Fluctuation in stock properties of north‐east Arctic cod related to long‐term environmental changes

Hjort (1914. Fluctuations in the great fisheries of northern Europe. Rapport et Procès-Verbaux des Réunions du Conseil Permanent International pour l'exploration de la Mer, XX: 1–228) identified two important aspects of the early life of fish as being important determinants of fluctuations in year-class strength: changes in nutrition and transport. He dismissed a third possible influence, changes in the abundance of the reproductive stock. Here, we describe how a recently discovered characteristic behaviour of age-structured populations termed cohort resonance, which does involve changes in adult abundance, can have a substantial effect on fluctuations in fished populations. Cohort resonance involves selectively greater sensitivity of age-structured populations to generational frequencies and to very low frequencies in the environmental signal influencing a population. This frequency-dependent selectivity has been shown to increase with fishing, as do the total amounts of variability in recruitment, egg production, and catch. Cohort resonance differs from other recent model mechanisms proposed to explain the observed increase in variability with fishing in that it does not require over-compensatory density-dependence. It stems from the compensatory ascending limb of the egg–recruit relationship, and is a characteristic of a stable population driven by a random environment. We demonstrate the differences in frequency selectivity and increases in variability with fishing among three different Pacific coast species with different longevity: coho salmon (Oncorhynchus kisutch; ∼3 years), Pacific hake (Merluccius productus; ∼25 years), and Pacific Ocean perch (Sebastes alutus; ∼90 years). The shortest lived, coho salmon is the most sensitive to environmental variability, but variability in egg production and catch both increase more rapidly with fishing in the longer-lived species. Understanding cohort resonance will aid in anticipation of predicted potential changes in the frequency content of the physical environment with changing climate (e.g. more frequent El Niños), and it provides a warning regarding the possible confounding of increasing sensitivity to slow change due to fishing with actual slow change of population parameters due to climate change. Our understanding of the role of cohort resonance in population variability will be enhanced by further identification of empirical examples. We describe some of the challenges in this effort.

Long run time series tests of constant steady-state growth

It is well known that fluctuations of species abundances observed in ecological time series emerge from an interplay between deterministic nonlinear dynamics and stochastic forces. Importantly, nonlinearity and stochasticity introduce significant challenges to the analysis of ecological time series, such as the inference of the effect of species interactions on community dynamics and forecasting of species abundances.Local linear fits with state‐space‐dependent kernel functions, known as S‐maps, provide an efficient method to infer Jacobian coefficients (a proxy for the local effect of species interactions) and to make reliable forecasts from nonlinear time series. Yet, while it has been shown that the S‐map outperforms existing methods for nonparametric inference and forecasting, the methodology is sensitive to process noise. To overcome this limitation, we integrate the S‐map with different regularization schemes.To validate our approach, we test our methodology against different levels of noise and nonlinearity using three standard population dynamics models. We show that an appropriate choice of the regularization scheme, alongside an accurate choice of the kernel functions, can significantly improve the in‐sample inference of Jacobian coefficients and the out‐of‐sample forecast of species abundances in the presence of process noise. We further validate our methodology using two empirical time series of marine microbial communities.Our results illustrate that the regularized S‐map is an efficient method for nonparametric inference and forecasting from noisy, nonlinear, ecological time series. Yet, attention must be paid on the regularization scheme and the structure of the kernel for whether inference or forecasting is the ultimate goal of a research study.

Present management of the north-east Arctic cod and haddock stocks is based on scientific assessments of stock abundance. The assessment procedures apply acoustic and bottom-trawl survey indices of abundance for tuning of the VPA (virtual population analysis). The reliability of the survey indices is strongly dependent on the precision and accuracy of thefish sampling conducted with trawls. Variation in bottom trawl catches and vertical distribution of cod and haddock from a small area were studied with the aim of finding environmental factors of explanatory importance for the observed variation. Data were obtained over a period of three days from two vessels trawling in parallel, one of which simultaneously collected acoustic information on vertical distribution. Variation in catch size and length composition was analysed and related to the observed vertical distribution. The results show that tidal currents aVect the vertical distribution of cod and haddock and, hence, the availability to the bottom trawl. On the other hand, light intensity apparently aVects the catch eYciency of the sampling trawl. Consequently, the abundance estimates from acoustic and bottom-trawl surveys are aVected by the interaction of diel rhythms in tidal current and light intensity. This illustrates the importance of monitoring the environmental factors that influence fish behaviour in order to increase the reliability of the abundance estimates. ? 1996 International Council for the Exploration of the Sea

Hart, A. M., Thomson, A. W., and Murphy, D. 2011. Environmental influences on stock abundance and fishing power in the silver-lipped pearl oyster fishery. – ICES Journal of Marine Science, 68: . Stock variability, fishing power, and the contributing environmental factors were examined for the Australian silver-lipped pearl oyster, Pinctada maxima. The approach was to compare the longer-term time-series of stock abundance derived from catch per unit effort against the shorter-term 0+ spat-settlement index to seek agreement on causal factors. Indices of stock abundance were established with generalized linear models that simultaneously examined the influence of technological and environmental factors, resulting in a predictive model with a 3-year forecast. The advent of global positioning systems caused a 30% increase in fishing power. A negative relationship between abundance and rainfall, and a positive relationship between abundance and temperature, was detected for both spat settlement and fishery abundance. Northerly winds (negative northings) from December to February significantly enhanced settlement, but easterly winds (negative eastings) in the main fishing month of May influenced fishing power positively. After standardizing for the effects of fishing power, a 150% increase in stock abundance of the exploited component of P. maxima stocks was detected between 2004 and 2009. A major contributor to this increase was an exceptionally high settlement of spat in 2005, associated with a rare combination of environmental conditions. Once this year class grows beyond the target size classes, abundance is predicted to decrease to the normal levels.

Inflation and output dynamics with state-dependent pricing decisions

Asynchronous fluctuations in abundance between species with similar ecological roles can stabilize food webs and support coexistence. Sardine (Sardinops spp.) and anchovy (Engraulis spp.) have long been used as an example of this pattern because low-frequency variation in catches of these species appears to occur out of phase, suggesting that fisheries and generalist predators could be buffered against shifts in productivity of a single species. Using landings data and biomass and recruitment estimates from five regions, we find that species do not have equivalent peak abundances, suggesting that high abundance in one species does not compensate for low abundance in the other. We find that globally there is a stronger pattern of asynchrony in landings compared to biomass, such that landings data have exaggerated the patterns of asynchrony. Finally, we show that power to detect decadal asynchrony is poor, requiring a time series more than twice the length of the period of fluctuation. These results indicate that it is unlikely that the dynamics of these two species are compensatory enough to buffer fisheries and predators from changes in abundance, and that the measurements of asynchrony have largely been a statistical artefact of using short time series and landings data to infer ecology.

Species interaction effects on populations can vary in both magnitude (i.e. strong vs. weak) and sign (positive, negative, or no effect). Context‐dependent effects of species interactions occur when the sign or strength of the interaction's effect on population growth rate changes across abiotic gradients. We know that species can vary substantially in the degree of context dependence they exhibit, even across similar abiotic gradients. However, few studies have characterised context dependence of co‐occurring species, limiting our ability to understand the implications of context dependence for species interaction effects on community composition. Using over three decades of data collected for 13 tallgrass prairie forbs at the Konza Prairie Biological Station, we parameterise density structured population models that predict population dynamics as functions of abiotic conditions and bison herbivory. We use these models to estimate the degree of context dependence in responses to bison herbivory for 13 species across three abiotic gradients: weather, fire frequency and soil type. All species showed significant context dependence for fire frequency in the same direction, though with variable magnitude, such that herbivory increased cover with more frequent fires. Context dependence with weather and soil type varied dramatically across species in both direction and magnitude. For example, herbivory effects on 3/13 species were stronger in wet conditions, but herbivory effects on 5/13 species were stronger in dry conditions. Thus, context dependence exhibited by individual species, as opposed to effects of abiotic conditions on the relative abundances of species, could generate much of the weather‐dependent effects of herbivory on community composition. Synthesis: Our work suggests that species can vary dramatically in the presence, direction and magnitude of context dependence, even when occurring in the same community and when considering the same species interaction (i.e. response to a herbivore). In addition, we find that context dependence could drive substantial variation in the effect of species interactions on community characteristics (e.g. composition) across multiple abiotic gradients.

Fluctuations in total landings and in catch per boat have characterized the Digby scallop fishery since it began in 1920. An analysis of records of the fishery indicates that, although changes in fishing methods have been partly responsible for early changes in catch, their influence in recent years has been small and changes in abundance have been primarily responsible for fluctuations in the fishery since it reached its full development in the mid-thirties.Changes in abundance are assessed from analyses of catch records, special "census-fishing" techniques, submarine photography and marking experiments. Estimates from the different methods correspond.Scallops are recruited into the catchable population as six-year-old year-classes. Abundance is high when these recruited year-classes are strong, but is low when they are weak. Abundance in any year is correlated with water temperature six years previously. Both abundance and the strength of individual year-classes are correlated with water temperatures which prevailed at the time the scallops were present as pelagic larvae.It is concluded that changes in the abundance of the catchable scallop stocks result from the combined action of temperature and circulation on the pelagic larvae. Low temperatures retard larval development, and are indicative of great exchange of the water in the Bay of Fundy with outside water masses. This apparently leads to heavy losses of the larvae from the Bay, poor sets on the parent beds, weak year-classes and low abundance of the catchable stocks of six years later. High temperatures speed larval development and are indicative of a closed Fundy circulation which holds the larvae in the vicinity of the parent beds. This leads to good sets, strong year-classes and high abundance of the catchable stocks six years later.

Interlocked use of multiple inland vendace, Coregonus albula (L.), stocks is a relatively new strategy to exploit asynchronous fluctuations in abundance of natural fish stocks between waterbodies. It combines vendace stocks from different lakes or parts of lakes into one interlocked stock to be managed across waterbody ownership boundaries. Management of interlocked stocks can be regarded as one form of portfolio management. Exploitation of interlocked stocks should reduce the interannual variation in yield, and thus ensuring fishers more constant income and the market more constant material flow. The strategy requires fishers to increase their mobility, to benefit from asynchronous fluctuations in abundance of vendace stocks between exploitable waters. A postal inquire addressed to Finnish commercial inland fishers examined whether existing property rights institutions’ and fishers’ harvesting policies were appropriate to establish interlocked use of multiple vendace stocks. Almost half of fishers had, to some extent, reaped benefit from a fishing strategy that included small‐scale mobility, which is consistent with the proposed strategy. By harvesting three or four lakes and stocks, fishers increased their yield compared with exploiting one fishing ground and one vendace stock. Public ownership provided fishers access to stocks nearer their place of residence than other ownership types.

We evaluate a 54-y survey time series for the Delaware Bay oyster beds in New Jersey waters to identify the characteristics of regime shifts in oyster populations and the influence of MSX and Dermo diseases on population stability. Oyster abundance was high during the 1970s through 1985. Oyster abundance was low at the inception of the time series in 1953, remained low through 1969, and has been low since 1985 and very low since 2000. Natural mortality was low in most years prior to the appearance of MSX in 1957. From 1957 through 1966, natural mortality generally remained above 10% annually and twice exceeded 20%. Natural mortality remained well below 15% during the 1970s and into the early 1980s when oyster abundance was continuously high. The largest mortality event in the time series, an MSX epizootic that resulted in the death of 47% of the stock, occurred in 1985. Mortality rose again with the incursion of Dermo in 1990 and has remained above 15% for most years since that time and frequently has exceeded 20%. The primary impact of MSX and Dermo diseases has been to raise natural mortality and ultimately to cause a dispersed stock to retreat into its habitat of refuge in the moderately low salinity reach of the bay. The time series of oyster abundance on the New Jersey oyster beds of Delaware Bay is dominated by two regime shifts, the 1970 abundance increase that was maintained for about 15 y thereafter, and the 1985 abundance decrease that continues through today. These two regime shifts ushered in long-term periods of apparent constancy in population dynamics. The 1985 regime shift was induced by the largest MSX epizootic on record that produced high mortalities throughout a population distributed broadly throughout its habitat range after 15 y of high abundance. A putative new regime commenced circa 2000 as a consequence of a series of Dermo epizootics. Mortalities routinely exceeded 20% of the population annually during this period, with the consequence of a greater degree of stock consolidation than any previous time in the 54-y record. Extreme consolidation of the stock would appear to be a characteristic of the population's response to Dermo disease. The 1970-1984 and post-1985 regimes each were ushered in by a confluence of events unique in the 54-y time series. Each was characterized by a period of relative stability in population abundance. However, the stability in total population abundance belies a more dynamic process of stock redistribution during both time intervals, demonstrating that the appearance of constancy in stock abundance is not necessarily a result of invariant stock dynamics. Rather, the Delaware Bay oyster time series suggests that regime shifts delimit periods during which differential, often offsetting, local trends impart similar abundance levels, and thus constancy at the level of the stock masks substantive changes in local population dynamics potentially fostering future catastrophic changes in population-level attributes. Understanding such regime shifts will likely determine the success of decadal management goals more so than measures designed to influence population abundance.

When populations decline in response to unfavorable environmental change, the dynamics of their population growth shift. In populations that normally exhibit high levels of variation in recruitment and abundance, as do many amphibians, declines may be difficult to identify from natural fluctuations in abundance. However, the onset of declines may be evident from changes in population growth rate in sufficiently long time series of population data. With data from 23 years of study of a population of Fowler's toad (Anaxyrus [ = Bufo] fowleri) at Long Point, Ontario (1989-2011), we sought to identify such a shift in dynamics. We tested for trends in abundance to detect a change point in population dynamics and then tested among competing population models to identify associated intrinsic and extrinsic factors. The most informative models of population growth included terms for toad abundance and the extent of an invasive marsh plant, the common reed (Phragmites australis), throughout the toads' marshland breeding areas. Our results showed density-dependent growth in the toad population from 1989 through 2002. After 2002, however, we found progressive population decline in the toads associated with the spread of common reeds and consequent loss of toad breeding habitat. This resulted in reduced recruitment and population growth despite the lack of significant loss of adult habitat. Our results underscore the value of using long-term time series to identify shifts in population dynamics coincident with the advent of population decline.

Quantitative models that predict stock abundance can inform stock assessments and adaptive management that allows for less stringent controls when abundance is high and environmental conditions are suitable, or tightening controls when abundance is low and environmental conditions are least suitable. Absolute estimates of stock abundance are difficult and expensive to obtain, but data from routine reporting in commercial fisheries logbooks can provide an indicator of stock status. Autoregressive integrated moving average (ARIMA) models were constructed using catch per unit effort (CPUE) from commercial fishing in Port Phillip Bay from 1978–79 to 2009–10. Univariate and multivariate models were compared for short-lived species (Sepioteuthis australis), and species represented by 1–2 year-classes (Sillaginodes punctatus) and 5–6 year-classes (Chrysophrys auratus). Simple transfer models incorporating environmental variables produced the best predictive models for all species. Multivariate ARIMA models are dependent on the availability of an appropriate time series of explanatory variables. This study demonstrates an application of time series methods to predict monthly CPUE that is relevant to fisheries for species that are short lived or vulnerable to fishing during short phases in their life history or where high intra-annual variation in stock abundance occurs through environmental variability.

This paper investigates the relationship between the hydroclimatic parameters (rainfall and flood index) and the catch, stock abundance and recruitment of the catfish, Chrysichthys nigrodigitatus (Bagridae) and the bonga, Ethmalosa fimbriata (Clupeidae) of southeastern Nigeria's inshore waters. For C. nigrodigitatus, most peaks in the mean biomass \((\Delta \% {\bar B})\) and recruitment \((\Delta \% {\bar R(n)})\) curves occurred during the 'wet' years, i.e., years for which the percentage deviations of rainfall \((\Delta \% \overline {{RF}} )\) and flood index \((\Delta \% \overline {{FI}} - 1)\) from their means remained above their averages. Catch and abundance respectively showed good positive linear correlation with the flood index. E. fimbriata behaved differently; some peaks in the \(\Delta \% {\bar B}\) and \(\Delta \% {\bar R(n)}\) curves occurred in the 'wet' and some in the 'dry' years; and there was no correlation between the annual catch of bonga and either the rainfall or the flood index. The hypothesis, that linear relationships exist between the interannual variations in the hydroregime and the yearly fluctuations in the catch and population structure of some coastal and estuarine fishes, holds true for the catfish, C. nigrodigitatus, but not for bonga, E. fimbriata.

This article considers two questions concerning fluctuations in fish stocks and catches: Are they possibly beneficial and would fishery managers opt for a strategy that enchances stock or catch quota variability, or both? For this analysis, three models are followed: in the first, variations in stock abundance are stochastic with no population dynamics; in the second, the stock in a certain time period depends on a stochastic recruitment to the stock and survivors from the previous time period; and the third is a year class model that has more complicated population dynamics but is more realistic.