BIOL3305 Fisheries Science: Foundation and Application PDF

Basics in fisheries biology School of Biological Sciences Dirk Zeller University of Western Australia Sea Around Us – Indian Ocean www.seaaroundus-io.org [email protected] ...

Basics in fisheries biology School of Biological Sciences Dirk Zeller University of Western Australia Sea Around Us – Indian Ocean www.seaaroundus-io.org [email protected]  Russell’s axiom* * Not the math version of Russell’s axiom (reducibility) Simple conceptual model of fisheries population dynamics first formalized by Edward S. Russell (1931) Biomass/abundance of a fish population usually changes from year to year, and this axiom conceptualizes the main drivers/factors accounting for such changes in a fisheries context Represents the dynamics of self-contained breeding populations, or stocks in fisheries parlance… basic unit in fisheries science Stock size can be measured in numbers (i.e., abundance) but usually in biomass (weight in tonnes)… “tonnes” versus “tons” versus “metric tons” Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Russell ES (1931) Journal du Conseil International pour l'Exploration de la Mer 6(1): 3-20  Russell’s axiom Some species may be a single stock… e.g., southern bluefin tuna (Thunnus maaccoyi), as apparently only one spawning ground Barramundi (Lates calcarifer) likely at least 4 separate stocks across northern Australia Common coral trout (Plectropomus leopardus) on the GBR may be one single stock due to substantial larval disbursal option ensuring genetic mixing, despite separate spawning aggregations on each reef Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Russell ES (1931) Journal du Conseil International pour l'Exploration de la Mer 6(1): 3-20  Russell’s axiom Parameters Stock: self-contained breeding population. Sometimes more arbitrarily defined, e.g., by ecosystem division or even statistical accounting area Recruitment: new fish entering the fishable stock Difference in definition between ecology and fisheries science Growth: increase in body weight (and length) of individual fish = somatic growth Natural mortality: death of fish by natural causes, e.g., predation, disease Catch: the part of the stock taken by fisheries (sometimes also called ‘yield’) Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Russell ES (1931) Journal du Conseil International pour l'Exploration de la Mer 6(1): 3-20  Russell’s axiom Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Some core assumptions No immigration or emigration into or out of the stock No serious degradation or loss of habitat, or other changes to environment or ecosystem for all stages of life cycle (including larval and nursery habitats)… static, equilibrium nature In general, the job of most fisheries biologists is to measure or estimate: Stock size Recruitment Growth Natural mortality Catch And then use this knowledge to provide advice and recommendations to fisheries managers on a sustainable level of Catch next year 5/27  Russell’s axiom Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Is such a simple formula useful in our high-tech world of quantitative computer modeling, differential equations and advanced statistics? But it is at the very heart of almost all of the most complex population dynamics models and mathematical calculations in fisheries science today. Why? Because it captures all of the essential features of fish population dynamics clearly and eloquently… even if simplified Use this equation in a simple simulation to demonstrate the basic tenet of fisheries science… Namely that fishing, by reducing stock sizes from the original unfished stock level, can actually generate ‘new’ fish population growth (i.e., stock productivity) that can then be caught sustainably by holding the stock at a level below that of the unfished population. Let’s see via a simple fishery simulation model  Russell’s axiom Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) S2 = S1 + (R + G) – (M + C) Assumptions: 1. Growth G, recruitment R and natural mortality M depend on stock size a) Growth G increases as stock size decreases (individual fish grow faster at lower density, e.g., less competition) b) Natural mortality M decreases as stock size decreases (fish die of natural causes at a lower rate at lower density, e.g., less cannibalism) c) Recruitment R initially increases, then decreases as stock size decreases (less cannibalism of new recruits as stock size declines initially from unfished levels, but once stock is thinned out fish have trouble findings mates and producing potential recruits) 3. Start: unfished stock with biomass of 10,000 t  Russell’s axiom S2 = S1 + (R + G) – (M + C) Growth, recruitment and natural mortality as % of stock size with unfished stock of 10,000 t Stock size (tonnes) G increases as stock size decreases (individual fish grow faster at lower density, e.g., less competition)  Russell’s axiom S2 = S1 + (R + G) – (M + C) Growth, recruitment and natural mortality as % of stock size with unfished stock of 10,000 t Stock size (tonnes) M decreases as stock size decreases (fish die at a lower rate at lower density, e.g., less cannibalism)  Russell’s axiom S2 = S1 + (R + G) – (M + C) Growth, recruitment and natural mortality as % of stock size with unfished stock of 10,000 t Stock size (tonnes) R initially increases, then decreases as stock size decreases (less cannibalism of new recruits as stock size declines initially from unfished levels, but once stock is thinned out fish have trouble findings mates and producing potential recruits)…. Density dependence of recruitment 10/27  Russell’s axiom S2 = S1 + (R + G) – (M + C) Assumptions (cont.): 4. We simulate 100 years of fishing, i.e., 10 years of fishing at each of 10 stock sizes, from unfished (10,000 t) to 1000 t (i.e., to 10%, heavily depleted) in steps of 1000 t of stock 5. Growth G, recruitment R and natural mortality M vary randomly within given limits a) 10% of annual G, 20% of annual M, 40% of annual R 6. Thus, no 2 years of fishing at any given stock sizes are ever the same, but vary around a given mean productivity of that stock size 7. Let’s solve this for ‘catch’ via the other 4 parameters  Russell’s axiom Simulation results: Parabolic relationship between stock size and sustainable catch Thus, if we reduce stock size from unfished levels, stock productivity (and thus sustainable catch if we only take that extra productivity generated by fishing) at first increases to a maximum, but thereafter decreases to low levels if we fish the stock too low “extra productivity generated by fishing” = surplus production Sustainable catch (tonnes) Stock size (tonnes)  Another way of looking at this Much of modern fisheries science based on surplus production A few terms: Nt ≈ Bt = stock size at time t K = unfished biomass = carrying capacity = B0 r = population growth rate… don’t mix up with R or G Logistic population growth curve Interacts with predators/prey/ecosystem: Accounts for M, competition, fecundity, individual growth etc. Schaefer (1954) Bulletin of the Inter-American Tropical Tuna Commission 1: 27-56 Reprinted as: Schaefer MB (1991) Bulletin of Mathematical Biology 53(1): 253-279  Another way of looking at this Much of modern fisheries science based on surplus production A few terms: Nt ≈ Bt = stock size at time t K = unfished biomass = carrying capacity = B0 (“virgin” biomass) r = population growth rate… don’t mix up with R Logistic population growth curve Maximum new biomass production at ~ ½ K (or ½ B0) Schaefer (1954) Bulletin of the Inter-American Tropical Tuna Commission 1: 27-56 Reprinted as: Schaefer MB (1991) Bulletin of Mathematical Biology 53(1): 253-279  Another way of looking at this Much of modern fisheries science based on surplus production A few terms: Nt ≈ Bt = stock size at time t K = unfished biomass = carrying capacity = B0 r = population growth rate… don’t mix up with R Logistic population growth curve Keeping Bt at ~ ½ K (½ B0) allows maximum surplus production (new biomass) to be available to fisheries Schaefer (1954) Bulletin of the Inter-American Tropical Tuna Commission 1: 27-56 15/27 Reprinted as: Schaefer MB (1991) Bulletin of Mathematical Biology 53(1): 253-279  Keeping Bt at ~ ½ K (½ B0) allows maximum surplus production (new biomass) to be available to fisheries We will deal further with surplus production and the ½ B0 issue in the lecture on basics of stock assessment  Russell’s axiom last point Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Ignore ‘catch’, i.e., non-exploited population Stock size next year = stock size this year + (recruitment + growth) – (natural mortality) or Stock size = (R + G) - M A species/stock with high natural mortality rate (the ‘-’ item reducing stock size) must have high recruitment and/or growth rates to maintain stock size…. Therefore such a stock must be more ‘productive’ than A species/stock with low natural mortality rate, which can have lower recruitment and/or growth rates to maintain stock size….  Russell’s axiom last point Stock size next year = stock size this year + (recruitment + growth) – (natural mortality + catch) Ignore ‘catch’, i.e., non-exploited population Stock size next year = stock size this year + (recruitment + growth) – (natural mortality) or Stock size = (R + G) - M A species/stock with high natural mortality rate (the ‘-’ item reducing stock size) must have high recruitment and/or growth rates to maintain stock size…. Think mice Therefore such a stock must be more ‘productive’ than A species/stock with low natural mortality rate, which can have lower recruitment and/or growth rates to maintain stock size…. Think elephants  Data How do we measure/estimate the 5 parts of this relationship? 1. Stock size 2. Growth 3. Recruitment 4. Natural mortality 5. Catch (fishing mortality)  Methods for collecting the needed data 1. Research surveys a) Using same type of gear as the fishery, but applied in a systematic, carefully measured and designed sampling strategy b) Measure number, size (length), weight, taxonomic composition, age c) Expensive, thus limited sampling compared to actual fishery d) How to correlate this to full stock can be a challenge 2. Fisheries data (‘industry’ data) a) Catch & fishing effort b) Logbooks, landing site (harbor) monitors, boat ramp surveys, onboard observes, electronic observers etc. c) Accuracy and reliability issues… human and fishers’ behaviour important d) Where, how, when, why and how hard fishers fish e) Works best if strong buy-in/support from fishing community f) Catch rates, i.e., Catch per Unit of Effort can be a good indicator of relative abundance changes over time… but not always 20/27  Methods for collecting the needed data 1. Research surveys a) Using same type of gear as the fishery, but applied in a systematic, carefully measured and designed sampling strategy b) Measure number, size (length), weight, taxonomic composition, age c) Expensive, thus limited sampling compared to actual fishery d) How to correlate this to full stock can be a challenge 2. Fisheries data (‘industry’ data) a) Catch & fishing effort b) Logbooks, landing site (harbor) monitors, boat ramp surveys, onboard observes, electronic observers etc. c) Accuracy and reliability issues… human and fishers’ behaviour important d) Where, how, when, why and how hard fishers fish e) Works best if strong buy-in/support from fishing community f) Catch rates, i.e., Catch per Unit of Effort can be a good indicator of relative abundance changes over time… but not always 20/27  Age Age of a fish is one of the most valuable/important data items you can have… why? Can measure how heavy and long fish in each age class are Growth rates by age class Number of fish per age class…. Mortality rates (chances of surviving to age class x?) Recruitment rates (how many fish enter the age 1 year class each year?) Maximum age Gives clues on stock productivity Tropical anchovy… 1 year Orange roughy… 100 years Think mice versus elephants…  Age structure of humans Shape of frequency distribution gives insights… Mortality rates… probability of reaching age 60 higher in developed countries Recruitment rates… greater proportion in youngest age class in developing countries  Age Shape of frequency distribution gives insights… Truncation of time series Fewer year classes in stock  Age Age of a fish is one of the most valuable data items you can have… gold standard But also difficult, time consuming and very expensive to obtain…. Common in developed countries, important species only Rare, limited or absent in developing countries Alternative non-age methods are more widespread in developing countries, such as catch and/or length-based methods… See sources below and references therein Froese et al. (2018) ICES Journal of Marine Science 75(6): 2004-2015 Froese et al. (2023) Acta Ichthyologica et Piscatoria 53: 173-189 25/27 Taylor and Mildenberger (2017) Fisheries Management and Ecology 24(4): 330-338  Methods for collecting the needed data 1. Research surveys 2. Fisheries data (‘industry’ data) Example where this went wrong (Lab) 25/27  Summary Introduction into fisheries biology and foundations of population dynamics Always think carefully about what your data and analyses represent Data uncertainty (fish versus trees in forest) But don’t hide behind uncertainty Be aware of, understand, address and account for assumptions Sea Around Us – Indian Ocean