepted product, LAP least accepted product Sanceda et al [22] indicating that the amounts of volatile acids in spoiled fish to which salt had been added were significantly lower than those in the control (no salt added) On the other hand, the intensity of salty, umami, sour, bitter taste, greasy mouthfeel, grainy mouthfeel, and bitter aftertaste was stronger in LAP than in MAP The deliciousness of bakasang may be contributed by free amino acids such as glutamic acid, alanin, isoleucine, and lysine [18] The amino nitrogen concentration reflects the amount of primary amino groups in fish sauce [23] Our results imply that LAP (more salt added) might have higher quantities of FAN than MAP does (lower salt added), although FAN levels were not significantly different across the products tested (Table 3) Contrary to our findings, Ijong and Ohta [1] reported that samples containing NaCl 10 g/kg showed higher values of FAN than those containing NaCl at 20 g/kg This contradictory pattern may have been caused by the difference in the period of fermentation LAP had a longer fermentation period than MAP A previous study of Jiang et al [4] indicated that total soluble nitrogen, formaldehyde nitrogen, and trichloroacetic acid (TCA) soluble peptides increased throughout the fermentation period The difference in saltiness can be explained based on the amount of salt added during the fermentation process More salt was added to the LAP (25% w/w) than the MAP (17.5% w/w) This result implies that the higher intensity of salty attributes in the LAP related well with the higher salt concentration that was added by producers during the fermentation process In addition, it is known that both perceived 123 194 saltiness and taste intensity depend on salt content in the product The salt content analysis of the products showed that the salt content of LAP was higher than MAP (Table 3) The grainy mouthfeel attributes might have been related to the viscosity and moisture contents of products The QDA results indicated that LAP (higher salt added) had a more grainy mouthfeel than MAP (lower salt added) It is likely that salt played an important role in protein hydrolysis during the fermentation process and impacted the viscosity of products Reducing salt content in fish sauce may increase the rate of protein breakdown [24] MAP showed lower viscosity than LAP and had higher moisture content than LAP (Table 3) indicating that MAP might contain higher quantities of soluble solid substances from protein breakdown Our study showed that the intensity of bitter aftertaste attributes in LAP was higher than that in MAP Hydrolysis of fish protein results in free amino acids, peptides, and ammonia [4] It is assumed that the difference in bitter aftertaste attributes might be due to the higher quantities of bitter peptide and bitter amino acids in the products, resulting from protein hydrolysis Valine, arginine, histidine, and amino acids with bitter taste impact were found in tuna sauce made from tuna viscera [25] As mentioned earlier, it is likely that salt plays an important role in fish protein breakdown High salt concentration (25%) inhibited peptidase activity and retarded protein hydrolysis [26, 27], suggesting that protein hydrolysis in LAP (higher added salt) was retarded due to the high concentration of salt The higher quantities of peptides and amino acids in MAP compared to LAP might be due to the hydrolysis process, particularly in an earlier period of fermentation It could be speculated that significant amounts of short chain peptides and free amino acids in MAP yielded a favorable taste, masking any bitter aftertaste, although the FAN content of LAP was slightly higher than that of MAP (Table 3) Regarding the physicochemical properties, water activity (aw) is one of the major factors that determines the microbial, chemical, and enzymatic stability of foods [28] The water activity value of our products, around 0.8, showed the possibility of enzymatic activity and microbial activity during storage Most spoilage bacteria, spoilage yeast, and spoilage molds not grow below aw = 0.91, 0.88, and 0.80, respectively [29] However, the microbiological analysis results strongly suggested that bakasang was safe for human consumption (Table 4) The PLS-R plots indicated that a high correlation coefficient was obtained between sensory and physicochemical parameters of bakasang Salty, umami, bitter, and bitter aftertaste showed an excellent correlation to moisture content, aw, salt content, and FAN (Fig 2a) It can be assumed that salt can control the protein hydrolysis (FAN) and influence salty, umami, bitter, and bitter aftertaste It 123 Fish Sci (2012) 78:187–195 has been reported that 60–80% of amino compounds in fish sauce are amino acids [21, 30] Amino acids contributed significantly to the taste of fish sauce Ijong and Ohta [1] reported that traditional bakasang contains acidic and basic amino acids such as aspartic acid, glutamic acid, and histidine Amino acids have their own tastes Lysine, alanine, glycine, serine, and threonine provide a sweet taste, while arginine, leucine, valine, phenylalanine, histidine, and isoleucine give a bitter taste, and glutamic and aspartic acids give a sour taste [23, 31] The umami taste is contributed by the high content of glutamic acid in fish sauce [23, 32, 33] Furthermore, the occurrence of greasy and grainy mouthfeel apparently also showed a relation to the taste of the product as well as to the physicochemical parameters indicated above It can be assumed that during fermentation salt can drive out water, which leads to a decreased aw of product, creates the texture of product, and influences the mouthfeel of product The presence of mouthfeel attributes might also possibly be due to the formation of soluble solids It was suggested by Beddows et al [34] that the soluble solids in fish sauce are most likely due to free amino acids and small peptides that are released during protein hydrolysis Therefore, further investigations on the relationship between soluble solids and mouthfeel attributes should be conducted Sulphury meaty, overripe cheese, sweaty, and ammonia were influenced by the moisture content, aw, salt content, and total volatile bases (TVB) (Fig 2b) The PLS-R plots indicated that the odors sulphury meaty, overripe cheese, sweaty, and ammonia correlated well with the moisture, aw, salt content, and TVB, although the correlation to TVB was weak (see Table 5) For the link between odor attributes and physicochemical parameters, it can be assumed that salt can suppress the existence of unwanted flavors owing to its ability to retard proteolysis activity, control the growth of microorganisms, and influence the volatile compounds that contribute to the odor of bakasang The volatile compounds contributing to the flavor of fish sauce were produced by nonenzymatic reactions of various components and enzymatic reactions of endogenous enzymes of fish origin and those of microorganisms surviving during fermentation [3] The used raw material, level of salt added, and period of fermentation affect the sensory, physicochemical, and microbiological quality of the product It is assumed that controlled handling of the materials and standardized fermentation process will lead to a better sensory flavor profile as well as to the improvement of the physicochemical and microbiological quality of the product In other words, for maintaining product quality, the home industries should be concerned with raw material, the amount of salt added, and the fermentation period Further investigations on how Fish Sci (2012) 78:187–195 flavor is formed during the fermentation stage should be conducted In addition, the model being developed herein should be considered a preliminary study correlating the sensory attributes of bakasang with its physicochemical properties It was developed using only two home industry products; therefore it may be useful for developing the manufacturing process but should not be considered for use in practical applications to predict the flavor of other samples The next step of this study will be to investigate the relation between sensory characteristics of various bakasang samples and physicochemical properties in order to develop a good model Acknowledgments The authors would like to thank the Directorate General of Indonesian Higher Education for financing this work as a part of Graduate Students Scholarships (2006–2008) to Silvana D Harikedua, sensory panel members (Food Science Graduate Students IPB 2006 and 2007) for participating in QDA sessions, Bakti Kumara (Foodex Inti Ingredients, Indonesia) for providing flavor references, Rebecca Bleibaum (Tragon, USA) for providing literature on QDA, and Suresh Kumar BV (CAMO Asia, India) for guidance in analyzing the data References Ijong FG, Ohta Y (1996) Physicochemical and microbiological changes associated with bakasang processing: a traditional Indonesian fish sauce J Sci Food Agric 71:69–74 Fukami K, Yaguramaki IS, Masuzawa T, Nabeta Y, Endo K, Shimoda N (2002) Identification of distinctive volatile compound in fish sauce J Agric Food Chem 50:5412–5416 Fukami K, Funatsu IY, Kawasaki K, Watabe S (2004) Improvement of fish sauce odor by treatment with bacteria isolated from the fish sauce mush (moromi) made from frigate mackerel J Food Sci 69:FMS45–49 Jiang JJ, Zeng QX, Zhu ZW, Zhang LY (2007) Chemical and sensory changes associated Yu-Lu fermentation process—a traditional Chinese fish sauce Food Chem 104:1629–1634 Chapman KW, Lawless HT, Boor KJ (2001) Quantitative descriptive analysis and principal component analysis for sensory characterization of ultrapasteurized milk Dairy Sci 84:12–20 Meilgaard M, Civille GV, Carr T (1999) Sensory evaluation techniques, 3rd ed CRC Press, Florida, pp 116–121 Lawless HT, Heymann H (1998) Sensory evaluation of food, principles and practices Chapman and Hall, New York Stone H, Sidel JL (2004) Sensory evaluation practices, 3rd ed Elsevier, San Diego, pp 201–245 Badan Standarisasi Nasional (1996) SNI 1-4271-1996 Fish sauce Indonesian National Standardization Committee, Jakarta 10 AOAC (2005) Official methods of analysis, 18th ed 950.46, 937.09, 935.57 Association of Official Analytical Chemists, Gaithersburg, MD 11 Badan Standarisasi Nasional (1998) SNI 1-4495-1998 Analysis methods of total volatile base (TVB) and trimethylamine (TMAS) by Conway Indonesian National Standardization Committee, Jakarta 12 Killinc B, Cakli S, Tolasa S, Dincer T (2006) Chemical, microbiological and sensory changes associated with fish sauce processing J Eur Food Res Tech 22:604–613 13 Yin LJ, Pan CL, Jiang ST (2002) New technology for producing paste-like fish products using lactic acid bacteria fermentation J Food Sci 67:3114–3118 195 14 Martin-Platero AM, Maqueda M, Valdivia E, Purswani J, Martinez-Bueno M (2009) Polyphasic study of microbial communities of two Spanish farmhouse goats milk cheeses from Sierra de Aracena Food Microbiol 26:294–304 15 Ersoy B, Aksan E, Ozeren A (2008) The effect of thawing methods on the quality of eels (Anguila anguila) Food Chem 111:377–380 16 Harrigan WF, Mc Chance ME (1998) Laboratory methods in food microbiology, 3rd ed Academic, New York 17 Kappes SM, Schmidt SJ, Lee SY (2007) Relationship between physical properties and sensory attributes of carbonated beverages J Food Sci 72:S1–S11 18 Ijong FG, Ohta Y (1995) Microflora and chemical assessment of an Indonesian traditional fermented fish sauce ‘‘bakasang’’ J Fac Appl Biol Sci 34:95–100 19 Peralta RR, Shimoda M, Osajima Y (1996) Further identification of volatile compounds in fish sauce J Agric Food Chem 44:3606–3610 20 Shimoda M, Peralta RR, Osajima Y (1995) Headspace gas analysis of fish sauce J Agric Food Chem 44:3601–3605 21 Yongsawatdigul J, Choi YJ, Udomporn S (2004) Biogenic amines formation in fish sauce prepared from fresh temperature abused Indian anchovy (Stolephorus indicus) J Food Sci 69:312–319 22 Sanceda NG, Suzuki E, Kurata T (2001) Development of normal and branched chain volatile acids during the fermentation process in the manufacture of fish sauce J Sci Food Agric 81:1013–1018 23 Tungkawachara S, Park JW, Choi YJ (2003) Biochemical properties and consumer acceptance of Pacific whiting fish sauce J Food Sci 68:855–860 24 Hjalmarsson HG, Park JW, Kristbergsson K (2007) Seasonal effects on the physicochemical characteristics of fish sauce made from capelin (Mallotus villosus) Food Chem 103:495–504 25 Cadwallader KR, Cha YJ (1998) Aroma active compounds in skipjack tuna sauce J Agric Food Chem 46:1123–1128 26 Gildberg A (1989) Accelerate fish sauce fermentation by initial alkalification at low salt concentration In: Miyachi S, Karube J, Ishida Y (eds) Current topic in marine biotechnology Fuji Technology, Tokyo, pp 101–104 27 Sikorski ZE, Gildberg A, Ruiter A (1995) Fish products In: Ruiter A (ed) Fish and fishery products Cab International, Wallingford, pp 315–346 28 Anihouvi V, Ayernor GS, Hounhouigan JD, Sakyi-Dawson E (2006) Quality characteristics of Lanhouin: a traditionally processed fermented fish product in the Republic of Benin Afr J Food Agric Nutr Dev 6:1–15 29 Jay JM (2000) Modern food microbiology, 6th ed Aspen Publishers, Gaithersburg, MD, pp 41–42 30 Park JN, Fukumoto Y, Fujita E, Tanaka T, Washio T, Otsuka S, Shimizu T, Watanabe K, Abe H (2001) Chemical composition of fish sauce produced in Southeast and East Asian countries J Food Comp Anal 14:113–125 31 Kato H, Rhue MR, Nishimura T (1998) Role of free amino acids and peptides in food taste In: Teranishi R, Buttery RG, Shahidi F (eds) Flavor chemistry: trends and developments ACS Symposium Series 388 American Chemical Society, Washington DC, pp 158–174 32 Komata Y (1990) Umami taste of seafoods Food Rev Int 6:457–487 33 Sanceda N, Kurata T, Arakawa N (1990) Overall quality and sensory acceptance of a lysine-fortified fish sauce J Food Sci 55:983–988 34 Beddows CG, Ardeshir AG, Daud WJ (1980) Development and origin of the volatile fatty acids in budu J Sci Food Agric 31:86–92 123 Fish Sci (2012) 78:197–206 DOI 10.1007/s12562-011-0433-9 ORIGINAL ARTICLE Food Science and Technology Species identification method for marine products of Seriola and related species Jun Iguchi • Yasuharu Takashima • Atsushi Namikoshi • Michiaki Yamashita Received: 27 July 2011 / Accepted: 23 October 2011 / Published online: 19 November 2011 Ó The Japanese Society of Fisheries Science 2011 Abstract The complete nucleotide sequences of mitochondrial DNA (mtDNA) from four Seriola spp (S quinqueradiata, S lalandi, S dumerili, and S rivoliana) were determined with the aim of developing a species identification analysis method for discriminating between commercially important Seriola spp and other related species In addition, the nucleotide sequences of the mitochondrial cytochrome b gene (Cytb) from five related but less expensive species in terms of market value (Seriolella brama, S caerulea, S punctata, Hyperoglyphe japonica, and Rachycentron canadum), which are often used as substitutes for Seriola spp., were determined Restriction enzyme sites were examined by comparing the nucleotide sequences, and species-specific primers were designed for PCR-based restriction fragment length polymorphism (RFLP) analysis Based on the results of the PCR amplification studies, the four Seriola spp and the five related species tested could be categorized into three groups J Iguchi (&) Á Y Takashima Food and Agricultural Materials Inspection Center, 2-1 Shintoshin, Chuo-ku, Saitama 330-9731, Japan e-mail: jun_iguchi@nm.famic.go.jp Y Takashima e-mail: yasuharu_takashima@nm.famic.go.jp A Namikoshi Food and Agricultural Materials Inspection Center, Kobe Regional Center, 1-3-7 Minatoshima-minamimachi, Chuo-ku, Kobe 651-0047, Japan e-mail: atsushi_namikoshi@nm.famic.go.jp M Yamashita National Research Institute of Fisheries Science, Fisheries Research Agency, 2-12-4 Fukuura, Kanazawa-ku, Yokohama 236-8648, Japan e-mail: mic@affrc.go.jp according to their PCR product pattern: a 373-bp product from the four Seriola spp., a 513-bp product from three Seriolella spp and H japonica, and a 204-bp product from R canadum In addition, RFLP analysis of the PCR products was able to differentiate these fish species Keywords Hyperoglyphe japonica Á PCR–RFLP Á Rachycentron canadum Á Seriola spp Á Seriolella spp Á Species identification Introduction The genus Seriola includes nine species that occur in tropical and temperate waters [1, 2] Of these, four Seriola species are commercially important in Japan, i.e., the yellowtail Seriola quinqueradiata, yellowtail amberjack Seriola lalandi, greater amberjack Seriola dumerili, and longfin yellowtail Seriola rivoliana In Japan, the annual catch of Seriola spp was 236000 t in 2010 [3] These fishes are eaten raw, as fillets, and marinated The appearance of the fillets of these four Seriola spp are similar to that of Seriolella spp., such as blue warehou Seriolella brama, white warehou Seriolella caerulea, and silver warehou Seriolella punctata, the Japanese butterfish Hyperoglyphe japonica, and cobia Rachycentron canadum Seriolella spp inhabit the tropical and temperate regions of the oceans of the Southern Hemisphere [4–6] and are imported into Japan from Australia and New Zealand, while H japonica and R canadum inhabit the waters of the peripheral regions of Japan These species are used commercially as lower priced substitutes for Seriola spp It is therefore important in terms os food labeling authentication to be able to distinguish between Seriola spp and other related species, including Seriolella spp., H japonica, and R canadum To 123 198 facilitate this process, we have developed a species identification technique Various species identification techniques have been developed and applied to surveys of commercial products by the Japanese government [7–9] A PCR-based restriction fragment length polymorphism (RFLP) technique using mitochondrial DNA (mtDNA) was developed for distinguishing European eel Anguilla anguilla products produced in China from those of the Japanese eel Anguilla japonica [7] Japanese jack mackerel Trachurus japonicus and Atlantic horse mackerel Trachurus trachurus can be identified in processed products by mtDNA analysis [8] Thus, mtDNA analysis has been applied in numerous techniques for species identification in processed products As mtDNA copy number is higher than that of the nuclear genome, the analysis of processed products in which DNA has been decomposed by the heating step during cooking is a useful approach In this study, a PCR–RFLP method using the cytochrome b gene (Cytb) of the partial mtDNA sequence was developed to distinguish among nine commercially important species i.e., four Seriola spp., three Seriolella spp., H japonica, and R canadum The complete mtDNA nucleotide sequences of the four Seriola spp were determined to identify an appropriate DNA region for PCR– RFLP analysis As the Cytb of the four Seriola spp had adequate substitutions and a useful DNA region for PCR– RFLP analysis, the mitochondrial Cytb nucleotide sequences of the five other species studied were also determined Materials and methods Fish samples Specimens of Seriola quinqueradiata (n = 24), S lalandi (n = 7), S dumerili (n = 11), S rivoliana (n = 9), H japonica (n = 4), and R canadum (n = 5), were obtained from different locations in Japan [R canadum (n = 5) was also obtained from Taiwan], and Seriolella brama (n = 9), S caerulea (n = 3), and S punctata (n = 9) were obtained from New Zealand The fishes were morphologically identified according to published procedures [4–6, 10] In addition, commercially processed products of S quinqueradiata, S caerulea, S punctata, and H japonica (n = 14) were collected from a Japanese market Preparation of DNA and PCR amplification DNA was purified from muscle samples using a DNeasy Tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions 123 Fish Sci (2012) 78:197–206 A total of 107 primers were used for sequencing the entire mtDNA of the four Seriola spp and Cytb of the partial mtDNA sequence for the four Seriola spp., three Seriolella spp., H japonica, and R canadum (Table 1) This primer set included 53 fish-versatile primers [11, 12], whereas 54 other primers were newly designed for this analysis of the four Seriola spp mtDNA Each PCR analysis was performed in a volume of 25 ll containing 18.6 ll of distilled water, 2.5 ll of 109 PCR buffer, 2.0 ll of dNTP Mix (5 mM), 0.3 ll of each primer (50 lM), 0.1 ll of U/ll ExTaq polymerase (Takara, Otsu, Japan), and 2.5 ll of template DNA containing approximately ng of DNA The PCR cycling program consisted of an initial denaturation step at 94°C for min, followed by 40 cycles at 94°C for 30 s, 50°C for 20 s, and 72°C for Sequencing of PCR products PCR products were purified using a DNA purification kit (Promega, Madison, WI) Direct sequencing of the PCR products was performed using a BigDye Terminator Cycle Sequencing Ready Reaction kit (Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions Labeled fragments were analyzed on a model ABI 3130 DNA sequencer (Life Technologies) PCR with species-specific primers and RFLP analysis To distinguish among the nine fish species, we designed a species-specific primer set corresponding to restriction sites within the Cytb of the partial mtDNA sequence with reference to the sequences of the nine fish species (Table 1) Each PCR analysis with the species-specific primer set was performed in a reaction volume of 20 ll containing 13.5 ll of distilled water, 2.0 ll of 109 PCR buffer, 1.6 ll of dNTP Mix (5 mM), 0.2 ll of each primer (50 lM), 0.1 ll of U/ll ExTaq polymerase HS (Takara, Shiga, Japan), and 2.0 ll of template DNA containing approximately ng of DNA The PCR cycling program consisted of an initial denaturation step at 94°C for min, followed by 35 cycles at 94°C for 30 s, 60°C for 30 s, and 72°C for The RFLP analysis was performed with PCR products of the four Seriola spp., three Seriolella spp., and H japonica The PCR products of the four Seriola spp were digested with HinfI (Fermentas, Burlington, ON, Canada) and those of the three Seriolella spp and H japonica were digested with FspBI (Fermentas) The reaction mixture contained 7.5 ll of the PCR products, 2.5 U/ll of the restriction enzyme, and 1.5 ll of universal buffer; the total volume was made up to 15 ll with distilled water The PCR products were digested at 37°C for h according to the manufacturer’s instructions, and the digested samples were Fish Sci (2012) 78:197–206 199 Table PCR and sequencing primers used in analysis of Seriola spp mitochondrial genomes Common primers a Primers designed for this study b Primer Nucleotide sequence Primera Nucleotide sequenceb L1083-12S ACAAACTGGCATTAGATAC L63-Phe CAA GCA CAA AGG TTT GGT CCT GAC L1803-16S AGTACCGCAAGGGAAAGCTCAAA L201-12S CCA CGA CAC CTT GCT TAG CCA CAC L2510-16S CGCCTGTTTACCAAAAACAT L620-12S GAG GAG CCT GTT CTA GAA CCG ATA ACC L2949-16S GGGATAACAGCGCAATC L1045-Val AAG CAT CTC CCT TAC ACC GAG AAG TC L5260-ND2 CTGGSTTTATGCCMAARTG L1654-16S AAT GTT TTG GTG GGC CTA AAA GCA GC L5698-Asn AGGCCTCGATCCTACAAAGKTTTAGTTAAC L3038-ND1 GCT CCT TCA ACC TAT TGC TGA CGG L5956-CO1 CACAAAGACATTGGCACCCT L3844-Ile TGAAGCCTAAGGGCCACTTTGAT L6730-CO1 TATATAGGAATRGTMTGAGC L4493-ND2 GCC CTR TTC CTY CAA MTT CAM CCY C L7255-CO1 GATGCCTACACMCTGTGAAA L4929-ND2 CCC TCC TCA GCC TCT ACT TCT ACC TC L7863-CO2 ATAGACGAAATTAATGACCC L6060-CO1 GTY TGA GCT GTY CTR ATY ACR GCY GT L8329-Lys AGCGTTGGCCTTTTAAGC L6254-CO1 ACTACTCAGGTAAAAAAGAACC L9220-CO3 AACGTTTAATGGCCCACCAAGC L6633-CO1 CAC TTC CAC TAC GTC CTA TCC ATA GGA GC L9514-CO3 TTCTGAGCCTTCTAYCA L7100-Ser CCA ACC ACA TAA CCG CTC TGC CAC L10201-ND3 TTTCACCCTCTRGGSTCTGCCCG L7718-CO2 GCA TCA AAG TTG ACG CAG TAC CAG L11424-ND4 TGACTTCCWAAAGCCCATGTAGA L8323-ATP AAA YCC AGG GGG CCA CAA ATG AGC L12321-Leu GCTCTTAGGAACCAAAAACTCTTGGTGCAA L9760-ND3 AAA CCC CGA CCA CGA GAA GCT CTC L13553-ND5 AACACMTATTAYCTWAACGC L10392-ND4 CTT CAG AGC CTT AAC CTH CTA CAA TGC L13731-ND5 TTAACCCWATCAAACGMCTWGCCTG L10709-ND4 CTA GCT TTC GGC GCA ACA GAA C L14735-Glu AACCACCGTTGTTATTCAAC L11170-ND4 CATCATCTTCGCACTCTGAGGCGT L15765-CYB ATTCTWACMTGAATTGGMGG L11474-ND4 AGC CTA GCC AAC TTA GCC CTH CCD CC L15927-Thr AGAGCGTCGGTCTTGTAAKCCG L12131-ND5 CAGCTTGCTCCCCCTTTTCTTATACCT L15411-CYB GATAAAATTYCATTCCACCC L12497-ND5 CCT ACA AGC CGT TCT TTA TAA CCG AGT C L15998-Pro AACTCTTACCMTTGGCTCCCAARGC L12935-ND5 AAT AGT TAC CAT CGG CCT CAA CCA AC H690-12S GCGGAGGCTTGCATGTGTA L13503-ND5 GAA CTG GCA TCA CTA ACG AAC AAA C H884-12S AACCGCGGTGGCTCGCACGAG L13833-ND6 CCA CGA CTA AGC CCA CGA GTT AAC TC H1358-12S CGACGGCGGTATATAGGC L15192-CYB CCT TAG TGA CCC CGC CAC ATA TCA H1903-16S GTAGCTCGTYTAGTTTCGGG L16061-Con TAT TCC TGG CAT TTG GTT CCT ACT TCA G H2590-16S ACAAGTGATTGCGCTACCTT H444-12S TCG TGG TTG GGC TTC ATA CCT TC H3084-16S AGATAGAAACTGACCTGGAT H569-12S CAC CAA ACA TCC GCC TGG GAA TTA C H3718-ND1 ACTTCGTATGAAATWGTTTG H866-12S CTC GCT TAC TGC TAA ATC CTC CTT CAG H3934-ND1 GCGTATTCTACGTTGAATCC H1523-16S GAA GCS KDG AGG CTG AAC TYY TAT C H4866-ND2 AAKGGKGCKAGTTTTTGTCA H1587-16S GTATCTTGTCTCAAAGGGGC H5334-ND2 CGKAGGTAGAACTAHAGGCT H2988-ND1 GCA CYT TYC GTT CGA TTA RGG TSA G H5669-Asn AACTGAGAGTTTGWAGGATCGAGCCC H3494-ND1 TCAACGTTAAAGCCTGAGAC H5937-CO1 TGGGTGCCAATGTCTTTGTG H4006-Met AGG RAG TGG TGT AGW GGA AGC AC H6371-CO1 TTGATTGCCCCKAGGATWGA H5830-CO1 TGA GCT TCT GAC TCC TTC CCC CTT C H6558-CO1 CCKCCWGCKGGGTCAAAGAA H6606-CO1 GGA CGA TRT CTA AAG ARG ART TGG CYA GRA C H6855-CO1 AGTCAGCTGAAKACTTTTAC H7569-CO2 GTC AAA CCC TAG GTC CTC GTA GTC TG H7480-Ser ATGTGGYTGGCTTGAAA H7957-Lys CAC CAA TCT TTA GCT TAA AAG GCT AAY GC H8168-CO2 CCGCAGATTTCWGAGCATTG H8243-ATP CCG AGG GGG TTG GAA ATA GAA TTC A H9639-CO3 CTGTGGTGAGCYCAKGT H8496-ATP TTW CGC ATT CCA ATA ATM ACK GTY GC H10019-Gly CAAGACKGKCTGATTGGAAG H8936-CO3 GAG TGG AAG TGG AAY CAR ATK GCK AG H10433-Arg AACCATGGWTTTTTGACCCGAAAT H10540-ND4 GCA AGA TCA GCC TGT TTC GGA G H10970-ND4 GATTATWAGKGGGAGWAGTCA H10911-ND4 GAT AGA GTG CCG GTG TTG TTT TGG AG H11618-ND4 TGGCTGACKGAKGAGTAGGC H11599-ND4 TGC TGT GAT TAG GGT GCC TGC TC H12145-His CTAGTGTTTTKGTTAAACTA H12230-ND5 GCT GAT GTT GAC ATC GAA GGT GAG H12632-ND5 GATCAGGTTACGTAKAGKGC H13005-ND5 CTT TAA AGA AAG CGT GGG TGC AGA TG H13069-ND5 GTGCTGGAGTGKAGTAGGGC H13251-ND5 TGA AAG AGG TTG CTA GGA GGG TTA GG H13396-ND5 CCTATTTTTCGGATGTGTTG H13439-ND5 GGT CAT GAT TGG GGT TTT TAA AGG 123 200 Fish Sci (2012) 78:197–206 Table continued Common primers a Primers designed for this study b Primer Nucleotide sequence Primera Nucleotide sequenceb H13727-ND5 GCGATKATGCTTCCTCAGGC H14144-ND6 GGT TGT GTT TGC ATA TTC GGC AG H14080-ND5 AGGTAKGTTTTGATTAKKCC H14515-CYB GAGAGCCGAAGTTTCATCATGCTGA H14718-Glu TTTTTGTAGTTCAATWACAACGGT H15347-CYB CGTTGTTTGGAGGTGTGAAGGATTGG H15149-CYB GGTGGCKCCTCAGAAGGACATTTGKCCTCA H15630-Pro TTAATTTAGAATCCTAGCTTTGG H16042-Con CCA CGA TTA TTG TCC CTG ACC ATC A Common-Lc TCC CCT GAG GAC AAA TAT CAT TCT GAG G Seriola-Hc ATG AAG TTG TCG GGG TCG CCR AG Seriolella & Hj-Hc CTA CCA TGA GGA TGA GAA TGG AGG C Rc-Hc GAA GGA ACA GGA GGT GAA TTA TGG TTG Codes R = A/G, Y = C/T, K = G/T, M = A/C, S = G/C, W = A/T a L and H denote light and heavy strands, respectively b Positions with mixed bases are labeled with their IUB coces c Species-specific primers then electrophoresed on a 3% agarose gel and stained with ethidium bromide Results Sequence analysis Complete mtDNA nucleotide sequences from the four Seriola spp (i.e., S quinqueradiata, S lalandi, S dumerili, and S rivoliana) were determined using one individual of each species (Table 2) and deposited in the DDBJ/GenBank databases with accession numbers AB517556, AB517557, AB517558, and AB517559, respectively The total length of the mtDNA of each of these four species was 16537 bp (S quinqueradiata), 16532 bp (S lalandi), 16530 bp (S dumerili), and 16530 bp (S rivoliana) All mtDNA of the four species contained two rRNA genes, 22 tRNA genes, 13 protein-coding genes, and a control region, similar to other vertebrate mtDNA All genes were encoded on the heavy strand (H-strand), with the exception of the ND6 gene and eight tRNA genes, and all genes were similar in length to those in other teleosts The tRNA genes ranged in size from 66 to 76 nucleotides (Table 2) The tRNA encoded by each gene was large enough to fold into the characteristic cloverleaf secondary structure The 12S and 16S rRNA genes were located between the genes for tRNAPhe and tRNALeu, separated by the tRNAVal gene These two rRNA genes were identified from the proposed secondary structures of the carp 12S rRNA and loach 16S rRNA genes [13] The 13 protein-coding genes were characterized by comparing them to the nucleotide sequences of other vertebrates [14–18] As in other teleost 123 fishes, all of the mitochondrial protein-coding genes began with an ATG start codon, with the exception of COI, which started with GTG (Table 2) Eight genes ended with TAA (ND1, ND2, COI, ATPase8, ATPase6, COIII, ND4L, and ND5) and TAG (ND3 and ND6), and the remainder had incomplete stop codons, T (COII, ND4, and Cytb) (Table 2) A comparison of nucleotide substitutions among the four Seriola spp., calculated as absolute pairwise nucleotide differences and nucleotide sequence identity, excluding noncoding regions, resulted in the detection of 853–1,411 nucleotide substitutions within the four Seriola spp (Table 3) S quinqueradiata was the most closely related species to S lalandi, and S dumerili was the most closely related species to S rivoliana The nucleotide sequence data of Cytb of three Seriolella spp (S brama, S caerulea, and S punctata), H japonica, and R canadum were determined and deposited in the DDBJ/GenBank databases with accession numbers AB281614 (S brama), AB262190 (S caerulea), AB262189 (S punctata), AB513919 (H japonica), AB292793 (Rachycentron canadum), respectively These data for Cytb from the partial mtDNA sequence were analyzed to determine whether they could be used to differentiate the nine fish species of this study Also, 1,001–1,054 bp of common sequences were observed in the comparison of the 1,141 bp of Cytb in the four Seriola spp (Table 4) However, only 837–881 bp of common sequences were detected between the four Seriola spp and the other five species Thus, on the basis of such nucleotide differences, the four Seriola spp (Carangidae) could be clearly differentiated from the other five fish species, i.e., the three Seriolella spp., H japonica (Centrolophidae), and R canadum (Rachycentridae) The Fish Sci (2012) 78:197–206 201 Table Location of gene/element in the mitochondrial genomes of four Seriola spp Gene/element Seriola quinqueradiata (AB517556)a S lalandi (AB517557) Position number Position number From tRNAPhe Size To From S dumerili (AB517558) Size To S rivoliana (AB517559) Position number From Size To Codon Position number From Size To 68 68 68 68 68 68 68 68 69 1023 955 69 1019 951 69 1018 950 69 1018 950 tRNAVal 1024 1095 72 1020 1091 72 1019 1090 72 1019 1090 72 16S rRNA 1096 2813 1718 1092 2808 1717 1091 2808 1718 1091 2809 1719 tRNALeu 2814 2888 75 2809 2883 75 2809 2883 75 2810 2884 75 ND1 2889 3863 975 2884 3858 975 2884 3858 975 2885 3859 975 tRNAIle 3868 3937 70 3863 3932 70 3863 3932 70 3864 3933 70 tRNAGln (L)b 3937 4007 71 3932 4002 71 3932 4002 71 3933 4003 71 tRNAMet 4007 4075 69 4002 4070 69 4002 4070 69 4003 4071 69 ND2 4076 5122 1047 4071 5117 1047 4071 5117 1047 4072 5118 1047 12S rRNA tRNA (UUR) Trp tRNAAla (L) 5122 5192 71 5117 5187 71 5117 5187 71 5118 5188 71 5194 5262 69 5189 5257 69 5189 5257 69 5190 5,258 69 Start Stop ATG TAA ATG TAA GTG TAA ATG T tRNAAsn (L) 5264 5336 73 5259 5331 73 5259 5331 73 5260 5332 73 OL tRNACys (L) 5337 5371 5370 5436 34 66 5332 5366 5365 5431 34 66 5332 5366 5365 5431 34 66 5333 5367 5366 5432 34 66 tRNATyr (L) 5437 5506 70 5432 5501 70 5432 5501 70 5433 5502 70 COI 5508 7058 1551 5503 7053 1551 5503 7053 1551 5504 7054 1551 7059 7129 71 7054 7124 71 7054 7124 71 7055 7125 71 tRNAAsp 7134 7204 71 7129 7199 71 7129 7199 71 7130 7200 71 COII 7213 7903 691 7208 7898 691 7208 7898 691 7209 7899 691 tRNALys 7904 7978 75 7899 7973 75 7899 7973 75 7900 7974 75 ATPase8 7980 8147 168 7975 8142 168 7975 8142 168 7976 8143 168 ATG TAA ATPase6 8138 8821 684 8133 8816 684 8133 8816 684 8134 8817 684 ATG TAA COIII 8821 9606 786 8816 9601 786 8816 9601 786 8817 9602 786 ATG TAA tRNAGly 9606 9676 71 9601 9671 71 9601 9671 71 9602 9672 71 ND3 9677 10027 351 9672 10022 351 9672 10022 351 9673 10023 351 ATG TAG tRNASer tRNA (UCN) (L) Arg 10026 10094 69 10021 10089 69 10021 10089 69 10022 10090 69 ND4L 10095 10391 297 10090 10386 297 10090 10386 297 10091 10387 297 ATG TAA ND4 tRNAHis 10385 11766 11765 11834 1381 69 10380 11761 11760 11829 1381 69 10380 11761 11760 11829 1381 69 10381 11762 11761 11830 1381 69 ATG T 11835 11901 67 11830 11896 67 11830 11896 67 11831 11897 67 tRNASer tRNA (AGY) Leu (CUN) 11906 11978 73 11901 11973 73 11901 11973 73 11902 11974 73 ND5 11979 13817 1839 11974 13812 1839 11974 13812 1839 11975 13813 1839 ATG TAA ND6 (L) 13814 14335 522 13809 14330 522 13809 14330 522 13810 14331 522 ATG TAG ATG T tRNAGlu (L) 14337 14405 69 14332 14400 69 14332 14400 69 14333 14401 69 Cytb 14410 15550 1141 14405 15545 1141 14406 15546 1141 14407 15547 1141 72 tRNA Thr 15551 15622 72 15546 15617 72 15547 15618 72 15548 15619 tRNAPro (L) 15622 15692 71 15617 15687 71 15618 15688 71 15619 15689 71 Control region 15693 16537 845 15688 16532 845 15689 16530 842 15690 16530 841 a DDBJ/EMBL/GenBank Accession Number b L indicates that the gene is encoded on the L strand 123 202 Fish Sci (2012) 78:197–206 Table The number of nucleotide differences (upper half of the matrix) and nucleotide sequence identity (lower half of the matrix) of the Seriola spp complete mitochondrial DNA sequences Seriola spp S quinqueradiata S lalandi S dumerili S rivoliana S quinqueradiata – 853 bp 1403 bp 1411 bp 1333 bp S lalandi 94.5% – 1325 bp S dumerili 91.0% 91.5% – S rivoliana 91.0% 91.5% 94.0% 939 bp – The table is based on the consensus of 15624 bases excluding non-coding regions Table The number of nucleotide differences (upper half of the matrix) and the nucleotide sequence identity (lower half of the matrix) of Cytb Fish species Seriola quinqueradiata S lalandi S dumerili S rivoliana Seriolella brama S caerulea S punctata Hyperoglyphe japonica Rachycentron canadum S quinqueradiata – 1054 bp 1001 bp 1001 bp 865 bp 878 bp 879 bp 861 bp 840 bp S lalandi 92.4% – 1017 bp 1007 bp 852 bp 863 bp 859 bp 861 bp 844 bp S dumerili S rivoliana 87.7% 87.7% 89.1% 88.3% – 96.2% 1057 bp – 869 bp 869 bp 875 bp 877 bp 872 bp 881 bp 863 bp 863 bp 841 bp 837 bp S brama 75.8% 74.7% 76.2% 76.2% – 1087 bp 1061 bp 998 bp 849 bp S caerulea 77.0% 75.6% 76.7% 76.9% 95.3% – 1071 bp 1008 bp 847 bp S punctata 77.0% 75.3% 76.4% 77.2% 93.0% 93.9% – 987 bp 849 bp H japonica 75.5% 75.5% 75.7% 75.7% 87.5% 88.4% 86.6% – 829 bp R canadum 73.6% 74.0% 73.7% 73.4% 74.4% 74.2% 74.4% 72.7% – The table is based on the consensus nucleotide sequences of 1141 bases four Seriola spp could be distinguished by restriction analysis using HinfI In addition, the three Seriolella spp and H japonica could be distinguished by restriction analysis using FspBI Species-specific PCR and RFLP analysis We designed a species-specific primer set that can be used to distinguish among three groups of fishes (i.e., four Seriola spp., three Seriolella spp., H japonica, and R canadum) based on the HinfI and FspBI restriction sites The species-specific primer set was designed to amplify the 373-bp product of the four Seriola spp., the 513-bp product of the three Seriolella spp and H japonica, and the 204-bp product of R canadum (Figs 1, 2a) In the four Seriola spp., HinfI digested the 373-bp product to yield fragments of 91, 130, and 152 bp in S quinqueradiata, 91 and 110 bp in S lalandi, 91 and 240 bp in S dumerili, and 91 and 282 bp in S rivoliana (Fig 2b; Table 5) Also, FspBI digested the 513-bp product from the three Seriolella spp and H japonica to yield fragments of 138 and 327 bp in S brama, 138 and 291 bp in S caerulea, 185 and 327 bp in S punctata, and 138 and 210 bp in H japonica (Fig 2c; Table 5) The fragments smaller than 80 bp were not characterized In total, 100 specimens of 123 nine fish species, including raw fish materials and commercial processed items (seasoned and grilled), were examined using the method developed here, and all specimens could be distinguished successfully (Table 5) The nucleotide sequences of the PCR products for the specimens were also determined and subsequently confirmed to have no nucleotide substitution on the HinfI and FspBI restriction sites Discussion In this study, the complete nucleotide sequences of mtDNAs from four Seriola spp were determined and the differences between these sequences characterized (Table 3) The smallest number of nucleotide substitutions was 853 sites between S quinqueradiata and S lalandi, indicating a close phylogenetic relationship among the four Seriola spp of Japanese origin characterized in this study Coulson et al performed a nucleotide sequence analysis of the complete mtDNAs from ten taxa of gadine codfishes and pollock and characterized the differences [19] Their analysis of the complete mtDNA nucleotide sequences, excluding missing or ambiguous sites, indicated 51–588 nucleotide substitutions and 14036 bp of common sequences in three species (Gadus morhua, G ogac, and G Fish Sci (2012) 78:197–206 Fig Partial nucleotide sequences of the mitochondrial cytochrome b gene (Cytb) for the four Seriola spp (S quinqueradiata, S lalandi, S dumerili, S rivoliana), three Seriolella spp (S brama, S caerulea, S punctata), Hyperoglyphe japonica, and Rachycentron canadum Dots indicate sequence identity with S quinqueradiata, arrows 203 indicate the primer set Common-L, Seriola-H, Seriolella & Hj-H, and Rc-H used for the PCR–restriction fragment length polymorphism (RFLP) analysis Restriction enzyme sites are underlined; the four Seriola spp were digested with HinfI and the three Seriolella spp and H japonica were digested with FspBI 123 204 Fish Sci (2012) 78:197–206 a M M 513 bp 373 bp 204 bp M b M 373 bp 282 bp 240 bp 152 bp 130 bp 91 bp c 110bp M M 513 bp 327 bp 185 bp 138 bp 291 bp 210 bp Fig Comparison of agarose gel electrophoresis patterns for PCR products amplified with species-specific primers to those obtained from the subsequent RFLP analysis for the four Seriola spp (S quinqueradiata, S lalandi, S dumerili, S rivoliana), three Seriolella spp (S brama, S caerulea, S punctata), H japonica, and R canadum a Typical patterns of PCR products with a combination of species-specific primers (Common-L, Seriola-H, Seriolella & Hj-H, and Rachycentron-H) Lanes: 1–9 S quinqueradiata, S lalandi, S dumerili, S rivoliana, S brama, S caerulea, S punctata, H japonica, and R canadum, respectively, M 100-bp DNA ladder marker b RFLP patterns of the four Seriola spp following digestion with the restriction enzyme HinfI Lanes: 1–4 S quinqueradiata, S lalandi, S dumerili, and S rivoliana, respectively, S quinqueradiata excluding restriction enzyme HinfI, M 100-bp DNA ladder marker c RFLP patterns of the three Seriolella spp and H japonica following digestion with the restriction enzyme FspBI Lanes: 1–4 S brama, S caerulea, S punctata, and H japonica, respectively, S brama excluding restriction enzyme FspBI, M 100-bp DNA ladder marker macrocephalus) Takashima et al analyzed the complete mtDNA nucleotide sequences of Trachurus trachurus and T japonicus; their results indicated 414 nucleotide substitutions (2.6% of the total) within the two species, excluding noncoding regions [8] Our four Seriola spp had more nucleotide substitutions than observed by these researchers in the three Gadus spp and two Trachurus spp.; it can 123 therefore be concluded that the four Seriola spp are the most distantly related within these three genera It is essential that rapid and authentic species identification techniques are available for the detection of fraudulent labeling with the aim of providing better utilization of processed foods originating from commercially important fish species PCR–RFLP-based species identification methods using mtDNA sequences have been developed for the explicit purpose of analyzing processed seafood from such commercially important fish species, such as eel, jack mackerel, salmon, and hake [7–9, 20] In this study, we developed a species identification technique to discriminate between Seriola spp and other related species, including Seriolella spp., H japonica, and R canadum for food labeling authentication according to the Law Concerning Standardization and Proper Labeling of Agricultural and Forestry Products (JAS Law) We designed a PCR primer set for PCR–RFLP analysis to differentiate between Seriola and related species based on sequence comparison of the mtDNA Cytb region (Fig 1) To design the primer set, we selected those nucleotide substitutions specific for each family RFLP patterns corresponding to phylogenetic relationships differentiated the four Seriola spp (Carangidae), three Seriolella spp., H japonica (Centrolophidae), and R canadum (Rachycentridae) To test the species specificity of the developed PCR– RFLP method, we confirmed that there was no apparent PCR amplification for the DNA samples of commercially important fish species, including jack mackerels and cods other than Seriola spp., except for red sea bream, which showed a 0.4-kbp PCR product In this latter case, the amplified PCR product of red sea bream had a different DNA length from those of the nine species that we characterized in this study and had different restriction sites in the corresponding Cytb regions (data not shown) Nucleotide sequence determination can also be used for the species identification of fish samples Nucleotide substitutions were identified in Cytb of the four Seriola spp., forming the basis for a new and useful PCR–RFLP method Although DNA samples prepared from the heat-processed products that were boiled or grilled may be partially decomposed, it will be possible to use the PCR primer set for short nucleotide sequences (204–513 bp) for the analysis of both raw fish materials and processed products Russell et al [9] established a method to identify ten salmon species using six restriction enzymes, and Quinteiro et al [20] established a method to identify 11 hake species using four restriction enzymes In this study, our PCR–RFLP method distinguished nine fish species by species-specific PCR and RFLP analysis using only two Fish Sci (2012) 78:197–206 Table PCR–restriction fragment length polymorphism analysis of fish products 205 Species Processing procedure Number of specimens Length of PCR product (bp) RFLP products (bp) Seriola quinqueradiata Raw fish 24 373 91, 130, 152 373 91, 130, 152 HinfI Marinated in miso Teriyaki 373 91, 130, 152 Seriola lalandi Raw fish 373 91, 110 Seriola dumerili Raw fish 11 373 91, 240 Seriola rivoliana Raw fish 373 91, 282 FspBI Seriolella brama Frozen fillet 513 Seriolella caerulea Frozen fillet 513 138, 291 Marinated in miso Grilled fillet 513 513 138, 291 138, 291 Frozen fillet 513 185, 327 Seriolella punctata Hyperoglyphe japonica Rachycentron canadum 138, 327 Marinated in miso 513 185, 327 Grilled fillet 513 185, 327 Raw fish 513 138, 210 Marinated in miso 513 138, 210 10 204 – Raw fish enzymes As such, the method reported here is a powerful tool to identify nine commercially important fish species and will be applicable for commercial product surveys by the Japanese government Acknowledgments This study was supported by a project grant (Development of Evaluation and Management Methods for Supply of Safe, Reliable and Functional Food and Farm Produce) from the National Food Research Institute and the Ministry of Agriculture, Forestry and Fisheries of Japan References Balanov AA (2008) On the species composition of fish of the genus Seriola (Carandigae) in the northwestern part of the sea of Japan J Ichthyol 48:415–421 Gushiken S (1983) Revision of the carangid fishes of Japan Galaxea 2:135–264 The Ministry of Agriculture, Forestry, Fisheries of Japan (MAFF) (2010) Statistics of Agriculture, Forestry and Fisheries in 2010 (in Japanese) MAFF, Tokyo Norman JR (1937) Coast fishes Part II The Patagonia region Discov Rep 16:115–116 Hadrich RL (1967) The Stromateoid fishes: systematics and a classification Bull Mus Comp Zool 135:69–72 Cousseau MB, Forciniti L, Ubaldi G (1993) Species of the genus Seriolella (Centrolophidae) in Southwest Atlantic Waters Jpn J Ichtyol 40(2):183–187 Sezaki K, Itoi S, Watabe S (2005) A simple method to distinguish two commercially valuable eel species in Japan Anguilla japonica and A anguilla using polymerase chain reaction strategy with a species-specific primer Fish Sci 71:414–421 Takashima Y, Morita T, Yamashita M (2006) Complete mitochondrial DNA sequence of Atlantic horse mackerel Trachurus 10 11 12 13 14 15 16 17 18 trachurus and molecular identification of two commercially important species T trachurus and T japonicus using PCRRFLP Fish Sci 72:1054–1065 Russell VJ, Hold GL, Pryde SE, Rehbein H, Quinteiro J, ReyMendez M, Sotelo CG, Pe´rez-Martı´n RI, Santos AT, Rosa C (2000) Use of restriction fragment length polymorphism to distinguish between salmon species J Agric Food Chem 48:2184–2188 Nakabo T (2000) Fishes of Japan with pictorial keys to the species, 2nd edn (in Japanese) Tokai University Press, Tokyo, pp 792–793, 960 Kocher TD, Thomas WK, Meyer A, Edwards SV, Pa¨a¨bo S, Villablanca FX, Wilson AC (1989) Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers Proc Natl Acad Sci USA 86:6196–6200 Miya M, Nishida M (2000) Use of mitogenomic information in teleostean molecular phylogenetics: a tree-based exploration under the maximum-parsimony optimality criterion Mol Phylogenet Evol 17:437–455 Gutell RR, Gray MW, Schnare MN (1993) A compilation of large subunit (23S and 23S-like) ribosome RNA structures Nucleic Acids Res 21:3055–3074 Bibb MJ, Van Etten RA, Wright CT, Walberg MW, Clayton DA (1981) Sequence and gene organization of mouse mitochondrial DNA Cell 26:167–180 Anderson S, de Bruijn MHL, Coulson AR, Eperon IC, Sanger F, Young IG (1982) Complete sequence of bovine mitochondrial DNA Conserved features of the mammalian mitochondrial genome J Mol Biol 156:683–717 Inoue JG, Miya M, Tsukamoto K, Nishida M (2000) Complete mitochondrial DNA sequence of the Japanese sardine Sardinops melanostictus Fish Sci 66:924–932 Inoue JG, Miya M, Aoyama J, Ishikawa S, Tsukamoto K, Nishida M (2001) Complete mitochondrial DNA sequence of the Japanese eel Anguilla japonica Fish Sci 67:118–125 Inoue JG, Miya M, Tsukamoto K, Nishida M (2001) Complete mitochondrial DNA sequence of the Japanese anchovy Engraulis japonicus Fish Sci 67:828–835 123 206 19 Coulson MW, Marshall HD, Pepin P, Carr SM (2006) Mitochondrial genomics of gadine fishes: implications for taxonomy and biogeographic origins from whole-genome data sets Genome 49:1115–1130 20 Quinteiro J, Vidal R, Izquierdo M, Sotelo CG, Chapela MJ, Pe´rez-Martı´n RI, Rehbein H, Hold GL, Russell VJ, Pryde SE, 123 Fish Sci (2012) 78:197–206 Rosa C, Santos AT, Rey-Me´ndez M (2001) Identification of Hake Species (Merluccius Genus) using sequencing and PCR-RFLP analysis of mitochondrial DNA control region sequences J Agric Food Chem 49:5108–5114 Fish Sci (2012) 78:207 DOI 10.1007/s12562-011-0434-8 ERRATUM Erratum to: Analysis of juvenile tuna movements as correlated random walk Minoru Kadota • Shinsuke Torisawa • Tsutomu Takagi • Kazuyoshi Komeyama Hiromu Fukuda • Published online: 25 November 2011 Ó The Japanese Society of Fisheries Science 2011 Erratum to: Fish Sci (2011) 77:993–998 DOI 10.1007/s12562-011-0415-y The original version of this article unfortunately contained a mistake The author Hiromu Fukuda was not included The online version of the original article can be found under doi:10.1007/s12562-011-0415-y M Kadota (&) Á S Torisawa Á T Takagi Á H Fukuda Department of Fisheries, Faculty of Agriculture, Kinki University, 3327-204 Naka-machi, Nara 631-8505, Japan e-mail: kadota@cims.nyu.edu K Komeyama The Education and Research Center for Marine Resources and Environment, Faculty of Fisheries, Kagoshima University, 4-50-20 Shimoarata, Kagoshima 890-0056, Japan 123 [...]... patterns 12 3 20 Fish Sci (2 012 ) 78 :15 –22 Fig 5 Time-series data of the hourly detection rates of each juveniles a YT05 at P 9, b YT 11 at P 7, c YT12 at P 7, d YT13 at P 7, e YT14 at P 7, and f YT16 at P6 ‘‘R’’ indicates recapture 10 0 (a) YT05 0 0:00 5 Sep 10 0 12 :00 0:00 6 Sep 12 :00 0:00 7 Sep 12 :00 0:00 8 Sep 12 :00 0:00 9 Sep 12 :00 0:00 14 Sep 12 :00 0:00 15 Sep 12 :00 0:00 16 Sep 12 :00 0:00 17 Sep 12 :00 12 :00... J Fish Manag 10 :11 1 11 3 10 Pollock KH, Hoening JM, Jones CM, Robson DS, Greene CJ (19 97) Catch rate estimation for roving and access point surveys N Am J Fish Manag 17 :11 19 11 Hiett RL, Worrall JW (19 77) Marine recreational fishermen’s ability to estimate catch and recall catch and effort over time Research report HSR-RR /13 -CD Human Sciences Research, McClean, VA 12 Chase DR, Harada M (19 84) Response... net Fish Sci (2 012 ) 78:23–32 Pennahia argentata 400 Codend 300 Cover net 200 10 0 0 15 18 21 24 27 30 33 36 39 Total length (cm) 32 Eopsetta grigorjewi Trichiurus lepturus 10 0 80 Catch in number Catch in number 12 0 Codend 60 Cover net 40 20 0 5 7 24 Codend Cover net 16 8 0 9 11 13 15 17 19 21 23 25 27 29 11 13 15 Anal length (cm) 50 Codend Cover net 40 30 20 10 21 23 25 8 10 12 14 16 18 20 22 Scomberomorusrus... bl)/ [1 ? exp(a ? bl)] Constant Re(l) = constant a -2.2 910 -2.7888 -0.8855 -1. 119 9 -3.2329 b 0.0325 0.00 01 0.000 01 0.00 01 0.0709 c 0 .18 45 0.0580 0.29 21 0.2466 0.2376 L50 (cm)a 70.5 S.R (cm)b 67.6 MLLc -963.06 MLL (full)d -956.94 AICe 19 30 .12 19 30.04 Model deviance 12 .247 14 .16 5 1. 108 1. 108 4.630 4.630 6.4 51 6.4 51 5.492 5.948 Degrees of freedom P value 14 0.586 15 0. 513 5 0.953 6 0.9 81 2 0.099 3 0.2 01 5... Degrees of freedom P value 14 0.586 15 0. 513 5 0.953 6 0.9 81 2 0.099 3 0.2 01 5 0.265 6 0.375 3 0 .13 9 4 0.203 27888.2 88546.2 219 72.2 -964.02 -80 .17 219 722.5 -80 .17 -79. 61 164.34 11 199.3 -12 2. 01 219 72.2 -12 2. 01 -11 9.69 16 2.34 248.02 45.6 -40.77 31. 0 -40.77 -37.55 246.02 85.55 -55 .15 -55.38 -52. 41 83.55 11 4. 31 112 .76 H0: Model fit a Length of 50% retention probability b Selection range defined as l75 (length... identified with automated listening stations Mar Biol 14 6:5 81 594 14 Dagorn L, Holland KN, Hallier JP, Taquet M, Moreno G, Sancho G, Itano DG, Aumeeruddy R, Girard C, Million J, Fonteneau A 12 3 22 15 16 17 18 19 Fish Sci (2 012 ) 78 :15 –22 (2006) Deep diving behavior observed in yellowfin tuna (Thunnus albacares) Aquat Living Resour 19 :85–88 Dagorn L, Holland KN, Itano DG (2007) Behavior of yellowfin (Thunnus... 57:869–880 22 Schroeder SA, Fulton DC, Currie L, Goeman T (2006) He said, she said: gender and angling specialization, motivations, ethics, and behaviors Hum Dimens Wildl 11 :3 01 315 23 Stoll JR, Ditton RB (2006) Understanding anglers’ willingness to pay under alternative management regimes Hum Dimens Wildl 11 :27–42 Fish Sci (2 012 ) 78 :15 –22 DOI 10 .10 07/s12562- 011 -04 31- y ORIGINAL ARTICLE Fisheries Association... participation J Leis Res 16 :322–329 12 3 Fish Sci (2 012 ) 78 :1 14 13 Claussen SE (19 98) Applied correspondence analysis: an introduction Sage, Thousand Oaks, CA 14 Green RH (19 93) Relating two sets of variables in environmental studies In: Patil GP, Rao CR (eds) Multivariate environmental statistics Elsevier Science, New York, pp 14 9 16 3 15 Ghertsos K, Luczak C, Dauvin JC (20 01) Identification of global... highlights Fisheries of the United States 2006 Fisheries Statistics Division, NMFS-F/STI 4 Silver Spring, MD 2 Terceiro M (2002) The summer flounder chronicles: science, politics, and litigation, 19 75–2000 Rev Fish Biol Fisher 11 :12 5 16 8 3 Terceiro M (2006) Stock assessment of summer flounder in 2006 NEFSC Ref Doc 06 -17 Northeast Fisheries Science Center, Gloucester, MA 12 3 14 4 Bochenek EA, Powell EN, DePersenaire... (&) Á K Lee Á S.-H Kim Fisheries System Engineering Division, Fundamental Research Department, National Fisheries Research and Development Institute, Busan 619 -70 5, Korea e-mail: cdpark1@nfrdi.go.kr Y Fujimori Graduate School of Fisheries Sciences, Hokkaido University, 3 -1- 1 Minato, Hakodate, Hokkaido 0 41- 8 61 1, Japan Introduction The recent increase in the population of jellyfish, especially giant jellyfish