Technical ReportPDF Available

The Quantification of Illegal, Unreported and Unregulated (IUU) Fishing in the Pacific Islands Region - a 2020 Update. A report prepared for the Pacific Islands Forum Fisheries Agency

Authors:
  • MRAG Asia Pacific
The Quantification of Illegal, Unreported and Unregulated (IUU)
Fishing in the Pacific Islands Region a 2020 Update
A report prepared for the Pacific Islands Forum Fisheries Agency
October 2021
About MRAG Asia Pacific
MRAG Asia Pacific is an independent fisheries and aquatic resource consulting company dedicated to the sustainable use
of natural resources through sound, integrated management practices and policies. We are part of the global MRAG group
with sister companies in Europe, North America and the Asia Pacific.
2/29 Woodstock Rd PO Box 732 P: +61 7 3371 1500
Toowong Qld 4066 Toowong Qld 4066 F: +61 7 3100 8035
Australia Australia E: info@mragasiapacific.com.au
This study was commissioned by the Pacific Islands Oceanic Fisheries Management Project II which is implemented by
the Pacific Island Forum Fisheries Agency and managed by the United National Development Programme and the Food
and Agriculture Organisation of the United Nations with funding from the Global Environment Facility.
This report was prepared by D. Souter, J. Lowe, C. Dixon, J. Potts, R. Banks and F. Blaha on behalf of MRAG Asia
Pacific. The views in the report represent those of the authors and do not necessarily represent the views of the above
organisations.
Suggested citation:
MRAG Asia Pacific (2021). The Quantification of Illegal, Unreported and Unregulated (IUU) Fishing in the Pacific Islands
Region a 2020 Update. 125 p.
Acknowledgements
Undertaking a study of this type requires the collection of data, information and insights from a wide range of people and
organisations. Particular thanks go staff from FFA member fisheries and maritime surveillance agencies who gave
generously of their time and knowledge during interviews, as well as providing important national-level information. Very
special thanks go to Peter Williams and Andrew Hunt at SPC who worked through innumerable data requests with patience
and good humour, as well as staff at the FFA Secretariat (‘Ana Taholo, Damian Johnston, Yohni Fepuleai, Ramesh Chand,
Bryan Scott) for providing regional level MCS data and insights. A big thanks also go to Tim White and Charlie Kilgour at
Global Fishing Watch who generously provided their AIS data and R code, and the WCPFC Secretariat who provided their
Transhipment Declaration data. Thanks also go to staff at FFA, SPC and PNAO for reviewing a draft of the report. Finally,
a very big thanks goes to Hugh Walton and Ana Taholo at FFA for overseeing the work.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
i
Contents
LIST OF FIGURES ..............................................................................................................................................III
LIST OF TABLES .............................................................................................................................................. IV
ACRONYMS AND ABBREVIATIONS ................................................................................................................. VI
EXECUTIVE SUMMARY .................................................................................................................................. VII
1 INTRODUCTION .................................................................................................................................... 1
2 APPROACH USED IN THIS STUDY ............................................................................................................. 2
2.1 APPROACHES TO ESTIMATING IUU FISHING ..................................................................................................... 2
2.2 OVERALL APPROACH ................................................................................................................................... 2
2.2.1 Identifying IUU risks .......................................................................................................................... 3
2.2.2 Determining ‘best estimate’ and minimum/maximum range .......................................................... 3
2.2.3 Assigning likely probability distribution ............................................................................................ 3
2.2.4 Monte Carlo simulations ................................................................................................................... 4
2.2.5 Quantifying ex-vessel value .............................................................................................................. 5
2.2.6 Changes to the 2020 model .............................................................................................................. 5
2.3 STUDY PERIOD ........................................................................................................................................... 5
2.4 INFORMATION COLLECTION, PLANNING AND GROUND-TRUTHING ......................................................................... 6
2.4.1 Planning workshop ........................................................................................................................... 6
2.4.2 Data collection .................................................................................................................................. 6
2.4.3 Ground-truthing workshop ............................................................................................................... 7
2.4.4 Out of scope issues ............................................................................................................................ 7
2.5 APPROACH TO QUANTIFYING INDIVIDUAL RISKS ................................................................................................ 7
2.5.1 Unlicensed fishing ............................................................................................................................. 8
2.5.2 Misreporting and non-reporting ..................................................................................................... 21
2.5.3 Non-compliance with other license conditions ............................................................................... 35
2.5.4 Post-harvest risks ............................................................................................................................ 43
3 ESTIMATES OF THE VOLUME AND VALUE OF IUU FISHING .................................................................... 55
3.1 OVERALL ESTIMATES ................................................................................................................................. 55
3.1.1 By risk type ...................................................................................................................................... 55
3.1.2 By sector ......................................................................................................................................... 56
3.1.3 By species ........................................................................................................................................ 57
3.2 PURSE SEINE FISHERY ................................................................................................................................ 59
3.2.1 Overall ............................................................................................................................................. 59
3.2.2 Unlicensed/unauthorised fishing .................................................................................................... 60
3.2.3 Misreporting ................................................................................................................................... 60
3.2.4 Non-compliance with other license conditions ............................................................................... 62
3.2.5 Post-harvest risks ............................................................................................................................ 63
3.3 TROPICAL LONGLINE FISHERY...................................................................................................................... 63
3.3.1 Overall ............................................................................................................................................. 63
3.3.2 Unlicensed/unauthorised fishing .................................................................................................... 64
3.3.3 Misreporting ................................................................................................................................... 65
3.3.4 Non-compliance with other license conditions ............................................................................... 67
3.3.5 Post-harvest risks ............................................................................................................................ 68
3.4 SOUTHERN LONGLINE FISHERY .................................................................................................................... 68
3.4.1 Overall ............................................................................................................................................. 68
3.4.2 Unlicensed/unauthorised fishing .................................................................................................... 70
3.4.3 Misreporting ................................................................................................................................... 70
3.4.4 Non-compliance with other license conditions ............................................................................... 72
3.4.5 Post-harvest risks ............................................................................................................................ 72
3.5 DOUBLE COUNTING .................................................................................................................................. 73
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
ii
4 WHAT ARE THE ‘REAL’ COSTS AND BENEFITS OF IUU FISHING? ............................................................. 74
4.1 WHAT IS THE REAL IMPACT ON PACIFIC ISLAND ECONOMIES? ............................................................................ 74
4.2 WHAT IS THE BENEFIT OF IUU FISHING TO VESSELS?........................................................................................ 76
5 ANALYSIS AND MAIN MESSAGES ........................................................................................................... 77
6 WHAT ADDITIONAL MEASURES CAN BE TAKEN TO BETTER UNDERSTAND AND ELIMINATE IUU FISHING?
83
6.1 LONGLINE ............................................................................................................................................... 83
6.1.1 Strengthening catch monitoring ..................................................................................................... 83
6.1.2 Transhipment regulation and monitoring ....................................................................................... 84
6.1.3 Non-compliance with other license conditions ............................................................................... 84
6.1.4 Unlicensed fishing ........................................................................................................................... 85
6.2 PURSE SEINE............................................................................................................................................ 85
6.2.1 Catch verification ............................................................................................................................ 85
6.2.2 FAD tracking and management ...................................................................................................... 86
7 REFERENCES .......................................................................................................................................... 87
ANNEX 1: TERMS OF REFERENCE .............................................................................................................................. 91
ANNEX 2: EX-VESSEL VALUES ................................................................................................................................... 95
ANNEX 3: IUU ACTIVITY DESCRIPTIONS .................................................................................................................... 96
ANNEX 4: WCPFC TRANSHIPMENT DECLARATION/AIS MATCHING PROCEDURE ............................................................... 98
ANNEX 5: POSSIBLE ADDITIONAL MEASURES TO STRENGTHEN MCS ARRANGEMENTS ........................................................ 104
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
iii
List of Figures
FIGURE 1: EXAMPLE DISTRIBUTIONS OF PROBABILITY ASSIGNED TO IUU ACTIVITY. TRIANGULAR DISTRIBUTIONS WERE USED WHERE IT
WAS MORE LIKELY THE ACTUAL LEVEL OF IUU ACTIVITY WAS CLOSER TO THE BEST ESTIMATE THAN EITHER THE MINIMUM OR
MAXIMUM VALUES. UNIFORM DISTRIBUTIONS WERE USED WHERE THE INFORMATION BASE WAS HIGHLY UNCERTAIN. ............ 4
FIGURE 2: EXAMPLE PROBABILITY DISTRIBUTION OUTPUTS FROM MONTE CARLO SIMULATIONS USING @RISK SOFTWARE. ............... 4
FIGURE 3: PROPORTION OF TOTAL FFA COMPLIANCE INDEX (CI) RATINGS BY SECTOR, 2017-2019, BASED ON A RANDOM SAMPLE OF
18,890 INDIVIDUAL CI RATINGS. ........................................................................................................................... 9
FIGURE 4: VMS POSITION DATA SAMPLED FOR VESSELS ON THE FFA REGISTER IN 2018. .......................................................... 9
FIGURE 5: COMPARISON BETWEEN NUMBER OF PURSE SEINE VESSELS BY FLAG ON THE FFA VR AND THE WCPFC RFV (AS AT
NOVEMBER/DECEMBER, 2018). NOTE THAT THESE NUMBERS DO NOT TAKE INTO ACCOUNT CHARTERING ARRANGEMENTS. . 13
FIGURE 6: COMPARISON BETWEEN NUMBER OF LONGLINE VESSELS BY FLAG ON THE FFA VR AND THE WCPFC RFV (AS AT
NOVEMBER/DECEMBER, 2018). NOTE THAT THESE NUMBERS DO NOT TAKE INTO ACCOUNT CHARTERING ARRANGEMENTS. . 14
FIGURE 7: DISTRIBUTION OF CATCHES OF SKJ BY SET TYPE IN THE IATTC AREA, 2019 (LEFT PANEL; IATTC, 2020); WCPFC/IATTC
BOUNDARIES (RIGHT PANEL). ............................................................................................................................... 16
FIGURE 8: AN EXAMPLE OF AN UNLICENSED GROUP SEINE OPERATION DETECTED IN PALAUS EEZ (SOURCE: PALAU MARITIME
OPERATIONS CENTRE) ....................................................................................................................................... 17
FIGURE 9: DISTRIBUTION OF AVERAGE ANNUAL CATCHES OF BET AND YFT IN THE IATTC AREA BY CHINESE, JAPANESE, KOREAN, AND
CHINESE TAIPEI LONGLINE VESSELS, 2014-2018 (IATTC, 2020). ............................................................................. 18
FIGURE 10: PROPORTION OF LONGLINE VESSELS LISTED ON THE IATTC REGIONAL VESSEL REGISTER, WHICH ARE ALSO LISTED ON THE
WCPFC RFV FOR FLEETS >2 VESSELS (AS AT APRIL, 2021). ..................................................................................... 18
FIGURE 11: MOTHERSHIP WITH ONE-PERSON PAKURAS DETECTED IN PALAUS EEZ (SOURCE: PALAU MARITIME OPERATIONS
CENTRE). ........................................................................................................................................................ 19
FIGURE 12: COMPARISONS OF WEIGHTS RECORDED IN LOGSHEETS AND OBSERVER REPORTS FOR INDIVIDUAL TARGET TUNA SPECIES
AND TOTAL WEIGHT AT THE TRIP LEVEL IN THE PURSE SEINE FISHERY, 2017-2019. (N=1930; BLACK LINES SHOW A 1:1
RELATIONSHIP BETWEEN LOGSHEET AND OBSERVER ESTIMATES. RED DATA POINTS ARE OUTSIDE OF THE 10% TOLERANCE
LEVEL.) ........................................................................................................................................................... 23
FIGURE 13: DECISION RULES USED TO CATEGORISE REPORTING OFFENCES IN THE PURSE SEINE FISHERY. ...................................... 24
FIGURE 14: COMPARISON BETWEEN NUMBERS OF TARGET TUNA SPECIES AND SWORDFISH REPORTED IN LOGSHEET (X-AXIS) AND
RECORDED AT UNLOADING (Y-AXIS) FOR MATCHED TRIPS IN THE TLL, 2017-2019. (N= 2513 TRIPS FOR TARGET TUNA SPECIES;
N=576 FOR SWORDFISH; BLACK LINES SHOW 1:1 RELATIONSHIP BETWEEN LOGSHEET AND UNLOADINGS; R = PEARSONS
CORRELATION COEFFICIENT) ................................................................................................................................ 27
FIGURE 15: COMPARISON BETWEEN NUMBERS OF TARGET TUNA SPECIES AND SWORDFISH REPORTED IN LOGSHEET (X-AXIS) AND
RECORDED AT UNLOADING (Y-AXIS) FOR MATCHED TRIPS IN THE SLL, 2017-2019. (N= 3714 TRIPS FOR TARGET TUNA SPECIES;
N=2664 FOR SWORDFISH; BLACK LINES SHOW 1:1 RELATIONSHIP BETWEEN LOGSHEET AND UNLOADINGS; R = PEARSONS
CORRELATION COEFFICIENT) ................................................................................................................................ 27
FIGURE 16: UNREPORTED FROZEN BET DETECTED BY A NEW ZEALAND HIGH SEAS BOARDING, 2017 (NZ, 2017). ...................... 28
FIGURE 17: COMPARISON BETWEEN NUMBERS OF OTHER SPECIES REPORTED IN LOGSHEET AND UNLOADINGS FOR MATCHED TRIPS IN
THE TLL (LEFT SIDE; N= 990 TRIPS) AND SLL (RIGHT SIDE; N=3665 TRIPS), 2017-2019. (BLACK LINES SHOW 1:1
RELATIONSHIP BETWEEN LOGSHEET AND UNLOADINGS; R = PEARSONS CORRELATION COEFFICIENT) .................................. 30
FIGURE 18: COMPARISON BETWEEN WEIGHT OF TARGET TUNA SPECIES AND SWORDFISH REPORTED IN LOGSHEET (X-AXIS) AND
RECORDED AT UNLOADING (Y-AXIS) FOR MATCHED TRIPS IN THE TLL, 2017-2019. (N= 2513 TRIPS FOR TARGET TUNA SPECIES;
N=576 FOR SWORDFISH; BLACK LINES SHOW 1:1 RELATIONSHIP BETWEEN LOGSHEET AND UNLOADINGS) NOTE THAT ONE
OUTLIER DATA POINT HAS BEEN REMOVED FROM EACH OF THE YFT AND BET GRAPHS AND THREE OUTLIERS HAVE BEEN
REMOVED FROM THE SWO GRAPH TO IMPROVE CLARITY. ......................................................................................... 32
FIGURE 19: COMPARISON BETWEEN WEIGHT OF TARGET TUNA SPECIES AND SWORDFISH REPORTED IN LOGSHEET (X-AXIS) AND
RECORDED AT UNLOADING (Y-AXIS) FOR MATCHED TRIPS IN THE SLL, 2017-2019. (N= 3714 TRIPS FOR TARGET TUNA SPECIES;
N=2664 FOR SWORDFISH; BLACK LINES SHOW 1:1 RELATIONSHIP BETWEEN LOGSHEET AND UNLOADINGS) NOTE THAT ONE
OUTLIER DATA POINT HAS BEEN REMOVED FROM THE BET GRAPH TO IMPROVE CLARITY. ................................................. 32
FIGURE 20: COMPARISON OF CATCH RATES FOR SPECIES DISCARDED IN THE TLL SECTOR (N=25 TRIPS). ...................................... 34
FIGURE 21: COMPARISON OF CATCH RATES FOR SPECIES DISCARDED IN THE SLL SECTOR (N=72 TRIPS). ...................................... 34
FIGURE 22: VMS TRACKS SHOWING (A) PURSE SEINE AND (B) LONGLINE VESSEL ACTIVITY ADJACENT TO CLOSED AREAS IN 2017-
2019. ............................................................................................................................................................ 37
FIGURE 23: FATE OF SHARKS IN THE PURSE SEINE SECTOR (TOP PANEL) AND THE LONGLINE SECTORS (BOTTOM PANEL) AS REPORTED BY
FFA MEMBER OBSERVERS, 1995-2019 (DATA SOURCE: SPC). PURSE SEINE OBSERVER COVERAGE HAS VARIED OVER THIS TIME
SERIES, BUT HAS BEEN ~ 100% FROM 2010 ONWARDS. IN THE LONGLINE SECTORS, OBSERVER COVERAGES RATES HAVE BEEN
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
iv
VERY LOW OVER THIS TIMESERIES, PARTICULARLY DURING 2009-2015 WHEN MORE OBSERVER EFFORT WAS DIRECTED
TOWARDS THE PURSE SEINE SECTOR. TO THAT END, CAUTION IS REQUIRED WHEN INTERPRETING THE LONGLINE SECTOR DATA.
..................................................................................................................................................................... 39
FIGURE 24: DISTRIBUTION OF AIS DETECTED ENCOUNTERS/LOITERS WITHIN THE STUDY AREA, 2017-2019 (DATA SOURCE: GFW) 45
FIGURE 25: GEOGRAPHIC DISTRIBUTION OF HIGH SEAS LONGLINE TRANSHIPMENT EVENTS WITHIN THE STUDY AREA REPORTED TO THE
WCPFC, 2017-2019. (DATA SOURCE: WCPFC) .................................................................................................. 46
FIGURE 26: REPORTED WCPFC TRANSHIPMENTS ABLE TO BE SUCCESSFULLY MATCHED TO AIS ENCOUNTERS/LOITERING EVENTS. ... 47
FIGURE 27: GEOGRAPHIC DISTRIBUTION OF REPORTED WCPFC TRANSHIPMENT EVENTS UNABLE TO BE MATCHED TO AIS EVENTS.
NOTE, TRANSHIPMENTS ABOVE 20ON ARE OUTSIDE OF THE STUDY AREA. ..................................................................... 48
FIGURE 28: EXAMPLE TRACKS FROM NOMINAL LONGLINE CARRIER (TOP) AND PURSE SEINE CARRIER (BOTTOM). .......................... 49
FIGURE 29: GEOGRAPHIC DISTRIBUTION OF AIS ENCOUNTERS/LOITERS UNABLE TO BE MATCHED TO REPORTED WCPFC
TRANSHIPMENTS. .............................................................................................................................................. 50
FIGURE 30: DURATION OF AIS ENCOUNTERS MATCHED TO REPORTED WCPFC TRANSHIPMENT EVENTS. .................................... 51
FIGURE 31: CONTRIBUTION OF EACH RISK CATEGORY TO TOTAL ESTIMATED IUU (A) VOLUME AND (B) VALUE IN IN PACIFIC ISLANDS
REGION TUNA FISHERIES. .................................................................................................................................... 56
FIGURE 32: CONTRIBUTION OF IUU (A) VOLUME AND (B) VALUE IN PACIFIC ISLANDS REGION TUNA FISHERIES BY MAIN SECTOR. ..... 57
FIGURE 33: PROPORTION OF EACH MAIN SPECIES IN THE OVERALL ESTIMATES OF IUU (A) VOLUME AND (B) VALUE IN PACIFIC ISLANDS
REGION TUNA FISHERIES. .................................................................................................................................... 58
FIGURE 34: CONTRIBUTION OF EACH RISK CATEGORY TO TOTAL ESTIMATED IUU VOLUMES IN THE PURSE SEINE SECTOR. ................ 59
FIGURE 35: TOTAL ESTIMATED VOLUME OF EACH SPECIES INVOLVED IN IUU ACTIVITY IN THE PURSE SEINE SECTOR. ....................... 59
FIGURE 36: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU UNLICENSED/UNAUTHORISED FISHING VOLUMES IN THE PS
SECTOR. .......................................................................................................................................................... 60
FIGURE 37: PROPORTION OF EACH RISK TYPE/SPECIES CATEGORY TO TOTAL (A) VOLUME AND (B) VALUE OF MISREPORTING IN THE
PURSE SEINE SECTOR. (UR = UNDERREPORTED; OR = OVERREPORTED; MISIDENT. = MISIDENTIFIED) ................................. 62
FIGURE 38: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU VALUE ($) ASSOCIATED WITH THE NON-COMPLIANCE WITH
LICENSE CONDITIONS RISKS IN THE PS SECTOR. ....................................................................................................... 62
FIGURE 39: CONTRIBUTION OF EACH RISK CATEGORY TO TOTAL ESTIMATED IUU VOLUMES IN THE TLL SECTOR. ........................... 64
FIGURE 40: TOTAL ESTIMATED VOLUME OF EACH SPECIES INVOLVED IN IUU ACTIVITY IN THE TLL SECTOR. .................................. 64
FIGURE 41: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU VOLUMES IN THE TLL SECTOR. .......................................... 65
FIGURE 42: CONTRIBUTION OF EACH SPECIES/FATE COMBINATION TO TOTAL (A) VOLUME AND (B) VALUE OF MISREPORTING IN THE
TLL SECTOR. .................................................................................................................................................... 67
FIGURE 43: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU VALUE ($) ASSOCIATED WITH THE NON-COMPLIANCE WITH
LICENSE CONDITIONS RISKS IN THE TLL SECTOR. ...................................................................................................... 67
FIGURE 44: CONTRIBUTION OF EACH RISK CATEGORY TO TOTAL ESTIMATED IUU VOLUMES IN THE SLL SECTOR. ........................... 69
FIGURE 45: TOTAL ESTIMATED VOLUME OF EACH SPECIES INVOLVED IN IUU ACTIVITY IN THE SLL SECTOR. .................................. 69
FIGURE 46: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU VOLUMES IN THE SLL SECTOR. .......................................... 70
FIGURE 47: CONTRIBUTION OF EACH SPECIES/FATE COMBINATION TO TOTAL (A) VOLUME AND (B) VALUE OF MISREPORTING IN THE
SLL SECTOR. .................................................................................................................................................... 71
FIGURE 48: CONTRIBUTION OF EACH RISK TO TOTAL ESTIMATED IUU VALUE ($) ASSOCIATED WITH THE NON-COMPLIANCE WITH
LICENSE CONDITIONS RISKS IN THE SLL SECTOR. ...................................................................................................... 72
List of Tables
TABLE 1: BEST ESTIMATE AND MIN/MAX RANGE FOR UNLICENSED FISHING ACTIVITY BY VESSELS ON THE FFA VR BY SIZE AND SECTOR
(BY AVERAGE NUMBER OF DAYS FISHING PER YEAR). PERCENTAGES IN PARENTHESES REPRESENT THE PROPORTION OF THE TOTAL
AVERAGE FISHING DAYS BY RELEVANT VESSELS IN THAT SECTOR. .................................................................................. 11
TABLE 2: BEST ESTIMATE AND MIN/MAX RANGE FOR UNLICENSED FISHING ACTIVITY BY VESSELS ON THE WCPFC RFV BUT NOT ON THE
FFA VR BY SIZE AND SECTOR (BY AVERAGE NUMBER OF DAYS FISHING PER YEAR). .......................................................... 15
TABLE 3: LEVEL OF DUAL IATTC/WCPFC AUTHORISATION AMONGST THE MAIN LONGLINE FLEETS OPERATING IN THE PACIFIC (AS AT
APRIL, 2021). ................................................................................................................................................. 18
TABLE 4: LEVEL OF CROSS-AUTHORISATION IN THE WCP-CA BY LONGLINE VESSELS AUTHORISED UNDER CCSBT RECORD OF
AUTHORISED VESSELS (AS AT JUNE 2021). ............................................................................................................ 20
TABLE 5: BEST ESTIMATE AND MIN/MAX RANGE FOR UNREGULATED FISHING ACTIVITY BY VESSEL SIZE AND SECTOR (BY AVERAGE
NUMBER OF DAYS FISHING PER YEAR). ................................................................................................................... 20
TABLE 6: AVERAGE RATES OF MISREPORTING FOR EACH CATEGORY ACROSS SAMPLE TRIPS (IN TONNES PER SET). .......................... 24
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
v
TABLE 7: TOTAL NUMBERS OF FISH REPORTED IN LOGSHEETS (LOG) AND UNLOADINGS REPORTS (UNL) ACROSS 2,513 MATCHED
TRIPS IN THE TLL SECTOR AND 3,714 MATCHED TRIPS IN THE SLL SECTOR. ................................................................... 28
TABLE 8: ESTIMATED UNDER-REPORTING RATES OF KEY LONGLINE SPECIES IN PERCENTAGE OF TOTAL CATCH. ............................... 31
TABLE 9: TOTAL WEIGHT OF FISH (IN METRIC TONNES) REPORTED IN LOGSHEETS (LOG) AND UNLOADINGS REPORTS (UNL) ACROSS
2,513 MATCHED TRIPS IN THE TLL SECTOR AND 3,714 MATCHED TRIPS IN THE SLL SECTOR. ........................................... 33
TABLE 10: RATES OF DISCARDING FOR KEY TARGET AND NON-TARGET SPECIES IN THE WCPO LONGLINE SECTOR. (SOURCE: SPC,
2020; RATES FOR BILLFISH AND OTHER SPECIES CARRIED OVER FROM SPC ADVICE FOR THE 2016 STUDY) .......................... 33
TABLE 11: ESTIMATED UNDER-REPORTING RATES OF KEY LONGLINE SPECIES IN PERCENTAGE OF TOTAL DISCARDS. ........................ 34
TABLE 12: BEST ESTIMATE AND MIN/MAX RANGE FOR ILLEGAL FAD FISHING IN THE PURSE SEINE SECTOR (BY NUMBER OF SETS DURING
THE FAD CLOSURE PERIOD LIKELY TO BE SETS ON FLOATING OBJECTS). ......................................................................... 36
TABLE 13: BEST ESTIMATE AND MIN/MAX RANGE FOR ILLEGAL FISHING ACTIVITY WITHIN CLOSED WATERS INSIDE FFA MEMBER ZONES
(BY NUMBER OF DAYS FISHING PER YEAR). .............................................................................................................. 38
TABLE 14: MIN/MAX PROPORTION OF DISCARDED SHARKS FINNED IN THE LONGLINE SECTORS. ................................................. 41
TABLE 15: BEST ESTIMATE AND MIN/MAX PROPORTION OF SHARKS TAKEN USING NON-PRESCRIBED GEAR. .................................. 42
TABLE 16: ENCOUNTERS AND LOITERING EVENTS DETECTED VIA AIS IN THE STUDY AREA, BY REGION AND DURATION, 2017-2019
(DATA SOURCE: GFW). (NOTE, THE ‘TOPICAL AREA MATCHES THE TLL BOUNDARIES; THE ‘SOUTHERN AREA MATCHES THE
SLL BOUNDARIES) ............................................................................................................................................. 45
TABLE 17: UNMATCHED AIS EVENTS ACCORDING TO SECTOR AND DURATION, 2017-2019. .................................................... 50
TABLE 18: AVERAGE VOLUME OF EACH SPECIES TRANSHIPPED IN EACH TIME BIN IN THE TLL AND SLL SECTORS. NOTE THAT
VOLUMES REPORTED IN DRESSED WEIGHTS (E.G. GILLED AND GUTTED, GUTTED AND HEADED) WERE RAISED TO WHOLE WEIGHTS
USING STANDARD CONVERSION FACTORS. .............................................................................................................. 52
TABLE 19: BEST ESTIMATE AND MIN/MAX RANGE FOR ILLEGAL TRANSHIPPING (BY AVERAGE ANNUAL NUMBER OF ILLEGAL
TRANSHIPMENTS). ............................................................................................................................................. 54
TABLE 20: ESTIMATED TOTAL IUU VOLUMES AND EX-VESSEL VALUE IN PACIFIC ISLANDS REGION TUNA FISHERIES, BY RISK CATEGORY.
..................................................................................................................................................................... 56
TABLE 21: ESTIMATED TOTAL IUU VOLUME AND VALUE IN EACH OF THE MAIN SECTORS. ......................................................... 57
TABLE 22: ESTIMATED TOTAL IUU VOLUMES AND EX-VESSEL VALUE OF EACH MAIN SPECIES IN PACIFIC ISLANDS REGION TUNA
FISHERIES. ....................................................................................................................................................... 58
TABLE 23: ESTIMATED TOTAL IUU VOLUMES IN THE PURSE SEINE SECTOR, BY RISK CATEGORY. .................................................. 59
TABLE 24: ESTIMATED TOTAL IUU VOLUMES IN THE PURSE SEINE SECTOR, BY SPECIES. ............................................................ 59
TABLE 25: ESTIMATED IUU VOLUMES ASSOCIATED WITH UNLICENSED/UNAUTHORISED FISHING IN THE PS SECTOR, BY RISK TYPE. ... 60
TABLE 26: ESTIMATED TOTAL MISREPORTING IN THE PURSE SEINE SECTOR, BY SPECIES AND FATE. .............................................. 61
TABLE 27: ESTIMATED TOTAL VOLUMES OF IUU PRODUCT ASSOCIATED WITH NON-COMPLIANCE WITH OTHER LICENSE CONDITIONS IN
THE PS SECTOR, BY RISK CATEGORY. ...................................................................................................................... 62
TABLE 28: ESTIMATED TOTAL IUU VOLUMES IN THE TLL SECTOR, BY RISK CATEGORY. ............................................................. 64
TABLE 29: ESTIMATED TOTAL IUU VOLUMES IN THE TLL SECTOR, BY SPECIES. ....................................................................... 64
TABLE 30: ESTIMATED TOTAL IUU VOLUMES IN THE TLL SECTOR, BY RISK CATEGORY. ............................................................. 65
TABLE 31: ESTIMATED TOTAL UNDER-REPORTING IN THE TLL SECTOR, BY SPECIES AND FATE. .................................................... 66
TABLE 32: ESTIMATED TOTAL VOLUMES OF IUU PRODUCT ASSOCIATED WITH NON-COMPLIANCE WITH OTHER LICENSE CONDITIONS IN
THE TLL SECTOR, BY RISK CATEGORY. .................................................................................................................... 67
TABLE 33: ESTIMATED TOTAL VOLUMES OF IUU PRODUCT INVOLVED IN POST-HARVEST IUU ACTIVITY IN THE TLL SECTOR. ............ 68
TABLE 34: ESTIMATED TOTAL IUU VOLUMES IN THE SLL SECTOR, BY RISK CATEGORY. ............................................................. 69
TABLE 35: ESTIMATED TOTAL IUU VOLUMES IN THE SLL SECTOR, BY SPECIES. ....................................................................... 69
TABLE 36: ESTIMATED TOTAL IUU VOLUMES IN THE SLL SECTOR, BY RISK CATEGORY. ............................................................. 70
TABLE 37: ESTIMATED TOTAL UNDER-REPORTING IN THE SLL SECTOR, BY SPECIES AND FATE. .................................................... 71
TABLE 38: ESTIMATED TOTAL VOLUMES OF IUU PRODUCT ASSOCIATED WITH NON-COMPLIANCE WITH OTHER LICENSE CONDITIONS IN
THE SLL SECTOR, BY RISK CATEGORY. .................................................................................................................... 72
TABLE 39: ESTIMATED TOTAL VOLUMES OF IUU PRODUCT INVOLVED IN POST-HARVEST IUU ACTIVITY IN THE SLL SECTOR. ............ 73
TABLE 40: ESTIMATED IUU EX-VESSEL VALUES COMPARED WITH POTENTIAL REAL REVENUE FORGONE BY PACIFIC ISLAND COUNTRIES
IN THE FORM OF RENT, OR ECONOMIC PROFIT, BY FLEET SECTOR. ................................................................................ 75
TABLE 41: SUMMARY OF CHANGES IN BEST ESTIMATE IIUU VOLUME BETWEEN THE 2016 AND 2020 STUDIES. ......................... 78
TABLE 42: EX-VESSEL MARKET VALUE BY SPECIES (US$/MT) ............................................................................................. 95
TABLE 43: IUU ACTIVITY DESCRIPTIONS ......................................................................................................................... 96
TABLE 44: POSSIBLE MEASURES TO STRENGTHEN MCS ARRANGEMENTS. ........................................................................... 104
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
vi
Acronyms and abbreviations
AI/ML
Artificial Intelligence/ Machine Learning
MT
Metric tonnes
AIS
Automatic identification system
NGO
Independent non-profit organization
ALB
Albacore tuna
NPM
Net profit margin
ALC
Automatic location communicator
OTH
Other species
BET
Bigeye tuna
PIRFO
Pacific Islands Regional Fisheries Observer
BIL
Billfish
PMSP
Pacific Maritime Security Program
BOBLME
Bay of Bengal Large Marine Ecosystem
PNA
Parties to the Nauru Agreement
CCM
Members and Cooperating Non-members
(of the WCPFC)
PNAO
Parties to the Nauru Agreement Office
CCSBT
Commission for the Conservation of
Southern Bluefin Tuna
PS
Purse seine
CDS
Catch Documentation Scheme
RFMO
Regional Fisheries Management
Organization
Ci
Confidence interval
RFV
WCPFC Record of Fishing Vessels
CI
Compliance Index
RIMF
Regional Information Management Facility
CMM
Conservation and management measure
RMCSS
FFA Regional Monitoring, Control and
Surveillance Strategy
EEZ
Exclusive economic zone
ROP
WCPFC Regional Observer Program
EM
Electronic monitoring
SHK
Sharks
EPO
Eastern Pacific Ocean
SKJ
Skipjack tuna
EU
European Union
SLL
Southern longline fishery
FAD
Fish aggregation device
SOLAS
International Convention for the Safety of
Life at Sea
FAO
UN Food and Agriculture Organisation
SPC
The Pacific Community
FFA
Forum Fisheries Agency
SSI
Species of special interest
FFA VR
FFA Vessel Register
TLL
Tropical longline fishery
FSMA
Federated States of Micronesia
Arrangement
ULT
Ultra Low Temperature
GFW
Global Fishing Watch
USMLT
US Multilateral Treaty
HMTCs
Harmonised Minimum Terms and
Conditions for Foreign Fishing Vessel
Access
UTC
Coordinated Universal Time
HSP
High seas pocket
VDS
Vessel Days Scheme
IATTC
Inter-American Tropical Tuna Commission
VMS
Vessel monitoring system
ICCAT
International Commission for the
Conservation of Atlantic Tunas
WCPFC
Western and Central Pacific Fisheries
Commission
IMO
International Maritime Organization
WCPF-CA
Western and Central Pacific Fisheries
Convention Area
IPOA-IUU
International Plan of Action for IUU
WCPO
Western and Central Pacific Ocean
IUU
Illegal, unreported and unregulated fishing
YFT
Yellowfin tuna
LL
Longline
MARPOL
Marine pollution
MCS
Monitoring, control and surveillance
MCSWG
Forum Fisheries Agency Monitoring,
Control and Surveillance Working Group
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
vii
Executive Summary
Illegal, unreported and unregulated (IUU) fishing is a recognised global problem that
undermines the integrity of responsible fisheries management arrangements and
results in lost value to coastal states (e.g. FAO, 2002; Agnew et al, 2009). The first
attempt at quantifying the value and volume of IUU fishing in tuna fisheries within
the Pacific Islands region was undertaken in 2016 using data from 2010-2015
(MRAG Asia Pacific, 2016). That study estimated the total volume of product either
harvested or transhipped involving IUU activity in Pacific tuna fisheries was
306,440t, with an ex-vessel value of $616.11m. Nevertheless, the authors noted
that the data and information underlying many of the estimates were highly
uncertain and that the outputs should be seen as a ‘first cut’.
In order to assess changes in the nature and extent of IUU fishing since that time,
this study was commissioned as part of the Global Environment Facility-funded
Pacific Islands Oceanic Fisheries Management Project II (OFMP II) to undertake a
‘2020 update’ of the original estimates. Broadly, the aim was to undertake an
‘apples vs apples’ update of the original estimates, using a consistent methodology
and taking into account the latest available information. The study period covered
the years 2017-2019. Importantly, this preceded any COVID-19 related impacts on
monitoring, control and surveillance (MCS) and IUU activity in the region.
Broadly, we used a ‘bottom up’ approach to quantify IUU fishing activity across key
IUU risks in four categories: (i) unlicensed/unauthorised fishing, (ii) misreporting,
(iii) non-compliance with other license conditions (e.g. shark finning) and (iv) post-
harvest risks (e.g. illegal transhipping). ‘Best estimate’ and minimum/maximum
range values were generated for each risk, taking into account the best available
information. Monte Carlo simulation was then used to produce probabilistic
estimates of IUU activity, taking into account probability distributions assigned
within the minimum and maximum range values. Using this approach, estimates of
IUU volume and value were developed for each of the three main fishing sectors -
purse seine (PS), tropical longline (TLL) and southern longline (SLL) and then
aggregated to produce an overall estimate for Pacific Islands region tuna fisheries.
While the same basic approach to estimating IUU was used between the 2016 and
2020 studies, a number of changes were made to the information underlying
estimation of individual risks. In some cases, this was driven by new information
becoming available (e.g. to estimate the scope for illegal transhipment), while in
other cases the information previously used to support estimates for the 2016 study
was no longer available. For some risks, these changes of information had
substantial impacts on the estimated volume and value between studies.
Our simulations suggest the best estimate total annual volume of product either
harvested or transhipped involving IUU activity in Pacific tuna fisheries during the
2017-19 period was 192,186t, with 90% confidence that the actual figure lies
within a range of 183,809t to 200,884t. Based on the expected species
composition and markets, the ex-vessel value of the best estimate figure is
$333.49m. The 90% confidence range is between $312.24m and $358.17m. For
context, the estimated IUU volume figure is around 6.5% of the total WCPFC
Convention Area (WCPFC-CA) catch in 2019.
This result is a considerable reduction from the ‘first cut’ estimates in the 2016 study
of 306,440t (276,546t to 338,475t) with a best estimate value of $616.11m
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
viii
($517.91m to $740.17m). The reduction was primarily driven by substantial
reductions in estimates for illegal transhipping and FAD fishing during the closure
period (in turn driven by the use of better and different information, respectively) as
well as the removal of the ‘unauthorised landings in foreign ports’ risk. Overall
figures were also influenced by changes in fishery dynamics (e.g. catch, effort, price).
Amongst the four categories of risk identified here, the largest contribution to the
overall IUU volume was made by misreporting, accounting for 89% of the total
volume. Importantly, much of this volume was driven by misreporting and
misidentifying target species in the purse seine sector for which challenges exist in
making accurate estimates of catch at sea. The various types of unlicensed fishing
collectively accounted for 5% of overall estimated IUU volume, while non-
compliance with license conditions and post-harvest offences accounted for 3%
each.
Of the three main sectors assessed, estimated volume of IUU product was highest in
the purse seine sector, accounting for 72% of overall volume. Nevertheless, much of
the estimated volume in this sector was driven by estimates for misreporting for
which mechanisms exist (through 100% observer coverage) to correct any errors in
catch reports and, given the nature of access arrangements under the VDS, it is likely
that economic rents associated with any misreporting would be captured anyway.
This result should be seen in that context. The tropical longline and southern
longline sectors accounted for 21% and 7% of the overall volume respectively. The
purse seine fishery also contributed to slightly under half the overall ex-vessel of IUU
product ($152.26m), although the higher market value of target species in the
longline fisheries meant that TLL sector made a proportionally higher contribution
by value (40%) than volume to overall estimates. The southern longline fishery had
the lowest overall estimates of IUU product value (14%).
Of the main target species, yellowfin (YFT) accounted for the highest volume of IUU
product, making up 33% of the total estimated IUU volume, and 25% of the ex-
vessel value. The total estimated IUU volume of YFT equated to around 9.4% of the
estimated total catch of YFT in the WCPFC-CA area during 2019. However, because
much of the YFT volume is driven by misreporting in the purse seine fishery which is
subject to 100% observer coverage, this should not result in ‘unaccounted for’ catch.
Skipjack (SKJ) accounted for the next highest volume, making up around 27% of
overall estimated volume, but only 20% of the overall ex-vessel value given its lower
market price relative to other species. The total estimated IUU volume of SKJ
equated to around 2.5% of the estimated total catch of SKJ in the WCPFC-CA area in
2019. Bigeye (BET) accounted for 17% of the overall estimated IUU volume, but 20%
of the ex-vessel value. The proportionally higher contribution to the ex-vessel value
total reflects the fact that much of the estimated IUU volume came from the
longline sector which achieves relatively high market prices. The total estimated IUU
volume of BET equates to around 24.3% of the estimated total catch of BET in the
WCPFC-CA area during 2019. Importantly, this does not necessarily mean that
24.3% of additional BET have been taken in addition to reported figures. For
example, some of the BET estimates relate to over-reporting in the purse seine
fishery. Albacore (ALB) accounted for 2% of the overall estimated IUU volume and
total ex-vessel IUU value. The total estimated ALB IUU volume equates to around
2.8% of the estimated total ALB catch in the WCPFC-CA area in 2019.
Apart from the headline volume and value estimates, there are a number of key
messages arising from the analysis:
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
ix
The reduction in estimates since 2016 is positive, but should be seen in
context - The overall volume and value of IUU estimated in this 2020 update
are a substantial reduction on those from 2016. Broadly, this is a very
positive result for the region and its MCS efforts, but should be seen in
context. The 2016 estimates were a ‘first cut’ with highly uncertain data
across a number of key risk areas. On that basis, estimates were kept
deliberately broad to account for high levels of uncertainty. For the 2020
study, new information became available to estimate some risks most
notably illegal transhipping and longline misreporting while information
previously used to quantify risks for the 2016 study were not available for
the current study period. Broadly, it was these changes in information base
that produced the biggest overall changes in volume and value estimates. In
addition, incorporating one new risk (exceeding effort limits) and removing
another (unauthorised landing of catch in foreign ports) together with
changes in fishing effort, catch rates and fish price also influenced overall
estimates. In practice, the 2020 estimates should be seen as the next
evolution in an ongoing process to refine approaches to quantify the nature
and scale of IUU in the Pacific region;
Cooperation works - While IUU fishing in its various guises will require
ongoing attention from FFA members, there is little doubt that the MCS
measures FFA members and their partners/regional secretariats have
implemented over recent decades have had a profound impact on both the
nature and volume of IUU fishing in the region. Cooperative regional MCS
measures such as the establishment of the FFA Vessel Register and Good
Standing requirement, the agreement of Harmonised Minimum Terms and
Conditions (HMTCs) for foreign fishing vessel access, the establishment of
the FFA Vessel Monitoring System (VMS), the development of common
regional data collection protocols and forms, the establishment of regional
Pacific Island Regional Fisheries Observer (PIRFO) standards and training for
observers, the Niue Treaty and Subsidiary Arrangement to facilitate
cooperation on MCS including information sharing and coordinated Regional
Operations, amongst others, have substantially strengthened the MCS
environment across all member zones compared to individual members
acting alone. The relatively low estimates of IUU activity in the FFA region
compared to many other parts of the world is practical evidence of the MCS
framework’s success;
Estimates continue to be dominated by the licensed fleet - A key outcome
of the 2016 study was that estimates of IUU volume and value were
dominated by the licensed fleet. The 2020 update shows a similar pattern
with unlicensed fishing accounting for only 5% of overall IUU activity;
Unlicensed fishing remains an issue at the margins - unlicensed fishing
continued to be an issue at the margins, both figuratively and literally.
Overall, evidence for unlicensed fishing by vessels on the FFA VR and/or
WCPFC RFV was very limited with no confirmed instances of unlicensed
fishing by these vessels detected during regional operations and few
national level detections/prosecutions during the study period. The main
exception to this is on the fringes of the FFA region, and in particular on the
western fringe adjacent to the domestic fleets of south east Asian countries,
where evidence of regular incursions was stronger;
Priorities for strengthening MCS measures are in the longline sectors
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
x
Of the two main gear types operating in the Pacific Islands region, the purse
seine fleet is subject to comparatively very strong MCS arrangements
including 100% observer coverage, a requirement to tranship in port and a
requirement for e-reporting under the Parties to the Nauru Agreement’s
(PNA) Vessel Days Scheme (VDS). Moreover, the majority of fishing effort
occurs in EEZs subject to strong coastal State MCS. In contrast, MCS
arrangements in place for the longline sector are weaker with lower
observer coverage, a far higher proportion of effort on the high seas, and a
higher proportion of the catch transhipped at sea which limits opportunities
for port State MCS measures. Particular focus should be on strengthening
measures to monitor and validate catch both on longline vessels and as it
moves through the supply chain. Given the shared nature of stocks in the
region, it is important that strong catch validation measures are applied
across the full footprint of stocks, including on the high seas;
Estimates of illegal transhipping have come down, but monitoring and
control remain a work in progress - The availability of WCPFC Transhipment
Declaration information together with Global Fishing Watch’s (GFW)
Automatic Identification System (AIS) dataset has provided considerably
better information on the scope for unauthorised transhipment than was
available to the 2016 study. Broadly, this has led to a substantial reduction
in overall estimates of volume and value. Nevertheless, important areas of
uncertainty remain in the at sea transhipment component of the longline
supply chain and monitoring and control remain a work in progress. In
particular, improvements are required to strengthen the implementation of
the observer program such that information provided by vessels on the
volume and species composition of fish transhipped can be validated against
independent observer estimates;
‘IUU’ is not straightforward while the formal definition of ‘IUU fishing’ in
the IPOA-IUU is relatively clear in theory, applying it for the purposes of
quantifying its nature and extent presents a range of practical challenges. In
addition to the inevitable uncertainties in the underlying data, resolving
what should, and shouldn’t, be considered in estimates frequently requires
judgements that can have a large impact on overall volume and value
figures;
Ex-vessel value is not a good indicator of actual loss to FFA members this
is because the full value of the catch is not returned to coastal states under
normal circumstances (only a proportion of total revenue is, typically
through access fees). A better benchmark of revenue forgone by Pacific
Island countries is likely to be the rent generated by vessels from IUU
activity, however even then the nature of access arrangements such as the
VDS mean that economic rents associated with many IUU activities (e.g.
misreporting) is likely to be captured anyway. Taking into account estimates
of profitability during the study period in the purse seine and longline
sectors, as well as the likelihood that rents associated with some risks
(notably misreporting in the purse seine sector) are likely to be captured
through the VDS, we estimate the rent associated with ex-vessel IUU value
to be $43.18m. This is a considerable reduction on the 2016 estimate
($152.67m), but may still overestimate actual loss. More accurate estimates
would require additional analysis of the unique circumstances of each IUU
risk.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
xi
As outlined in the 2016 study, considerable efforts have been taken at the national,
sub-regional (FFA/SPC/PNA) and regional levels (WCPFC) to mitigate IUU fishing in
Pacific tuna fisheries. Moreover, a range of additional MCS measures have been
taken since then (e.g. establishment of the Pacific Maritime Security Program -
PSMP, strengthening of longline unloadings monitoring coverage in FFA member
ports) which have better informed the 2020 update estimates and contributed to
the lower overall estimates.
Nevertheless, ongoing uncertainties in relation to a number of key risk areas
highlight priority areas for future MCS development. In the longline sectors, the
priority is to strengthen measures to monitor and validate catch of licensed vessels
throughout the supply chain. Despite good improvements in some areas (e.g.
unloadings coverage in FFA ports), current monitoring arrangements remain limited
for some fleets. Measures that could be taken to strengthen monitoring include
strengthening observer coverage (for those longline fleets not meeting the 5%
WCPFC benchmark, as well as FFA domestic fleets), more active cross-verification of
independent data sources to identify reporting discrepancies (e.g. logsheet Vs
unloading, etc), an enhanced focus on investigating reporting offences, wider use of
electronic reporting and monitoring, and the development of an effective catch
documentation scheme (CDS) for key species. In addition, more effective monitoring
and control of at-sea transhipment is required including strengthening
arrangements for the implementation of the transhipment observer program.
In the purse seine sector, notwithstanding recent complications arising from COVID-
19 restrictions, the MCS arrangements in place are considerably stronger than those
for longline. Priorities include continuing efforts to validate estimates of catch
composition and monitoring and control of FAD usage.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
1
1 Introduction
Illegal, unreported and unregulated (IUU) fishing is a recognised global problem that undermines the
integrity of responsible fisheries management arrangements and results in lost value to coastal
states (e.g. FAO, 2002; Agnew et al, 2009). Previous studies have shown that the effects of IUU
fishing are often hardest felt in developing coastal states heavily reliant on fishing for income (e.g.
MRAG, 2005).
Quantifying the nature and extent of IUU fishing is important in gauging potential losses suffered by
coastal states, addressing uncertainties in stock assessments and planning effective monitoring
control and surveillance (MCS) responses. However, by its very nature IUU fishing is secretive and
difficult to estimate with accuracy (FAO, 2002; Le Gallic and Cox, 2006).
The first attempt at quantifying the value and volume of IUU fishing in tuna fisheries within the
Pacific Islands region
1
was undertaken in 2016 using data from 2010-2015 (MRAG Asia Pacific, 2016).
That study estimated the total volume of product either harvested or transhipped involving IUU
activity in Pacific tuna fisheries was 306,440t, with an ex-vessel value of $616.11m. Nevertheless,
the authors noted that the data and information underlying many of the estimates were highly
uncertain and that the outputs should be seen as a ‘first cut’. A key output from the study was the
development of a framework for the estimation of IUU in the region that can be updated over time
as information improves and circumstances change.
In order to assess changes in the nature and extent of IUU fishing since that time, this study was
commissioned as part of the Global Environment Facility-funded Pacific Islands Oceanic Fisheries
Management Project II (OFMP II) to undertake a ‘2020 update’ of the original estimates. Broadly, the
aim was to undertake an ‘apples vs apples’ update of the original estimates, using a consistent
methodology and taking into account the latest available information. The Terms of Reference for
the study are included at Annex 1. Consistent with the 2016 study, we have adopted the definition
of IUU fishing set out in the FAO International Plan of Action (IPOA-IUU) (FAO, 2001) (Box 1).
Broadly, the report is organised into six main parts. Following this introduction, section 2 sets out
the overall approach used for the study, as well as the data and information used to quantify each
individual IUU activity. Section 3 provides estimates of the volume and value of IUU fishing for each
main IUU activity within the three main fishing sectors examined (purse seine, tropical longline,
southern longline). Section 4 looks at the estimated actual costs of IUU fishing to Pacific Island
economies, while section 5 sets out the main messages arising from the analysis. Finally, section 6
looks at additional measures that may be taken to further reduce IUU fishing in the Pacific Islands
region.
BOX 1: WHAT IS IUU FISHING?
Illegal fishing refers to fishing activities:
(1) conducted by national or foreign vessels in waters under the jurisdiction of a State, without
the permission of that State, or in contravention of its laws and regulations;
(2) conducted by vessels flying the flag of States that are parties to a relevant regional fisheries
management organization but operate in contravention of the conservation and management
measures adopted by that organization and by which the States are bound, or relevant provisions
of the applicable international law; or
1
The Pacific Islands Region is considered to comprise the EEZs of FFA’s 15 Pacific Island member countries and
adjacent high seas areas in the tropics.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
2
(3) in violation of national laws or international obligations, including those undertaken by
cooperating States to a relevant regional fisheries management organization.
Unreported fishing refers to fishing activities:
(1) which have not been reported, or have been misreported, to the relevant national authority, in
contravention of national laws and regulations; or
(2) undertaken in the area of competence of a relevant regional fisheries management
organization which have not been reported or have been misreported, in contravention of the
reporting procedures of that organization.
Unregulated fishing refers to fishing activities:
(1) in the area of application of a relevant regional fisheries management organization that are
conducted by vessels without nationality, or by those flying the flag of a State not party to that
organization, or by a fishing entity, in a manner that is not consistent with or contravenes the
conservation and management measures of that organization; or
(2) in areas or for fish stocks in relation to which there are no applicable conservation or
management measures and where such fishing activities are conducted in a manner inconsistent
with State responsibilities for the conservation of living marine resources under international law.
2 Approach used in this study
2.1 Approaches to estimating IUU fishing
Approaches to quantifying IUU fishing can generally be grouped into two categories ‘top down’
and ‘bottom up’ (see for example, MRAG, 2005; FAO, 2018). Top-down approaches typically use a
fixed proportion (or range) of the catch, which is estimated to be IUU to arrive at an overall estimate
of IUU catch volume and value. For example, Pauly & McLean (2003) provide estimates of
unreported catch as a proportion of the total global reported catch in the range of 25-30%. Top-
down approaches are convenient in that they can be applied to produce direct global or regional
estimates of IUU catch, but should be applied with caution, given the nature and extent of IUU
fishing may vary substantially from country to country, region to region and fishery to fishery.
The bottom-up approach involves analysis of more detailed information at a local scale in an effort
to build a more accurate picture of IUU fishing activity and particularly the variation in vulnerability
to such activity from state to state, or fishery to fishery. Estimates obtained in this way are added
together to develop an overall estimate of IUU catch. The challenge with this approach is that it is
time consuming and information is often very patchy and hard to collect. There are therefore many
gaps to fill that require analytical methodologies of varying degrees of complexity. Even when these
are used, it is still possible that some types of IUU catches will be missed, and also that some may be
‘double-counted’. Nevertheless, depending on the nature and level of information available,
bottom-up approaches arguably have the potential to provide more accurate estimates of IUU
activity.
2.2 Overall approach
The overall approach we used for this study was the same as that used in MRAG Asia Pacific (2016).
Broadly, we used a bottom-up approach, which aimed to arrive at regional-scale estimates of the
volume and value of IUU fishing by first breaking down the ‘IUU problem’ into discrete quantifiable
units, and then aggregating these up to produce a regional scale estimate. The approach is similar in
part to the ‘anchor points’ approach described in Ainsworth and Pitcher (2005) (and later used by
Agnew et al, 2009, for their global IUU study) in that we assigned ‘best estimates’ and minimum and
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
3
maximum (‘min/max’) ranges of known IUU activities, and then used Monte Carlo simulations to
determine the likelihood that IUU fishing would be within a certain range. However, the approach
differs in that the intent is to estimate a ‘typical’ annual level of IUU activity within a specified
timeframe, rather than produce a historical timeseries.
The approach used is consistent with advanced copies of volumes released to date of FAO’s
Technical guidelines on methodologies and indicators for the estimation of the magnitude and
impact of illegal, unreported and unregulated (IUU) fishing (FAO, 2018, 2021).
Generating estimates of the volume, species composition and value of IUU fishing involved five main
steps:
2.2.1 Identifying IUU risks
The first step in the analysis involved identifying the main IUU risks (e.g. unlicensed fishing,
misreporting, illegal transhipping, etc.) in each of the main fishing sectors (purse seine - PS, tropical
longline - TLL, southern longline - SLL). These were discussed at the project Planning Workshop (see
2.4.1) and largely carried over from the 2016 study, although one IUU risk was added (exceeding
effort limits) and one was removed (landing catch in unauthorised foreign ports). Exceeding effort
limits was added because evidence was available from the current study period that regionally
agreed quantitative effort limits had been exceeded. Landing catch in unauthorised foreign ports
was removed because the underlying nature of landing requirements were often either opaque,
difficult to access (e.g. within confidential state agreements) and/or not actively enforced.
2.2.2 Determining ‘best estimate’ and minimum/maximum range
The next step involved identifying the information available to support estimates of IUU activity, and
then using that information to determine a ‘best estimate’ level of activity and the most plausible
min/max values. As in the 2016 study, the quality and nature of the information available varied
considerably between risks. Relatively precise ‘best estimates’ could be assigned to some risks
based on the nature of the available information (e.g. misreporting in the purse seine fishery which
is subject to 100% observer coverage), while others were more subjective (e.g. unlicensed fishing).
Given the highly variable nature of the IUU risks, a basic calculation to quantify likely IUU volume
and species composition was tailored to each risk based on the information available. As a general
rule, an uncertain quantity (e.g. the number of days fished by vessels on the FFA Vessel Register in
EEZs for which they were unlicensed) was multiplied by a known quantity (e.g. the average catch
rate and species composition per day in the relevant sector) to constitute the basic equation for
each risk. ‘Best estimates’ and min/max ranges could then be assigned to the uncertain value based
on the nature and quality of information available. Min/max ranges took into account the
uncertainty in the available information base (i.e. risks with more certain information had narrower
ranges, risks with limited information had larger ranges).
2.2.3 Assigning likely probability distribution
Once ‘best estimate’ and min/max values had been assigned, a likely probability distribution of IUU
activity within this range was determined. In general, triangular distributions were used where
there was a reasonable level of confidence that the actual level of IUU activity was likely to be closer
to the ‘best estimate’ than either the minimum or maximum value (Figure 1). This is consistent with
the approach used by Pitcher et al (2002), Ainsworth and Pitcher (2005), and Agnew et al (2009).
Uniform distributions were used where the information base was highly uncertain (e.g. shark finning
in the LL sectors), although in general, we attempted to avoid these.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
4
Figure 1: Example distributions of probability assigned to IUU activity. Triangular distributions were used where it was
more likely the actual level of IUU activity was closer to the best estimate than either the minimum or maximum values.
Uniform distributions were used where the information base was highly uncertain.
2.2.4 Monte Carlo simulations
We then used Monte Carlo simulation (using ‘@Risk’ software) to define the relative probability that
IUU volumes were within certain ranges, based on the best estimate and min/max values as well as
the probability distribution assigned (Figure 2). Monte Carlo simulation is a widely-used analytical
technique for calculating probability distributions of possible outcomes by performing a large
number of runs (in the case 10,000), in which variables are sampled randomly from within a
specified distribution. Monte Carlo simulation has previously been used by a number of authors in
attempts to estimate IUU activity (e.g. Pitcher et al, 2002; Ainsworth and Pitcher, 2005; Agnew et al,
2009). The approach has a number of benefits over ‘single point’ or deterministic models in that it
produces probabilistic results. In the context of this study, the simulations produce a probability
that IUU for a given risk will be greater than a certain value.
(a)
(b)
Figure 2: Example probability distribution outputs from Monte Carlo simulations using @Risk software.
Taking our example distributions in Figure 1, where Monte Carlo sampling was performed across a
triangular distribution, values around the best estimate have a higher probability of occurrence than
values sampled closer to the minimum or maximum range. In this way, the ‘best estimate’ value
would be given higher weight in the ultimate probability distribution. However, where we thought
there was no better chance that the actual IUU value was around the ‘best estimate’ than the
minimum or maximum values a uniform distribution was assigned. In these cases, Monte Carlo
sampling resulted in each value within the minimum and maximum range having an equal
probability of occurrence (Figure 1) and given equal weight in the ultimate probability distribution.
To that end, uncertainty is factored into the estimates in three ways:
The width of the min/max range in most cases, the narrower the min/max range around
the best estimate, the more certain the inputs;
The probability distribution chosen for the Monte Carlo simulations for risks in which there
was a reasonable level of confidence that the actual level of IUU was likely to be closer to
the ‘best estimate’ than the minimum or maximum values, a triangular distribution was
chosen; and
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
5
The probability distribution produced by the Monte Carlo simulation this gives an estimate
of the likelihood that the actual IUU estimate will be above or below chosen benchmarks.
2.2.5 Quantifying ex-vessel value
Based on the likely volumes and species compositions associated with each risk, we were able to
calculate the likely ex-vessel value of IUU fishing activity. For each species in the three main sectors
(PS, TLL, SLL), we assumed likely markets (sashimi grade for longline yellowfin; canning grade for
purse seine yellowfin) and were able to generate likely market values based on known trade and
market data (Annex 2). From this we calculated the likely ex-vessel value of IUU fish across each
main sector and collectively. In cases where species were misidentified in logsheets, only the
marginal difference in value between the actual species and the species reported was taken into
account.
Notwithstanding that, despite being the most widely used figure to convey IUU catch values, simple
ex-vessel or market values are not necessarily an accurate reflection of loss of value added to Pacific
Island economies. This is discussed in more detail in Section 4.
2.2.6 Changes to the 2020 model
While the same basic approach to estimating IUU was used between the 2016 and 2020 studies, a
number of changes were made to the information underlying estimation of individual risks. These
were largely driven by either (i) better information becoming available since the 2016 study (e.g. the
availability of the Global Fishing Watch [GFW] Automatic Identification System [AIS] dataset and
WCPFC Transhipment Declaration dataset to assist in estimating scope for illegal transhipment) or
(ii) the unavailability of information/analysis used for the 2016 study (e.g. predictive analysis of
purse seine set type based on catch composition used in the 2016 study to assist in estimation of
illegal FAD fishing during the closure period). For some risks, these changes of information had
substantial impacts on the estimated volume and value between studies. The nature of changes
made and implications are discussed in association with each risk in section 2.5.
In addition, for some risks in the 2020 model, Monte Carlo simulations were allowed to sample from
a distribution informed by bootstrapped outcomes on the ‘right hand side’ of the equation to better
reflect uncertainty in IUU catches. For example, for unlicensed fishing we assume that the catch
taken during an unlicensed fishing day is likely to be the broadly the same as that taken by a licensed
fisher using the same gear in the same area. In the 2016 model, to estimate the extent of unlicensed
fishing, we multiplied an unknown amount (i.e. the estimated number of days unlicensed fishing) by
a static ‘known’ amount (i.e. the average catch and species composition per day for that sector). In
the 2020 model, we used the same approach to estimate the unknown amount (i.e. the estimated
number of days unlicensed fishing), but allowed Monte Carlo simulations to sample from a
distribution of catch volume and species composition informed by the bootstrapped mean and 95%
confidence intervals of catch data reported during the study period. This approach recognises that
catch and species composition will vary by vessel, location and time and was mainly used for
calculations involving catch rates per day (e.g. unlicensed fishing, fishing inside closed waters).
2.3 Study period
The data and information used for this study covered the period 2017 to 2019. Consistent with the
2016 study, we have not attempted to estimate total IUU in a single snapshot year within this
period. Rather, our overall estimates should be considered ‘typical’ annual levels of IUU fishing
across each category for the period encompassed by the study (2017-2019). In general, where
consistent, comparable data series have been available to estimate individual IUU activities, we have
averaged these across the three year period. The specific amount of IUU occurring in each year will
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
6
vary according to inter-annual factors such as overall level of effort, regulatory changes, MCS
coverage, etc.
Importantly, the time period covers the years immediately prior to the COVID-19 pandemic which
constrained FFA members’ capacity to deliver many MCS activities (e.g. observer coverage, at sea
boarding and inspections, port inspections). Accordingly, the estimates do not pick up the impacts
of COVID-19 related MCS impacts on IUU.
2.4 Information collection, planning and ground-truthing
The approach and methodology described above was supplemented with a number of initiatives
designed to identify, collect and ground-truth relevant information. These included:
2.4.1 Planning workshop
We commenced the study with a Planning Workshop on 9th October, 2020, involving members of the
project team, FFA, SPC and PNA Office staff. The main purpose of the workshop was to agree the
methodology to be used for the study, as well as to agree the scope. The main outcomes were that:
the “Pacific Islands Region” will be defined consistent with the 2016 study – i.e. as the area
below 20oN, east of 130oE and north of the southern boundary of the WCPFC Convention
area, and east to the eastern boundary of the WCPFC Convention boundary (including the
area of ‘overlap’ with IATTC Convention area). The area includes EEZs of FFA member states
(excluding Australia and New Zealand) and areas of high seas
2
;
the area will exclude the Indonesian and Philippines EEZs;
IUU estimates will be made at the level of the three main fishery sectors PS, TLL and SLL;
the boundary between the TLL and SLL will be 10oS (i.e. the TLL is the area within the Pacific
Islands Region between 10oS and 20oN; the SLL is the area within the Pacific Islands Region
between 10oS and the southern boundary of the WCPFC-CA);
data and information used should cover the period 2017-2019; and
the pole and line fishery would not be included in IUU estimates.
The workshop agreed that the IUU risks used in the 2016 study were still relevant in the 2020 study
and covered the main IUU activities. These risks were separated into four main categories:
1. unlicensed and unauthorised fishing;
2. misreporting and non-reporting;
3. non-compliance with other license conditions (e.g. use of unauthorised gear); and
4. post-harvest risks (e.g. illegal transhipping).
A description of each of these risks is provided at Annex 3.
2.4.2 Data collection
The study was undertaken during the COVID-19 pandemic, so all data collection was undertaken
remotely. Data from FFA members were collected through surveys, as well as video conferences to
work through survey responses and seek member input on IUU experiences within their EEZs.
Responses to the project survey were received from fisheries agencies within Cook Islands, FSM, Fiji,
Nauru, Palau, PNG, RMI, Samoa, Solomon Islands, Tokelau, Tonga, Tuvalu and Vanuatu. Video
conference discussions were held with representatives from Cook Islands, Fiji, FSM, Nauru, Palau,
RMI, Solomon Islands, Tuvalu, Vanuatu, Australia and New Zealand. FFA member surveys focused
2
Note that data provided by SPC for the study was only that available to FFA members, and did not include
data for non-FFA members such as the French Territories. Where public domain aggregate WCPFC data was
used to inform risks, some data for these areas may be included.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
7
largely on information typically held only at the national level (e.g. outcomes of dockside
inspections, prosecutions, surface patrols, etc.).
Data held regionally were provided by relevant regional agencies including FFA (VMS and
aerial/surface surveillance data), SPC (logsheet, observer and unloadings data) and WCPFC (high seas
transhipment data). GFW also provided access to their AIS data for the purposes of cross-
referencing against reported WCPFC Transhipment Declarations.
BOX 2: MONITORING AND SURVEILLANCE IN THE PACIFIC ISLANDS REGION
Attempts to quantify the volume and value of IUU fishing of the type undertaken here that is, to
attempt to estimate all main forms of IUU fishing across each main sector, using a data driven,
bottom up approach require relatively good information across a wide range of different areas.
The fact that such an exercise is possible in the FFA region is largely due to Pacific Island countries’
investments over time in strong MCS mechanisms and commitment to regional coordination,
cooperation and information sharing. While there are undoubtedly areas in which monitoring can
be improved, the region’s establishment of measures such as the FFA Vessel Register (VR),
Harmonised Minimum Terms and Conditions for Access by Fishing Vessels (HMTCs), the FFA
Vessel Monitoring System (VMS), standardised forms including logsheets, observer and unloading
forms, 100% purse seine observer coverage, standardised regional observer training under the
Pacific Islands Regional Fisheries Observer (PIRFO) Standard and regionally coordinated fisheries
operations, together with centralised coordination and analysis of information through FFA, SPC
and PNA has the Pacific Islands region relatively well-placed to detect, analyse and combat IUU.
These measures, together with the fact that much of the fishing (particularly in the purse seine
sector) occurs in EEZs, rather than the high seas, means the FFA region is better placed to
undertake this type of study than many other regions internationally.
2.4.3 Ground-truthing workshop
The information used by the project team to estimate each risk, together with the basic equations
and proposed minimum, best estimate and maximum range values were ‘ground-truthed’ at a
workshop with FFA, SPC and PNA Office staff on 25th June, 2021. Proposed range values were
agreed or refined based on attendees’ practical experience with Pacific tuna MCS and IUU issues.
This process ensured that any obvious errors or misinterpretations were picked up and added a layer
of validation to the outputs.
2.4.4 Out of scope issues
Consistent with the 2016 study, there are a number of other activities that are illegal under national
law or license conditions, but did not directly result in ‘unaccounted for’ fish and were considered
out of scope. These included, for example, breaches of marine pollution (MARPOL) regulations,
interactions with species of special interest (SSIs) and illegal bunkering. Nevertheless, we note that
these issues remain important violations and in some cases have been implicated in the trade of
illegally harvested fish products (e.g. there have been anecdotal concerns about bunker vessels
facilitating the movement of shark fins to market).
2.5 Approach to quantifying individual risks
As described in 2.4.1, ‘in scope’ IUU risks were categorised into four basic types of activity:
1. Unlicensed fishing;
2. Misreporting (including under-reporting and misidentification);
3. Non-compliance with other license conditions; and
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
8
4. Post-harvest risks.
Approaches to quantifying each individual risk within these categories, including the information
available, the basic equation to calculate the level of activity and best estimate and min/max ranges,
are outlined below.
2.5.1 Unlicensed fishing
Three different classes of unlicensed/unauthorised fishing have been assessed:
1. Unlicensed fishing by vessels on the FFA VR;
2. Unlicensed/unauthorised fishing by vessels on the WCPFC Record of Fishing Vessels (RFV),
but not on the FFA VR; and
3. Unregulated fishing (i.e. by vessels flagged to non-WCPFC CCMs, or vessels not authorised
on the WCPFC RFV).
Classes were separated based on the information available to quantify the IUU risk. For example,
vessels in the first class are all licensed in at least one FFA member EEZ and are required to have a
functioning automatic location communicator (ALC) reporting to the FFA VMS. Vessels in the second
class are authorised to fish on the high seas within the WCPFC-CA, but not in FFA member EEZs.
These vessels are required to have a functioning ALC reporting to the WCPFC VMS. Vessels in the
third class do not report to any VMS for which FFA or WCPFC have data sharing privileges and are
effectively ‘dark’ from that point of view (albeit they may be tracked by AIS in some cases).
2.5.1.1 Unlicensed fishing by vessels on the FFA Register
This risk relates to the possibility of a vessel licensed in one FFA member EEZ fishing in another zone
for which it has no license (i.e. ‘border hopping’). A number of different data sources were used to
arrive at ‘best estimate’ and min/max range estimates for this risk including:
VMS data for each of the three sectors (PS, TLL, SLL), combined with ‘compliance index (CI)’
data for each vessel derived from FFA’s Regional Fisheries Surveillance Centre (RFSC);
Aerial and surface surveillance data both from regional operations during the study period
as well as surveillance activities carried out at the national level;
Observer reporting; and
Information from FFA member surveys and interviews.
‘Best estimates’ and min/max ranges were assigned by estimating a proportion of the overall
number of fishing days undertaken by vessels on the FFA VR in each of the three sectors likely to be
fished in an EEZ for which the vessel had no license. These were then multiplied by the average daily
catch rate and species composition for that sector to arrive at overall estimates of IUU volume and
species composition.
Consistent with the 2016 study, ‘best estimate’ and min/max range figures were primarily assigned
using VMS and FFA CI data. As part of their regional surveillance function, the FFA RFSC assigns each
vessel for which it has visibility (i.e. all vessels on the FFA Register and vessels on the WCPFC RFV
operating within FFA member EEZs, as well as vessels visible using the AIS) a CI. The CI is a number
between 0 to -5 and is based on the vessel’s compliance status. Vessels rated 0 are deemed to be a
very low risk of undertaking IUU activity; vessels rated -4 or -5 are considered to be at very high risk
of undertaking IUU activity, or have been involved in confirmed IUU activity. A CI of -3 is most
frequently assigned to vessels currently located in zones for which they have no license. While many
vessels with a -3 CI are simply making innocent passage through an EEZ (which is relatively easily
visually detected by using VMS data), this group of vessels (together with the -4 and -5 CIs) are at
higher risk of unlicensed fishing and gave us a ‘starting point’ to adjust based on aerial surveillance
and other information.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
9
To collect a relatively random sample of data, a snapshot of CI data for all vessels across all FFA
member EEZs was provided by FFA for one day of each week during the period January 2017 to
December 2019. This produced 6,263, 7,650 and 4,977 individual CI vessel ratings for the PS, TLL
and SLL sectors, respectively. Of these, the proportion assigned to each CI category was identified
(Figure 3), and from this, the proportion of these vessels likely to be engaged in unlicensed fishing
activity was estimated based on expert judgement. Final best estimates and min/max ranges were
also informed by aerial and surface surveillance, and observer information, where available.
Figure 3: Proportion of total FFA compliance index (CI) ratings by sector, 2017-2019, based on a random sample of 18,890
individual CI ratings.
Figure 4: VMS position data sampled for vessels on the FFA Register in 2018.
Purse seine
In the purse seine sector, the ground-truthing workshop agreed that the level of unlicensed fishing
activity was likely to be negligible. During the study period this sector included around 250-260
vessels, fishing an average of 42,176 days per year inside FFA member EEZs. Vessels in two fleets
the Federated States of Micronesia Arrangement
3
(FSMA) and the US Multilateral Treaty (USMLT)
are authorised to fish in each of the main EEZs and accounted for around 1/3 of the fleet during the
3
Parties to the FSMA include FSM, Kiribati, Nauru, Palau, PNG, RMI, Solomon Islands and Tuvalu.
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
0-1 -2 -3 -4 -5
% of CI ratings
PS TLL SLL
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
10
study period (~90-100 vessels out of ~255). Given these vessels are authorised to fish in each of the
main purse seine EEZs, they are unlikely to engage in unlicensed fishing in the context of this risk.
No unlicensed fishing activity has been detected by aerial surveillance assets during the study
period, vessels are polled hourly by VMS under the PNA’s Vessel Days Scheme (VDS) arrangements
and all vessels are subject to 100% observer coverage, which is likely to act as a strong deterrent to
non-compliance. While a relatively high proportion (36%) of sampled vessels had a CI score of -3
and lower, the overwhelming majority of these are likely to have been engaged in innocent passage
to transhipment ports.
In the 2016 study, the best estimate and min/max range were set based on expert judgement, taking
account of available information. Delegates at the 2016 ground truthing workshop agreed that the
minimum estimate should be set at 0% to account for the possibility all fishing is done in zones for
which the vessel has a license, while the maximum figure should account for the possibility of slightly
higher levels of unlicensed activity (0.2% of the average total days fished) and the best estimate
should be set closer to the minimum (0.05%). The same basic approach was taken in the current
study, however given the absence of detected IUU activity and strong MCS coverage of the purse
seine sector, the ground-truthing workshop agreed that estimates for the 2020 study should be set
at lower levels. Accordingly, minimum, best estimate and maximum figures were set at 0%, 0.02%
and 0.1% of average total fishing days respectively.
Tropical longline
In the tropical longline sector, the inherent risks of unlicensed fishing are likely to be higher, given
the larger number of vessels (650-700 longliners on the FFA VR during the study period), higher
levels of effort (average of 51,177 fishing days in FFA EEZs during 2017-2019) and the operational
nature of the gear which can drift many tens of nautical miles in a set. MCS arrangements for the
TLL sector are generally weaker than those in the PS sector, given very low rates of observer
coverage, less frequent VMS polling, and there are arguably fewer financial disincentives to
unlicensed fishing. Moreover, longline vessels have typically had slightly higher rates of non-
reporting to the FFA VMS than the PS sector and several instances of ALCs not reporting or having
wiring rigged such that ALCs could be switched on and off
4
were detected during FFA coordinated
regional operations and Pacific Maritime Security Program (PMSP) aerial patrols during the study
period, which leaves open the possibility of unlicensed activity.
Nevertheless, actual evidence of unlicensed fishing during the study period is very limited. For
example, aerial surveillance assets operating within 12 FFA led regional operations during the study
period covered a combined surveillance footprint of 21,001,047nm2, with no confirmed unlicensed
fishing vessels detected. Similarly, surface surveillance assets within the same regional operations
steamed a combined distance of 124,976nm for no confirmed detections or apprehensions of
unlicensed fishing vessels. Moreover, of the FFA members providing information for the study, none
had prosecuted a vessel for unlicensed fishing during the study period and, while all recognised that
the risk of unlicensed fishing required constant vigilance, actual levels of unlicensed fishing were
likely to be substantially less than they were several decades ago (before key MCS measures such as
the FFA VMS, FFA Regional Register, aerial and surface surveillance programs, etc).
CI information showed that around 42% of days at sea during 2017-2019 were spent in EEZs for
which vessels had no license, albeit the vast majority of these days are likely to be in innocent
passage. Vessels considered at highest risk of IUU i.e., those with CIs of -4 and -5 - accounted for
0.5% and 0.2% of total polls in the TLL sector, respectively.
In the 2016 study, the best estimate and min/max range for the longline sectors was set based on
expert judgement, using the same process as that described for the purse seine sector above. The
4
E.g. during Operation Tui Moana 2018.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
11
minimum estimate of unlicensed fishing was set at 0.1% of the days fished in zones (given the
absence of widespread evidence of non-compliance detected through regional operations), while
the maximum range was set at 1.5% of days given the weaker monitoring arrangements in the fleet.
The best estimate was set slightly towards the lower end of the min/max range. In this study, the
ground truthing workshop agreed that the absence of any confirmed activity despite significant
surveillance coverage, together with the relatively strong monitoring of FFA fleets and the fact many
have licenses in multiple zones meant the proportions could be reduced. Accordingly, minimum,
best estimate and maximum values were set at 0.05%, 0.2% and 0.5%, respectively.
Southern longline
The information base for the SLL sector was similar to the TLL sector in that no instances of
unlicensed fishing were detected during regional operations in the study period, although MCS
arrangements are weaker than the PS sector and observer coverage remains very low. SLL vessels
fished an average of 32,596 days inside FFA EEZs during the study period. VMS and CI data showed
around 41% of longline days in the SLL area were spent in zones for which the vessel had no license,
although the majority of these days were likely to have been transiting to and from key SLL ports
such as Suva. Vessels with CIs of -4 and -5 accounted for 0.5% and 0.4% of polls, respectively. Of the
FFA members contributing information for the study, none reported prosecutions of longline vessels
on the FFA VR for unlicensed fishing. This was fewer than the 2016 study when Fiji reported a total
of 17 incidents of possible unlicensed fishing (picked up through port inspections during the study
period), and the Solomon Islands reported 22 incidents of possible unlicensed fishing (detected by
aerial/surface surveillance) (MRAG Asia Pacific, 2016).
The ground-truthing workshop agreed there was no obvious evidence to suggest the rate of
unlicensed activity was likely to be different between the TLL and SLL sectors, and the proportions of
overall days used for the min/max range and best estimate were the same as for TLL.
Table 1: Best estimate and min/max range for unlicensed fishing activity by vessels on the FFA VR by size and sector (by
average number of days fishing per year). Percentages in parentheses represent the proportion of the total average fishing
days by relevant vessels in that sector.
Sector
Min
BE
Max
Dist.
Purse seine
0 (0%)
8 (0.02%)
40 (0.1%)
Triangular
Tropical Longline
26 (0.05%)
102 (0.2%)
255 (0.5%)
Triangular
Southern longline
16 (0.05%)
65 (0.2%)
163 (0.5%)
Triangular
2.5.1.2 Unlicensed/unauthorised fishing by vessels on the WCPFC RFV, but not on the FFA
Register
This risk broadly relates to the possibility of WCPFC high seas vessels fishing in FFA EEZs for which
they have no license. Vessels on the WCPFC RFV report to the WCPFC VMS and are not routinely
visible to FFA members on the high seas, although members may apply to view vessels within 100nm
of the EEZ boundary. Members may also apply to have their zone included in the WCPFC VMS,
which allows them to view WCPFC vessels while in their zone, although not all FFA members have
reportedly taken up this option to date. At the regional level, FFA also requests the full suite of
WCPFC VMS data for the area of operation during regional operations. However, high seas VMS
data is required to be destroyed within seven days following the completion of MCS activity, so the
capacity to undertake analysis of high seas vessel activity is somewhat limited.
Similar to the 2016 study, the ground-truthing workshop agreed that the main risks in this category
were likely to be incursions on the fringes of FFA member EEZs by vessels ordinarily fishing the high
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
12
seas, or when transiting FFA member EEZs, and on the western fringes of the FFA area by WCPFC
authorised vessels that are ordinarily fishing in their own domestic waters or high seas pockets.
The information available to quantify unlicensed fishing by high seas vessels is largely limited to
some VMS information (particularly during regional operations), aerial and surface surveillance data
and expert judgement by practitioners in relevant areas. The basic calculation used to quantify
activity in each sector was the number of unlicensed fishing days * the average catch volume and
species composition for each sector per day. Average catch volume and species composition for the
longline sectors was calculated based on catches within FFA member EEZs during the study period.
For the purse seine sector, average catch rate and species composition was calculated based on
catches in the western part of the fishery, between 130oE and 150oE, using WCPFC public domain
data. Small vessels in the purse seine sector were assumed to have catch rates 1/3 of ‘average’
vessels.
Purse seine
In the purse seine sector, estimates of IUU activity in the 2016 study were split into ‘average’ and
‘small’ vessel categories to account for likely differences in catching capacity. The 2016 ground-
truthing workshop agreed that the amount of unlicensed fishing activity by ‘average’ high seas
vessels was likely to be very low, although agreed to include a small number of estimated IUU days
to account for the possibility of some incursions. During the 2017-19 period, the overall level of high
seas activity was relatively low (an average of 7,503 fishing days on the high seas with 42,176 in zone
during 2017-2019; SPC, 2021), and no unlicensed fishing activity was detected by aerial surveillance
during the study period. PS vessels are subject to 100% observer coverage and considerable
financial disincentives exist if caught illegally fishing. While some northern WCPFC CCMs (e.g. China,
Japan and Korea) have substantial proportions of their fleets on the WCPFC RFV but not on the FFA
VR, all of the vessels which fish in the industrial purse seine fishery in and around the FFA member
EEZs will be on the FFA VR. On that basis, the 2021 ground truthing workshop agreed there was
negligible opportunity for the ‘RFV only’ vessels from these States to fish in FFA EEZs. Accordingly,
the current study assumed no unlicensed fishing activity in this sector.
In the small vessel category, the main risk is likely to be from incursions on the western fringes of the
FFA area. Of the fleets operating in this area, the Philippines has the largest number of vessels on
the WCPFC RFV but not licensed in FFA member EEZs (Figure 5). This includes vessels authorised to
fish in high seas pocket 1 (HSP1) under CMM 18-01
5
. Under CMM 18-01, Philippines vessels were
limited to a high seas cap of 4,659 days and 36 fishing vessels. SPC (2021) indicates that actual
fishing effort by Philippines flagged vessels in HSP1 averaged around 2,700 days in the 2017-19
period, while the Indonesian and Philippines fleets fished a combined average of around 2,304 days
within their domestic EEZs (outside archipelagic waters) during 2017-2019. Philippines advises that
all vessels fishing in HSP1 have 100% observer coverage (Anon, 2020). They also advise that vessels
fishing in HSP1 are required to operate VMS, although in 2020, the Philippines was assessed for the
fourth consecutive year as priority non-compliant against the requirement under WCPFC CMM 14-
02 to ensure fishing vessels on the high seas are fitted with an ALC which meets the WCPFC VMS
requirements (WCPFC, 2020a). The extent to which this applied to vessels fishing HSP1 is not
known.
While anecdotal reports of incursions exist (e.g. Palau were of the view that group seine vessels
heading through the Palau EEZ to HSP1 would stop if they saw a school of fish), direct evidence of
illegal activity is very limited. The 2021 ground truthing workshop agreed that estimates to account
for some level of illegal activity should be maintained, although the best estimate and maximum
figures should be reduced given the absence of confirmed detections, the reduction in the number
5
During the 2017-2019 study period, the relevant tropical tuna CMMs were CMM 2016-01, CMM 2017-01 and
CMM 2018-01. Each included a provision for a limited number of Philippines flagged vessels to fish in HSP1.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
13
of vessels from some CCMs (e.g. Indonesia) on the RFV and the application of observers to the main
fleet operating in HSP1.
Figure 5: Comparison between number of purse seine vessels by flag on the FFA VR and the WCPFC RFV (as at
November/December, 2018). Note that these numbers do not take into account chartering arrangements.
Tropical longline
Unlike purse seine, a considerable proportion of longline fishing effort occurs on the high seas (an
average of 103,597 fishing days in the high seas portion of TLL area during 2017-2019 versus 51,177
in FFA EEZs), and a large number of vessels are authorised on the WCPFC RFV but not on the FFA VR
(2584 vessels on the WCPFC RFV
6
with 684 on the FFA VR) (Figure 6). The evidence to support
estimates of unlicensed activity is both limited and mixed. While no unlicensed fishing activity was
detected by high seas vessels using aerial surveillance during 2017-2019, numerous instances have
been detected during FFA-led regional operations and/or PMSP flights where vessels on the WCPFC
RFV have been detected inside FFA members EEZs, but not reporting to the WCPFC VMS. For
example:
WCPFC vessels not reporting to the WCPFC VMS were detected in Palau’s EEZ on multiple
flights under the PMSP in 2019, while in another case the name of the vessel on the WCPFC
VMS was not consistent with the sighted vessel;
On other PMSP flights, a WCPFC longline vessel was detected inside Tokelau’s EEZ and not
reporting to the WCPFC VMS in 2018, while a similar incident was observed in FSM’s EEZ in
2019;
In another instance, a longline vessel not licensed in FSM’s EEZ was sighted and appeared to
be towing a line from the stern.
Of the FFA members interviewed for the study, RMI provided anecdotal evidence of longlines
floating into their EEZs from adjacent high seas vessels.
6
Not all vessels on the RFV may be active in the study area. Some may fish in and around their own EEZs (e.g.
US vessels, Chinese Taipei vessels).
0
10
20
30
40
50
60
70
80
90
100
Australia
China
Ecuador
El Salvador
EU
Indonesia
Japan
Kiribati
Korea
RMI
FSM
Nauru
NZ
PNG
Philippines
Solomon Is
Spain
Taiwan
Tuvalu
USA
Vanuatu
Number of vessels
FFA VR WCPFC RFV
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
14
Figure 6: Comparison between number of longline vessels by flag on the FFA VR and the WCPFC RFV (as at
November/December, 2018). Note that these numbers do not take into account chartering arrangements.
The main opportunities for infringement in the TLL sector are likely to be on the fringes of FFA
member zones. Longline gear is inherently higher risk than PS gear, given its operational nature
allows it to drift many tens of nautical miles in a set and there is relatively limited (apart from VMS)
consistent MCS coverage of large parts of the high seas longline fleet (e.g. very limited observer
coverage; very limited aerial/surface surveillance coverage). In 2016, the ground-truthing workshop
agreed that the minimum estimate should be set at a low level, although the min/max range should
be set broadly to acknowledge the high levels of uncertainty in the information base and to account
for the possibility of higher rates of unlicensed fishing. To that end, estimates were set at 0.1%
(min), 0.5% (best estimate) and 1.5% (max.) of total high seas fishing days, with a triangular
distribution used for MC simulations (Table 2).
For this study, the evidence base was similar although there were fewer confirmed cases of
unlicensed fishing during the study period. To that end, minimum and best estimate values were
maintained, but the maximum value was reduced to 1% of total high seas fishing days.
Southern longline
Similar to the TLL sector, the ground truthing workshop considered the main risk of infringement
was on the fringes of FFA member zones. However, overall effort is lower in the SLL sector and
unlike the TLL sector, the majority of effort is within EEZs (an average of 32,596 fishing days in FFA
EEZs versus 27,725 on the high seas during 2017-2019). Accordingly, the 2021 ground truthing
workshop agreed that estimates should account for a lower absolute level of unlicensed fishing
activity. The information base for the SLL is similar in nature to the TLL, in that no unlicensed vessels
were detected by aerial surveillance between 2017 and 2019. Of the FFA members contributing
information for the study, only the Cook Islands reported any confirmed instances of unlicensed
fishing by vessels on the RFV (one vessel in 2019). FFA members interviewed for the study noted it
was possible some longlines float into EEZs and there was uncertainty about the extent, but there
were no sanctions of high seas vessels for unlicensed fishing during the study period.
The workshop agreed that the best estimate and min/max range figures should be set at the same
proportions as those adopted for the TLL sector.
0
200
400
600
800
1000
1200
Number of vessels
FFA VR WCPFC RFV
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
15
Table 2: Best estimate and min/max range for unlicensed fishing activity by vessels on the WCPFC RFV but not on the FFA
VR by size and sector (by average number of days fishing per year).
Sector
Min
BE
Max
Dist.
Purse seine
Small vessel
0
50
250
Triangular
Tropical Longline
104 (0.1%)
518 (0.5%)
1036 (1.0%)
Triangular
Southern longline
28 (0.1%)
139 (0.5%)
277 (1.0%)
Triangular
2.5.1.3 Unregulated fishing
For the purposes of this analysis, ‘unregulated’ fishing was defined as fishing by vessels flagged to
States that are not cooperating members of the WCPFC, or by vessels flagged to cooperating
members but which are not on the WCPFC RFV (and are therefore not authorised to fishing on the
high seas inside the WCPFC-CA). The latter category is probably an extension of the standard IPOA-
IUU definition of unregulated fishing, but these vessels were included because they have the same
‘visibility’ as other unregulated vessels and will not be reporting to any regional VMS.
Unsurprisingly, data availability for unregulated fishing was particularly weak. To that end, estimates
for this risk were largely based on expert judgement using the information available including aerial
and surface surveillance reports, observer sightings, previous risk assessments and anecdotal
information.
Purse seine
In the purse seine sector, no unregulated vessels have been detected by aerial surveillance in recent
years. The main risks appear to be incursions by Eastern Pacific Ocean (EPO) vessels on the eastern
fringes of the WCPFC-CA and smaller domestic southeast Asian vessels on the western fringes (e.g.
small purse seiners with support vessels fishing in Palau/PNG). Aerial surveillance coverage of both
these areas, particularly in the east, is extremely limited. With no near-real time data sharing
arrangement between WCPFC and IATTC (and member states), VMS data for EPO vessels is
unavailable and any incursions by EPO vessels into the WCPFC area (and vice versa) are likely to go
undetected.
In general, the opportunity for truly unregulated purse seine vessels (i.e. those flagged to parties
other than cooperating members of the WCPFC) to fish in the WCPO undetected is likely to be
negligible. This is because the main purse seine fishing countries are members of the WCPFC, and
industry intelligence tracking both existing purse seine vessels and those under construction is
detailed to the point that there is likely to be few, if any, ‘unaccounted for’ vessels.
The 2021 ground-truthing workshop agreed that the main opportunities for unregulated purse seine
fishing came through incursions by EPO licensed vessels into the WCPFC-CA and by smaller, domestic
purse seiners from southeast Asia on the western fringes. The basic calculation used to estimate
activity was number of days unlicensed fishing * average catch rates and species composition per
day in the relevant sector. Because of the likely differences in capacity between EPO vessels and
smaller southeast Asian vessels, estimates were made for ‘average’ vessels and ‘small’ vessels.
Catch rates for small vessels were assumed to be 1/3 of average vessels.
Estimates of unregulated activity for average vessels in the 2016 study were set at very low levels,
with higher figures for smaller vessels. The 2016 workshop considered it possible that no
unregulated fishing occurred, although it was more plausible that there were a small number of
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
16
incursions (best estimate of 50 days, with a max. of 200 days) from EPO vessels. For the 2020 study,
the ground-truthing workshop noted that although the majority of the catch of YFT and SKJ came
from the eastern part of the IATTC area, considerable fishing for SKJ was undertaken near the border
of the convention areas (150oW) (Figure 7). Much of this fishing in this vicinity is based on FAD-
associated sets and given the prevailing equatorial current in the region flows westerly, there may
be incentive for vessels to follow drifting FADs that show large biomasses on sonar buoys across the
boundary. To that end, an allowance for some level of unauthorised incursions was maintained,
although the evidence base remains weak. The best estimate value was reduced to account for the
absence of confirmed unauthorised activity.
Figure 7: Distribution of catches of SKJ by set type in the IATTC area, 2019 (left panel; IATTC, 2020); WCPFC/IATTC
boundaries (right panel).
On the western fringes of the study area, the level of incursions was likely to be higher, given the
large number of small southeast Asian vessels that fish in this vicinity and the frequent anecdotal
reports of incursions in EEZs such as Palau and PNG. Tangible evidence of unlicensed fishing was
also higher for this sector than many others. Palau advised that they regularly see illegal small group
seine operations from the Philippines inside their EEZ (Figure 8). These operations typically consist
of an 80-100 ft catching vessel, a carrier vessel and two to three light boats to attract fish. FADs are
deployed to attract fish, with the catching vessel called when a school is located. Palau advised that
on a ‘busy’ day, a patrol may see up to five group seine operations, although some patrols report no
sightings. Most of the vessels they have apprehended have between 20-100 t of product on board,
although some can have >100 t. Because the vessels are not on either the FFA VR or WCPFC RFV, no
VMS or AIS data are available for these vessels, so they are difficult to detect. Palau advised that
much of the activity occurs in the northwest of the Palau EEZ, bordering the Philippines EEZ and the
high seas pocket east of the Philippines. They also advised that aerial surveillance on its own was
not effective, with vessels typically heading back into the Philippines EEZ upon sighting the aircraft.
Surface patrol data from the Palau Maritime Operations Centre shows 32 FADs were destroyed by
patrol vessels during 2017-2019
7
. Palau also advised that they have historically also had small group
seine operations from Indonesia undertaking incursions into its EEZ, although numbers had reduced
over the past five years because Palau has stationed rangers on its southwestern islands. Moreover,
Indonesia introduced a moratorium in 2015 on foreign-built vessels, so the purse seine fleet largely
comprises small vessels that rarely venture beyond their archipelagic waters (P. Williams, pers.
comm.).
7
Palau advised that group seine operations previously used metal FADs which were able to be sunk, but had
more recently changed to plastic floats which were harder to sink.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
17
Figure 8: An example of an unlicensed group seine operation detected in Palau’s EEZ (Source: Palau Maritime Operations
Centre)
While the actual number of illegal fishing days during the study period was uncertain, the evidence
indicates that some level of unregulated fishing occurred. The minimum estimate was maintained at
50 days after taking into account Palau’s advice that sightings during patrols were at least ‘regular’,
best estimates were set at 175 days based on Palau’s advice and the maximum figure was
maintained at 500 days to account for the possibility of a much higher level of unregulated activity,
including in neighbouring EEZs.
Tropical longline
Like the purse seine fishery, the main opportunities for unregulated fishing in the TLL sector came
from incursions on the eastern and western fringes of the study area, although there has historically
been some level of unregulated fishing by vessels not flagged to a WCPFC CCM in the longline sector.
For example, two vessels flagged to Georgia were added to the WCPFC IUU list in 2010
8
.
Opportunities for unregulated fishing are probably greater in the longline sector, given the larger
number of vessels, the absence of high levels of observer coverage, and greater difficulty tracking
movements between ocean basins amongst the fleet. Nevertheless, like the purse seine sector, the
information available to support quantitative estimates of unregulated fishing is very limited.
The main opportunity for unregulated fishing on the eastern fringes is by vessels authorised to fish in
the IATTC area but not in the WCPFC area. While a substantial amount of fishing occurs on the
boundary of the WCPFC-IATTC areas (Figure 9), a considerable proportion of the main fleets
operating in the area are authorised on both the IATTC and WCPFC regional vessel registers (Table 3;
Figure 10). This limits the scope for any unregulated fishing. Given the absence of strong aerial
surveillance or VMS evidence for vessels not on the WCPFC RFV, a small allowance has been
included to account for the possibility of some level of unregulated activity, but have been set at
relatively low levels (Table 5).
8
https://www.wcpfc.int/doc/wcpfc-iuu-vessel-list
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
18
Figure 9: Distribution of average annual catches of BET and YFT in the IATTC area by Chinese, Japanese, Korean, and
Chinese Taipei longline vessels, 2014-2018 (IATTC, 2020).
Table 3: Level of dual IATTC/WCPFC authorisation
amongst the main longline fleets operating in the
Pacific (as at April, 2021).
Flag
Vessels on
IATTC RVR
Also on
WCPFC RFV
Belize
1
Chile
1
China
418
341
Chinese Taipei
137
137
Costa Rica
2
Ecuador
22
EU (Germany)
1
EU (Portugal)
12
11
EU (Spain)
92
33
Japan
187
170
Korea
103
100
Mexico
40
1
Nicaragua
1
Panama
16
Peru
1
United States
38
36
Vanuatu
48
47
Figure 10: Proportion of longline vessels listed on the IATTC Regional
Vessel Register, which are also listed on the WCPFC RFV for fleets >2
vessels (as at April, 2021).
On the western fringes, the main opportunity for unregulated fishing is by smaller domestic vessels
of southeast Asian nations. For the purposes of this analysis, we have included all smaller vessels
that use lines to catch tuna (e.g. handliners, pump boats, outriggers and bancas) (e.g. Figure 11).
Anecdotal evidence suggests that incursions in the Palau EEZ and in the western part of the PNG EEZ
are relatively frequent, and there have been a number of seizures and sightings of unlicensed vessels
during the study period (by vessels targeting both tuna and demersal species). For example, during
Operation Kurukuru ‘17 the Palau patrol vessel Remelik issued an infringement to a vessel they
found allegedly secured to a FAD, while during Kurukuru ‘19 the PNG patrol vessel Moresby
apprehended three Indonesian wooden boats. These wooden boats were handed to the Papua New
Guinea National Fisheries Authority for further investigation. During Operation Rai Balang ‘18 the
Moresby detected an Indonesian FFV in the ‘dogleg’ area and the Remelik apprehended an outrigger
moored to a FAD. During Operation Island Chief ‘18 the Palau patrol vessel Kedam detected a
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
% dual authorised
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
19
Philippines outrigger in the Palau EEZ. In addition, an aerial surveillance flight under the PMSP
detected an illegal mother boat with multiple dinghies in the southwestern sector of Palau’s EEZ that
borders Indonesia during 2019.
During interviews for this study, Palau advised that most IUU activity using lines was related to
Philippines outriggers (bancas) using hand lines to target tuna, although small illegal vessels were
also observed in the south east pocket of the Palau EEZ close to the Indonesian border. They noted
the profile of Philippines vessels had changed in the past five years with fewer larger outrigger type
vessels and more metal motherships with fleets of smaller one-person bancas (called ‘pakuras’ in the
Philippines) (Figure 11). Motherships may be >24m, while smaller pakuras are typically 6-8ft in
length. SPC advised that, based on landings data from the Philippines, catch rates for similar types
of vessels averaged around 1-3 fish per day, with yellowfin comprising almost all of the catch (96-
98%) (e.g. NFRDI et al, 2020).
Figure 11: Mothership with one-person pakuras detected in Palau’s EEZ (Source: Palau Maritime Operations Centre).
PNG also reported observations of unlicensed fishing in its EEZ close to the Indonesian border, both
north and south, mostly by pump boats/type III vessels. Pump boats are typically small outrigger
vessels of wooden construction, with fishing usually done by handlines and other line gear. ‘Type III’
is a name given to slightly larger, low slung wooden vessels, typically from Indonesia. Fishing is often
done using longlines. In 2019, 20 vessels were detected during surface patrols, with four type III
vessels prosecuted for fishing for tuna without a license. The remaining 16 vessels were subject to
administrative sanction.
The 2021 ground-truthing workshop agreed that estimates should be set at levels which account for
the possibility of higher levels of activity, with a ‘best estimate’ of 500 days unregulated fishing
(equating to slightly >1 unregulated vessel fishing in the western fringes of the FFA area for each day
of the year), with a minimum of 200 days and a maximum of 1500 (4-5 unregulated vessels for each
days of the year).
Southern longline
In the southern longline sector, there is negligible scope for incursions in the west. The main
opportunities for incursions are in the east and to a lesser extent in the south. Given the SLL sector
has no borders with EEZs with small scale, potentially ‘unregulated’ fleets, small scale vessel
incursions were likely to be negligible. To that end, unlike the TLL sector, we have made no
provision for unregulated fishing by ‘small’ vessels. Estimates are made based on ‘average’ vessels
typical in the SLL sector.
There have been a number of incidents in which vessels flagged to WCPFC CCMs, but which were not
on the RFV, being sanctioned for fishing inside the WCPFC-CA during the study period. For example,
a New Zealand high seas patrol in 2017 detected five vessels in the vicinity of the Louisville Ridge to
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
20
the east of the New Zealand EEZ, but not on the RFV (New Zealand, 2017). The absence of WCPFC
authorisation was later confirmed by their flag State and the vessels sanctioned. In addition, two
Chinese flagged vessels were fined for fishing unlicensed within the WCPFC-CA adjacent to New
Zealand’s EEZ in 2017
9
. These vessels were also fined for misidentifying southern bluefin tuna as
bigeye tuna.
The experience of the Georgian-flagged vessels on the WCPFC IUU List indicates some scope for
fishing by non WCPFC CCM flagged vessels , although the numbers involved are likely to be
negligible. No unregulated vessels were detected by aerial surveillance during the study period,
while the majority of the major fleets operating in adjacent jurisdictions (IATTC, CCSBT) are also
authorised on the WCPFC RFV (with the exception of Chinese Taipei, although these vessels may fish
in other ocean basins (Table 4). Many of the CCSBT authorised vessels not on the WCPFC RFV (e.g.
Indonesian, EU and South African fleets) are likely to be fishing in other areas (e.g. Indian Ocean and
Atlantic Ocean).
Table 4: Level of cross-authorisation in the WCP-CA by longline vessels authorised under CCSBT Record of Authorised Vessels
(as at June 2021).
Flag
CCSBT
Also WCPFC
%
Australia
31
30
97%
Indonesia
176
0
0%
Japan
83
73
88%
Korea
11
9
82%
NZ
16
1
6%
Portugal
25
10
40%
Spain
77
23
30%
Chinese Taipei
94
4
4%
Given the confirmed instances of unregulated fishing during the study period, the minimum estimate
was set at 50 days to account for these vessels, the best estimate set at 100 days to account for the
likelihood that other vessels may have gone undetected and 200 days to account for a higher level of
activity (Table 5).
Table 5: Best estimate and min/max range for unregulated fishing activity by vessel size and sector (by average number of
days fishing per year).
Sector
Min
BE
Max
Dist.
Purse seine
Average vessel
0
20
200
Triangular
Small vessel
50
175
500
Triangular
Tropical longline
Average vessel
20
100
300
Triangular
Small vessel
200
500
1500
Triangular
Southern longline
Average vessel
50
100
200
Triangular
9
https://www.nzherald.co.nz/nz/chinese-vessels-fined-825000-and-deregistered-for-tuna-offences-off-nz-
coast/P2OETZSFY7BLHM5FPKFLHFDKBY/
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
21
2.5.2 Misreporting and non-reporting
Vessels authorised to fish in the Pacific Islands region have an obligation to report catch and effort
accurately under a range of provisions. At the WCPFC level, the Convention (Annex 3, Article 5)
requires that fishing operators “record and report vessel position, catch of target and non-target
species, fishing effort and other relevant fisheries data in accordance with the standards for
collection of such data set out in Annex I of the Agreement.” Moreover, CMM 13-05 requires each
CCM to ensure that “the master of each vessel flying its flag in the Convention Area shall complete
an accurate written or electronic log of every day that it spends at sea on the high seas of the
Convention Area”. The details of operational reporting requirements are set out in Annex 1.3 to 1.6
of the Scientific Data to be Provided to the Commission (WCPFC, 2012) and include obligations to
report a range of target and non-target species, as well as a range of operational details (e.g.
number of hooks per set for longlines, set type for purse seine and location). At the FFA member
level, the HMTCs require that vessel operators keep daily records of all catch and bycatch species,
including all catch discarded at sea and all bycatch transhipped or unloaded offshore, and submit
final versions of these reports to licensing countries within 45 days of trip completion (FFA, 2019).
Purse seine
In general terms, analysis of reporting behaviour by licensed vessels is best undertaken by
comparing reported catch against the best available independent source/s of the same data.
Consistent with the 2016 study, the best available data source to examine reporting behaviour in the
purse seine sector comes from independent observer estimates. Since 2009, purse seine vessels
have been subject to 100% observer coverage so there is a considerable body of independent data
on vessel catch. Other potential comparative data sources such as cannery out-turn reports show
promise (if collected under the appropriate conditions), but were not available for this study
10
.
To allow for an examination of reporting patterns, SPC provided set-by-set and trip level logsheet
and observer data for matched trips within FFA EEZs during the study period. In the case of the trip
level data, trips in which the number of sets reported on the logsheet and by the observer did not
match exactly were excluded. This left a total of 1930 trips for comparison purposes.
Estimating the extent of misreporting in the context of IUU in the purse seine sector is challenging
because large a volume or mixed species catch can be taken at one time, and records of catch and
species composition made by both the vessel and observer are estimates of weights made at sea.
While observer estimates of weight and species composition are made using a standardised
methodology (volume of brail * number and fullness of brails; grab samples to assist in determining
species composition) and following standardised PIRFO training, no scales are used, so comparisons
of logsheet data with that found in observer reports remain estimates vs estimates. With that in
mind, and consistent with the 2016 study, two measures were taken to minimise the impact of any
estimation errors:
First, data used for comparisons were at the TRIP level given most purse seine trips can
have between 15-40 sets, using data at the trip level should help to dampen out any set level
variability in estimates and provide a higher probability of any reporting differences being
persistent;
10
Under an initiative of the International Seafood Sustainability Foundation (ISSF) and participating companies,
comprehensive cannery receipts data has been provided to the WCPFC for trips involving WCPFC purse seine
catch by more than 20 processing companies since around 2013. See Williams (2020) for details. While this
data shows promise as an independent means of validating logsheet reporting, the terms under which data
was shared did not allow for access by this study.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
22
Second, a 10% tolerance level was applied to all comparisons i.e. if the logsheet estimate
of catch volume was within 10% of the observer estimate, we assumed this was a
reasonable attempt at reporting accurately. If the logsheet estimate was >10% different to
the observer estimate, this was included in estimates of misreporting.
Importantly, the 10% figure is a nominal value based on an often used ‘rule of thumb’ in the region
when examining cases of potential misreporting. There is no tolerance level specified in regional
management measures, although there is a clear recognition amongst all parties that estimates
made at sea will necessarily involve some degree of imprecision. To that end, the 10% tolerance
applied here is essentially a ‘judgement call’ for the purposes of estimation. Applying a different
tolerance level e.g. 0% or 20% - would result in different outcomes. Results of the misreporting
analysis should be seen in that context. Given the volumes involved in the purse seine sector, the
choice of tolerance level can have important implications for overall estimates of regional IUU
volume and value.
The other important assumption made in examining evidence of misreporting was that the
observer’s estimate was correct. While Pacific Island observers are all trained and certified
according to a common PIRFO standard and estimate catch weight and species compositions
according to standardised methodologies (see above), ultimately there is no guarantee their
estimates of weight or species composition will be more accurate than the estimates of experienced
skippers. Moreover, the challenges associated with the ‘grab’ sampling technique used by observers
to estimate species composition have been well-documented (e.g. Lawson, 2009, 2010, 2012;
Peatman et al, 2020). Nevertheless, observer data represents the best independent information
available and preliminary comparisons show observer data is more closely aligned with cannery out-
turn reports than logsheet data (Williams, 2020).
In examining misreporting, estimates of catch recorded by vessels and observers can vary in both
volume and species composition, and there are multiple possible ‘types’ of misreporting. For
example, where fish reported by an observer do not appear to be reported anywhere in the vessel
logsheet, these may be classified as ‘not reported’. Where a fish of species A appears to have been
reported as species B, this may be classified as ‘misidentified’. How they’re classified does not
affect the volume of catch misreported (the total volume of catch misreported is captured,
irrespective of how they’re categorised), but it does affect the potential economic loss involved. For
example, 10t of YFT not reported may result in the full ‘loss’ of the value of the fish. If the same 10t
of YFT was reported in the vessel logsheet as SKJ, only the marginal difference between YFT and SKJ
is lost (in theory).
Capturing these differences has particular relevance in the purse seine sector given challenges
associated with reporting of species composition in logsheets (e.g. Lawson, 2014, Peatman et al,
2017). Data provided by SPC at the trip level appeared to highlight these challenges. Across the
1930 matching trips, a relatively high proportion of logsheet estimates of total volume (i.e. across all
species) were within the 10% level of tolerance compared to the observer, but this reduced
progressively with individual species estimates for SKJ, YFT and BET (Figure 12). On average,
logsheets overestimated the catch of SKJ while underestimating the catch of YFT and BET, although
each species had instances of both underreporting and overreporting.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
23
Figure 12: Comparisons of weights recorded in logsheets and observer reports for individual target tuna species and total
weight at the TRIP level in the purse seine fishery, 2017-2019. (n=1930; Black lines show a 1:1 relationship between
logsheet and observer estimates. Red data points are outside of the 10% tolerance level.)
To ensure consistency in the analysis during the 2016 study, a framework of decision rules was
applied to categorise instances of misreporting in the PS sector. A similar approach was undertaken
in this study, except that we have refined the decision rules slightly to separate out cases of over-
reporting (i.e. where the logsheet recorded catches >10% higher than the observer) from cases of
under-reporting (Figure 13). While the majority of cases in which the logsheet record was >10%
different to the observer estimate involved under-reporting in the logsheet, there were also a
proportion of cases in which the logsheet appeared to over-report catch (particularly for SKJ). To
that end, a separate decision rule was added to separate out cases of over-reporting from under-
reporting.
Importantly, where an instance of misreporting was >10% and triggered a decision rule, the full
difference between the observer and vessel estimates was classified according to the type (under-
reported, over-reported, misidentified). That is, if the difference between the observer and vessel
was 11%, the full amount was recorded, not simply the excess over 10%.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
24
Figure 13: Decision rules used to categorise reporting offences in the purse seine fishery.
Based on the decision rules, the misreporting rate per set for each of the misreporting categories
was calculated across the 1930 trips (Table 6). Given that trips had varying numbers of sets, logbook
and observer data were subject to random resampling with replacement and were bootstrapped
10,000 times to generate mean misreporting rates (+/- 95% confidence intervals) for each species.
These bootstrapped misreporting rates and associated 95%Cis were then scaled up across the study
area based on the total number of sets during the study period to determine the total volume
misreported for each species.
Non-target species (OTH) comprised only a small proportion of the total PS catch. Catches of OTH
were very rarely over-reported by more than 10%, and where this occurred we assumed that the
vessel’s estimate was likely to have been correct (e.g. the observer may have missed a small number
of other species). To that end, no allowance was made for over-reporting.
Misreporting of discards was also estimated for trips where observer estimates of discarded catch
were more than 10% higher than those in logsheets. Although over-reporting of discards in the
logsheet compared to the observer report did occur, this was not considered ‘unreported’ for the
purposes of this analysis. To that end, only instances of underreporting >10% were included in
estimates.
Table 6: Average rates of misreporting for each category across sample trips (in tonnes per set).
Category
SKJ
YFT
BET
OTH
Dist.
Retained catch (t/set)
Under-reported
0.1398
0.5395
0.2337
0.0235
Over-reported
0.5781
0.1663
0.1287
0
Misidentified
0
0.3815
0.1722
0
Discarded catch (t/set)
Under-reported
0.0675
0.0082
0.0014
0.0649
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
25
Longline
In the longline sectors, there are several ways in which catches could be misreported in logsheets:
Under- or misreporting the number of each species taken it is the number, rather than
weight, of fish used in regional stock assessments, so accurate reporting of this figure is
particularly important. Given fish are landed individually, there is little excuse to get the
numbers wrong;
Misreporting of catch weight on the standard SPC/FFA Regional Longline Logsheet, vessels
are required to record the number of retained and discarded fish, as well as an estimated
weight of retained fish. Even where numbers are reported accurately, weights may be
under- (or over-) reported (either deliberately or inadvertently since weight reported on the
logsheet is a visual estimate). Incentives to under-report would be strong where access fees
are structured around catch volumes or value, or where other quantitative catch limits
apply. Underestimation of weights would be particularly important where compliance
against catch-based limits (e.g. the WCPFC BET limits) was determined against estimated
weights in the absence of independent verification;
Misidentifying species this is perhaps less of an issue in the longline sector than the purse
seine sector given fish are caught individually and are generally of a larger size (and
therefore easier to identify accurately), but may still be a risk, particularly where limits apply
on specific species (e.g. in the case of BET, there may be a temptation to report smaller BET
as YFT to stay under BET catch limits).
Given the very limited observer coverage across much of the longline fleet, and the absence of other
means of independently verifying catch and effort, the information base upon which to examine
misreporting is relatively weak compared to the purse seine sector.
For target tuna species retained by the vessel, the best information available against which to
compare logsheet reporting is unloadings records. These are records of individually weighed fish
taken by independent monitors (or by fishing companies generally overseen by independent
monitors) of all fish offloaded from vessels at port. Unloadings coverage of longline vessels in FFA
member ports has improved substantially in recent years and is now relatively high (~70%).
However, a key weakness of the unloadings dataset is that it covers a portion of the fleet only i.e.
those vessels returning to FFA member ports to unload. To that extent it will over-represent the
domestic and domestically-based fleets at the expense of long-range distant water fleets, for which
there were few available independent data for comparative purposes. In the context of this study,
this means that the data being used are representative of fleets subject to more intensive MCS
coverage, and arguably have greater incentives for compliance. Another limitation is that
monitoring may be largely focused on target tuna species, with less coverage of other species.
While there are data available for a small number of observer trips on longline vessels, coverage
within FFA member waters remains very low for many fleets and there are challenges involved in
making ‘apples vs apples’ comparisons between observer reports and logsheets in the longline
sectors. In particular, because setting and hauling takes place over many hours in the longline fleet
(hauling typically takes ~10-12hrs) observers are unable to monitor every hook that comes aboard.
While observer data can be used to support comparisons of catch rates between the logsheet and
observer (e.g. catch/hook), assumptions need to be made that catch rates of observed and
unobserved hooks are consistent (which may not always hold true). Electronic monitoring (EM) has
the potential to address some of these challenges, although coverage in the WCPO longline sector
remains limited to a few relatively small scale trials at this stage. To that end, unloading data
represents the best current opportunity to make ‘apples vs apples’ comparisons with logsheet data
for retained species (at least for the portion of the fleet unloading in FFA member ports).
For species discarded by the vessel at sea, the information base is particularly weak, with the best
available information coming from limited observer coverage and small scale EM trials. While EM
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
26
information is able to provide some estimate of total number and composition of species discarded,
for the reasons described above, comparisons between observer and logsheet discard reporting can
only usefully be done based on catch rates. Given the relatively small scale nature of EM trials at
this stage, observer catch rates were used as the primary dataset to examine reporting of discards
by vessels, although estimates were triangulated against available EM trial results.
The basic equation used to estimate IUU volumes was the difference between the volumes reported
in vessels logsheets and the estimated actual catch, taking into account the estimated levels of
misreporting. Rates of misreporting for both retained species and discards were calculated using the
data described below. Estimates were produced for each of five main species groups ALB, BET,
YFT, BIL and OTH across both retained and discarded catch.
Given the limitations in the sample data, and the lack of independent comparative data for large
sections of the WCPO LL fleet (particularly for high seas vessels), substantial assumptions have had
to be made in these estimates and outcomes should be seen in that context.
Retained species
To support estimates of retained target tuna species reporting, SPC provided logsheet and
unloadings data for all ‘matched’ trips in which unloadings were monitored in FFA member ports.
After removing those with obvious uncertainties (e.g. trips that didn’t pass automatic data quality
checks, trips in which zero values were recorded for either the logsheet or unloadings), data was
available for a total of 6,227 individual trips across the 2017-2019 study period. This was a
substantial improvement on the dataset available for the 2016 study for which only 564 trips were
available, solely from one port (Suva).
Within the 2017-2019 dataset, trips were separated into TLL and SLL trips based on the location of
fishing activity in the logsheet. Of the 6,227 trips, 3,714 were in the SLL area and 2,513 trips were in
the TLL area. The key ports in which monitoring was undertaken for the SLL trips were Suva, Fiji,
Apia, Samoa, Nukualofa, Tonga, and Pago Pago, American Samoa. For TLL trips, the key ports
included Majuro, RMI, Pohnpei, FSM, Koror, Palau and Noro, Solomon Is.
In the TLL sector, comparisons of the numbers of target tuna recorded in logsheet vs unloadings
revealed a relatively tight relationship for ALB, a slightly looser relationship for YFT and a looser
relationship still for BET (Figure 14). Overall, ALB and BET were under-reported in logsheets
compared to unloadings records by 4.8% and 7.1% respectively, while YFT was over-reported in
logsheets compared to unloads by 5.9% (Table 7). A broadly similar pattern emerged in the SLL
sector, with a tight relationship between logsheets and unloads for ALB and progressively weaker
relationships for YFT and BET (Figure 15). In the SLL sector, all species were under-reported in
logsheets compared to unloads, with ALB, YFT and BET under-reported by 1.5%, 8.8% and 13.4%
respectively.
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
27
Figure 14: Comparison between numbers of target tuna species and swordfish reported in logsheet (x-axis) and recorded at
unloading (y-axis) for matched trips in the TLL, 2017-2019. (n= 2513 trips for target tuna species; n=576 for swordfish; black
lines show 1:1 relationship between logsheet and unloadings; r = Pearson’s correlation coefficient)
Figure 15: Comparison between numbers of target tuna species and swordfish reported in logsheet (x-axis) and recorded at
unloading (y-axis) for matched trips in the SLL, 2017-2019. (n= 3714 trips for target tuna species; n=2664 for swordfish;
black lines show 1:1 relationship between logsheet and unloadings; r = Pearson’s correlation coefficient)
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
28
Table 7: Total numbers of fish reported in logsheets (LOG) and unloadings reports (UNL) across 2,513 matched trips in the
TLL sector and 3,714 matched trips in the SLL sector.
TLL
SLL
Species
LOG
UNL
%
under/over
LOG
UNL
%
under/over
ALB
195529
205488
4.8%
3397482
3448094
1.5%
YFT
349049
329663
-5.9%
793006
869979
8.8%
BET
180430
194197
7.1%
149363
172547
13.4%
While there have generally been very few prosecutions for misreporting in the longline sector
amongst FFA members, some instances of suspected or confirmed misreporting have been detected.
For example, Samoa reported suspected under-reporting of BET amongst a number of Chinese
Taipei longline vessels in 2019, while the Solomon Is sanctioned a vessel for misreporting in 2017.
During Operation Rai Balang ’18 the FSS Micronesia apprehended a Chinese-flagged longliner for
misreporting catch, while during Operation Kurukuru ’19 she apprehended another vessel for
misreporting. In addition to these in zone offences, New Zealand high seas boarding and inspection
detected a serious case of under-reporting in 2017 (New Zealand, 2017). In this case, the master of
the vessel initially stated that there was no BET on board, but a subsequent search of the vessel
holds revealed 5t of unreported BET (Figure 16). During subsequent questioning, the master
admitted that he was keeping the BET separate ‘for his own greed’ with a separate running record of
BET catches found in a notebook on the bridge.
Figure 16: Unreported frozen BET detected by a New Zealand high seas boarding, 2017 (NZ, 2017).
For both the TLL and SLL sectors we have used the outcomes of the logsheet vs unloadings
comparisons to inform our best estimate values for misreporting and set our min/max range based
on plausible scenarios to account for uncertainty in fleets not covered by unloadings data. Broadly:
the best estimate value has been set at the value of the overall difference between
logsheets vs unloadings. This assumes that patterns of reporting behaviour in fleets covered
by unloadings data are broadly reflective of the full fishery;
the minimum value has been set assuming that the unloadings results are reflective of the
in-zone fleet and the remainder of the fleet reports all fish accurately. This is perhaps
conservative given the evidence of misreporting described above and the incentives to
under-report for some species (e.g. BET) with catch limits. Nevertheless, it reflects catch
reporting to the WCPFC indicating quantitative catch limits are rarely exceeded. Minimum
The Quantification of IUU Fishing in the Pacific Islands Region a 2020 Update
29
values have been set taking into account the proportion of effort in zone and on the high
seas in each sector;
the maximum estimated value has been set to account for the possibility that misreporting
behaviour in fleets not covered by unloadings are up to 50% higher than those that are.
Maximum values have been set taking into account the proportion of effort in zone and on
the high seas in each sector.
For BET, this approach produces estimates which are broadly in line with the results of EM trials in
FSM, RMI and Palau, in which BET were under-reported in logsheets between 2-5% across 1052 sets
(Brown et al, 2021). For ALB, the estimates are below the results of the same EM trial, which
showed numbers of ALB detected by EM were up to three times higher than that reported in
logsheets (albeit the EM trials were undertaken in the TLL sector where ALB catch is comparatively
rarer and the total numbers of fish in the trial were not provided). To that extent, the ALB figure for
TLL may be an under-estimate and highlights the considerable uncertainties in longline datasets.
The main challenge to the approach described above is for YFT in the TLL sector for which logsheet
records exceeded unloadings. One possible explanation for this is that BET have been misidentified
as YFT, either inadvertently or deliberately (e.g. to comply with BET catch limits). However, vessels
landing to FFA member ports are typically flagged (or chartered) to Pacific Island states which are
not subject to BET limits and our analysis of TLL trips showed only 5 trips (0.2%) in which YFT is larger
in logsheets than unloadings and BET is smaller in logsheets than unloadings. Another possible
explanation is that YFT may be sold locally by some vessels and not included in unloadings records.
The YFT results are contrary to the outcomes of the EM trial analysis described above in which YFT
were under-reported in logsheets by up to 30% when compared to EM (Brown et al, 2021).
Observer data in the TLL sector was limited to only 20 trips across the three years, with the
attendant complications of not being able to monitor all sets, so was of limited value as a
comparison.
In the absence of better independent estimates for YFT, we have set the minimum value at the
difference in the unloadings comparison (-5.9%) to account for the possibility that YFT is over-
reported, applied a conservative best estimate value of 0% given the conflict between the
unloadings and EM results, and a larger maximum value of 20% to account for the fact that at least
some evidence exists through EM trials of high rates of underreporting. We appreciate these
estimates are highly uncertain however and should be updated as better data becomes available.
For billfish and other species, the data upon which to make valid comparisons with logsheets was
weaker. Most monitoring of unloadings is focused only on target tuna species, with less attention
paid to billfish and other species, the latter of which are often sold direct into local markets. For
these species, we removed all trips for which the unloadings monitor reported zero catch from the
unloadings dataset. With swordfish the only billfish included in the unloadings dataset, we used the
reporting of this species as a proxy for all billfish reporting.
For swordfish in the TLL area, removing trips for which the unloadings monitor reported zero catch
reduced the number of monitored trips from 2513 to 576. In the SLL area, removing the zero
unloadings trips reduced the number of monitored trips from 3715 to 2664 indicating SWO were
more comprehensively monitored in SLL ports. Taking only the trips in which the unloadings
monitor reported some catch
11
, the relationship between the t