Changes in the Structure of Demersal Fish ...

Changes in the Structure of Demersal Fish Communities of the South East Australian Continental Shelf from 1915 to 1961

by

Neil L. Klaer, B.Sc. M.App.Sc. Applied Ecology Research Group University of Canberra Canberra ACT 2602

A thesis submitted in fulfilment of the requirements for a degree of Doctor of Philosophy at the University of Canberra

May 2006

a

Certificate of Authorship of Thesis

Except as specially indicated in footnotes, quotations and the bibliography, I certify that I am the sole author of the thesis submitted today entitled Changes in the Structure of Demersal Fish Communities of the South East Australian Continental Shelf from 1915 to 1961

in terms of the Statement of Requirements for a Thesis issued by the University Higher Degrees and Scholarships Committee.

Signature of Author ______________________________

Date

______________________________

i

Acknowledgements

There are a large number of people who have made substantial contributions to this thesis. Professor Robert Kearney (University of Canberra) was my primary supervisor for this project, providing support, encouragement and review of my work over the past five years despite the difficulties in supervising work being done externally in Hobart. Dr Tony Smith (CSIRO) was my secondary supervisor, providing support generally for my work, and review of drafts of this thesis.

Mr Richard Mansfield, a former owner of Red Funnel Trawlers Pty Ltd (one of the main steam trawler companies operating in the SE Fishery) supplied company catch records that helped considerably with the interpretation of details in the per-haul data already held. He also helped with first-hand knowledge of operating procedures in the 1950s and early 1960s. Mr Norm Colless, who was an apprentice on the steam trawl vessels after World War II, also provided valuable information on the operating procedures of the vessels.

Richard Tilzey (Bureau of Rural Sciences), Kevin Rowling and Ken Graham (NSW Fisheries) assisted with investigations of the location and documentation of steam trawl data held at NSW Fisheries. Mark Bravington (CSIRO) gave useful advice on the design and interpretation of the log-linear model analyses, and Rob Campbell, Peter Thompson (CSIRO) and Andre Punt (University of Washington) provided useful comments on the chapter on abundance indices. Keith Jones from the South Australian Research and Development Institute helped locate unpublished information on the biology of Chinaman leatherjackets. Ron Thresher (CSIRO) provided the sunspot, SOI, and latitude of the sub-tropical ridge mean annual data series. Ian Barnes-Keoghan (Australian Bureau of Meteorology) provided access to standardised data sets of mean annual rainfall and temperature by collection station. Beth Fulton and Cathy Bulman (CSIRO) assisted with parameter values and comments on modelling using Ecopath.

Parts of this work detailed in Chapters 3 and 4 and the Appendix were carried out under a project funded by the Fisheries Research and Development Corporation. The History of Marine Animal Populations sub-project of the Census of Marine Life project of the

ii

US Sloan Foundation assisted with funding the work presented in Chapters 5, 6, 7 and 8. The remainder of the work has received support from CSIRO as part of the assessment work for the SE Fishery, and within a more comprehensive SE Australian marine ecosystem project.

Last but not least, I would like to thank my 10-year-old daughter Nicole who particularly wanted to be mentioned in my thesis as she understood it to be an important piece of work.

iii

Abstract

Haul-by-haul steam trawler catch and effort data for 1918–23, 1937–43 and 1952–57, which covers a large portion of the history of steam trawling in the Australian South East Fishery, were examined in detail for the first time. There were 64,371 haul records in total. The catch-rate for all retained catch combined shows a strong decline overall, with a brief recovery during World War II, probably due to increased retention of previously discarded species. The fishing fleet moved to more distant fishing grounds and deeper waters as the catch-rate declined. The catch-rates of the main commercial species followed a similar pattern in a number of regions within the fishery. The catchrate of the primary target species – tiger flathead (Neoplatycephalus richardsoni) – dropped considerably from the early, very high, catch-rates. Chinaman leatherjacket (Nelusetta ayraudi) and latchet (Pterygotrigla polyommata) – species that were apparently abundant in the early years of the fishery – virtually disappeared from catches in later years. The appearance of greater catches of jackass morwong (Nemadactylus macropterus), redfish (Centroberyx affinis), and shark/skate during the war and afterwards was probably due to increased retention of catches of these species. The disappearance of certain species from the catch may be due to high fishing pressure alone, or to a combination of fishing pressure, changes in the shelf habitat possibly caused by the trawl gear, and environmental fluctuations.

Catch-rates in weight per haul per species were standardised to annual indices of abundance using a log-linear model. Standardised annual index trends for flathead, latchet and leatherjacket indicate a strong to severe decline over the period covered by the data. All species showed seasonal patterns, but the peak season varied depending on the species. The distribution of standardised catch-rate by area also differed greatly by species, and no single area showed consistent differences across all species. Day trawls caught more flathead, redfish and latchet, while night trawls caught more morwong and leatherjacket. Moon phase had less influence on catch-rates than the other factors examined. Correlation of annual index trends with a number of annual mean environmental factors was examined and no strong correlations were found.

iv

Annual catches of the major commercial trawl species on the SE Australian shelf were estimated from recorded total trawl catches, catch species composition from subsamples and estimates of the rate of discarding. These annual catches, standardised indices of abundance and biological population parameters were used in single-species stock reduction models to estimate absolute biomass trends. Biological population parameters and the biomass estimates were used to calculate management reference point fishing mortality rates F0.1, Fspr30 and Fmsy. Results showed that simple plausible population models can be constructed that account for catches over the long period of time from 1915 to 1961.

Simple mass-balance ecosystem models were built for the demersal community of the SE Australian shelf for 1915 and 1961 using the Ecopath software. Model inputs were consistent with a more comprehensive SE marine ecosystem model in development by CSIRO. The models demonstrate that biomass estimates produced by the single species stock reduction models can be consistently integrated into simple plausible massbalance ecosystem models.

Modern stock assessments for the main commercial species in this fishery today mostly used data collected since about 1985. Abundance indices and total catch estimates from this study have been used in the most recent assessments for tiger flathead and morwong, allowing construction of the exploitation history for these species spanning almost 100 years. Use of the historical information has increased confidence in the estimates of the modern stock assessments – particularly management reference points, and has allowed us to quantify changes in fish abundance that have simply been documented anecdotally in the past.

v

Table of Contents Certificate of Authorship of Thesis .........................................................................................i Acknowledgements .................................................................................................................ii Abstract..................................................................................................................................iv List of Figures ........................................................................................................................ix List of Tables..........................................................................................................................xi

CHAPTER 1 - Introduction ....................................................................................... 1 CHAPTER 2 – Sources of available data................................................................... 5 CHAPTER 3 – Conversion of CSIRO/NSW historical data ..................................... 7 3.1 Introduction.......................................................................................................................7 3.2 Data from 1918-23.............................................................................................................8 3.2.1 Raw data description ........................................................................................................8 3.2.2 Loading procedure............................................................................................................9 3.2.3 Data quality......................................................................................................................9 3.2.4 Data summary ................................................................................................................ 13 3.3 Data from 1937-43........................................................................................................... 15 3.3.1 Raw data description ...................................................................................................... 15 3.3.2 Loading procedure..........................................................................................................16 3.3.3 Data quality.................................................................................................................... 16 3.3.4 Data summary ................................................................................................................ 19 3.4 Data from 1951-57........................................................................................................... 22 3.4.1 Raw data description ...................................................................................................... 22 3.4.2 Loading procedure..........................................................................................................23 3.4.3 Data quality.................................................................................................................... 23 3.4.4 Data summary ................................................................................................................ 24 3.5 Final data formats ........................................................................................................... 26

CHAPTER 4 – Conversion of Red Funnel Trawlers historical data ...................... 27 4.1 Introduction..................................................................................................................... 27 4.2 Skipper Logbooks 1952-61.............................................................................................. 28 4.2.1 Raw data description ...................................................................................................... 28 4.2.2 Loading procedure..........................................................................................................29 4.2.3 Data quality.................................................................................................................... 32 4.2.4 Data summary ................................................................................................................ 35 4.3 Landing records 1938-59................................................................................................. 38 4.3.1 Raw data description ...................................................................................................... 38 4.3.2 Loading procedure..........................................................................................................40 4.3.3 Data quality.................................................................................................................... 41 4.3.4 Data summary ................................................................................................................ 45 4.4 Radio Reports 1946-57 .................................................................................................... 49 4.4.1 Raw data description ...................................................................................................... 49 4.4.2 Loading procedure..........................................................................................................50 4.4.3 Data quality.................................................................................................................... 52 4.4.4 Data summary ................................................................................................................ 57 4.5 Final data formats ........................................................................................................... 60

vi

CHAPTER 5 – Trends in catch rates and species composition............................... 65 5.1 Introduction..................................................................................................................... 65 5.2 Methods ........................................................................................................................... 68 5.2.1 Data summary ................................................................................................................ 68 5.2.2 Mean catch-rate of all landed commercial species........................................................... 69 5.2.3 Catch-rate by main commercial species by area .............................................................. 69 5.3 Results ............................................................................................................................. 71 5.3.1 Data summary ................................................................................................................ 71 5.3.2 Mean catch-rate of all landed commercial species........................................................... 74 5.3.3 Catch-rate by main commercial species by area .............................................................. 79 5.4 Discussion ........................................................................................................................ 85

CHAPTER 6 – Abundance indices........................................................................... 90 6.1 Introduction..................................................................................................................... 90 6.2 Methods ........................................................................................................................... 90 6.2.1 Input data .......................................................................................................................90 6.2.2 Astronomical Events ...................................................................................................... 91 6.2.3 Conversion of continuous variables into factors .............................................................. 92 6.2.4 Sensitivity tests .............................................................................................................. 93 6.2.5 Log-linear model (LLM) ................................................................................................ 93 6.2.6 Standardised annual abundance index correlation with available long-term environmental data......................................................................................................................................... 94 6.3 Results ............................................................................................................................. 96 6.3.1 LLM results for all retained catch ................................................................................... 96 6.3.2 Species-specific LLM results........................................................................................ 102 6.3.3 Species-aggregated LLM results................................................................................... 104 6.3.4 Standardised annual abundance index correlation with available long-term environmental data....................................................................................................................................... 105 6.4 Discussion ...................................................................................................................... 106 6.4.1 Model design and possible biases in the methods and data processing........................... 106 6.4.2 Changes in fishing practices through time..................................................................... 108 6.4.3 Conclusions.................................................................................................................. 111

CHAPTER 7 – Population modelling..................................................................... 113 7.1 Introduction................................................................................................................... 113 7.2 Methods ......................................................................................................................... 113 7.2.1 Catch history per main commercial species................................................................... 113 7.2.2 Stock reduction modelling............................................................................................ 118 7.2.3 Management Reference Points...................................................................................... 121 7.3 Results ........................................................................................................................... 124 7.3.1 Stock reduction modelling............................................................................................ 124 7.3.2 Management reference points ....................................................................................... 128 7.4 Discussion ...................................................................................................................... 129

CHAPTER 8 – Ecosystem modelling ..................................................................... 132 8.1 Introduction................................................................................................................... 132 8.2 Change in the SE shelf ecosystem prior to 1915 ........................................................... 132 8.2.1 Natural fluctuations ...................................................................................................... 133 8.2.2 Human extraction of marine organisms......................................................................... 133 8.2.3 Human-assisted invasions of exotic species .................................................................. 137 8.2.4 Human alteration to the physical environment .............................................................. 138 8.3 Methods ......................................................................................................................... 139 8.4 Results ........................................................................................................................... 141 8.5 Discussion ...................................................................................................................... 145

vii

CHAPTER 9 – General discussion......................................................................... 147 9.1 Implications of the results for current management and further research ................. 147 9.2 International significance of this work ......................................................................... 153

References ............................................................................................................... 155 Appendix 1 –Red Funnel Trawlers conversion tables ........................................... 165

viii

List of Figures Figure 2.1. Summary of historical per-haul data available for the SE Australian shelf region by source (see text for explanation of sources). .......................................................................5 Figure 3.1. Catch positions - 1918-23......................................................................................10 Figure 3.2. Estimated catch positions - 1937-43.......................................................................17 Figure 3.3. Estimated catch positions - 1951-57.......................................................................23 Figure 4.1. Estimated catch positions (200 and 1000 m isobaths also shown)...........................34 Figure 4.2. Comparison of catches per individual trip from skipper logbooks and corresponding landings by species. .........................................................................................................44 Figure 4.3. Estimated radio report positions.............................................................................56 Figure 5.1. Photograph of the steam trawler Moona – one of the last to operate in New South Wales waters....................................................................................................................66 Figure 5.2. Early trawl grounds in shelf waters after Colefax (1934) (shaded blocks), and spatial strata used for the analyses in this study (between the horizontal lines). ............................67 Figure 5.3. CPUE for steam trawlers only for all landed catch (from Houston 1955)................75 Figure 5.4. CPUE for retained commercial catch by year. CPUE from haul-by-haul records in kg/h are shown as points on solid lines, and historical CPUE as presented by Houston (1955) in trawler-ton-months is shown as dashed lines. ....................................................75 Figure 5.5. Mean latitude fished by year..................................................................................77 Figure 5.6. Mean latitude fished by year and month.................................................................78 Figure 5.7. Mean latitude fished by month over all years. ........................................................78 Figure 5.8. Mean depth fished by year.....................................................................................78 Figure 5.9. Proportion of total hours trawled for hauls with location information (as per Figure 5.2) that were made in each area.......................................................................................79 Figure 5.10. Contribution per species to the total commercial CPUE by year for area D. .........80 Figure 5.11. Contribution per species to the total commercial CPUE by year for area G. .........81 Figure 5.12. Contribution per species to the total commercial CPUE by year for area H. .........82 Figure 5.13. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952– 57 in area D. ....................................................................................................................83 Figure 5.14. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952– 57 in area G. ....................................................................................................................83 Figure 5.15. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952– 57 in area H. ....................................................................................................................84 Figure 5.16. Contribution per species to the total commercial CPUE by year for area H in depths of less than 100 m. ...........................................................................................................85 Figure 6.1. Fitted vs residual values and normal quantile-quantile plot for the all retained catch LLM. ...............................................................................................................................97 Figure 6.2. LLM results for year. Transformed (antilog) values are shown, as well as 95% confidence intervals. The percentage of all records having zero or 1000 kg per haul) in comparison with later steam trawler catch rates (approximately 300 kg per haul). This may be due to incorrect conversion of catch weight from the original records. The original records are reported to no longer exist, so checking is probably impossible. A common feature of many developing fisheries is high catch rates initially, so the data may also be correct. 12

3.2.4 Data summary Total catch by year and vessel Total catch and number of hauls per year and vessel for the 1918-23 data are given in Table 3.5 and Table 3.6. Table 3.5. Total retained catch and number of hauls by year. Year 1918 1919 1920 1921 1922 1923

Total Catch (kg) 2,633,491 4,646,646 2,925,533 6,080,484 8,507,776 1,225,367

Hauls 1,759 2,324 1,110 2,769 3,587 7,492

Table 3.6. Catch and number of hauls by vessel. Vessel Brolga Dibbiu Dureenbee Goonambee Goorangai Gunundaal Koraaga

Retained wt (kg) 2,520,637 1,246,122 584,111 1,392,307 1,190,492 1,100,573 1,905,516

Discard wt (kg) 470,800 272,459 58,878 329,855 380,481 65,214 309,609

Hauls 2,660 1,554 688 1,574 1,437 1,911 2,217

Catch by depth Total catches by depth interval (0=0-20) are given in Table 3.7. Depths have been converted from fathoms to metres. A small number of hauls (17) had a recorded depth of greater than 600 m. It was assumed in this case that the depth was recorded in feet instead of fathoms, and the recorded depth was converted accordingly. The absence of catches in depths between 200 and 600 m supported this assumption.

13

Table 3.7. Total retained and discarded catch weight by depth interval. Depth (m) 0 20 40 60 80 100 120 140 160 180 200

Retained wt (kg) 4,659 3,852 802,340 3,505,460 1,697,237 2,686,399 1,146,112 39,223 4,854 6,168 816

Discard wt (kg) 1,444 1,179 157,597 666,309 253,383 485,909 295,946 7,033 1,360 1,361 16

Hauls 9 6 906 3,309 2,308 3,847 1,549 59 6 9 1

Catch by year and species Total retained catch by year per species is presented in Table 3.8. Table 3.8. Total retained catch (kg) by species by year. Code BA DO FL GU JD LJ MI MO NA PE SA SK SN TH

Species unknown barracouta dory tiger flathead latchet john dory chinaman leatherjacket mixed jackass morwong redfish ocean perch unknown rays snapper thetis fish

1918 4,264 28,188 159 578,315 192,494 20,381 103,416 4,642 14,431 1,084 893 5,406 58,652 1,465 0

14

1919 1920 1921 1922 1923 14,605 2,603 10,261 32,201 3,764 93,656 10,383 44,722 38,735 476 816 0 0 0 0 992,030 485,606 1,483,128 2,113,319 311,286 337,118 192,128 228,569 215,015 21,624 11,827 8,334 7,119 225 0 282,033 375,329 472,844 682,457 104,413 2,954 23,582 45,691 51,056 1,264 0 992 3,493 456 0 2,409 3,800 2,481 1,800 0 0 0 0 0 0 5,831 2,046 0 0 0 118,863 28,970 27,918 17,581 157 1,595 4,416 204 1,339 0 0 0 445 0 0

3.3 Data from 1937-43 3.3.1 Raw data description The field description accompanying the data was as follows: Table 3.9. 1937-43 Data description. Field Width trip information vessel-year 24 vessel code 2 trip number 3 depart date 6 depart time 4 return date 6 return time 4 No. of hauls 2 No. Of species 2 last trip 1 species 1 2 sp 1 catch 8 species 2 2 sp 2 catch 8 . . . . species 12 2 sp 12 catch 8 haul information haul No. 2 date 6 initial time 4 final time 4 area name 6 area code 2 initial depth 3 final depth 3 species 1 2 species 1 catch 6 . . species 8 2 species 8 catch 6 Notes:

Position Type

Comments

1-24 A 25-26 N 27-29 N 30-35 N 36-39 N 40-45 N 46-49 N 50-51 N 52-53 N 54 N 55-56 N 57-64 N 65-66 N 67-74 N . . 165-166 N 167-174 N

*

175-176 N 177-182 N 183-186 N 187-190 N 191-196 N 197-198 A 199-201 N 202-204 N 205-206 N 207-212 N . 261-262 N 263-268 N

(yymmdd) (24 hr clock) (yymmdd) (24 hr clock) # (blank,0,1) % * (baskets) * (baskets)

* (baskets)

(yymmdd) (24 hr clock) (24 hr clock)

(fathoms) (fathoms) * (baskets) * (baskets)

A - Alphabetic N - Numeric * - Refer to code lists # - No. Of species recorded in species catch (1) - (12) % - ‘1’ denotes last trip for the year 1 basket = 70 lb

15

3.3.2 Loading procedure The data were converted to the format described in section 3.5 using dBase III+ databases and purpose-built procedures using the Borland Clipper compiler.

3.3.3 Data quality A total of 31,266 individual haul records were available for the 1937-43 period. A summary of the completeness of important fields is given in Table 3.10. Table 3.10. Data completeness 1937-43. Field depth fished species catch wt position vessel name date

Records 13,634 31,266 0 31,266 31,266

% of total 43.61 100.00 0.00 100.00 100.00

Note:- depth applies to initial or final depth - position is both latitude and longitude - 100% of records had an encoded fishing ground

Catch positions No hauls had latitude or longitude recorded. A six-character fishing ground code was given, and these were interpreted as shown in Table 3.11 based on similarity with names of geographic features in the region. The positions of most of the fishing grounds were located using a computerised gazetteer. All ground positions were plotted on a map showing the coastline and 200 m and 1000 m depth contours. A table was then constructed that contained the position of the ground at the coast or island (depth 0), and the positions where a line from that point crosses the 200 m and 1000 m contours most directly towards the continental shelf (Table 3.11). For each haul, it was then possible to assign a position in latitude and longitude according to the fishing ground and a linear interpolation of the average depth of the shot. Where no depth was given, 100 m was used.

16

Figure 3.2 shows a map of the resulting estimated catch positions. Figure 3.2. Estimated catch positions - 1937-43.

Notes as for Figure 3.1.

17

Table 3.11. Fishing ground codes and assumed locations. Code BABEL BASSPT BATE BERMAG BIRD BOTANY BRGHTN BROKEN BRRNJY BRUSH CATHRN CHRLTT CLCLFF CPFRST CPGREN CPHOWE CPHWKE CPSMKY CRNLLA CROWDY DSASTR EDEN EVRARD GABO GOALEN HAYSTK HCKING HELENS JERVIS KEMBLA KIAMA KRGORO LAKENT LILVAL LONGPT MARIA MARION MARLEY MCQRIE MONTAG MOON MORNA MORUYA MOWRRY MRMBLA NORAH NWCSTL NWZLND OHARA PINES RED HD SHLHVN SISTER STPHNS SYDHDS TATHRA TLLGTS TNCRRY WABUNG WARDEN WATMOL WRECK

Ground Babel Is Bass Point Batemans Bay Bermagui Bird Is Botany Bay Broughton Is Broken Bay Barrenjoey Pt Brush Is Catherine Hill Bay Charlott Hd Coalcliff Cape Forster Cape Grenfell Cape Howe Cape Hawke Smoky Cape Cronulla Crowdy Hd Disaster Bay Eden Cape Everard Gabo Is Goalen Haystack Rock Port Hacking St Helens Jervis Bay Port Kembla Kiama Korogoro Is Lakes Entrance Lily Vale Long Point Maria Is Marion Bay Marley Beach Port Macquarie Montague Is Moon Bay Unknown1 Moruya Mowarry Pt Merimbula Norah Hd Newcastle New Zealand O'Hara Hd The Pines Red Hd Shoalhaven Hd Unknown2 Port Stephens Sydney Hd Tathra Tollgate Is Tuncurry Wybung Head Warden Wattamolla Wreck Bay

Lat 0 m 3957 3436 3544 3626 3314 3400 3237 3336 3335 3532 3309 3220 3415 3211 3454 3730 3213 3055 3402 3151 3716 3704 3748 3734 3631 4212 3404 4120 3507 3429 3440 3103 3752 3412 3345 4240 4248 3407 3127 3615 3642 0 3555 3709 3654 3317 3256 3800 3534 3158 3515 3451 0 3242 3351 3644 3545 3211 3312 3523 3408 3511

Long 0 m Lat 200 m 14820 3957 15005 3436 15015 3544 15004 3626 15136 3314 15114 3400 15219 3237 15119 3336 15120 3335 15025 3532 15138 3309 15233 3220 15059 3415 15231 3211 15036 3454 14959 3730 15234 3213 15305 3055 15111 3402 15245 3151 14958 3716 14955 3704 14916 3807 14955 3734 15003 3631 14804 4212 15106 3404 14815 4120 15046 3507 15055 3429 15051 3440 15302 3103 14800 3820 15100 3412 15115 3345 14808 4240 14800 4248 15108 3407 15255 3127 15014 3615 14959 3642 0 0 15008 3555 15000 3709 14956 3654 15135 3317 15146 3256 17035 3800 15023 3534 15236 3158 15033 3515 15045 3451 0 0 15210 3242 15118 3351 14959 3644 15016 3545 15230 3211 15140 3312 15030 3523 15108 3408 15038 3511

Long 200 m 14852 15111 15050 15015 15214 15133 15240 15152 15154 15043 15217 15254 15124 15254 15103 15014 15256 15312 15135 15303 15017 15017 14922 15013 15020 14832 15131 14837 15058 15116 15110 15312 14831 15124 15145 14824 14821 15127 15308 15018 15016 0 15028 15017 15016 15211 15233 17035 15041 15300 15054 15106 0 15241 15142 15016 15034 15256 15214 15047 15128 15055

Lat 1000 m Long 1000 m 3957 14858 3436 15120 3544 15056 3626 15020 3314 15224 3400 15151 3237 15251 3336 15208 3335 15208 3532 15047 3309 15229 3220 15301 3415 15135 3211 15305 3454 15111 3730 15018 3213 15305 3055 15319 3402 15152 3151 15319 3716 15021 3704 15021 3820 14926 3734 15018 3631 15025 4212 14837 3404 15149 4120 14848 3507 15106 3429 15126 3440 15120 3103 15319 3829 14844 3412 15137 3345 15205 4240 14831 4248 14827 3407 15142 3127 15315 3615 15023 3642 15021 0 0 3555 15031 3709 15021 3654 15020 3317 15219 3256 15239 3800 17035 3534 15045 3158 15312 3515 15100 3451 15113 0 0 3242 15250 3351 15205 3644 15020 3545 15038 3211 15305 3312 15224 3523 15053 3408 15143 3511 15103

Note: Positions shown here are in degrees and minutes, with minutes as the last two digits.

18

Species identification Two letter species codes were used in the data, and were interpreted in the same way as described for the 1918-23 data (Table 3.12). Table 3.12. Species codes in the 1937-43 data, and assumed identification. Name Code FL MI MO LJ NA GU SH JD SA SN LA DO KG SK PE RA

CSIRO code

Wt (kg)

Records

296001 999999 377003 465006 258003 288006 18000 264004 1 353001 288006 264000 337006 31000 287001 31000

6,033,406 2,033,521 928,455 499,216 431,481 199,339 114,240 107,522 51,127 14,767 13,705 4,978 1,461 1,296 222 64

30,318 13,229 5,442 3,701 1,600 2,492 1,049 1,671 941 35 334 88 4 34 1 2

tiger flathead mixed jackass morwong chinaman leatherjacket redfish latchet shark john dory unknown snapper latchet dory kingfish rays ocean perch rays

3.3.4 Data summary Total catch by year and vessel

Total catch and number of hauls by year and vessel for the 1937-43 data are shown in Table 3.13 and Table 3.14. Discarded catch was not recorded. Table 3.13. Total retained catch and number of hauls by year. Year 1937 1938 1939 1940 1941 1942 1943

Total catch (kg) 1,950,643 3,055,925 3,066,765 730,804 878,350 617,398 128,428

Hauls 6,476 9,222 9,337 2,192 2,416 1,396 227

19

Table 3.14. Catch and number of hauls by vessel. Vessel Alfie Cam Bareamul Beryl 2 Dureenbee Goolgwai Goonambee Goorangai Korowa Mary Cam Olive Cam Samuel Benbow

Retained wt 888,574 1,892,268 828,382 1,498,070 521,114 1,026,538 847,519 530,306 1,457,932 816,626 120,984

Hauls 2,928 4,804 2,577 4,738 1,589 3,189 2,709 1,515 4,300 2,547 370

Catch by depth Total catches by depth interval (0=0-20) are given in Table 3.15. Depths have been converted from fathoms to metres. All recorded depths were within the ranges shown and no adjustments for mis-recording were made. Table 3.15. Total retained catch weight by depth interval. Depth (m) 0 20 40 60 80 100 120 140 160 180 200 220 240

Retained wt (kg) 826 1,937 57,084 187,007 714,170 905,706 1,468,218 970,727 85,727 50,733 40,584 2,572 159

Hauls 3 6 213 756 2,465 2,883 4,106 2,572 251 181 182 10 1

20

Catch by year and species Total retained catch by year per species is presented in Table 3.16. Table 3.16. Total retained catch (kg) by species by year. Code DO FL GU JD KG LA LJ MI MO NA PE RA SA SH SK SN

Species 1937 1938 1939 1940 1941 1942 1943 dory 0 1,435 0 480 1,311 1,752 0 tiger flathead 1,132,089 1,942,816 1,952,715 410,179 348,055 186,122 61,430 latchet 32 76,900 48,766 19,437 29,670 20,481 4,053 john dory 9,817 27,347 21,943 14,946 21,538 8,644 3,287 kingfish 0 0 0 0 223 1,238 0 latchet 0 2,855 7,046 2,412 112 416 864 chinaman leatherjacket 2,574 54,762 160,777 67,494 96,066 92,371 25,172 mixed 553,751 582,461 520,140 144,368 159,410 72,529 862 jackass morwong 209,782 322,307 249,183 35,861 82,573 28,398 351 redfish 43,514 40,287 75,653 21,753 118,086 129,925 2,263 ocean perch 0 0 0 0 0 222 0 rays 0 64 0 0 0 0 0 unknown SA 0 5,828 16,989 13,742 11,673 2,895 0 shark 0 320 0 574 10,150 72,937 30,259 rays 0 256 656 96 288 0 0 snapper 0 0 14,767 0 0 0 0

21

3.4 Data from 1951-57 3.4.1 Raw data description The field description accompanying the data was as follows: Table 3.17. 1951-57 Data description. Field trip information vessel-year vessel code trip number depart date depart time return date return time No. of hauls No. Of species last trip species 1 sp 1 catch species 2 sp 2 catch . . species 10 sp 10 catch haul information haul No. date initial time final time area name area code initial depth final depth species 1 species 1 catch . species 8 species 8 catch Notes:

Width Position Type Comments 24 2 3 6 4 6 4 2 2 1 2 8 2 8 . .

.

1-24 25-26 27-29 30-35 36-39 40-45 46-49 50-51 52-53 54 55-56 57-64 65-66 67-74 . . 2 145-146 8 147-154

A N N N N N N N N N N N N N

# (blank,0,1) % * (baskets) * (baskets)

N N

* (baskets)

2 6 4 4 6 2 3 3 2 6

N N N N N A N N N N

(fathoms) (fathoms) * (baskets)

N N

* (baskets)

155-156 157-162 163-166 167-170 171-176 177-178 179-181 182-184 185-186 187-192 . 2 241-242 6 243-248

A - Alphabetic N - Numeric * - Refer to code lists # - No. Of species recorded in species catch (1) - (10) % - ‘1’ denotes last trip for the year 1 basket = 70lb

22

*

(yymmdd) (24 hr clock) (yymmdd) (24 hr clock)

(yymmdd) (24 hr clock) (24 hr clock)

3.4.2 Loading procedure The data were converted to the format described in section 3.5 using dBase III+ databases and purpose-built procedures using the Borland Clipper compiler.

3.4.3 Data quality A total of 21,120 individual haul records were available for the 1951-57 period. A summary of the completeness of important fields is given in Table 3.18. Table 3.18. Data completeness 1951-57. Field depth fished species catch wt position vessel name date

Records 19,345 21,120 0 21,120 21,120

% of total 91.60 100.00 0.00 100.00 100.00

Note:- depth applies to initial or final depth - position is both latitude and longitude - 100% of records had an encoded fishing ground

Catch positions No hauls had latitude or longitude recorded. The six-character fishing ground code and depth information was converted to estimated catch positions as described for the 1937-43. Figure 3.3 shows a map of the resulting estimated catch positions. Figure 3.3. Estimated catch positions - 1951-57.

Notes as for Figure 3.1.

23

Species identification Two letter species codes were used in the data, and were interpreted in the same way as described for the 1918-23 data. Table 3.19. Species codes in the 1951-57 data, and assumed identification. Code DO FL GU JD JW LA LJ MA MI MO NA RA SA SD SH SN

Name dory tiger flathead latchet john dory jewfish latchet chinaman leatherjacket mackerel mixed jackass morwong redfish rays unknown SA silver dory shark snapper

CSIRO code wt (kg) 264000 27,890 296001 1,118,890 288006 77,380 264004 21,470 354001 3,496 288006 4,009 465006 206,651 337002 6,097 999999 699,981 377003 2,863,055 258003 1,135,721 31000 160,826 1 735 264002 2,605 18000 406,775 353001 17,175

records 309 13,672 890 415 15 31 1,110 6 9,124 15,483 5,460 1,818 17 13 5,913 360

3.4.4 Data summary Total catch by year and vessel

Total catch and number of hauls by year and vessel for the 1951-57 data are given in Table 3.20 and Table 3.21. Discarded catch was not recorded. Table 3.20. Total retained catch and number of hauls by year. Year 1951 1952 1953 1954 1955 1956 1957

Total catch (kg) 2,128 1,259,720 1,427,085 1,091,219 1,117,610 1,000,446 845,776

Hauls 6 3,309 4,701 3,398 3,282 3,328 3,096

24

Table 3.21. Catch and number of hauls by vessel. Vessel Alfie Cam Goolgwai Korowa Maldanna Mary Cam Matong Moona Mulloka

Retained wt 39,192 410,235 691,860 1,716,452 44,883 1,634,927 1,464,698 741,737

Hauls 131 1,380 2,026 4,910 171 4,934 5,197 2,371

Catch by depth Total catches by depth interval (40=40-60) are given in Table 3.22. Depths have been converted from fathoms to metres. All recorded depths were within the ranges shown and no adjustments for mis-recording were made. Table 3.22. Total retained catch weight by depth interval. Depth (m) 40 60 80 100 120 140 160 180 200 220

Retained wt (kg) 4,671 280,297 700,832 1,025,131 2,426,664 1,618,564 98,845 11,278 6,416 572

Hauls 22 992 2,316 3,157 7,610 4,908 279 40 20 1

Catch by year and species Total retained catch by year per species is presented in Table 3.23. Table 3.23. Total retained catch by species by year. Code DO FL GU JD JW LA LJ MA MI MO NA RA SA SD SH SN

Species dory tiger flathead latchet john dory jewfish latchet chinaman leatherjacket mackerel mixed jackass morwong redfish rays unknown sa silver dory shark snapper

1951 0 1,398 0 0 0 0 0 0 160 478 0 0 0 0 96 0

1952 192 299,378 19,749 8,610 0 672 80,614 6,097 150,782 398,859 180,896 17,775 0 2,605 93,230 1,727

25

1953 1954 1955 1956 1957 1,306 16,404 9,988 0 0 239,326 193,149 150,900 155,555 79,184 10,439 4,650 10,240 11,917 20,385 5,295 2,268 2,043 1,464 1,790 0 0 0 2,732 764 3,337 0 0 0 0 43,271 13,199 22,826 31,387 15,354 0 0 0 0 0 149,603 97,434 90,045 110,132 101,825 497,891 481,549 562,068 489,696 432,514 348,210 207,601 179,204 114,507 105,303 47,601 35,834 32,651 19,270 7,695 735 0 0 0 0 0 0 0 0 0 80,013 39,814 54,966 61,085 77,571 1,912 669 4,121 4,111 4,635

3.5 Final data formats All data were converted into the following format. This is suitable for loading into the Australian Fishing Zone Information System (AFZIS) without the need for further data manipulation. The files are currently in dBase III+ databases, but they can easily be converted to text tables or delimited ASCII files as required. Structure for database: E:ops.dbf Field Field Name 1 BOAT_NAME 2 DATE 3 HAUL_NO 4 OP_NO 5 ST_TIME 6 EN_TIME 7 ST_DEP 8 EN_DEP 9 LAT 10 LONG 11 EFFORT 12 RET_WT 13 DIS_WT 14 TOT_WT ** Total **

Type Character Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric Numeric

Width 15 6 2 2 4 4 4 4 6 6 6 8 8 8 84

Dec

2 2 2

Structure for database: E:sp.dbf Field Field Name 1 BOAT_NAME 2 DATE 3 HAUL_NO 4 OP_NO 5 SPECIES 6 SP_CODE 7 WT ** Total **

Type Character Numeric Numeric Numeric Character Numeric Numeric

Width 15 6 2 2 2 6 8 42

Dec

Notes: All measures were converted to metric with depth in metres, and weights in kilograms. Positions are in decimal degrees. Effort is measured in decimal hours fished. HAUL_NO is a sequential number for a haul within a single fishing trip. Date is given as a 6-digit number with year being the first 2, month being the middle 2 and day being the last 2. Times are given as a 4-digit number in hours and minutes with the last 2 digits being minutes. OP_NO the sequential number of the fishing operation within a single day of fishing. A fishing operation is uniquely identified by the boat name plus the date plus the operation number. These fields also appear in the species catch table to allow catches to be associated with the correct fishing operation. All species codes have been standardised to 6 digit CSIRO codes. Only the 1918-23 data contains information about discards, so for other time periods DIS_WT is always given as 0.

26

CHAPTER 4 – Conversion of Red Funnel Trawlers historical data

4.1 Introduction

The Red Funnel Trawler company ceased fishing operations during the 1960s. In 1992, a search by the Australian Securities Commission revealed that Red Funnel Trawlers Pty Ltd was still registered - although not now operating in the fishing industry. Mr Richard Mansfield was listed as a director and he confirmed that old company records were stored in one of his warehouses, and that he would be happy to let them be examined.

Mr Mansfield presented three original skipper logbooks dating from 1952-61 at a meeting in Sydney soon after. Other records were still in storage and difficult to retrieve at the time as he did not wish to disturb a business that was renting the premises. He also mentioned that the warehouse was flooded out at one stage and a number of skipper logbooks and the like were destroyed. In May 1993, all available records were sent to BRS by Mr Mansfield. In June, Ms Anne Shepherd of D & S Datafix was contracted to commence data entry. Data entry and error checking was largely completed towards the end of 1993.

The original logbooks, ledgers and radio report sheets referred to in this chapter have been archived in the Mitchell Library, Macquarie St, Sydney, NSW.

This chapter describes the processing carried out to convert the raw computerised records to a form suitable for entry into AFZIS, and also summarises some important aspects of the data.

27

4.2 Skipper Logbooks 1952-61 4.2.1 Raw data description Three vessel logbooks were available. These were in cardboard covered books 37 cm by 36 cm and were all carbon copies of the original hand written records - the originals were presumably handed to the main office on completion of a fishing trip. Two of the books were from the vessel Moona and covered the periods 12.01.5407.10.57 and 09.10.57-29.01.61. The other was from the vessel Matong for the period 27.12.52-07.02.56. The data were originally entered into a computer as text files formatted in the same way as the pages of the logbooks. All data were entered except for skipper remarks. Each page contained the information for a trip and is described as follows: Table 4.1. Skipper logbook raw data description. Field Description header vessel vessel name sailed date of start of trip returned date of completion of trip no. days length of trip in days skipper skipper name trip no. sequential trip number for this vessel haul details date date of haul haul haul number for this trip time shot time of shot time hauled time hauled flathead wt (baskets) of flathead caught morwong wt (baskets) of morwong caught redfish wt (baskets) of redfish caught john dory wt (baskets) of john dory caught gurnard wt (baskets) of gurnard caught leatherjackets wt (baskets) of leatherjackets caught hake wt (baskets) of hake caught mixed wt (baskets) of mixed species caught jewfish* wt (baskets) of jewfish caught snapper* wt (baskets) of snapper caught dogfish* wt (baskets) of dogfish caught shark* wt (baskets) of shark caught grounds worked description of fishing ground depth fishing depth (fathoms)

Type char dd.mm.yy dd.mm.yy num char num dd.mm.yy num hh.mm hh.mm num num num num num num num num num num num num char num

* indicates fields added by skipper - these do not occur consistently throughout the records

28

4.2.2 Loading procedure The records were loaded into a database of similar structure to the raw data using a dBase III+ program. On loading into the database the following errors were reported and corrected with reference to the original records: •

date sailed invalid

•

date returned invalid

•

calculated days of trip not equal to number recorded

•

calculated trip days more than 20 or less than 0

•

shot start time invalid

•

shot end time invalid

•

hours trawled invalid

•

shot end time mismatch with start time

•

shot length greater than 10 hours

•

number of hauls in one day greater than 7

•

operation date not within trip dates

•

no vessel name

•

haul number not in sequence

•

no fishing ground recorded

•

minimum depth missing or invalid

•

maximum depth missing or invalid

Numeric fields were determined to be invalid if they contained nonsense values (e.g. 24 hour times less than 0 or greater than 2400), or non-numeric characters. The depth field contained either a single number, or a range such as ‘50-60’. As a range was often recorded, both a lower and upper depth was included in the database, with both assigned the same value if a single depth was originally recorded.

Fishing position was always recorded in terms of a fishing ground and possibly a distance in nautical miles or a compass bearing - e.g. ‘Everard 20’, ‘Babel Is WxS’, ‘Gabo Wide’. The position of an individual haul was also often recorded as a ditto relating to previous record, or just as an additional distance or compass bearing. A reference database was created containing every unique character string occurring in the position field - a total of 484 records. Each of these was then manually determined 29

to be either a reference to a fishing ground, or additional information to a previous record. On loading, the records were matched and converted so that a correct standardised fishing ground was assigned to each record. It was necessary to ignore the additional information at this stage, as conversion of it into useful information for analysis would require a full knowledge of the meaning of the skipper’s shorthand used at the time.

The positions of most of the fishing grounds were located using a computerised gazetteer. All ground positions were plotted on a map showing the coastline and 200 m and 1000 m depth contours. A table was then constructed containing the position of the ground at the coast or island (depth 0), and the positions where a line from that point crosses the 200 m and 1000 m contours most directly towards the continental shelf (Table 4.2). For each haul, it was then possible to assign a position in latitude and longitude according to the fishing ground and a linear interpolation of the average depth of the shot.

30

Table 4.2. Fishing grounds and designated positions (degrees and minutes). Ground Babel Is Barrenjoey Pt Bermagui Bird Is Botany Bay Broughton Is Brush Is Cape Everard Cape Hawke Cape Howe Charlott Hd Coalcliff Crowdy Hd Disaster Bay Eden Gabo Is Green Cape Jervis Bay Kiama Merimbula Montague Is Moruya Mowarry Pt Newcastle Norah Hd O'Hara Hd Port Kembla Port Stephens Red Hd Shoalhaven Hd Sydney Hd Tathra The Pines Tollgate Is Wattamolla Wreck Bay

Hauls 458 12 6 4 26 8 4 1,038 65 6 1 76 80 160 522 2,130 352 61 87 203 411 73 35 20 64 75 86 17 58 12 126 801 162 172 50 3

Total wt (kg) Lat 0 m Lon 0 m Lat 200 m Lon 200 m 129,247 39 57 148 20 39 57 148 52 3,653 33 35 151 20 33 35 151 54 2,731 36 26 150 04 36 26 150 16 731 33 14 151 36 33 14 152 14 9,339 34 00 151 14 34 00 151 33 1,558 32 37 152 19 32 37 152 44 1,271 35 32 150 25 35 32 150 43 303,159 37 48 149 16 38 07 149 22 22,423 32 13 152 34 32 13 152 56 445 37 30 149 59 37 30 150 14 191 32 20 152 33 32 20 152 54 14,364 34 15 150 59 34 15 151 24 20,454 31 51 152 45 31 51 153 03 46,592 37 16 149 58 37 16 150 17 154,320 37 04 149 55 37 04 150 17 637,907 37 34 149 55 37 34 150 13 125,588 37 16 150 03 37 16 150 17 19,442 35 07 150 46 35 07 150 58 18,525 34 40 150 51 34 40 151 10 62,732 36 54 149 56 36 54 150 16 169,025 36 15 150 14 36 15 150 18 26,044 35 55 150 08 35 55 150 28 12,770 37 09 150 00 37 09 150 17 4,670 32 56 151 46 32 56 152 33 15,821 33 17 151 35 33 17 152 11 33,567 35 34 150 23 35 34 150 41 19,542 34 29 150 55 34 29 151 16 4,510 32 42 152 10 32 42 152 41 11,409 35 15 150 33 35 15 150 54 2,542 34 51 150 45 34 51 151 06 40,628 33 51 151 18 33 51 151 42 252,700 36 44 149 59 36 44 150 16 43,838 36 01 150 09 36 01 150 24 42,824 35 45 150 16 35 45 150 34 8,137 34 08 151 08 34 08 151 28 762 35 11 150 38 35 11 150 55

Lat 1000 m 39 57 33 35 36 26 33 14 34 00 32 37 35 32 38 20 32 13 37 30 32 20 34 15 31 51 37 16 37 04 37 34 37 16 35 07 34 40 36 54 36 15 35 55 37 09 32 56 33 17 35 34 34 29 32 42 35 15 34 51 33 51 36 44 36 01 35 45 34 08 35 11

In the ‘grounds worked’ column, the skipper often (but not always) noted if the gear was fouled or damaged during the shot, or if the weather was bad. Logical fields called ‘valid’ and ‘weather’ were also created to reflect these comments - where valid = False for fouled gear, and weather = False for bad weather.

The spelling of vessel and skipper names was also standardised using additional tables of the alternative spellings present in the raw data. Position assignment, spelling standardisation, and correction and standardisation of species names were also carried out by the loading program.

Standard databases (trip, fishing operation and species caught) of a structure suitable for entry onto AFZIS were then created from the loaded database. Catch weights were

31

Lon 1000 m 148 58 152 08 150 20 152 24 151 51 152 52 150 47 149 26 153 05 150 18 153 01 151 35 153 19 150 21 150 21 150 18 150 21 151 06 151 20 150 20 150 23 150 31 150 21 152 39 152 19 150 45 151 26 152 50 151 00 151 13 152 05 150 20 150 28 150 38 151 43 151 03

converted to kg from baskets (1 basket = 70 lb = 70 * 0.45359237 kg), and depths in fathoms were converted to metres (1 fm = 1.8288 m).

4.2.3 Data quality As these records were those recorded by Red Funnel Trawlers Pty Ltd for their own purposes, it is likely that the data are of high quality from this point of view. Unfortunately, the company was mainly interested in recording catches of their main saleable target species, which means that there were no records of discards or species of minor commercial importance such as ocean perch, barracouta or skate. However, the catch records for the main target species that were pre-printed on the logbooks (flathead, morwong, redfish, john dory, gurnard and hake) should be accurate. Even for those species, an unknown level of high-grading may have occurred. It is likely for species such as shark and dogfish that the level of retained catch would be driven by catches of other more valuable species or market conditions at the time.

The level of coverage of information in important fields in the data is shown in Table 4.3. Note that the vessel name, skipper name, date of haul, depth fished, retained catch weight and fishing ground information are available for all records, and that the information has been checked and corrected against the originals where necessary.

For about 23% of hauls, a range of depths was recorded. Latitude and longitude for each haul was derived using the fishing ground and depth information as previously described. For grounds on the east coast, the position should be accurate to +/- 30 nm (0.5 degree) at worst. For Cape Everard and Babel Is, it is likely that the accuracy is about +/- 120 nm (2 degrees).

Where the skipper commented on fouled or damaged gear, an invalid shot has been recorded. For hauls where the shot duration is less than 1 hour (likely to be an invalid trawl), 12 records are recorded as valid, and 57 invalid. This suggests that a large proportion of the invalid trawls have been flagged.

About 5% of records have been flagged as being carried out in bad weather. The accuracy of this information is unknown.

32

Table 4.3. Data coverage for important fields. Field Records % of total vessel name 7,464 100.00 skipper name 7,464 100.00 date 7,464 100.00 depth fished 7,464 100.00 depth range 1,763 23.62 species catch wt 7,464 100.00 fishing ground 7,464 100.00 lat/long * 7,464 100.00 invalid shot 593 7.94 bad weather 401 5.37 * calculated from other information as described

Catch positions A map of the estimated positions of all hauls is given in Figure 4.1. Red Funnel Trawlers were based in Sydney, so all trips were made from there. It is notable that the grounds identified (particularly off the New South Wales coast) are fairly close together suggesting that the position information is reasonably accurate. The map also shows the 200 and 1000 m isobaths as obtained from the Commonwealth Hydrographers office. The lines are only suitable for large-scale maps such as this, and as such, are not precise.

33

Figure 4.1. Estimated catch positions (200 and 1000 m isobaths also shown).

Species identification The only means for identifying species were the names printed in the logbooks (see Table 4.1). In most cases there can be no doubt about the identity of the species concerned. It was assumed that ‘morwong’ refers to jackass morwong and not rubberlip morwong. ‘Hake’ was assumed to refer to gemfish and ‘leatherjacket’ was assumed to mean chinaman leatherjacket. The flathead catch was certainly composed of mostly tiger flathead as it has in recent years, but probably includes some catches of other flathead species such as the sand flathead.

34

Table 4.4. Assumed species identification and CSIRO species codes. Name CSIRO code Wt (kg) Records jackass morwong 377003 1,056,021 5,499 tiger flathead 296001 364,601 4,525 redfish 258003 350,987 1,714 mixed 999999 223,144 3,238 shark 18000 138,780 1,797 gemfish 439002 60,328 670 chinaman leatherjacket 465006 41,181 182 gurnard 288004 19,205 119 john dory 264004 6,632 114 jewfish 354001 2,732 6 snapper 353001 1,662 41 dogfish 20000 1,020 11 4.2.4 Data summary Catch and effort summary by year and vessel Catch and effort summaries from the vessel logs of the Matong and Moona are given in Table 4.5 and Table 4.6, and totals in Table 4.7. Table 4.5. Catch and effort summary – Matong. Year

Trips

1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 Total

1 32 26 27 1 0 0 0 0 0 87

Days fished 12 324 310 310 12 0 0 0 0 0 968

Hauls 15 884 777 925 69 0 0 0 0 0 2,670

Hours trawled 56.75 3,455.84 3,184.51 3,397.62 250.67 0.00 0.00 0.00 0.00 0.00 10,345.39

Catch (kg) 6,098 283,412 256,101 328,483 18,041 0 0 0 0 0 892,135

Days Hauls Hauls Hours per Catch per per trip per trip per day haul hour 12.00 15.00 1.25 3.78 107.45 10.13 27.63 2.73 3.91 82.01 11.92 29.88 2.51 4.10 80.42 11.48 34.26 2.98 3.67 96.68 12.00 69.00 5.75 3.63 71.97

11.13

30.69

2.76

3.87

86.24

Table 4.6. Catch and effort summary – Moona. Year

Trips

1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 Total

0 0 27 24 27 24 29 0 24 1 156

Days fished 0 0 314 278 296 271 284 0 262 11 1,716

Hauls 0 0 871 844 875 741 799 0 607 57 4,794

Hours trawled 0.00 0.00 3,375.77 3,358.94 3,499.50 3,039.25 3,314.59 0.00 2,525.08 245.50 19,358.63

Catch (kg) 0 0 268,794 254,349 215,691 184,076 211,747 0 224,244 12,425 1,371,326

35

Days Hauls Hauls Hours per Catch per per trip per trip per day haul hour

11.63 11.58 10.96 11.29 9.79

32.26 35.17 32.41 30.88 27.55

2.77 3.04 2.96 2.73 2.81

3.88 3.98 4.00 4.10 4.15

79.62 75.72 61.63 60.57 63.88

10.92 11.00 11.00

25.29 57.00 30.73

2.32 5.18 2.79

4.16 4.31 4.04

88.81 50.61 70.84

Table 4.7. Catch and effort summary – Total. Year 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 Total

Trips

Days fished

Hauls

Hours trawled

Catch (kg)

1 32 53 51 28 24 29 0 24 1 243

12 324 624 588 308 271 284 0 262 11 2,684

15 884 1,648 1,769 944 741 799 0 607 57 7,464

56.75 3,455.84 6,560.28 6,756.56 3,750.17 3,039.25 3,314.59 0.00 2,525.08 245.50 29,704.02

6,098 283,412 524,895 582,832 233,732 184,076 211,747 0 224,244 12,425 2,263,461

12.00 10.13 11.77 11.53 11.00 11.29 9.79

15.00 27.63 31.09 34.69 33.71 30.88 27.55

1.25 2.73 2.64 3.01 3.06 2.73 2.81

Hours per haul 3.78 3.91 3.98 3.82 3.97 4.10 4.15

10.92 11.00 11.05

25.29 57.00 30.72

2.32 5.18 2.78

4.16 4.31 3.98

Days Hauls Hauls per trip per trip per day

Catch per hour 107.45 82.01 80.01 86.26 62.33 60.57 63.88 88.81 50.61 76.20

The number of trips undertaken by vessel and skipper are shown in Table 4.8. Table 4.8. Skipper summary by vessel and number of trips. Vessel Skipper Trips Matong J. Low 48 J. Barber 16 E. Featherstone 10 G. Stewart 7 W. Gimpton 3 L. Sansom 2 W. Wilson 1 Moona C. Olsson 151 J. Jurrell 2 L. Mullarky 2 W. Wilson 1

Catch by depth Total catches by depth interval (20=20-40 m) are given in Table 4.9. Depths have been converted from fathoms to metres. All recorded depths were within the range 20240 m and no adjustments for mis-recording were made. 72% of all hauls were made between 120 and 160 m, and 43% were made between 120 and 140 m.

36

Table 4.9. Total retained catch weight by depth interval. Depth Retained wt (kg) Hauls 20 5,749 17 40 11,245 27 60 95,233 335 80 183,119 655 100 269,619 918 120 977,109 3,216 140 653,276 2,060 160 64,137 210 180 987 11 200 2,987 13 220 0 2

Catch by year and species

Table 4.10 shows the total retained catch by species and year. Morwong was consistently the major species caught in all years, and comprised 47% of the catch overall. Flathead and redfish were caught in similar weights in most years and when combined accounted for 32% of the overall catch. Notable also is the 60 t of gemfish caught during the period. Species reported as ‘mixed’ accounted for almost 10% of the overall catch. Present day trawlers catch in excess of 70 commercial species, so a reasonably high proportion of mixed species such as this is not surprising. Table 4.10. Total retained catch (kg) by species and year. Species

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

Total

jackass morwong tiger flathead redfish mixed shark gemfish ch leatherjacket gurnard john dory jewfish snapper dogfish

4,732 415 0 447 508 0 0 0 0 0 0 0

106,944 47,964 50,743 43,242 19,571 8,266 5,882 0 990 0 192 0

239,705 99,495 96,519 42,407 16,254 26,571 3,658 0 894 0 0 0

287,087 87,043 101,943 48,609 23,255 15,676 11,976 3,722 2,359 0 896 1,020

110,582 58,135 21,009 29,574 2,550 893 6,829 191 1,238 2,732 254 0

94,133 23,757 30,131 22,182 8,912 4,649 0 0 511 0 0 0

74,060 16,282 38,267 14,823 40,596 1,339 11,216 14,689 480 0 320 0

0 0 0 0 0 0 0 0 0 0 0 0

130,833 28,640 12,375 20,999 26,366 2,934 1,620 603 160 0 0 0

7,945 2,870 0 861 768 0 0 0 0 0 0 0

1,056,021 364,601 350,987 223,144 138,780 60,328 41,181 19,205 6,632 2,732 1,662 1,020

Total

6,102

283,794

525,503

583,586

233,987

184,275

212,072

0

224,530 12,444

2,266,293

37

4.3 Landing records 1938-59 4.3.1 Raw data description Two sales analysis ledgers from the Red Funnel Trawler Company were available. These both had covers made of canvas covered steel, with pages measuring 48 cm by 30 cm. All records were hand written using ink. The first ledger was 14 cm thick excluding 2 cm for the covers and contained approximately 1040 sheets that recorded daily unloadings by vessel and species for the period 01.12.38-30.06.57. The second was 2 cm thick and contained 19 sheets of daily unloadings for the period 01.07.5731.12.58, 12 sheets of monthly sales summaries by species and price received for the period 1947-1958, 4 sheets of financial year sales summaries by species and price received for the period 1939-1958, and 45 sheets of daily sales summaries for the period 01.07.51-31.12.58.

The daily unloading data were originally entered into a computer as Excel spreadsheets formatted to resemble pages of the ledgers. All data were entered except for summary information and prices received. Each page contained the information for a number of unloadings and is described in Table 4.11.

38

Table 4.11. Landings raw data description. Field Description vessel date coutta dogs flake flake special flathead flathead BB flathead BO flathead MIXED flathead MED flathead SML flathead X gurnard hake leatherjackets jewfish john dory latchet lobster mixed morwong morwong SM redfish rubber lip sand sawfish snapper shark shark HG shark LIV skate box total

vessel landing date of unloading wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (pounds) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) wt (baskets) - head and gutted wt (baskets) - livers wt (baskets) total wt (baskets)

Type char dd.mm.yy num num num num num num num num num num num num num num num num num num num num num num num num num num num num num num num

Due to memory and processing limitations of the computer used for data entry, 31 separate Excel files were created.

Through time, the main species and grades recorded on a page varied, presumably as the proportions in the total catch changed, or when the company wished to record different grades for analysis. Consequently, not all species and grades appearing in Table 4.11 were used for all records. Main species such as flathead, morwong and redfish were always recorded separately, but the grades may have varied through time. It has been assumed that when a grade was not recorded for a species, the record referred to whole fish. All recorded weights were in baskets or boxes except for “lobster” which was recorded in pounds.

39

Flathead was usually recorded in a number of grades including X (extra large?), BB (broken bellies), BO (?), MED (medium) and SML (small). The true meanings for these grade codes are yet to be determined.

Sharks were often headed and gutted (HG), filleted (flake or special flake), or just livers retained (LIV). As far as can be determined at present, these are the only gradings that do not appear to refer to whole fish.

4.3.2 Loading procedure Each excel file was exported as a dBase III+ database, and the records were combined into a single database using the dBase III+ program. On loading into the database the following errors were reported and corrected with reference to the original records: •

no vessel name

•

date invalid

•

year not between 1930 and 1965

•

invalid species weights

•

calculated total weight for unloading not equal to entered total

Numeric fields were determined to be invalid if they contained non-numeric characters. The check of calculated and recorded total weights was particularly useful for eliminating data entry errors, and also a number of miscalculations in the original records.

A second stage of processing converted the single database into two database files for entry onto AFZIS. A Clipper/dBase III+ program was used to create files of the structure described later. During this second stage, baskets/boxes were converted to kg (1 basket = 70*0.45359237 kg), the spelling of vessel names was corrected, species names were standardised and assigned CSIRO code numbers, and grade codes were converted to a separate field.

Single vessels were often unloaded on two or three consecutive days. Second stage processing also combines unloadings for a vessel which occurred up to three days

40

apart into a single landing, and stores the start and end date. Due to this aggregation, a landing record should reflect catches from a single fishing trip.

4.3.3 Data quality As for the skipper logbooks, these landings records are those recorded by Red Funnel Trawlers Pty Ltd for their own purposes. It is likely that the data are of high quality from this point of view. Hand writing in the original ledgers was neat and legible, so problems with readability were minimal for computer data entry (compared to the Red Funnel skipper logbooks).

The only important fields for an unloading were the vessel name, and the unloading date. Valid values for both of these fields were obtained for all 5,572 unloadings (all errors detected by the loading procedure were corrected with reference to the original records).

Aggregation of unloading records using the assumption that an unloading from a vessel within two days of the last was from the same landing produced 2,610 landings. Note that a single landing may possibly span more than three days using this assumption when a series of unloadings occurred at intervals of two days or less. An examination of the number of landings and unloadings by the duration of landing is given in Table 4.12. Most landings were made over one or two days and the maximum duration for a landing was five days according to this method. Table 4.12. Number of landings and unloadings by duration in days. Duration (days) 1 2 3 4 5

Landings 926 1,055 414 165 50

Unloadings 1,148 2,212 1,277 680 253

Detailed landing records were only kept for fish which were sold on unloading, so there is little information on fish which were brought to port but subsequently dumped due to damage or spoilage (except for some notes in the ledgers which were not computerised). Some information on this aspect can be gathered by comparison of catch per trip from the skipper logbooks and corresponding landings records for the same vessels and trips, as shown in Table 4.13 and Figure 4.2. For this comparison, the 41

landing corresponding to a trip was assumed to be a landing by a vessel that commenced within three days of the end of a trip as recorded in the skipper logbooks. A corresponding landing was found for all trips recorded in the logbooks except those that occurred after 1959 (the last year for which landings records were available). A total of 219 trips were matched with landings.

Total landings exceeded logbook catch records by 23%, suggesting either underrecording on the skipper logbooks, or over-recording on landing records. Differences varied considerably among vessels - 6% for the Matong, and 36% for the Moona. In particular, the catch records of one skipper on the Moona were considerably less than shown on landing records (30-47% annually). Comparison of catch and landings records by species show that there is close comparison for flathead, morwong and redfish (the main target species), and that most difference is shown in records for other species (see Figure 4.2). This strongly suggests that the skippers kept good logbook records for the main target species, but other species were considerably under-reported. It follows that the landings information contains a more comprehensive record of total catch than the skipper logbooks.

The extremely close correlation between skipper logbook catch records by trip and the landings information for the main target species reinforces the high quality and accuracy of these data, and also suggests that spoilage and dumping of these species prior to sale was minimal. It also provides good evidence that aggregation of the logbook data by trip and the unloadings into individual landings was carried out in a consistent manner.

42

Table 4.13. Comparison of catch per trip from skipper logbooks and corresponding landings by vessel, skipper and year. Vessel

Skipper

Matong E.Featherstone G.Stewart J.Barber J.Barber J.Low J.Low J.Low L.Sansom L.Sansom W.Gimpton W.Wilson Subtotal Moona

C.Olsson C.Olsson C.Olsson C.Olsson C.Olsson J.Jurrell W.Wilson Subtotal

Year Landings Flathead Log 10 15,304 53 55 7 7,880 55 14 15,897 56 2 3,789 53 22 31,833 54 22 35,532 55 4 16,152 54 1 1,085 55 1 415 54 3 5,924 55 1 512 134,323 54 55 56 57 58 58 55

26 23 28 23 28 2 1

Total

Vessel

Skipper

Matong E.Featherstone G.Stewart J.Barber J.Barber J.Low J.Low J.Low L.Sansom L.Sansom W.Gimpton W.Wilson Subtotal Moona

Total

C.Olsson C.Olsson C.Olsson C.Olsson C.Olsson J.Jurrell W.Wilson Subtotal

Land Diff% 15,526 1 5,771 -27 14,455 -9 3,485 -8 31,108 -2 34,133 -4 15,257 -6 1,008 -7 318 -23 5,668 -4 611 19 127,339 -5

Morwong Log 63,153 60,454 67,773 8,773 46,962 109,698 12,783 4,480 5,180 13,161 10,641 403,058

26 23 28 23 28 2 1

Redfish Log 12,550 19,546 22,287 64 38,193 23,833 2,448 3,971 287 8,548 9,373 141,100

Land Diff% 13,844 10 20,019 2 22,051 -1 159 148 40,817 7 21,726 -9 1,429 -42 5,271 33 397 38 8,311 -3 8,549 -9 142,572 1

58,132 43,382 55,907 23,311 15,385 1,343 1,308 198,768

62,820 46,833 61,161 23,599 18,210 1,357 913 214,894

8 8 9 1 18 1 -30 8

113,673 115,797 109,435 92,164 72,686 3,375 7,087 514,217

127,911 129,895 131,332 106,653 89,928 3,858 7,493 597,070

13 12 20 16 24 14 6 16

60,167 46,157 21,264 30,131 34,227 4,262 1,526 197,734

62,439 41,483 26,925 36,070 41,134 5,207 1,048 214,307

4 -10 27 20 20 22 -31 8

333,091

342,233

3

917,275

1,011,427

10

338,834

356,879

5

Year Landing Other species All species s Log Land Diff% Log 53 10 19,121 23,329 22 110,128 55 7 21,815 26,338 21 109,695 55 14 40,354 48,048 19 146,311 56 2 5,869 7,223 23 18,495 53 22 58,257 66,384 14 175,245 54 22 46,974 63,479 35 216,037 55 4 7,579 10,796 42 38,962 54 1 1,816 3,247 79 11,352 55 1 4,168 4,445 7 10,050 54 3 5,583 7,732 38 33,216 55 1 3,417 5,168 51 23,943 214,953 266,189 24 893,434 54 55 56 57 58 58 55

Land Diff% 72,330 15 59,082 -2 66,456 -2 9,002 3 52,096 11 111,027 1 11,415 -11 2,858 -36 5,430 5 13,915 6 10,748 1 414,357 3

Land Diff% 125,029 14 111,210 1 151,010 3 19,868 7 190,406 9 230,365 7 38,896 0 12,383 9 10,589 5 35,625 7 25,076 5 950,457 6

37,067 27,783 39,414 35,871 81,672 2,396 1,439 225,642

95,667 90,587 112,770 91,968 115,956 3,556 3,167 513,672

158 226 186 156 42 48 120 128

269,039 233,119 226,020 181,477 203,970 11,376 11,360 1,136,361

348,837 308,799 332,188 258,290 265,228 13,979 12,621 1,539,943

30 32 47 42 30 23 11 36

440,595

779,860

77

2,029,795

2,490,399

23

43

Figure 4.2. Comparison of catches per individual trip from skipper logbooks and

corresponding landings by species. (a) Flathead

(b) Morwong

12,000

18,000

8,000

12,000

4,000

6,000

0

0

0

4,000

8,000

12,000

0

Log book ca tch (kg)

6,000

12,000

18,000


(c) Redfish

(d) Other species

20,000

12,000

15,000

9,000

10,000

6,000

5,000

3,000

0

0 0

5,000 10,000 15,000 20,000 Log book ca tch (kg)

30,000

20,000

10,000

0 10,000

20,000

3,000 6,000 9,000 12,000 Log book ca tch (kg)

(e) All species

0

0

30,000


44

4.3.4 Data summary

Summary statistics for number of landings by vessel and year, total weight landed by vessel and year and weight landed by species and year are given in Table 4.14, Table 4.15 and Table 4.16. Number of landings by vessel and year Table 4.14. Landings by vessel and year (kg). Name Moona Bareamul Maldanna Matong Korowa Goolgwai Durraween Mulloka Millimumul Nanagai Gippsland Winbirra Purchases Misc Total

1938 0 5 0 0 4 3 4 0 2 3 0 0 0 0 21

1939 0 45 0 0 31 35 46 0 45 35 0 0 0 0 237

1940 0 51 0 0 0 0 29 0 50 42 0 0 0 2 174

1941 0 51 0 0 0 0 0 0 13 42 32 14 0 0 152

1942 0 50 0 0 0 0 0 0 0 36 0 1 0 0 87

1943 0 46 0 0 0 0 0 0 0 0 0 0 1 0 47

1944 0 39 0 0 0 0 0 0 0 0 0 0 0 1 40

1945 0 41 0 0 0 0 0 0 0 0 0 0 0 0 41

Name Moona Bareamul Maldanna Matong Korowa Goolgwai Durraween Mulloka Millimumul Nanagai Gippsland Winbirra Purchases Misc Total

1946 0 38 6 0 34 0 6 0 0 0 0 0 0 0 84

1947 23 32 31 14 29 12 29 0 0 0 0 0 0 0 170

1948 36 3 33 34 31 33 34 0 0 0 0 0 0 0 204

1949 30 0 27 34 27 27 22 0 0 0 0 0 0 0 167

1950 34 0 30 34 29 30 32 0 0 0 0 0 0 0 189

1951 27 0 30 32 33 30 7 0 0 0 0 0 0 0 159

1952 35 0 32 31 32 4 0 0 0 0 0 0 0 0 134

1953 28 0 31 32 29 33 0 0 0 0 0 0 0 0 153


1954 28 0 29 28 27 26 0 0 0 0 0 0 0 0 138

1955 25 0 24 28 5 11 0 21 0 0 0 0 0 0 114

1956 28 0 26 24 0 0 0 27 0 0 0 0 0 0 105

1957 23 0 27 27 0 0 0 24 0 0 0 0 0 0 101

1958 32 0 20 18 0 0 0 21 0 0 0 0 0 0 91

1959 0 0 0 1 0 1 0 0 0 0 0 0 0 0 2

45

Total 349 401 346 337 311 245 209 93 110 158 32 15 1 3 2,610

Weight landed by vessel and year Table 4.15. Landings by vessel and year (kg). Name Moona Bareamul Maldanna Matong Korowa Goolgwai Durraween Mulloka Millimumul Nanagai Gippsland Winbirra Purchases Misc Total

1938 0 30,307 0 0 36,721 26,782 31,791 0 30,156 11,264 0 0 0 0 167,021

1939 0 406,705 0 0 289,272 316,903 463,040 0 445,267 102,002 0 0 0 0 2,023,189

1940 0 448,910 0 0 0 0 256,250 0 467,374 92,000 0 0 0 0 1,264,534


1946 0 468,064 93,802 0 430,074 0 91,405 0 0 0 0 0 0 0 1,083,345

1947 351,489 272,134 320,872 150,677 332,255 135,198 373,151 0 0 0 0 0 0 0 1,935,776


1954 358,482 0 284,136 296,233 307,291 272,316 0 0 0 0 0 0 0 0 1,518,458

1955 327,747 0 289,216 346,043 38,935 113,607 0 291,947 0 0 0 0 0 0 1,407,495

1941 0 462,817 0 0 0 0 0 0 121,926 85,824 57,692 5,469 0 0 733,728

1942 0 464,214 0 0 0 0 0 0 0 60,725 0 437 0 0 525,376

1943 0 480,955 0 0 0 0 0 0 0 0 0 0 318 0 481,273

1944 0 480,011 0 0 0 0 0 0 0 0 0 0 0 -127 479,884

1945 0 420,437 0 0 0 0 0 0 0 0 0 0 0 0 420,437

1948 469,874 17,051 467,128 466,985 425,501 333,763 359,228 0 0 0 0 0 0 0 2,539,530

1949 1950 351,108 316,943 0 0 307,354 297,122 374,937 318,237 309,053 326,850 303,068 349,290 182,912 307,275 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1,828,432 1,915,717

1951 268,594 0 323,841 314,157 335,168 263,640 56,795 0 0 0 0 0 0 0

1952 412,499 0 391,702 312,339 367,666 32,466 0 0 0 0 0 0 0 0

1,562,195

1,516,672

1953 339,249 0 352,672 315,435 308,767 324,770 0 0 0 0 0 0 0 0 1,640,893

1956 332,184 0 315,268 269,522 0 0 0 323,127 0 0 0 0 0 0 1,240,101

1957 258,290 0 249,209 284,255 0 0 0 271,888 0 0 0 0 0 0 1,063,642

1958 279,778 0 174,530 157,789 0 0 0 194,255 0 0 0 0 0 0 806,352

1959 0 0 0 1,365 0 1,810 0 0 0 0 0 0 0 0 3,175

Total 4,066,237 3,951,605 3,866,852 3,607,974 3,507,553 2,473,613 2,121,847 1,081,217 1,064,723 351,815 57,692 5,906 318 -127 26,157,225

The meaning of “Purchases” and “Misc” in these records is unclear, but may refer to fish obtained that were not caught by Company vessels. Negative values in the “Misc” column were in the records, but were not explained.

46

Weight landed by species and year Table 4.16. Landings by species and year (kg). Code

Name

377003 296001 258003 18000 465006 288004 999999 439002 264004 888888 288006 20000 353001 31000 439001 296003 377002 354001 777777

ja morwong tiger flathead redfish shark ch leatherjacket gurnard mixed gemfish john dory sawfish latchet dogfish snapper rays barracouta sand flathead rl morwong jewfish lobster Total

Code

Name

377003 296001 258003 18000 465006 288004 999999 439002 264004 888888 288006 20000 353001 31000 439001 296003 377002 354001 777777


Code

Name

377003 296001 258003 18000 465006 288004 999999 439002 264004 888888 288006 20000 353001 31000 439001 296003 377002 354001 777777


1938

1939

1940

1941

1942

1943

1944

1945

11,304 95,548 10,915 230 17,281 6,938 1,064 0 9,430 0 1,357 4,652 1,826 619 1,318 4,183 357 0 0 167,022

181,301 1,149,594 84,292 2,961 231,746 110,043 18,606 0 66,448 0 31,466 60,590 10,042 16,400 6,438 49,659 3,604 0 0 2,023,190

44,516 589,069 140,818 15,733 206,480 76,561 16,003 0 51,461 0 18,924 54,311 3,421 15,542 6,382 23,353 1,961 0 0 1,264,535

50,334 307,188 80,855 18,257 142,699 42,364 8,192 0 22,107 0 5,064 32,140 1,492 11,732 4,025 6,612 667 0 0 733,728

15,669 155,217 81,808 117,338 84,554 20,305 8,970 0 15,812 0 5,445 0 1,230 14,376 1,373 2,532 746 0 0 525,375

1,945 219,514 21,670 123,283 43,412 18,281 3,469 0 12,709 0 6,660 0 1,135 25,719 3,104 0 373 0 0 481,274

32,125 179,674 25,219 101,390 57,462 31,164 6,787 0 5,310 0 7,430 0 2,334 17,646 13,256 0 87 0 0 479,884

56,454 210,933 28,537 44,468 33,974 21,655 6,358 0 2,627 0 10,780 0 1,469 492 2,556 0 135 0 0 420,438

1946

1947

1948

1949

1950

1951

1952

1953

359,133 304,838 194,359 64,781 53,454 41,198 40,745 0 3,397 0 12,550 0 3,255 32 5,533 0 71 0 0 1,083,346

746,461 377,660 446,449 146,366 20,123 120,378 61,225 0 5,025 0 2,461 0 2,524 802 6,152 0 151 0 0 1,935,777

684,911 410,054 1,040,003 122,815 70,988 65,400 118,687 0 4,874 0 770 0 8,708 1,215 10,518 0 587 0 0 2,539,530

474,938 151,367 821,402 113,519 77,069 48,826 63,741 27,108 8,986 4,826 56 0 18,940 14,177 2,993 0 484 0 0 1,828,432

383,129 161,012 880,063 90,468 187,524 73,552 41,086 28,553 16,892 32,910 1,413 0 17,209 0 1,905 0 0 0 0 1,915,716

378,549 267,077 188,350 156,424 296,059 66,583 80,260 68,313 11,145 32,323 40 0 13,296 0 3,778 0 0 0 1,585 1,563,782

437,575 336,828 201,257 109,932 114,567 37,086 121,513 55,517 22,925 32,720 27,489 0 12,740 0 6,525 0 0 0 315 1,516,989

551,833 239,533 333,065 140,897 80,077 27,140 95,302 97,660 7,422 39,054 17,114 0 7,120 0 4,675 0 0 0 20 1,640,912

1954

1955

1956

1957

1958

1959

Total

581,798 227,491 285,827 122,505 27,155 34,839 91,992 90,897 6,692 34,093 6,795 0 2,635 0 5,739 0 0 0 0 1,518,458

646,627 177,650 195,653 136,992 35,451 28,426 60,907 73,878 4,620 30,624 7,462 0 8,009 0 1,199 0 0 0 0 1,407,498

550,023 162,441 132,165 136,642 42,865 25,782 43,214 88,229 4,914 34,577 5,183 0 8,390 0 2,413 0 0 3,262 24 1,240,124

482,360 89,555 122,632 151,939 32,466 55,668 31,013 35,982 7,406 27,362 14,161 0 11,494 0 1,603 0 0 0 0 1,063,641

311,807 54,938 144,533 142,596 20,202 40,563 23,512 24,115 3,532 22,155 10,645 0 6,993 0 762 0 0 0 0 806,353

1,588 635 0 222 381 349 0 0 0 0 0 0 0 0 0 0 0 0 0 3,175

6,984,380 5,867,816 5,459,872 2,059,758 1,875,989 993,101 942,646 590,252 293,734 290,644 193,265 151,693 144,262 118,752 92,247 86,339 9,223 3,262 1,944 26,159,179

47

Prices received 1939-52 Monthly and annual summaries of weight and price received by species were also recorded in the back of the smaller landings ledger. Weight was converted from baskets to kg (1 basket = 70 lb), and prices received were converted from pounds, shillings and pence to dollars and cents (1 pound = 2 dollars). A summary is given in Table 4.17. Table 4.17. Average price (cents/kg per financial year). Species Flathead Morwong J Dory Latchet Gurnard Leatherjackets Redfish Snapper Skates/rays Rubberlip Barracouta Mixed Flake/sharks Total

1939 9.01 5.67 9.25 4.59 1.78 2.10 2.46 13.86 1.66 4.99 1.45 3.43 3.07 6.92

1940 10.53 8.74 9.90 4.55 2.18 2.34 3.09 18.38 1.80 8.12 1.66 3.54 3.32 7.87

1941 13.58 9.77 11.01 5.33 3.02 2.90 4.13 20.02 2.10 8.28 3.09 4.46 4.26 9.04

1942 16.21 12.45 11.86 6.06 4.17 3.94 5.90 30.13 2.41 10.41 4.35 4.99 5.04 10.18

1943 21.50 14.91 20.82 7.85 6.21 7.27 8.98 34.66 3.19 4.12 3.32 6.61 6.24 15.28

1944 27.01 18.31 24.99 8.08 8.07 10.09 11.30 31.02 3.51 5.16 5.89 8.29 6.87 20.71

1945 22.64 18.42 19.73 8.30 8.31 10.24 11.56 38.27 6.70 24.33 7.88 9.29 8.07 17.19

Species Flathead Morwong J Dory Latchet Gurnard Leatherjackets Redfish Snapper Skates/rays Rubberlip Barracouta Mixed Flake/sharks Total

1946 20.81 18.43 24.15 9.02 8.21 10.26 13.84 41.61 28.74 19.61 8.27 13.01 8.45 17.35

1947 21.79 17.21 24.73 8.01 8.12 9.82 11.60 37.64 6.21 20.67 6.17 14.19 8.61 16.32

1948 25.99 17.15 23.96 8.19 8.18 9.10 11.16 36.17 4.21 16.94 5.62 14.67 8.92 17.15

1949 28.30 18.42 24.65 23.36 8.18 9.85 12.23 41.03 6.43 17.98 6.18 15.41 8.53 15.53

1950 30.24 20.05 23.58

1951 34.96 25.41 29.47

8.09 8.76 11.64 42.42

10.48 9.97 16.46 44.91

1952 39.93 26.54 34.20 16.53 15.86 11.37 18.11 48.04

6.18 14.56 8.14 15.04

8.81 15.37 10.79 19.99

7.20 13.59 12.57 24.40

‘Rubberlip’ probably refers to blue morwong (Nemadactylus douglasi). The species names in Table 4.17 are as written in the Company records, and no standardisation of the names has been attempted.

48

4.4 Radio Reports 1946-57 4.4.1 Raw data description Radio reports were in a roll of 137 large loose-leaf sheets of paper measuring 70 cm by 50 cm. Records were hand written in blue or black ink, and legibility ranged from very good on most, to faded but legible on a small number. Most sheets were water damaged at the edges, and less than 10 were more extensively damaged.

Each page was printed with the title VESSELS MOVEMENTS, and contained the records for a single month. The month (e.g. September 1957) was hand written beneath the title. The page was ruled to allow one line per day, and the days of the month were numbered from the top to the bottom of the page. Individual entries were made in columns headed by the name of a vessel. A single entry was composed of a position, comments on weather conditions, and catch in baskets of flathead, morwong and mixed or other species. Positions were given as a fishing ground e.g. ‘Eden close’, ‘Everard 20’ etc. Comments on weather conditions were limited to ‘bad’, ‘gale’, ‘vis. nil’ etc. On leaving port ‘sailed’ was entered as the position, and ‘arrived’ was entered on return.

Data were originally entered into a computer as Excel spreadsheets formatted in a similar manner to the original records. All data were entered as written as follows: Table 4.18. Radio reports raw data description. Field Description header vessel vessel name report details date date of radio report position fishing ground weather description of weather conditions flathead wt (baskets) of flathead caught morwong wt (baskets) of morwong caught mixed wt (baskets) of other species caught

49

Type char dd.mm.yy char char num num num

4.4.2 Loading procedure The records were first converted from multi-column Excel spreadsheets to a single column dBase III+ database. Processing was then carried out in two stages - (a) error detection, correction and field conversion and (b) re-formatting. Error detection, correction and field conversion A dBase III+ program was used for this purpose. The following errors were reported and corrected with reference to the original records: •

date of report invalid

•

catch of flathead, morwong or mixed species invalid

•

catch of flathead, morwong or mixed species more than 300 baskets

•

date of sailing not recorded

•

date of arrival not recorded

Numeric fields were determined to be invalid if they contained non-numeric characters. If dates of sailing or return were not noted in the original records they were assumed to have been the day before the first radio report, or the day after the last as appropriate.

Original records of vessel names, positions and weather conditions were converted to standard values using the lookup tables described below. The lookup tables were created by manually assigning standard values to summaries of the contents of fields in the raw data. It was not possible to assign a position in latitude and longitude to the radio reports, but a full description of the fishing ground and area of the fishing ground was retained. This full description includes a field ground which is a 6 character description for a fishing ground (e.g. BABEL for Babel Is, BERMAG for Bermagui), a three character field direction which indicates an area fished within the ground (e.g. S for south, WID for wide, CL for close), and a 3 digit field miles which also indicates an area fished within a fishing ground. The meaning of various combinations of the direction and miles codes are only partly understood, and were the ‘short-hand’ used by skippers to indicate a specific area at the time.

50

For more precise estimation of positions reported, it was desirable to derive an estimate of the depth fished at each position. This was achieved by matching mean depths fished by fishing grounds summarised from the skipper logbook with the standardised radio reported position. The ‘short-hand’ position information could be used for this, even though the meanings were not completely understood. A code with values from 1 to 4 indicating the reliability of the depth estimate was determined as follows:

1. the ground, direction and miles were matched and at least 10 records were present for that position in the skipper logbook;

2. as above, but less than 10 records were present in the skipper logbook;

3. only the ground and direction could be matched from the skipper logbook;

4. only the ground could be matched from the skipper logbook.

Data re-formatting The above procedure produced a single database with one record per radio report, and also records indicating the start and end of each trip. This database was reformatted into a single trip database and report database using a dBase III+ program. These two databases were in a form suitable for loading onto AFZIS. A position in latitude and longitude was assigned where possible, using the same estimation procedure used for the skipper logbook data - i.e. position derived from and estimated longitude of the fishing ground, and latitude estimated by interpolation using depth fished.

51

4.4.3 Data quality A comparison of annual catch by species, year, vessel and skipper from radio reports and the skipper logbook is shown in Table 4.19. This indicates that catches are generally under-reported on the radio reports by 15-20% depending on the species. Possible explanations may be that radio reports are not always made daily (some are missed), catches at the end of a trip may not appear, or that not all catch was reported by radio due to competition with other fishing companies. Radio reports therefore, do not record total catch as well as the logbook or landings records.

52

Table 4.19. Radio reported catch compared to skipper logbook Vessel

Skipper

Year

Trips

Matong

E.Featherstone E.Featherstone G.Stewart J.Barber J.Barber J.Low J.Low J.Low L.Sansom L.Sansom W.Gimpton W.Wilson Subtotal

52 53 55 55 56 53 54 55 54 55 54 55

1 9 7 15 1 23 22 3 1 1 3 1 87

Flathead Log 1,497 13,807 7,880 17,492 2,194 33,648 38,168 11,701 1,085 415 5,924 512 134,323

Moona

C.Olsson C.Olsson C.Olsson C.Olsson W.Wilson Subtotal

54 55 56 57 55

27 23 27 15 1 93

64,867 38,938 53,616 19,409 1,308 178,138

54,200 29,942 43,817 14,860 857 143,675

-16.44 -23.10 -18.28 -23.44 -34.46 -19.35

119,838 117,894 101,173 78,048 7,087 424,040

100,652 105,097 88,936 71,949 5,557 372,191

-16.01 -10.85 -12.10 -7.81 -21.60 -12.23

180

312,461

249,217

-20.24

827,098

706,915

-14.53

RRep 1,397 19,305 22,607 42,039 1,556 68,297 39,658 7,684 1,969 3,651 11,081 11,970 231,214

Diff% -4.57 -36.09 -45.34 -36.45 -35.79 -30.81 -43.18 -12.51 -65.98 -18.04 -21.58 -6.41 -35.06

All species Log 8,329 101,799 109,695 156,217 8,589 181,551 218,350 30,343 11,352 10,050 33,216 23,943 893,434

RRep 8,097 78,426 72,044 119,576 8,065 130,213 161,551 26,449 4,763 9,303 29,942 23,052 671,480

Diff% -2.79 -22.96 -34.32 -23.46 -6.10 -28.28 -26.01 -12.83 -58.05 -7.43 -9.86 -3.72 -24.84

Total Vessel

Skipper

Year

Trips

Other species Log 1,464 30,207 41,361 66,151 2,423 98,712 69,789 8,783 5,787 4,455 14,131 12,790 356,053

RRep 1,334 12,066 4,477 13,463 2,572 24,703 29,243 11,145 603 286 5,080 572 105,542

Diff% -10.92 -12.61 -43.19 -23.04 17.22 -26.59 -23.38 -4.75 -44.40 -31.14 -14.24 11.63 -21.43

Morwong Log 5,368 57,785 60,454 72,574 3,972 49,191 110,393 9,859 4,480 5,180 13,161 10,641 403,058

RRep 5,366 47,056 44,960 64,074 3,937 37,213 92,651 7,620 2,191 5,366 13,780 10,510 334,724

Diff% -0.04 -18.57 -25.63 -11.71 -0.88 -24.35 -16.07 -22.71 -51.10 3.59 4.70 -1.23 -16.95

Matong

E.Featherstone E.Featherstone G.Stewart J.Barber J.Barber J.Low J.Low J.Low L.Sansom L.Sansom W.Gimpton W.Wilson Subtotal

52 53 55 55 56 53 54 55 54 55 54 55

1 9 7 15 1 23 22 3 1 1 3 1 87

Moona

C.Olsson C.Olsson C.Olsson C.Olsson W.Wilson Subtotal

54 55 56 57 55

27 23 27 15 1 93

98,061 74,932 58,859 30,729 2,965 265,546

90,047 74,743 42,293 28,100 1,397 236,580

-8.17 -0.25 -28.15 -8.56 -52.88 -10.91

282,766 231,764 213,648 128,186 11,360 867,724

244,899 209,782 175,046 114,909 7,811 752,446

-13.39 -9.48 -18.07 -10.36 -31.24 -13.29

180

621,599

467,794

-24.74

1,761,158

1,423,926

-19.15

Total

53

Landings records do not contain information on the positions where catches were made. These radio reports provide the only information that may indicate catch position for most of the landings records. To evaluate the quality of the radio report position information, cross checking was carried out with the skipper logbook. Positions are given for each fishing operation in the skipper logbook, and also with each radio report. For a skipper logbook trip, a count was kept of the number of times a fishing position appeared or not in the radio reports for the same trip (Table 4.20). For those grounds fished more than 25 times in the skipper logbook, all except Port Kembla were matched more than 50% of the time. It may be that grounds close to Sydney, such as Port Kembla, Botany Bay and Wattamolla are under-represented in the radio reports as these were often fished at the end of a trip before returning to port. The two most heavily fished grounds, Gabo Is and Cape Everard, were almost completely matched (98% and 94%). This is probably because vessels fishing these distant grounds usually remained in the area for a number of days and so were likely to make corresponding radio reports. It is notable however, that some popular grounds such as Tathra and Eden were not well matched (54% and 58%). Overall, the agreement between the two data sets was 79%, indicating that the radio reports provide useful supplementary position information for the landings records.

54

Table 4.20. Grounds fished by skipper logbook compared with radio reports. Ground Gabo Is Cape Everard Babel Is Tathra Shoalhaven Hd The Pines Green Cape Merimbula Bermagui Montague Is O'Hara Hd Tollgate Is Brush Is Kiama Cape Hawke Crowdy Hd Red Hd Disaster Bay Cape Howe Moruya Eden Bird Is Newcastle Port Stephens Charlott Hd Norah Hd Botany Bay Port Kembla Mowarry Pt Jervis Bay Coalcliff Wattamolla

Code GABO EVRARD BABEL TATHRA SHLHVN PINES GREEN MRMBLA BERMAG MONTAG OHARA TLLGTS BRUSH KIAMA CPHWKE CROWDY RED HD DSASTR CPHOWE MORUYA EDEN BIRD NWCSTL STPHNS CHRLTT NORAH BOTANY KEMBLA MOWRRY JERVIS CLCLFF WATMOL

Total

Match 1,538 814 401 338 9 84 220 118 6 262 64 93 1 17 45 43 33 79 0 51 239 0 3 11 0 21 0 25 12 27 18 2

No Match 34 57 0 288 3 46 40 72 0 85 11 36 3 44 14 34 10 49 6 22 174 4 2 5 1 4 14 31 23 21 28 29

% Match 97.84 93.46 100.00 53.99 75.00 64.62 84.62 62.11 100.00 75.50 85.33 72.09 25.00 27.87 76.27 55.84 76.74 61.72 0.00 69.86 57.87 0.00 60.00 68.75 0.00 84.00 0.00 44.64 34.29 56.25 39.13 6.45

4,574

1,190

79.35

As the radio reports consistently record both the departure and return date for each vessel trip, this provides the most reliable source of fishing effort information in terms of days fished for the landed catch.

Unfortunately, radio report information is only available for the period 1946-57, whereas landings are available from 1938-59. Radio report positions A map of the estimated positions of each radio report is given in Figure 4.3. Interpolation of the longitude according to estimated depth did not produce a great deal of position variation, so the map appears to contain far fewer than the 8,016 records which were plotted. In comparison with the map of skipper logbook catch positions given in Figure 4.1, most areas are represented.

55

Figure 4.3. Estimated radio report positions.

56

4.4.4 Data summary

Report summary by year and vessel

The following tables summarise information contained in the radio reports. As previously indicated, catch weight is under-reported so catch and effort information should be interpreted with caution. Table 4.21. Number of radio reports. Vessel Bareamul Durraween Goolgwai Korowa Maldanna Matong Moona Mulloka Total

1946 138 22 0 152 35 0 0 0 347

1947 113 154 52 121 139 67 117 0 763

1948 7 140 142 129 143 161 169 0 891

1949 0 86 147 122 136 136 133 0 760

1950 0 120 143 127 138 143 134 0 805

1951 0 35 101 128 119 124 126 0 633

Vessel Bareamul Durraween Goolgwai Korowa Maldanna Matong Moona Mulloka Total

1952 0 0 9 146 171 150 172 0 648

1953 0 0 156 141 149 159 170 0 775

1954 0 0 151 156 119 154 173 0 753

1955 0 0 68 21 130 165 163 130 677

1956 0 0 0 0 137 123 150 160 570

1957 0 0 0 0 103 103 107 81 394

Total 258 557 969 1,243 1,519 1,485 1,614 371 8,016

Table 4.22. Total catch by radio report (kg). Vessel Bareamul Durraween Goolgwai Korowa Maldanna Matong Moona Mulloka Total

1946 286,144 71,218 0 307,164 85,348 0 0 0 749,874

1947 178,824 271,443 86,269 219,085 214,386 109,034 242,835 0 1,321,876

1948 11,462 212,004 231,023 268,522 311,768 315,451 329,358 0 1,679,588

1949 0 103,097 195,684 159,456 195,875 228,674 191,017 0 1,073,803

1950 0 187,873 228,324 209,115 214,449 200,352 201,336 0 1,241,449

1951 0 39,753 151,327 206,194 200,415 219,149 204,956 0 1,021,794


1952 0 0 14,225 209,591 293,669 245,407 284,366 0 1,047,258

1953 0 0 208,321 189,493 228,642 210,830 203,019 0 1,040,305

1954 0 0 207,051 199,240 174,347 199,971 240,930 0 1,021,539

1955 0 0 93,413 24,925 207,083 255,853 217,561 206,130 1,004,965

1956 0 0 0 0 215,084 172,188 187,111 210,163 784,546

1957 0 0 0 0 126,783 126,879 128,720 94,016 476,398

57

Total 476,431 885,389 1,415,638 1,992,784 2,467,849 2,283,786 2,431,208 510,309 12,463,394

Table 4.23. Radio report trips. Vessel Bareamul Durraween Goolgwai Korowa Maldanna Matong Moona Mulloka Total

1946 35 7 0 35 7 0 0 0 84

1947 30 28 11 27 28 14 23 0 161

1948 3 33 34 29 34 34 33 0 200

1949 0 21 28 29 28 34 29 0 169

1950 0 31 30 26 29 33 35 0 184

1951 0 7 28 33 30 33 29 0 160


1952 0 0 3 31 32 30 33 0 129

1953 0 0 30 29 31 32 29 0 151

1954 0 0 28 27 29 28 27 0 139

1955 0 0 11 4 25 28 25 22 115

1956 0 0 0 0 25 23 27 26 101

1957 0 0 0 0 20 21 17 19 77

Total 68 127 203 270 318 310 307 67 1,670

Table 4.24. Radio report days fished. Vessel Bareamul Durraween Goolgwai Korowa Maldanna Matong Moona Mulloka Total

1946 250 57 0 260 55 0 0 0 622

1947 203 261 102 238 250 132 206 0 1,392

1948 16 279 288 256 287 290 282 0 1,698

1949 0 172 237 248 257 234 246 0 1,394

1950 0 283 252 235 258 292 281 0 1,601

1951 0 78 272 297 262 287 256 0 1,452


1952 0 0 30 278 286 264 295 0 1,153

1953 0 0 280 289 281 292 300 0 1,442

1954 0 0 293 282 294 297 289 0 1,455

1955 0 0 119 40 256 289 266 225 1,195

1956 0 0 0 0 252 240 268 275 1,035

1957 0 0 0 0 207 221 189 194 811

58

Total 469 1,130 1,873 2,423 2,945 2,838 2,878 694 15,250

Table 4.25. Radio report catch and effort summary. Year

Trips

1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 Total

84 161 200 169 184 160 129 151 139 115 101 77 1,670

Days fished Catch (kg) 622 1,392 1,698 1,394 1,601 1,452 1,153 1,442 1,455 1,195 1,035 811 15,250

Flathead 228,896 270,681 327,675 104,049 109,955 219,371 242,772 181,237 185,778 136,214 133,039 53,977 2,193,645

Morwong 270,650 540,981 493,735 349,552 278,714 249,503 335,518 387,463 456,332 540,791 401,720 289,256 4,594,215

59

Mixed 275,095 496,593 859,449 659,129 806,964 571,622 455,538 475,605 389,209 340,249 212,925 133,166 5,675,543

Total 774,641 1,308,256 1,680,859 1,112,730 1,195,633 1,040,496 1,033,828 1,044,306 1,031,319 1,017,254 747,684 476,399 12,463,403

Days per trip

Catch per day (kg)

7.40 8.65 8.49 8.25 8.70 9.07 8.94 9.55 10.47 10.39 10.25 10.53 9.13

1,245 940 990 798 747 717 897 724 709 851 722 587 817

4.5 Final data formats

Standard data structure for skipper records All data were converted into the following format. This is suitable for loading into the AFZIS system without the need for further data manipulation. The files are currently in dBase III+ databases, but they can easily be converted to text tables or delimited ASCII files as required. Measurement units and methods for storing dates, times positions etc are compatible with those described in section 3.5 of Chapter 3. Structure for database: E:trip.dbf Number of data records: 243 Field Field Name Type Width 1 BOAT_NAME Character 15 2 SKIPPER Character 15 3 TRIP_NO Numeric 3 4 ST_DATE Numeric 6 5 EN_DATE Numeric 6 6 DAYS Numeric 2 7 OPS Numeric 3 ** Total ** 51 Structure for database: E:ops.dbf Number of data records: 7464 Field Field Name Type Width 1 BOAT_NAME Character 15 2 DATE Numeric 6 3 TRIP_NO Numeric 3 4 HAUL_NO Numeric 2 5 OP_NO Numeric 2 6 ST_TIME Numeric 4 7 EN_TIME Numeric 4 8 ST_DEP Numeric 4 9 EN_DEP Numeric 4 10 GROUND Character 6 11 LAT Numeric 6 12 LONG Numeric 6 13 EFFORT Numeric 6 14 RET_WT Numeric 8 15 DIS_WT Numeric 8 16 TOT_WT Numeric 8 17 VALID Logical 1 18 WEATHER Logical 1 ** Total ** 95 Structure for database: E:sp.dbf Number of data records: 17916 Field Field Name Type Width 1 BOAT_NAME Character 15 2 DATE Numeric 6 3 HAUL_NO Numeric 2 4 OP_NO Numeric 2 5 SPECIES Character 2 6 SP_CODE Numeric 6 7 WT Numeric 8 ** Total ** 42

Dec

Dec

2 2 2

Dec

60

Standard data structure for landing records

All data were converted into the following format. This is suitable for loading into the AFZIS system without the need for further data manipulation. The files are currently in dBase III+ databases, but they can easily be converted to text tables or delimited ASCII files as required. Landings have been converted to two files. Land.dbf contains one record per landing or unloading, with a start and end date for unloading as this may have taken a number of days. Landsp.dbf contains one record per species and grade landed which can be linked to land.dbf using boat_name and st_date.

Structure for database: E:land.dbf Number of data records: 2610 Field Field Name Type Width 1 BOAT_NAME Character 15 2 ST_DATE Numeric 6 3 EN_DATE Numeric 6 4 TOTWT Numeric 12 5 UNLOADS Numeric 2 ** Total ** 42 Structure for database: E:landsp.dbf Number of data records: 30337 Field Field Name Type Width 1 BOAT_NAME Character 15 2 ST_DATE Numeric 6 3 SPECIES Character 15 4 SP_CODE Numeric 6 5 GRADE Character 6 6 WT Numeric 12 ** Total ** 61

61

Dec

2

Dec

2

Field conversions made to radio report records Vessel names Record# 1 2 3 4 5 6 7 8 9 10 11 12

FROM BAREAMUL BAREAMUL DARAWEEN DURRAWEEN GOOLGWAI GOOLGWAi KOROWA MALDANNA MATONG MOONA MULDANNA MULLOKA

TO BAREAMUL BAREAMUL DURRAWEEN DURRAWEEN GOOLGWAI GOOLGWAI KOROWA MALDANNA MATONG MOONA MALDANNA MULLOKA

Positions Positions as recorded were converted to match a standard set of fishing grounds extracted from the skipper logbooks (TableA1.1 in Appendix 1). A position may be in three parts (1) a code indicating the name of a fishing ground, (2) a description identifying a region within the fishing ground - e.g. North, South, Wide, Close and (2) a distance in miles that was also used to identify an area within a ground. All positions were summarised to unique entries (shown here as the field position), and then manually interpreted individually to correspond with grounds extracted from the skipper logbooks as given in TableA1.1 in Appendix 1 (shown here as code, ground and miles). As the vessel name appears in the position column in primary raw database, these have been identified using the code ‘VES’. Other records that did not indicate a true position were when the vessel sailed ‘SAILED’, arrived ‘ARRIVE’, was in port ‘PORT’ or travelling ‘TRAV’. A ground code that did not indicate an area within a ground was when the vessel reported that it was anchored ‘ANC’ (usually when sheltering from bad weather).

62

Weather details

Weather details were finally only stored as good (true) or bad (false) after data processing. Table A1.2 in Appendix 1 lists the interpretations made for all comments as recorded in the original records. A blank was interpreted as good weather.

Standard data structure for radio reports All data were converted into the following format. This is suitable for loading into the AFZIS system without the need for further data manipulation. Structure for database: E:trip.dbf Number of data records: 1669 Field Field Name Type Width 1 BOAT_NAME Character 15 2 ST_DATE Numeric 6 3 EN_DATE Numeric 6 4 DAYS Numeric 3 5 N_OPS Numeric 3 6 TRIP_NO Numeric 5 7 FLAT Numeric 10 8 MORW Numeric 10 9 MIX Numeric 10 ** Total ** 69 Structure for database: E:dayreps.dbf Number of data records: 8016 Field Field Name Type Width 1 BOAT_NAME Character 15 2 TRIP_NO Numeric 5 3 DATE Numeric 6 4 GROUND Character 6 5 DIR Character 3 6 MILES Numeric 5 7 WEATHER Logical 1 8 LAT Numeric 6 9 LONG Numeric 6 10 DEPTH Numeric 7 11 DEPTHR Numeric 1 12 FLAT Numeric 10 13 MORW Numeric 10 14 MIX Numeric 10 ** Total ** 92

63

Dec

2 2 2

Dec

2 2

2 2 2

Fishing ground information from skipper logbook Table A1.3 in Appendix 1 gives a summary of fishing grounds and mean depth in metres extracted from the skipper logbooks. This summary was used to provide information on depth fished for the radio reports. The column depthe indicates the standard error of the mean depth (where this could not be calculated due to insufficient samples, a value of 9999999999.00 is shown).

64

CHAPTER 5 – Trends in catch rates and species composition

5.1 Introduction All available per-haul steam trawler catch records were compiled into a single database (see Chapter 3). This data set makes it possible to compare catch-rates over a considerable period of the early development and expansion of the fishery. It also enables examination of catch-rates in some detail by species, depth and area fished that has never been possible previously. Steam trawler catch-rates have not been examined in detail since Blackburn in 1978, who relied on earlier calculations made by Colefax (1938), Fairbridge (1948) and Houston (1955). None of these workers had the benefit of modern computers to arrange and analyse large data sets. This chapter examines per-haul records of catches by steam trawlers in the period 1918 to 1957, which covers most of the period that those vessels operated. Although the tonnage of the steam trawler vessels varied considerably, they all consistently used the same type of fishing gear, operated from the same home port area, and used the same fishing routine. The fishing gear used was similar across the fleet in any year, and was also similar across years, except in 1925 and 1926 when the Vigneron-Dahl otter trawl modification was introduced. A fishing trip was generally about 4–6 days (Colefax 1934), and hauls of between about 3 and 5 hours’ duration were made around the clock while on the fishing grounds (Roughley 1916, R Mansfield pers. comm.). A photograph of the Moona – one of the last steam trawlers to operate in the SEF – is shown in Figure 5.1. Estimates of the total catch of the steam trawlers, the total catch of the trawl fishery and the number of steam trawlers that operated in most years are given in Table 5.1.

65

Figure 5.1. Photograph of the steam trawler Moona – one of the last to operate in New South Wales waters.

Note: This vessel was formerly a mine sweeper in the New Zealand Navy, and fished from 1947 to 1958 for Red Funnel Trawlers Pty Ltd (R Mansfield, pers. comm.).

66

Figure 5.2. Early trawl grounds in shelf waters after Colefax (1934) (shaded blocks), and spatial strata used for the analyses in this study (between the horizontal lines).

67

5.2 Methods

5.2.1 Data summary Haul-by-haul steam trawler catch and effort data were available for the time periods 1918–23, 1937–43 and 1952–57. They totalled 64,371 haul records. The data were converted to a standard format, the areas fished were geographically located, and the species codes interpreted as detailed in Chapter 3. The cleaned haul-by-haul data have been used directly here. There are no haul-by-haul records available for the periods 1915–17, 1924–36, 1944–1951 and 1958–61.

Basic annual statistics were compiled from the data to show total hauls per vessel and year, as well as total catch, discarded catch (where recorded), and catch by primary commercial species landed.

To show spatial features of the data, regions within which to aggregate records had to be defined. As the early data, in particular, were not accurately located (see Chapter 3), a suitable scale was considered to be broad fishing regions. The original records for 191823 contained latitude and longitude positions that were no more accurate than to about the nearest degree. Data for the periods 1937–43 and 1952–57 were only located according to grounds fished as shown as individual areas in Figure 5.2, and it was sometimes not possible to decide to which adjacent grounds the records referred (see Chapter 3). The spatial grouping was therefore chosen as groups of identified adjacent fishing grounds to limit possible mis-placement of records between spatial groups, while allowing for possible errors within groupings. Areas chosen that best grouped fishing grounds were separated by the following lines of latitude (in decimal degrees): 33.6ºS, 33.2ºS, 33.9ºS, 34.5ºS, 35.0ºS, 35.6ºS, 36.7ºS, 37.6ºS and 38.5ºS. The defined areas were allocated identifying letters A to J (Figure 5.2).

68

5.2.2 Mean catch-rate of all landed commercial species There were 2,918 hauls that had a value of zero for retained catch, and no reason was given for these records. To avoid the inclusion of trawls that were invalid because of gear failures in catch-rate calculations, those that had no retained catch were excluded, which left 61,453 hauls for catch-rate analysis. This exclusion was supported by the 1918–23 records, which also recorded discarded catch. Of the 279 haul records in this time period with no retained catch, 268 also had no discarded catch.

Mean catch per unit effort (CPUE) was calculated by summing the total catch and dividing by the total number of hours trawled. The fishing procedure on steam trawlers seldom varied. The tow time of most trawls was recorded: 84% of these times were between 3 and 5 hours.

5.2.3 Catch-rate by main commercial species by area Mean CPUE was calculated per species, calendar year and area as defined above, and results displayed for the most heavily fished regions that had a continuous history of fishing. An area and year was deemed to be sufficiently sampled when there were at least 200 hours trawling (about 50 hauls) in a stratum. Only these results are presented to avoid spurious conclusions because of low sampling levels.

As the depth fished may have a strong influence on the species composition of the catch, the percentage of hours trawled at different 20 m depth strata were also examined for these same areas for different periods in the fishery. Depth fished was given for 99% of records for the period 1918–23, 44% of records for 1937–43 and 92% of records for 1952–57 (Chapter 3). For examination of patterns by depth fished, only records that had a recorded depth were included.

For area aggregations, it was important to allocate fishing effort as well as possible to each of the broad fishing areas defined above. Hauls in the 1937–43 and 1952–57 data that were reported to have been made at the location coded as GABO (grounds associated with Gabo Is.) were assumed to have been made in area I, even though these hauls would be placed in area H using the latitude cut-off of 37.6ºS; it is most likely

69

they actually occurred south of that line. Therefore, the 8,797 hauls in these years that were reported to have been made in the Gabo Is. Grounds were assigned to area I.

The CPUE in catch-per-hour trawled by species was compared for the three areas that were most heavily fished over the history of the fishery: (1) the ‘home’ area D off Botany Bay, (2) area G around Montague Is., and (3) area H off Eden (see Figure 3.2). Although area I was heavily fished from 1943 onwards, it was not examined in detail because it provided no information over the 1918–23 period.

Species were encoded in the original records as two letter identifiers (Chapter 3). There can be little doubt about the true identity of most of the species codes because of the small number of commercial target species, and published accounts of the composition of the steam trawler catches (e.g. Roughly 1916, Colefax 1934, Fairbridge 1948, Houston 1955, Blackburn 1978). It was not possible to separate shark and skate in the original records based on the two letter species codes however (code ‘SK’ could be either), so that is why the combined category of shark and skate is used here. The leatherjacket was considered most likely to be chinaman leatherjacket (Nelusetta ayraudi) and latchet catches likely to be Pterygotrigla polyommata from examining Roughley’s (1916) information about early trawl catches (Chapter 3). Other leatherjacket species (F. Monacanthidae) may have been present, and some catch recorded as latchet was possibly red gurnard (Chelidonichthys kumu). The common names of latchet and gurnard were apparently used interchangeably at times in the commercial catch records, and the two fish are similar in appearance.

70

5.3 Results

5.3.1 Data summary Estimates of the total catch of all landed fish taken from the SEF, the total landed catches of steam trawlers alone, and the number of steam trawlers operating for the entire history of steam trawling from 1915–61 are given in Table 5.1.

The number of hauls by vessel and year are given in Table 5.2. A large proportion of the total fleet is represented in the haul-by-haul data set for the periods covered. Some individual vessels made more than 1,000 hauls in a year, as the Bareamul did during 1939 and 1940 – an extraordinary number considering the time required for unloading and moving to and from the fishing grounds. Some of the vessels spanned the major periods of available data. The Dureenbee, for example, fished from 1919 to 1942. The demise of this vessel in 1942 was related by Loney (1993): “A [Japanese] submarine attacked the trawler Dureenbee, 223 tons, on August 3, killing two seamen and a brother of the master, Captain Reid. Extensively damaged above the waterline by shell and machine gun fire, with her funnel shot away and all superstructure either smashed or burnt out, she drifted for several hours before going ashore one mile north of Moruya Heads.” Other vessels that spanned long periods of available data are the Goolgwai (28 years), Goonambee (22 years), Goorangi (21 years), Korowa (27 years), and Mary Cam (27 years).

71

Table 5.1. Estimated annual total commercial landed catch for the New South Wales South East Fishery fishery, and for steam trawlers the total catch, number of vessels and the proportion of the total catch taken by steam trawl each year. Year

Total catch (tonnes)

1915-16 1916-17 1917-18 1918-19 1919-20 1920-21 1921-22 1922-23 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937-38 1938-39 1939-40 1940-41 1941-42 1942-43 1943-44 1944-45 1945-46 1946-47 1947-48 1948-49 1949-50 1950-51 1951-52 1952-53 1953-54 1955 1956 1957 1958 1959 1960 1961

740 743 861 1,372 2,533 2,507 2,452 1,478 1,301 2,405 3,490 3,838 4,841 5,800 6,839 6,450 5,187 5,325 4,410 4,493 4,896 6,275 6,222 6,391 5,619 3,648 3,113 1,016 1,032 4,568 5,802 7,360 6,885 6,419 5,857 5,094 4,767 5,571 – 4,620 3,924 4,560 3,456 3,228 3,720 1,296

Steam trawlers Catch (t)

Vessels

% of total catch

740 743 861 1,372 2,533 2,507 2,452 1,478 1,301 2,405 3,490 3,838 4,841 5,800 6,665 5,884 5,181 4,665 4,335 4,019 4,433 4,781 5,129 5,149 3,845 1,749 961 566 488 841 1,900 3,867 4,530 4,549 3,990 3,690 3,412 3,465 3,541 1,407 1,240 1,064 806 – 224 12

3 3 3 3 4-7 7 7 7 – 5 5-8 9 10 11 17 16 16 15 14 14 13 13 14 14 7-14 4 3 1 1 1-4 4-5 6-11 11-12 12 11-12 11 11 10 10 6-8 4 4 4 2 1 1

100 100 100 100 100 100 100 100 100 100 100 100 100 100 97 91 100 88 98 89 91 76 82 81 68 48 31 56 47 18 33 53 66 71 68 72 72 62 – 30 32 23 23 – 6 1

Notes: This table includes a mixture of records by calendar and financial year. Total catch 1915-16 to 1947-48 from Fairbridge (1948) taking the greater of Marine Services Board or NSW State Fisheries Department annual catch statistics; Total catch 1948-49 to 1952-53 from Houston (1954); Total catch 1955-1961 from best estimated total catch of flathead (includes some landings from Victoria), morwong and redfish combined plus 20% for additional species from Tilzey (1994); Steam trawler catch 1915-16 to 1953-54 from Houston (1954); Steam trawler catch 1955-1958 from Red Funnel Trawler company landings in Chapter 4, Steam trawler catch 1960-1961 from Red Funnel Trawler company vessel logbooks in Chapter 4, Number of steam trawlers working 1915-16 to 1953-54 from Houston (1954), Number of steam trawlers working in 1955 from Blackburn (1978) given as 7.5 vessels in 1954/55, Number of steam trawlers working in 1956 to 1961 from Red Funnel Trawler company landings and vessel logbooks in Chapter 4 and also in Blackburn (1978).

72

Table 5.2. Number of hauls by vessel and year. Vessel

Year 1918

19

20

21

22

23

Alfie Cam Bareamul 541 507 216 626 603 108

Dibbiu

158 132 544 639 91 174 397

41

42

52

53

54

55

56

57

127

2,915

1 764

792 984

227 139 452 516

72

936 992

676

729 506 225

2,432 1,488

947 892 1,124 893 940

Koraaga

4,714 2,601

46

459 577

221 894 370

5,110 57 850 383

2,847

194

4,545 4,010

54 659 125

1,874

607 111

2,178 668

817

Maldanna Mary Cam

43

685

208 198 620 462

Gunundaal Korowa

40 222

15

Goolgwai Goorangi

39 899

839 908

Brolga

Goonambee

37

793 1,149 1,042 829 675 226

Beryl 2

Dureenbee

38

742 925

Total

877 985

990

223 646 306

768 791 401

3,445

851 813 804 681 772 696

4,617

160

4,187

Matong

691 824 802 895 697 792

4,701

Moona

865 891 832 789 813 643

4,833

706 872 699

2,277

Mulloka Olive Cam Samuel Benbow

750 921

661 145

2,332 213

Note: These records cover most vessels operating during 1918–23, about half those operating in 1937–40, all vessels during 1941–43, about half from 1952–55 and most from 1956–57.

Total annual retained catch by principal commercial species or species group is given in Table 5.3. The amount of catch discarded was also recorded for the very early period of the fishery, so those figures have also been shown. The records show that flathead, latchet and chinaman leatherjacket dominated early catches from 1918–23. Flathead was the main target species from 1937–43, with more morwong and redfish and much smaller catches of latchet and leatherjacket. During 1952–57, catches of morwong and redfish surpassed flathead catches, and catches of latchet and leatherjacket remained low. Catches of shark and skate were at a low level throughout; they were apparently seldom landed in 1923 and 1937–40. A relatively low and consistent catch of other species throughout shows that the main identified species were the primary commercial target species for the fishery – at least in terms of landed catch weight.

73

358

Table 5.3. Total number of hauls, hours trawled and catch per commercial species from SEF haul-by-haul records for the period 1918 to 1957. Year

1918 1919 1920 1921 1922 1923 1937 1938 1939 1940 1941 1942 1943 1952 1953 1954 1955 1956 1957

Hauls

1,749 2,313 1,108 2,768 3,581 492 6,470 9,221 9,326 2,192 2,416 1,396 227 3,309 4,699 3,398 3,282 3,328 3,096

Hours

6,956 9,060 3,873 10,193 13,511 1,858 21,687 31,893 33,241 7,921 8,839 5,222 909 13,356 19,015 13,611 12,763 13,559 12,757

Catch (tonnes) Retained Discard Flathead Morwong Redfish Latchet Leatherjacket 1,010 177 578 14 1 192 103 1,858 236 992 0 2 337 282 1,136 215 486 1 4 192 375 2,326 434 1,483 3 2 229 473 3,149 710 2,113 0 2 215 682 443 113 311 0 0 22 104 1,950 – 1,132 210 44 0 3 3,054 – 1,943 322 40 80 55 3,064 – 1,953 249 76 56 161 731 – 410 36 22 22 67 878 – 348 83 118 30 96 617 – 186 28 130 21 92 128 – 61 0 2 5 25 1,260 – 299 399 181 20 81 1,427 – 239 498 348 14 43 1,091 – 193 482 208 5 13 1,118 – 151 562 179 10 23 1,000 – 156 490 115 12 31 846 – 79 433 105 20 15

Shark/ Other Skate 59 65 119 131 29 51 28 108 18 124 0 6 0 564 1 617 1 574 1 174 10 194 73 87 30 4 111 170 128 159 76 117 88 106 80 118 85 5

5.3.2 Mean catch-rate of all landed commercial species A time series of CPUE values for the total landed catch of steam trawlers from 1924 to 1954 was produced by Houston (1955), who based it in part on earlier work by Fairbridge (1948) and is reproduced here as Figure 5.3. The CPUE was calculated by dividing the annual total steam trawler landed catch by the total tonnage of steam trawlers that operated in the fishery each month, based on monthly returns to the Government Marine Services Board. Sections of the series that are comparable to periods where haul-by-haul data are available are shown in Figure 5.4 as dashed lines. It is evident that the haul-by-haul calculation of steam trawler CPUE paints a very different picture to that of the catch per trawler-ton-month values.

74

2.5 2 1.5 1 0.5

1952-53

1950-51

1948-49

1946-47

1944-45

1942-43

1940-41

1938-39

1936

1934

1932

1930

1928

1926

0 1924

Tons per trawler-ton-month

Figure 5.3. CPUE for steam trawlers only for all landed catch (from Houston 1955).

Year

Figure 5.4. CPUE for retained commercial catch by year. CPUE from haul-by-haul records in kg/h are shown as points on solid lines, and historical CPUE as presented by Houston (1955) in trawler-ton-months is shown as dashed lines.

CPUE (kg/h)

300

2

250 200

1.5

150

1

100 0.5

50 0

Tons per trawler-ton-month

2.5

350

0 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Year

Note: Because of differences in the mode of reporting from calendar year to financial year, the 1918-19 data from Houston (1955) have been compared with 1918 calendar year data. Houston reported that the 1922-23 and 1923-24 data were incomplete, so data for those years do not appear here.

75

The results of calculating haul-by-haul CPUE by dividing the total retained catch in each year by (a) total hauls, (b) total hauls that had a retained catch, (c) total hours fished and (d) total hours fished for those hauls with retained catch all showed a pattern very similar to that shown in Figure 5.4. It can be concluded, then, that the difference in the CPUE calculated here and that calculated earlier by Fairbridge (1948) and Houston (1955) is because the data sets and effort measures were different, rather than because of some property of the method used to calculate the mean CPUE. Fairbridge (1948) commented that he did not have haul-by-haul records for the period 1937–43 used here, and Houston (1955) makes no mention of such data. The haul-by-haul records were apparently extracted subsequently from company records that were not available to researchers at the time. The earlier CPUE series for 1915–16 to 1922–23 were calculated and presented in tabular form by Houston (1955), but were not presented as part of the graphical representation of the series as shown in Figure 5.4 because he did not consider that catch-rates from the development period of the fishery would give a true representation of what fish were available.

The haul-by-haul picture of retained catch CPUE suggests a period of learning until about 1920 and then a levelling off. The CPUE by 1937 is less than 50% of the CPUE achieved in the early 1920s. From 1937 to 1940 the CPUE is reasonably steady, followed by a clear increase of more than 50% in the CPUE from 1940 to 1943. By 1952 the CPUE had returned to levels comparable to those before the war, and remained at about the same level to 1957.

The mean latitudes fished each year are shown in Figure 5.5. Because the home port of the steam trawlers was Sydney or Newcastle for the entire time they operated, and these ports are at the northern end of the latitudinal range of the fishery, the mean latitude fished is also a good indicator of how far, on average, the trawlers travelled to the fishing grounds. In the years 1918–23 much of the fishing was in the grounds closer to the home ports, even though the southern grounds such as those near Eden were known and fished at times. Fishing appears to have occurred in more southern areas by 1937 and 1938, but to have then moved north again during the war, probably because of increased retention and even targeting of species that were previously discarded, and also the general wartime risk to shipping. By 1952 the mean latitude fished was near the mean fished just before the war, and was even further south by 1954–57. The pattern of 76

mean latitude fished has much in common with the mean CPUE in Figure 5.4, indicating a movement further afield as CPUE declined.

Figure 5.5. Mean latitude fished by year.

Degrees South

34

35

36

37

38 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Year

Note: The mean latitude fished does not necessarily indicate the actual position of the most heavily fished areas. Actual fishing in any year normally occurred in many areas within the range of about 30–40° south.

The monthly mean latitude fished is given in Figure 5.6, which shows a repeating seasonal movement, with northerly fishing at a peak in about October, and southerly in February. The seasonal movement was apparently strongest in 1952 and 1953, but weakened thereafter. Monthly mean latitude fished by month across all years is shown in Figure 5.7, and the strong seasonal pattern is evident.

Mean depth fished each year is shown in Figure 5.8. This shows that the vessels fished at depths averaging between 75 and 100 m from 1918–23, and between 110-130 m in 1937–42 and 1952–57. The overall pattern is again similar to that of the CPUE, indicating a movement to deeper water as CPUE declined.

77

Figure 5.6. Mean latitude fished by year and month. 32

Degrees South

33 34 35 36 37 38 39 40 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Month

Figure 5.7. Mean latitude fished by month over all years.

Degrees South

35

36

37

38 1

2

3

4

5

6

7

8

9

10

11

12

Month

Figure 5.8. Mean depth fished by year. 60 70

Depth (m)

80 90 100 110 120 130 140 18 19 20 21 22 23

37 38 39 40 41 42

52 53 54 55 56 57

Year

78

5.3.3 Catch-rate by main commercial species by area The proportion of hours fished in different areas each year is shown in Figure 5.9. Only hauls that had position information (and a non-zero retained catch) were included, which was 44% of hauls (n = 5,279) from 1918–23, 95% of hauls from 1937–43 (n = 29,746) and 95% of hauls from 1952–57 (n = 19,965).

In the period 1918–23, fishing was mainly confined to either area D (the Botany Bay area near Sydney), or area H (off Eden). There was a little fishing in area C (north of Sydney) and area G (around Montague Is.). In most of these early years, the main fishing area was area D.

Figure 5.9. Proportion of total hours trawled for hauls with location information (as per Figure 5.2) that were made in each area. 100%

Percentage of hours trawled

A B

80%

C D

60%

E F 40%

G H

20%

I J

0% 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Year

In the period 1937–43 the fishing effort was much more evenly distributed over more regions, with the most hours trawled in area I off Cape Everard. During the war there was relatively less fishing far south of Eden and more in areas D and F further north, particularly by 1943.

It appears that the most southerly areas of I (Cape Everard) and J (Babel Is) were discovered at some time between 1923 and 1937, a period for which there are no haulby-haul data. By 1937 just less than half of the trawling was in these distant grounds.

79

During the period 1952–57, the areas fished remained relatively consistent throughout, with nearly half of the trawling directed at areas I and J far to the south, as in the years before the war. Compared to the pre-war period, there was relatively more fishing in area H off Eden, and relatively less fishing in areas B/C/D/E/F closer to the home port. Area G around Montague Is. was consistently fished at a relatively low level throughout most years from 1918 to 1957.

Figure 5.10. Contribution per species to the total commercial CPUE by year for area D. 300 Flathead

250

Morw ong

CPUE (kg/h)

Redfish

200

Latchet Leatherjacket

150

Shark/Skate Other

100 50 0 1918

1920

1922

1938

1940

1942

1953

1955

1957

Year

The CPUE per species for area D off Botany Bay is shown in Figure 5.10. Very high catch-rates of more than 100 kg per hour of flathead are shown in the period 1918–23. This corresponds well with Colefax’s (1934) documentation of large catches of flathead in these years in the ‘home’ grounds. Catches in this early period also included considerable amounts of leatherjacket and latchet. The relative catch-rates per species remained reasonably constant between 1918 and 1923 with flathead contributing most to catch rates over that period, and then leatherjacket. During this period the total CPUE was lowest in 1918 at about 150 kg per hour and then fluctuated in the range of 180 to 290 kg per hour from 1919 to 1923. By 1937, the catch-rate of all species combined was about one third of the earlier levels. This was apparently due to a considerable reduction in the catch-rate of flathead, and much reduced catches of leatherjacket and latchet. During the period 1937–43, the flathead catch-rate did not recover in this area, but the

80

overall catch-rate increased because of increases in the catch-rates of leatherjacket and shark/skate, as well as the appearance of catches of redfish. The period 1952–57 shows the catch-rate of flathead declining to virtually nil, an initially higher catch-rate of redfish followed by a decline, and an appearance of morwong in the catches. There was a strong overall decline between 1918 and 1957 in the catch-rates of species that were apparently the primary targets in the early period of the fishery – namely flathead, leatherjacket and latchet. At the same time, the catch-rate of some species that were of minor importance during the early period – namely redfish and morwong – appears to have increased.

Figure 5.11. Contribution per species to the total commercial CPUE by year for area G. 300 Flathead

250

Morw ong

CPUE (kg/h)

Redfish

200

Latchet Leatherjacket

150

Shark/Skate Other

100 50 0 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Year

Area G around Montague Is shows a broadly similar pattern to that of area D for the years in which there were enough hours trawled to allow comparison (Figure 5.11). The early catch-rates for flathead were close to and possibly greater than, those in area D, but the catch-rates of leatherjacket may have been smaller. Latchet also appeared in catches in area D during this early period. Between 1937 and 1943 the overall catch-rate was sustained or even increased, apparently mostly due to an increased catch-rate of redfish. The overall catch-rates from 1952–57 were close to the pre-war level, and higher than that achieved in area D. The relative catch-rate per species was similar to that of 1942, but much different to 1937, where redfish and morwong were scarcely represented.

81

Figure 5.12. Contribution per species to the total commercial CPUE by year for area H. 400 Flathead

350

Morw ong

CPUE (kg/h)

300

Redfish Latchet

250

Leatherjacket

200

Shark/Skate

150

Other

100 50 0 1918

1920

1922

1938

1940

1942

1953

1955

1957

Year

Area H shows a similar pattern to areas D and G, except perhaps that the overall catchrate in 1918–23 was the highest of all areas, and that catch-rates of leatherjacket and latchet were comparatively higher in this area during that early period (Figure 5.12). By 1937, CPUE for leatherjacket and latchet had dropped to low levels, flathead had declined by more than 50% and morwong had increased. Morwong were taken at a low catch-rate in this area during 1918–23 as well. Maintenance of the overall catch-rate from 1940–42 was apparently due to an increase in the catch-rate of redfish alone, in the same manner as area G. By 1952 the overall catch-rate was about the same as the 1937– 42 period, but the catch-rate for flathead appears to have declined further, while the morwong catch-rate increased. The catch-rate per species was fairly stable from 1952– 57.

82

Figure 5.13. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952–57 in area D.


50 1918-23

40

1937-43 1952-57

30 20 10 0 0

20

40

60

80 100 120 140 160 180 200 220 240 Depth (m)

Figure 5.14. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952–57 in area G.


60 1918-23

50

1937-43 1952-57

40 30 20 10 0 0

20

40

60

80

100 120 140 160 180 200 220 240 Depth (m)

Note: 0 = 0 to 20 m.

83

Figure 5.15. Percentage of hours trawled by depth for the periods 1918–23, 1937–43 and 1952–57 in area H. Percentage of hours trawled

90 80 1918-23

70

1937-43

60

1952-57

50 40 30 20 10 0 0

20

40

60

80

100 120 140 160 180 200 220 240 Depth (m)

From Figure 5.8 we know that fishing was carried out in progressively deeper waters over the history of steam trawling. As fish species may have narrow depth distributions, it would be expected that a change in fishing depth would change the species composition of catches. To examine what difference a change in depth might have on the comparative catch-rate per species, histograms of the percentage of hours trawled at 20 m depth intervals for each time period and fishing area are shown in Figure 5.13 to Figure 5.15. Figure 5.13 shows that, in area D, the vessels in the 1918–23 period fished mainly between 80 and 120 m, in 1952–57 between 120 and 160 m, and in 1937–43 at intermediate depths. Depths fished in area G did not change much from 100–140 m throughout the years (Figure 5.14). In area H, there is a considerable difference between the depths fished in the early and late periods: mainly less than 80 m in 1918–23 and from 60 to 160 m in 1937–43 and 1952–57 (Figure 5.15).

CPUE by year per species for area H in depths of less than 100 m is given in Figure 5.16. For 1918–22 and 1937–39 the pattern is similar to that in Figure 5.12. For the 1952–57 period, the overall catch-rate is similar, but there is a difference in the pattern shown for redfish (which were almost absent), and a small increase in catch-rate of leatherjacket. Morwong was still strongly represented in the more shallow hauls of 1952–57. These results indicate that the appearance of redfish in catches after 1940 may be related to the increase in depth fished, and similarly, that some of the decline in the

84

catch-rate of leatherjacket may perhaps also be attributed to the fishing depth change. However, change in depth fished does not seem to account for the increase in the morwong catch-rate.

Figure 5.16. Contribution per species to the total commercial CPUE by year for area H in depths of less than 100 m. 450 Flathead

CPUE (kg/h)

400

Morw ong

350

Redf ish

300

Latchet

250

Leatherjacket Shark/Skate

200

Other

150 100 50 0 18 19 20 21 22 23

37 38 39 40 41 42 43

52 53 54 55 56 57

Year

Note: Redfish CPUE has been included, but the values are small (< 3 kg/h in all years) and do not clearly show at the scale of this figure.

5.4 Discussion The overall trend in CPUE shows a decline over the history of steam trawling. There was certainly a period at the beginning of the fishery when the CPUE was probably lower than it could have been because of the inexperience of new crews, and a lack of knowledge about the fishing grounds and fish habits. Colefax (1934) said that “there seems to have been an extraordinary variation in the efficiency of the different crews or gear, some vessels regularly returning with much higher catches than certain of the others, even though all were working over the same ground. In addition, considerable loss seems to have been experienced through the use of faulty equipment and the inability of the crews in certain cases to restore the latter to efficient working order.” In 1925 and 1926, a more efficient Vigneron-Dahl extension to the otter trawl gear was 85

introduced. According to Hickling (1931), this increased the efficiency of the gear by one-third on average (Fairbridge 1948). There would have been many other changes to fishing practices, vessels or gear over the history of the steam trawling period that would have increased fishing efficiency. These changes, if they could be accounted for, would suggest that the apparent long-term decline in the abundance of commercial species in the trawl grounds was even greater than that evident in the data reported in this Chapter.

There has been considerable examination of fisheries around the world, and also of the SEF (Fairbridge 1948) to determine how the large reduction in fishing effort during World War II changed catch rates. Observed changes in catch rate during this time have been used as an indicator of the state of the fishery in terms of exploitation in the years before the war. Haul-by-haul results here show that overall CPUE increased by 1943 by about 50% over pre-war levels. However, trends per principal target species show that much of the increase in the overall catch-rate from 1941 to 1943 is explained by an increase in catches of redfish and morwong.

Catch per trawler-ton-month in 1937–43 (Figure 5.4) was of a similar order to 1918–21. However, the catch per hour trawled CPUE during 1937–43 was much lower than that of 1920–23. The new haul-by-haul data suggest that the fishery may have been heavily exploited just before World War II, and that there was apparently no recovery during the war (to 1943 at least) of the species mainly targeted before that time – flathead, leatherjacket and latchet.

Catch and catch-rates per species in the per-haul records refer to landed retained catch only. The interpretation of trends over long time periods may be complicated by changes in what species or sizes of fish were acceptable to the market at different times. For example, small sizes of certain species may become saleable, or when large catches are made, only larger, more valuable sizes might be retained. Fairbridge (1948) noted that the catches of flathead in the early period were probably representative of the true abundance of that species, but that some of the species that were marketable by 1948 were probably wasted earlier on. Morwong, redfish, latchet and barracouta became more acceptable through time. He also states that morwong was actively sought when flathead was “off”. These changes in market acceptance could well explain the 86

appearance of “new” species in the catches as seen in Figure 5.10 to Figure 5.12 and Figure 5.16. The increase in shark/skate catch during the war years is apparent, as well as the much increased catch-rate of redfish and, to some extent, morwong. Redfish and morwong did not apparently appeal to the market in the 1918–23 period, as the figures in Table 5.3 also suggest. During the period 1952–57, fishing effort in the order of 3,000 hauls caught about 600 tonnes of redfish and morwong. The total discarded catch in 1922 with 3,581 hauls was 710 tonnes; it is therefore likely that a considerable portion of this discarded catch was redfish and morwong.

It is much more difficult to explain the disappearance of the early marketable species from the catches. Leatherjacket and latchet were, after flathead, the main targets in the early years of the fishery. Colefax (1934) says that leatherjacket figured prominently in catches in the Eden ground in these years, almost to the total exclusion of flathead, while latchet were also taken in enormous quantities. He also noted that, in addition to the reduction in the flathead catch in later years, there was a corresponding reduction in the quantity of mixed fish (leatherjacket, latchet, John dory, morwong etc.) taken. Although leatherjacket may be less likely to be caught in deeper waters, changes in areas and depth fished do not seem to be sufficient explanation for an apparently large decline in the abundance of leatherjacket, and also of latchet in the shelf trawl grounds of the SEF.

Changes in the abundance of flathead in particular have been well studied over the years, as this was the principal target species for the fishery, at least over the time when steam trawlers operated. The results here show that the fishing fleet moved further afield and into deeper waters as catch-rates declined. In some areas, the local depletion of flathead was much greater than indicated by aggregated catch-rates for the whole fishery. In the early period of good fishing in the Botany Bay ground (area D; Figure 5.10), the bulk of the catch was flathead that were “very large and bursting with roe” Colefax (1934). The timing of the heavy fishing period was predictable from early September to early December, and vessels fishing then often returned to port with full holds well before the normal trip time of 4-6 days. At the time, the heavy catch period was referred to as the “Botany Glut”. This glut period became a yearly expectation, but failed to occur by 1926. By the 1930s, the Botany Bay ground was practically useless for trawling purposes, and was visited by boats only at the end of a trip (Colefax 1934). 87

It is clear that the overall catch-rate of flathead was sustained by the movement of vessels to new grounds or deeper waters. There were cases of localised, almost complete, depletion but there are no indications of cases of localised recovery of flathead catches.

There is little doubt that the abundance of the target species flathead, and possibly leatherjacket and latchet, was considerably decreased due to the heavy fishing effort on the shelf, to the point of over-exploitation in some years (Colefax 1934, Fairbridge 1948, Houston 1954). This in itself may explain apparent changes in the relative abundance of species in the catches over time, as each species has different biological characteristics that would result in different population responses to increased fishing mortality.

It is known that mobile fishing gear such as the otter trawl can change seabed habitats. In heavily fished areas of the United States side of the Gulf of Maine, habitat complexity was reduced by direct removal of biogenic (e.g. sponges, hydrozoans, bryozoans, amphipod tubes, holothurians, shell aggregates) and sedimentary (e.g. sand waves, depressions) structures (Auster et al. 1996). Also, the removal of organisms that create structures, such as crabs and scallops, indirectly reduced habitat complexity (Auster et al. 1996). The physical effect is caused by the ground rope, chains and bobbins, sweeps, doors and any chaffing mats or parts of the net bag contacting the bottom (Jones 1992).

The first demonstrated link between habitat change and changes in trawl fish catch composition was for the Australian northwest shelf fishery by Sainsbury (1987, 1988). Research trawl data showed that fauna associated with open sand benthos increased in the catch as the fishery developed, and that fauna associated with dense emergent benthos (e.g. sponges, gorgonians, alcyonarians) decreased. In other trawl fisheries such as the Spanish Galician shelf and slope, the abundance of non-commercial species has apparently increased as the abundance of primary commercial species has declined. These changes were attributed to fishing pressure and also environmental fluctuations affecting production in the region (Fariña et al. 1997).

88

The changes observed in the relative abundance of the principal commercial species on the SEF shelf during the steam-trawling period may therefore have various causes, some of which may not be easily reversed with a reduction in fishing. Theoretical models to explain the observed changes could be developed using additional information on species composition, bottom structure and sediments from recent scientific surveys in the region.

89

CHAPTER 6 – Abundance indices

6.1 Introduction

Changes in species composition through time as indicated by catch-rates were examined in Chapter 5. However, un-standardised catch-rates are difficult to interpret as indicators of relative abundance because changes in catch-rates can be confounded with changes in other aspects of fishing operations through time, such as species targeting and retention, or depth and seasons fished. This chapter attempts to account for factors such as these to produce standardised indices that could be evaluated as possible indices of annual abundance for the main commercial species.

Environmental factors that have irregular cycles such as El Niño may also have a strong influence on the abundance or availability of certain marine species. This chapter also examines a broad range of correlations of such environmental factors with the standardised annual abundance indices of each species to determine whether variation in these factors is a major driver of apparent abundance.

6.2 Methods

6.2.1 Input data

As analyses required detailed records of the set and haul times, as well as the area and depth fished, records not containing complete information for these details were excluded as shown in Table 6.1. As the duration of the trawl was the primary effort measure, records with suspiciously long or short trawl time (those less than 1 or greater than 6 hours) were also removed from the analysis. Of the 61,453 hauls available for analysis, 36,225 (59%) remained for analysis of all retained catch.

90

Table 6.1. Reasons for removal of haul records from analyses. Invalid

Reason

Number

No retained catch Hours trawled less than 1 or greater than 6 Area not known Depth not known

2,918 149 6,463 18,544

Separate analyses were conducted for the main species. For individual species, records were included in an analysis if greater than 20 kg of that species was caught in a haul. This rule excludes hauls where a species may have been caught but not retained, caught but not recorded, or trawls that were made in places or times where the species does not occur. All of the original trawl records had catches measured in baskets of fish (Chapter 3), and part-baskets were not often recorded. A basket of fish weighs approximately 30 kg, so 20 kg was chosen as an amount of fish that was likely to have been recorded. Exclusion of records that have less than a small catch of the species in question has been a common approach in recent years when standardising logbook catch rate data for the South East Fishery (SEF). Numbers of records remaining for analysis varied with species (Table 6.2).

Table 6.2. Valid records remaining for analysis per species and time period. Valid for analysis

Species All retained Flathead Morwong Redfish Latchet Leatherjacket

1918-23 5,247 4,485 99 27 2,099 2,395

1937-42 12,880 12,276 2,321 555 56 372

1952-57 18,114 12,081 14,098 4,810 838 968

Total 36,241 28,842 16,518 5,392 2,993 3,735

6.2.2 Astronomical Events

The time of first and last light, or nautical twilight were calculated from the position (latitude and longitude), the date, and algorithms by Doggett et al. (1990). Moon phase was calculated using the date, and modified versions of algorithms given in Press et al. (1988).

91

6.2.3 Conversion of continuous variables into factors

To simplify the interpretation of log-linear model (LLM) results, continuous factors were converted to a small set of discrete class values (Table 6.3). Moon phase continuous sinusoidal values from +1.0 (full) to –1.0 (new) were assigned to three classes to indicate full, mid-phase and new. Using the time of nautical dawn and dusk, trawls were determined to be day trawls if both setting and hauling occurred during the day, and night trawls if both setting and hauling occurred at night. Trawls at other times were deemed to be mixed. Exclusion of trawls with a duration of greater than 6 hours ensures that entire periods of night or daylight cannot occur within a trawl. The month of the trawl was used to determine which of 4 quarters to assign. The depth of the trawl was assigned to 3 classes indicating shallow (120 m). The depth divisions were chosen to equally divide the range of depths containing most of the data. Only 2% of hauls had a recorded depth less than 40 m or greater than 160 m. Area, year and individual vessel were also all discrete sets of classes.

Table 6.3. Vessel, time and environmental factors included in the LLM analyses and discrete classes defined for each factor. Factor Year Vessel Quarter

Area Depth

Moon phase

Time of trawl

Classes 1918-1957 Various 1 2 3 4 A-J Shallow Mid Deep New Mid Full Night Mixed Day

Explanation One value per year One value per vessel Q1 months 1-3, Q2 4-6, Q3 7-9, Q4 1012

Defined by lines of latitude (Figure 5.2) Shallow < 80 m, mid 80-120 m, deep > 120m New if phase value less than –0.5, mid if between –0.5 and 0.5 and full if greater than 0.5 Depends on whether both set and haul occur between morning and evening nautical twilight

92

6.2.4 Sensitivity tests

A number of somewhat arbitrary decisions have been made about what data to include in analyses and how to convert continuous measurements into discrete factors. The number of classes for any factor was kept to a minimum in the knowledge that the data are unbalanced – observations were not randomly distributed with respect to factors of interest. A minimum of 3 classes for any continuous measurement was also considered necessary to allow some interpretation of the results.

Sensitivity of the results to alternative decisions was tested by comparing the resulting annual index trend for each species with those produced using the base assumptions. Alternative decisions tested were (a) using 10 and 30 kg instead of 20 kg as a minimum catch for selecting records for each species, (b) using month instead of quarter, (c) using 6 depth classes (140 m) instead of 3. In each case, the resulting annual trends for each species were nearly identical to those that used the base assumptions.

6.2.5 Log-linear model (LLM)

Multiplicative models have been usefully applied to a number of fisheries to standardise catch-rates (e.g. Robson 1966, Kimura 1981). Kimura (1988) proposed a standard loglinear model framework to simultaneously analyse relative fishery abundance indices available from different survey techniques over different areas and years.

93

A LLM to describe the components of trawl catch-rate examined here is as follows:

log(r) = µ + Y + V + Q + A + D + M + T + ε where r µ Y V Q A D M T ε

= catch-rate in kg/hr = intercept = year factor = vessel factor = quarter factor = area factor = depth factor = moon phase factor = time of trawl factor = error term

The model described above with all terms will be referred to as the full model. In this model, quantification of the year factor provides a standardised index of annual abundance.

One main purpose of this study was to quantify the direct effect of single factors (quarter, area, depth, moon phase and time of day) on the catch rate of each species. It is possible to also examine interactions among these factors. Models were examined that included interactions that were judged to be plausible and potentially important – area*quarter, depth*quarter, moon*day and depth*day. Results for each species for models that included those interactions did not show greatly improved explanatory power in terms of explained variance – flathead +2%, morwong +0%, redfish +2%, leatherjacket +4%, latchet +3%. As there was not strong evidence that the inclusion of interactions was important, and for simplicity in interpretation and presentation of results, interactions were not examined further.

6.2.6 Standardised annual abundance index correlation with available long-term environmental data

Six different annual environmental data series were used to determine whether there was an obvious correlation with annual standardised abundance indices from the LLM. Annual indices for sunspots, southern oscillation index and the latitude of the subtropical ridge of high pressure were obtained from Thresher (2002, personal

94

communication). A sunspot index was included for examination because long-term data are available, and sunspot-correlated variability in the region is likely to have substantial effects on the climate and ecology of the region (Thresher 2002). The latitude of the sub-tropical ridge is correlated with variation in Southern Hemisphere zonal westerly winds that have affected the climate of the region, also having flow-on effects to coastal ecology (Harris et al. 1988, Thresher 1994).

Standardised annual rainfall indices for Jervis Bay and Gabo Island as well as Australian mean temperature were obtained from data published on the Internet by the Australian Bureau of Meteorology. These two sites were chosen because they are widely spaced at the centre and south of the region, have long-term records and are not located near major cities where the climate may be influenced by urban development. Parametric coefficient of determination r2 values were used to examine the correlation of environmental indices with standardised indices of abundance produced by the LLM. It is possible that environmental changes have a delayed effect on observed abundance as indicated by catch-rates of commercial size fish – for example, where effects are on survival of eggs, larvae or young fish. To examine such delays each environmental index was also offset by an additional 3 years (as this is the estimated age at recruitment to the fishery for flathead, leatherjacket and latchet – see Chapter 7).

95

6.3 Results

6.3.1 LLM results for all retained catch

All factors examined were found to have highly significant (P