Abalone spatial- and age-structured assessment model and projections for Zones A, B, C and D

FISHERIES/215/AUG/SWG-AB/15 Abalone spatial- and age-structured assessment model and projections for Zones A, B, C and D A. Brandão and D.S. Butterworth Marine Resource Assessment & Management Group (MARAM) Department of Mathematics and Applied Mathematics University of Cape Town Rondebosch 771, Cape Town August 215 Abstract The 213 assessment of abalone in Zones A-D is updated to take new data into account; these include results from the recent full population survey. Projections are shown for different scenarios for the future commercial and poaching catches in Zone A and Zone B. Current poaching levels (average of 214 and 215, estimated as 378t for Zone A, 293t for Zone B, 67t for Zone C and 139t for Zone D) if continued, would not be sustainable. Introduction This document provides results from fitting the spatial- and age-structured production model (ASPM) for abalone for Zones/Subareas A, B, CNP, CP and D in combination, using the new data that have become available since the previous assessment (Brandão and Butterworth, 213). These new data include those from a repetition in 215 of the full population survey carried out in 22 to provide information on the recruitment for Zones B and C. The extra year of catch-at-age data has been included in the assessment model. As there are now two indices of abundance available from this full population survey, the assessment model has been modified to include these indices in the model fitting procedure. Data The series that have been updated, compared to those used in Brandão and Butterworth (213), for the analyses that follow are (note that throughout this document the convention is that, for example, the year 28 refers to the Model-year running from October 27 to September 28): CPUE: new values from updated GLMM standardisation for Zones A and B only to include data until 214 (Brandão and Butterworth, 215a). Commercial catches for Zones A and B until 214 (TAC assumed taken in 215). 1

FISHERIES/215/AUG/SWG-AB/15 Commercial catch-at-age data: data provided for Zones A and B to include data until 214 (Maharaj and Mackenzie, 215a). Poaching trend: new values from updated analyses of policing effort and the number of confiscations for 28 to 215 (Brandão and Butterworth, 215b). Poaching catch-at-age data: Zones A to D (until 214). FIAS abundance indices for 213 in Zone C and 214 for Zones B and D (Maharaj and Mackenzie, 215b). FIAS catch-at-age data for 213 in Zone C and 214 for Zones B and D (Maharaj and Mackenzie, 215b). Extractive survey indices for 22 and 215 (Maharaj and Mackenzie, 215c). Extractive survey catch-at-age data for 215. Methodological Changes The full details of the spatial- and age-structured production model used for assessing abalone are provided in Brandão and Butterworth (29) as well as in Plagányi and Butterworth (21). The Basecase model described in those two documents has been modified in some generally minor ways that are described in Brandão and Butterworth (211). Brandão and Butterworth (213) describes how the standard errors associated with the poaching indices are incorporated in the assessment model. In the assessment presented in this paper, the indices of abundance obtained from the full population surveys carried out in 22 and 215 are now also incorporated into the model fitting procedure. Results Results have been obtained for the new Basecase model for the updated data as well as for the case when an Allee effect is taken into account. Results for the new Basecase model are reported in Tables 1 and 2 for some key statistics. The model fits to various data are given in the Appendix A. Spawning biomass with projections for all Zones are shown in Figure 1 and annual poaching estimates (by number) for Zones A and B in Figure 2. These Tables and Figures include comparisons with the results of the previous assessment of Brandão and Butterworth (213), referenced as Previous. Figure 3 plots the comparison of the depletion projections for new Basecase model and the model that takes an Allee effect into account. Projections Figure 4 shows spawning biomass projections for the new Basecase model for four scenarios for future commercial and poaching catches listed below. Figure 5 shows these projections when an Allee effect is taken into account. Poaching only (average of 214 and 215 levels) 25 ton commercial catch only for each of Zone A and B 2

FISHERIES/215/AUG/SWG-AB/15 Both poaching and commercial catches at the above levels The poaching reduction necessary to keep the biomass at its current level. Figure 6 shows the intended future poaching levels, which are assumed to remain at the current estimated level (average of 214 and 215), and the actual removals made by the model because of the model restriction that does not allow the fully selected fishing proportion to be greater than 95%. Thus the model builds in a factor to allow for the fact that as abundance declines, it would not be possible to sustain current poaching removals. Discussion The new Basecase results are similar to those from the previous assessment conducted in 213 (Previous). Results show that the current estimates of the extent by which the abalone resource depletion in Zones A and B in 213 are slightly less than was the case in the previous assessment, but also that the downward trend in abundance in these Zones as well as in Zones C and D has continued from 213 to 215 (Table 1 and Figure 1). These downward trends are projected to continue into the future under current poaching levels. There has been an estimated increase in poaching levels from the 213 season for both Zones A and B (Figure 2), though the estimate for the level of poaching in 213 is now lower than estimated in the previous assessment. The current level of poaching (the average of 214 and 215, is estimated as 378t for Zone A, 293t for Zone B, 67t for Zone C and 139t for Zone D) (Table 2). Future trends are unsurprisingly more pessimistic under the Allee effect, as might be expected (Figure 3). Figures 4 and 5 show that the current level of poaching is NOT sustainable if maintained in the future. The fits to the age structure from the full population surveys (Figure A.1) are not satisfactory, particularly for Zone C. This may in part reflect inaccuracies in the growth curve assumed for low ages, but also points to a need to re-examine the way the recruitment decrease arising from the lobster-urchin effect in Zone C is modelled. Nevertheless indications from Figure A.1 are that for both Zones B and C, recent recruitments are lower than the model is assuming. 3

FISHERIES/215/AUG/SWG-AB/15 References Brandão, A. and Butterworth, D.S. 29. Results for the Reference-case abalone spatial- and agestructured assessment model for Zones A, B, C and in 29. Marine and Coastal Management document: MCM/29/OCT/SWG-AB/8. Brandão, A. and Butterworth, D.S. 211. Results for the abalone spatial- and age-structured assessment model for Zones A, B, C and D in 211. FISHERIES/211/AUG/SWG-AB/11. Brandão, A. and Butterworth, D.S. 213. Final results for the abalone spatial- and age-structured assessment model for Zones A, B, C and D in 213. FISHERIES/213/AUG/SWG-AB/12. Brandão, A. and Butterworth, D.S. 215a. Updated GLMM standardisation of the commercial abalone CPUE for Zones A-D over the period 198-214. FISHERIES/215/AUG/SWG-AB/2. Brandão, A. and Butterworth, D.S. 215b. Updated trends in policing effort and the number of confiscations for abalone including compliance data until March 215. FISHERIES/215/AUG/SWG-AB/8. Maharaj, G. and Mackenzie, A. 215a. Summary of commercial data updates for the abalone fishery in Zones A-D in 215. FISHERIES/215/AUG/SWG-AB/1. Maharaj, G. and Mackenzie, A. 215b. Progress with inshore FIAS density and catch-at-age data for Zones A to D. FISHERIES/215/AUG/SWG-AB/3. Maharaj, G. and Mackenzie, A. 215c. Repeat joint industry-daff full population extractive survey of the abalone resource in 215. FISHERIES/215/AUG/SWG-AB/4. Plagányi, É.E. and Butterworth, D.S. 21. A spatial- and age-structured assessment model to estimate the impact of illegal fishing and ecosystem change on the South African abalone Haliotis midae resource. African Journal of Marine Science, 32(2):27-236. 4

FISHERIES/215/AUG/SWG-AB/15 Table 1. Best fit estimates for the pre-exploitation spawning biomass sp B, current depletion and the depletion at the end of the projection period for the new Basecase. Projections assume future poaching levels at their current estimated values (average of 214 and 215). For comparison, results for the previous Basecase ( Previous ) from the assessment of Brandão and Butterworth (213) are also given. sp sp sp B y sp sp B B By B Model A B CNP CP D y A B CNP CP D y A B CNP CP D Previous 759 5911 2633 4686 7313 213.262.239.88.5.195 233.121.136.25.9.82 New Basecase 7813 63 2548 4734 7811 213.271.256.17.7.21 233.67.4.32.13.57 215.227.225.88.52.19 235.58.3.29.11.52 5

FISHERIES/215/AUG/SWG-AB/15 Table 2. Estimates of the current (215) poaching levels (in terms of biomass), the average of the last five years of the proportion of confiscations to estimated poaching numbers and the minimum values of the negative of the log-likelihood function (-ln L) for the new Basecase. For comparison, results for the Previous 213 assessment are also given (the poaching estimates given for that assessment are those estimated at that time for 213). Note that all contributions from catch-at-age data to -ln L have been multiplied by.1 as an ad hoc adjustment to compensate for likely positive correlation in these data. The log-likelihood values are not comparable (because the data fitted previously differ from the current new Basecase) and are therefore shown in square brackets. Average proportion of confiscation to Poaching MT -ln L poaching over the last 5 years Model A B CNP CP D A B C D A B CNP CP D Total Previous (for 213) New Basecase (for 215) 42.1 291. 48.1 72. 117.6 18.3% 32.4% 5.6% 6.8% [-78.1-81.3-56.2-5.2-55.3-317.5] 395.8 288.8 44.4 73.5 132.2 19.5% 35.% 5.7% 3.7% -85.4-86.2-56.2-48.2-51.5-322.8 6

FISHERIES/215/AUG/SWG-AB/ Zone A Zone B 1 Previous Basecase 65 Previous Basecase 9 8 55 7 45 6 35 5 4 25 Spawning biomass (MT) (inshore + offshore) 3 2 1 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 Sub-area CNP 3 Previous Basecase 25 2 15 5 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235-5 Sub-area CP 5 Previous Basecase 45 4 35 3 8 7 6 5 Zone D Previous Basecase 15 25 4 1 5 2 15 1 5 3 2 1 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 Figure 1. Total (inshore + offshore) spawning biomass trajectories shown for Zones A to D for the new Basecase model compared to the Previous results obtained in the 213 assessment. Note that the 2-yr projections shown (after the vertical bar) represent scenarios under which future poaching levels are assumed to remain at the current estimated level (average of 212 and 213 for the Previous model and the average of 214 and 215 for the Basecase ) and future commercial catches are set to 5 tons for the Previous model and 25 tons for the Basecase s in each of Zones A and B only. 7

198 1982 1984 1986 1988 199 1992 1994 1996 1998 2 22 24 26 28 21 212 214 198 1982 1984 1986 1988 199 1992 1994 1996 1998 2 22 24 26 28 21 212 214 Numbers poached Numbers poached FISHERIES/215/AUG/SWG-AB/ Zone A Zone B 25 2 Previous 15 Basecase 1 5 16 14 12 1 8 6 4 2 Previous Basecase Figure 2. Comparison of model-predicted numbers of abalone poached for Zones A and B for the new Basecase model and those obtained in the 213 assessment ( Previous ). 8

FISHERIES/215/AUG/SWG-AB/ a) Zone A b) Zone B 1 New Basecase Allee 1 New Basecase Allee.9.9.8.8.7.7 Spawning biomass (MT) deepletion (inshore + offshore).6.5.4.3.2.1 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 c) Zone CNP 1 New Basecase Allee.9.8.7.6.5.4.6.5.4.3.2.1 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 d) Zone CP 1 New Basecase Allee.9.8.7.6.5.4 1.9.8.7.6.5.4 e) Zone D New Basecase Allee.3.3.3.2.2.2.1.1.1 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 Figure 3. Total (inshore + offshore) depletion trajectories shown for Zones A to D for the new Basecase model compared to sensitivity test 3 (Allee effect). Note that the 2-yr projections shown (after the vertical bar) represent scenarios under which future poaching levels are assumed to remain at the current estimated level (average of 214 and 215) and future commercial catches are set to 25 tons in each of Zones A and B only. 9

FISHERIES/215/AUG/SWG-AB/ Spawning biomass (MT) (inshore + offshore) 12 1 8 6 4 2 35 3 25 2 15 a) Zone A Poaching only TAC only Poaching+TAC 58% poaching reduction 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 a) Zone A Poaching only TAC only Poaching+TAC 58% poaching reduction 1 28 21 212 214 216 218 7 6 5 4 3 2 1 2 18 16 14 12 1 b) Zone B Poaching only TAC only Poaching+TAC 49% poaching reduction 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 8 b) Zone B Poaching only TAC only Poaching+TAC 49% poaching reduction 6 28 21 212 214 216 218 Figure 4. Total (inshore + offshore) spawning biomass trajectories shown for Zones A and B for the new Basecase model. The 2-yr projections shown (after the vertical bar) represent four different scenarios for future commercial and poaching catches. Unless a zero amount is assigned, future poaching levels are assumed to remain at the current estimated level (average of 214 and 215) and future commercial catches in each of these two Zones are set to the current TAC of 25 tons. The bottom plots zoom in on a shorter period to be able to distinguish the curves more clearly. In each plot, the required reduction in poaching necessary to keep the resource stable at its present level under the current TAC is also shown, with the required reduction indicated in the legend. 1

FISHERIES/215/AUG/SWG-AB/ Spawning biomass (MT) (inshore + offshore) 12 1 8 6 4 2 a) Zone A Poaching only TAC only Poaching+TAC 81% poaching reduction 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 7 6 5 4 3 2 1 b) Zone B Poaching only TAC only Poaching+TAC 98% poaching reduction 1951 1957 1963 1969 1975 1981 1987 1993 1999 25 211 217 223 229 235 Figure 5. Total (inshore + offshore) spawning biomass trajectories shown for Zones A and B for the new Basecase model taking an Allee effect into account. The 2- yr projections shown (after the vertical bar) represent four different scenarios for future commercial and poaching catches. Unless a zero amount is assigned, future poaching levels are assumed to remain at the current estimated level (average of 214 and 215) and future commercial catches are set to the current TAC of 25 tons. In each plot, the required reduction in poaching necessary to keep the resource stable at its present level under the current TAC is also shown, with the required reduction indicated in the legend. 11

Numbers FISHERIES/215/AUG/SWG-AB/ 2 18 16 14 12 1 8 6 4 2 a) Zone A Future poaching Poaching removals 216 221 226 231 2 18 16 14 12 1 8 6 4 2 b) Zone B Future poaching Poaching removals 216 221 226 231 4 a) Zone A Future poaching Poaching removals 35 b) Zone B Future poaching Poaching removals 35 3 3 25 2 15 1 25 2 15 1 5 5 216 221 226 231 216 221 226 231 Figure 6. Future poaching levels (in terms of numbers (top) and biomass (bottom)), as assumed to remain at the current estimated level (average of 214 and 215), and the actual removals made by the model because of the model restriction that does not allow the fully selected fishing proportion to be greater than 95%. Thus the model builds in a factor to allow for the fact that as abundance declines, it would not be possible to sustain current poaching removals. Results shown are for the new Basecase. 12

FISHERIES/215/AUG/SWG-AB/ Appendix A Diagnostic plots and others for the ASPM model for Zones A to D Legal and illegal catches as well as the commercial exploitable biomass are shown in Figure A1. Diagnostic plots for CPUE for Zones A and B are shown in Figure A2, selectivity functions for the commercial, recreational, poaching, FIAS, old surveys and full population surveys in Figure A3, FIAS data in Figure A4, and full population survey indices in Figure A5. The fits to the CPUE data for Zones A and B (Figure A2), fits to the FIAS data (Figure A3) and fits to the full population survey indices (Figure A4) are reasonable. Fits to the catch-at-age proportions are shown in Figures A6 to A11 for the various components of available data. The fits to the catch-at-age proportions are good with the exception of the full population survey (Figure A1). For Zone B the high proportion of 4 year olds is not fitted well and a higher proportion of 2 year olds is estimated for 215 than observed. For Zone C a higher proportion of younger and lower proportion of older abalone are estimated than observed. 13

Catch (t) Catch (t) FISHERIES/215/AUG/SWG-AB/ a) Zone A Illegal Recreational Commercial Exploitable biomass b) Zone B Illegal Recreational Commercial Exploitable biomass 12 6 8 5 1 8 6 4 2 5 4 3 2 1 7 6 5 4 3 2 1 45 4 35 3 25 2 15 1 5 Exploitable biomss (t) 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 21 26 211 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 21 26 211 c) Zone C Illegal Recreational Commercial Exploitable biomass d) Zone D Illegal Recreational Commercial Exploitable biomass 9 6 8 6 8 7 6 5 4 3 2 1 5 4 3 2 1 7 6 5 4 3 2 1 5 4 3 2 1 Exploitable biomss (t) 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 21 26 211 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 21 26 211 Figure A1. Estimated commercial exploitable biomass for the new Basecase model for Zones A to D (right hand axis) in tons and total catches (commercial + recreational + estimated illegal) for the Zones (left hand axis). 14

Selectivity Selectivity FISHERIES/215/AUG/SWG-AB/ 1.4 1.2 1.8.6.4.2 1 2 3 4 5 6 7 8 9 1 11 12 13 14 15 Age (yrs) comm rec poa 1.2 1.8.6.4.2 1 2 3 4 5 6 7 8 9 1 11 12 13 14 15 Age (yrs) fias old inds Figure A2. Plots of the new Basecase model selectivity functions estimated for the commercial (comm), recreational (rec) and poaching (poa) fishery sectors, and for FIAS (fias) and the old 198 s surveys (old). A uniform value is assumed for the full population surveys (inds) because of the extractive nature of the sampling methodology used. 15

CPUE CPUE FISHERIES/215/AUG/SWG-AB/ Zone A 1.4 1.2 1.8.6.4 A (obs).2 A (pred) 198 199 2 21 Zone B 1.8 1.6 1.4 1.2 1.8.6.4 B (obs).2 B (pred) 198 199 2 21 Figure A3. Comparisons between the standardised CPUE (obs) and model-predicted CPUE values for the new Basecase for Zones A and B. 16

Index of abunddance Index of abunddance Index of abunddance Index of abunddance Index of abunddance FISHERIES/215/AUG/SWG-AB/ Zone A Zone B Zone CNP 7 A (obs) 6 A (pred) 5 4 3 2 1 1994 1997 2 23 26 29 212 215 12 B (obs) 1 B (pred) 8 6 4 2 1994 1997 2 23 26 29 212 215 5 45 CNP (obs) 4 CNP (pred) 35 3 25 2 15 1 5 1994 1997 2 23 26 29 212 215 Zone CP Zone D 5 4 CP (obs) CP (pred) 6 5 4 D (obs) D (pred) 3 3 2 2 1 1 1994 1997 2 23 26 29 212 215 1994 1997 2 23 26 29 212 215 Figure A4. Comparison of observed FIAS and model-predicted trends for the new Basecase model for Zones A to D. Note that the 95% confidence intervals shown have been computed as: estimate*exp(±1.96*cv). 17

Index of abunddance Index of abunddance FISHERIES/215/AUG/SWG-AB/ Zone B Zone C 45 4 35 3 25 2 15 1 5 B (obs) B (pred) 8 7 6 5 4 3 2 1 C (obs) C (pred) 21 24 27 21 213 216 21 24 27 21 213 216 Figure A5. Comparison of observed extractive survey and model-predicted values and trends for the new Basecase model for Zones B and C. Note that the 95% confidence intervals shown have been computed as: estimate*exp(±1.96*cv). 18

FISHERIES/215/AUG/SWG-AB/.35.3.25.2.15.1.5 Zone A.3.25.2.15.1.5 Zone B 9 1 11 12 13 14 15 9 1 11 12 13 14 15.4.35.3.25.2.15.1.5 Subaerea CNP 9 1 11 12 13 14 15.35.3.25.2.15.1.5 Subarea CP 9 1 11 12 13 14 15.45.4.35.3.25.2.15.1.5 Zone D 9 1 11 12 13 14 15 Figure A6. Comparison for the average over all the years between observed and model predicted catch-at-age proportions for the commercial sector for Zones A to D for the new Basecase model. 19

FISHERIES/215/AUG/SWG-AB/.3.25 Zone A.3.25 Zone B.2.2.15.15.1.1.5.5 9 1 11 12 13 14 15 9 1 11 12 13 14 15.3.25.2.15.1.5 Zone C.25.2.15.1.5 Zone D 9 1 11 12 13 14 15 9 1 11 12 13 14 15 Figure A7. Comparison for the average over all the years between observed and model predicted catch-at-age proportions for the recreational sector for Zones A to D for the new Basecase model..35.3.25.2.15.1.5 Zone A.3.25.2.15.1.5 Zone B 5 6 7 8 9 1 11 12 5 6 7 8 9 1 11 12.3.25.2.15.1.5 Zone C 5 6 7 8 9 1 11 12.45.4.35.3.25.2.15.1.5 Zone D 5 6 7 8 9 1 11 12 Figure A8. Comparison for the average over all the years between observed and model predicted catch-at-age proportions for the poaching sector for Zones A to D for the new Basecase model. 2

FISHERIES/215/AUG/SWG-AB/.35.3.25.2 Zone A.25.2.15 Zone B.15.1.5.1.5 5 6 7 8 9 1 11 12 5 6 7 8 9 1 11 12.25.2.15.1.5 Subarea CNP.3.25.2.15.1.5 Subarea CP 5 6 7 8 9 1 11 12 5 6 7 8 9 1 11 12.25.2.15.1.5 Zone D 5 6 7 8 9 1 11 12 Figure A9. Comparison for the average over all the years between observed and model predicted catch-at-age proportions for the FIAS inshore survey data for Zones A to D for the new Basecase model. 21

Proportion Proportion Proportion Proportion FISHERIES/215/AUG/SWG-AB/ Zone B - 22 Zone C - 22.35.8.3.25.2.15.1.5.7.6.5.4.3.2.1 2 3 4 5 6 7 8 9 1 11 12 2 3 4 5 6 7 8 9 1 11 12 Zone B - 215 Zone C - 215.6.8.5.4.7.6.5.3.4.2.1.3.2.1 2 3 4 5 6 7 8 9 1 11 12 2 3 4 5 6 7 8 9 1 11 12 Figure A1. Comparison between observed and model predicted catch-at-age proportions for the full population survey data for Zones B and C for the new Basecase model. 22

Offshore proportion Offshore proportion Offshore proportion Inshore proportion Inshore proportion Inshore proportion Offshore proportion Offshore proportion Inshore proportion Inshore proportion FISHERIES/215/AUG/SWG-AB/ Zone A - 1986 Zone A - 1987.25.2.15.1.5 2 3 4 5 6 7 8 9 1 11 12.2.18.16.14.12.1.8.6.4.2 2 3 4 5 6 7 8 9 1 11 12.35.35.3.3.25.25.2.2.15.1 7.15.1.5.5 3 4 5 6 7 8 9 1 11 12 13 14 15 3 4 5 6 7 8 9 1 11 12 13 14 15 Zone B - 1982 Zone CP - 1981 Zone D - 1983.2.18.16.14.12.1.8.6.4.2 2 3 4 5 6 7 8 9 1 11 12.2.18.16.14.12.1.8.6.4.2 2 3 4 5 6 7 8 9 1 11 12.18.16.14.12.1.8.6.4.2 2 3 4 5 6 7 8 9 1 11 12.2.2.35.15.15.3.25.1.1.2.15.5.5.1.5 3 4 5 6 7 8 9 1 11 12 13 14 15 3 4 5 6 7 8 9 1 11 12 13 14 15 3 4 5 6 7 8 9 1 11 12 13 14 15 Figure A11. Comparison between observed and model predicted catch-at-age proportions for the old inshore and offshore survey data for Zones A to D for the new Basecase model. 23