Fine scale spatial pattern of Pacific sardine (Sardinops sagax) and northern anchovy (Engraulis mordax) eggs

Pacific sardine (Sardinops sagax) and northern anchovy (Engraulis mordax) eggs exhibited different spatial structure on the scale of 0.75–2.5 km in two egg patches sampled in the Southern California Bight in April 2000. Plankton samples were collected at 4‐min intervals with a Continuous Underway Fish Egg Sampler (CUFES) on 5 × 5 km grids centered on surface drifters. Variograms were calculated for sardine and anchovy eggs in Lagrangian coordinates, using abundances of individual developmental stages grouped into daily cohorts. Model variograms for sardine eggs have a low nugget effect, about 10% of the total variance, indicating high autocorrelation between adjacent samples. In contrast, model variograms for anchovy eggs have a high nugget effect of 50–100%, indicating that most of the variance at the scales sampled is spatially unstructured. The difference between observed spatial patterns of sardine and anchovy eggs on this scale may reflect the behavior of the spawning adults: larger, faster, more abundant fish may organize into larger schools with greater structure and mobility that create smoother egg distributions. Size and mobility vary with population size in clupeoids. The current high abundance of sardines and low abundance of anchovy off California agree with the greater autocorrelation of sardine egg samples and the observed tendency for locations of anchovy spawning to be more persistent on the temporal scale of days to weeks. Thus the spatial pattern of eggs and the persistence of spawning areas are suggested to depend on species, population size and age structure, spawning intensity and characteristic physical scales of the spawning habitat.     

Vertical distributions of Japanese sardine (Sardinops melanostictus) eggs: comparison of observations and a wind-forced Lagrangian mixing model

To understand how Japanese sardine eggs vertically disperse in water, we acquired a series of depth‐resolved samples of eggs while tracking a drogued buoy to follow the water mass in which the eggs were spawned. This survey was conducted in early March of 1991, when the sardine population was abundant and actively spawning in the Pacific off Kyushu. Buoyancy of the eggs was estimated by measuring the specific gravity, and was 1–4 × 10^?3 less than that of the water in which they were spawned, giving a rising speed of 1–2 mm s^?1. In an attempt to diagnose the dynamics underlying the observed vertical profiles of the eggs, a one‐dimensional egg distribution model incorporating the egg rising speed and wind‐ and depth‐dependent vertical diffusivity was formulated. Observed eggs were mainly in the surface mixed layer, and their vertical distribution changed with wind‐induced surface mixing: during strong winds, the eggs were distributed relatively evenly over the upper 60–75 m, whereas during weak winds and shallow pycnocline, the eggs formed a subsurface peak around 25 m depth. The model accurately reproduced these variations in the vertical distributions of eggs, suggesting that the gradients of egg concentration are sensitive to the strength of wind and the structure of underlying hydrography. Further tests with the model showed the importance of positive buoyancy: neutrally buoyant eggs formed subsurface maxima below the food‐rich euphotic zone. The balance of vertical mixing and floating, which allows the eggs to remain in the euphotic zone but away from the near‐surface waters, may be an adaptation to minimize surface‐enhanced predation, while allowing the eggs to hatch in the food‐rich euphotic zone.     

A continuous, underway fish egg sampler

We describe a method to sample the highly contagious distribution of pelagic fish eggs. CUFES, the continuous, underway fish egg sampler, consists of a submersible pump, concentrator, electronics and sample collector. This system operates continuously and under nearly all sea conditions, providing a real-time estimate of the volumetric abundance of pelagic fish eggs at pump depth, usually 3 m. CUFES-derived estimates of volumetric abundance agree well with those from nets towed at pump depth and with areal abundance estimated from vertically integrated plankton tows. CUFES has been used successfully to sample the eggs of menhaden, pinfish, sardine, and anchovy off the coasts of the eastern and western United States and South Africa. Two large patches of eggs of the Atlantic menhaden were sampled off North Carolina in winter 1993–94, had a linear scale of 5–10 km, and were found in waters between the Gulf Stream and mid-shelf front. Spawning location may be related to bathymetry. CUFES is now being used to estimate spawner biomass by the daily egg production method. An optical plankton counter provided accurate estimates of the number of Atlantic menhaden eggs sample by CUFES.     

Alternating dominance of postlarval sardine and anchovy caught by coastal fishery in relation to the Kuroshio meander in the Enshu-nada Sea

In the mid 1970s, the fishery catch of postlarval Japanese anchovy (Engraulis japonica) in a shelf region of the Enshu‐nada Sea, off the central Pacific coast of Japan, started to decline corresponding to a rapid increase of postlarval sardine (Sardinops melanostictus). In late 1980s, sardine started to decline, and it was replaced by anchovy in the 1990s. This alternating dominance of postlarval sardine and anchovy corresponded to the alternation in egg abundance of these two species in the spawning habitat of this sea. It was also noteworthy that during the period of sardine decline, sardine spawning occurred in April–May, a delay of two months compared with spawning in the late 1970s. The implication of oceanographic changes in the spawning habitat for the alternating dominance of sardine and anchovy eggs was explored using time‐series data obtained in 1975–1998, focusing on the effect of the Kuroshio meander. Large meanders of the Kuroshio may have enhanced the onshore intrusion of the warm water into the shelf region and contributed to an increase in temperature in the spawning habitat. This might favour sardine, because its egg abundance in the shelf region was more dependent on the temperature in early spring than was that of anchovy. In addition, enhanced onshore intrusion could contribute to transport of sardine larvae from upstream spawning grounds of the Kuroshio region. On the other hand, anchovy egg abundance was more closely related to lower transparency at the shelf edge, which may indicate the prevalence and prolonged residence of the coastal water, and therefore higher food availability, frequently accompanying non‐meandering Kuroshio. The expansion/shrinkage of the spawning habitat of sardine and anchovy in the shelf region, apparently responding to the change in the Kuroshio, possibly makes the alternation in dominance of postlarval sardine and anchovy most prominent in the Enshu‐nada Sea, in combination with changes in the abundance of spawning adults, which occurred almost simultaneously in the overall Kuroshio region. The implication of this rather regional feature for the alternating dominance of sardine and anchovy populations on a larger spatial scale is also discussed.     

Relative importance of gulf and shelf waters for spawning and recruitment of Australian anchovy, Engraulis australis, in South Australia

Gonosomatic indices and egg and larval densities observed from 1986 to 2001 suggest that the peak spawning season of the Australian anchovy (Engraulis australis) in South Australia occurs during January to March (summer and autumn). This coincides with the spawning season of sardine (Sardinops sagax) and the period when productivity in shelf waters is enhanced by upwelling. Anchovy eggs were abundant throughout gulf and shelf waters, but the highest densities occurred in the northern parts of Spencer Gulf and Gulf St Vincent where sea surface temperatures (SST) were 24–26°C. In contrast, larvae >10 mm total length (TL) were found mainly in shelf waters near upwelling zones where SSTs were relatively low (<20°C) and levels of chlorophyll a (chl a) relatively high. Larvae >15 mm TL were collected only from shelf waters near upwelling zones. The high levels of larval abundance in the upwelling zones may reflect higher levels of recruitment to later stages in these areas compared with the gulfs. The sardine spawns mainly in shelf waters; few eggs and no larvae were collected from the northern gulfs. The abundance of anchovy eggs and larvae in shelf waters increased when sardine abundance was reduced by large‐scale mortality events, and decreased as the sardine numbers subsequently recovered. We hypothesize that the upwelling zones provide optimal conditions for the survival of larval anchovy in South Australia, but that anchovy can only utilize these zones effectively when the sardine population is low. At other times, northern gulf waters of South Australia may provide a refuge for the anchovy that the sardine cannot utilize.     

Transport and environmental temperature variability of eggs and larvae of the Japanese anchovy (Engraulis japonicus) and Japanese sardine (Sardinops melanostictus) in the western North Pacific estimated via numerical particle-tracking experiments

Numerical particle-tracking experiments were performed to investigate the transport and variability in environmental temperature experienced by eggs and larvae of Pacific stocks of the Japanese anchovy ( Engraulis japonicus ) and Japanese sardine ( Sardinops melanostictus ) using high-resolution outputs of the Ocean General Circulation Model for the Earth Simulator (OFES) and the observed distributions of eggs collected from 1978 to 2004. The modeled anchovy individuals tend to be trapped in coastal waters or transported to the Kuroshio–Oyashio transition region. In contrast, a large proportion of the sardines are transported to the Kuroshio Extension. The egg density-weighted mean environmental temperature until day 30 of the experiment was 20–24°C for the anchovy and 17–20°C for the sardine, which can be explained by spawning areas and seasons, and interannual oceanic variability. Regression analyses revealed that the contribution of environmental temperature to the logarithm of recruitment per spawning (expected to have a negative relationship with the mean mortality coefficient) was significant for both the anchovy and sardine, especially until day 30, which can be regarded as the initial stages of their life cycles. The relationship was quadratic for the anchovy, with an optimal temperature of 21–22°C, and linear for the sardine, with a negative coefficient. Differences in habitat areas and temperature responses between the sardine and anchovy are suggested to be important factors in controlling the dramatic out-of-phase fluctuations of these species.     

Variability in the spawning habitat of Pacific sardine (Sardinops sagax) off southern and central California

The spatial pattern of sardine spawning as revealed by the presence of sardine eggs is examined in relation to sea surface temperature (SST) and mean volume backscatter strength (MVBS) measured by a 150 kHz acoustic Doppler current profiler (ADCP) during four spring surveys off central and southern California in 1996–99. Studies in other regions have shown that MVBS provides an excellent measure of zooplankton distribution and density. Zooplankton biomass as measured by survey net tows correlates well with concurrently measured MVBS. The high along‐track resolution of egg counts provided by the Continuous Underway Fish Egg Sampler (CUFES) is a good match to the ADCP‐based data. Large interannual differences in the pattern and density of sardine eggs are clearly related to the concurrently observed patterns of surface temperature and MVBS. The strong spatial relationship between sardine eggs and MVBS is particularly evident because of the large contrast in zooplankton biomass between the 1998 El Niño and 1999 La Niña. The inshore distribution of sardine spawning appears to be limited by the low temperatures of freshly upwelled waters, although the value of the limiting temperature varies between years. Often there is an abrupt offshore decrease in MVBS that is coincident with the offshore boundary of sardine eggs. Possible reasons for this association of sardine eggs and high zooplankton biomass include an evolved strategy that promotes improved opportunity of an adequate food supply for subsequent larval development, and/or adult nutrient requirements for serial spawning. Hence, the distribution of these parameters can be used as an aid for delineating the boundaries of sardine spawning habitat.     

Anchovy (Engraulis ringens) and sardine (Sardinops sagax) spatial dynamics and aggregation patterns in the Humboldt Current ecosystem, Peru, from 1983–2003

Three indexes of spatial aggregation are developed and used to examine the aggregation pattern of sardine (Sardinops sagax) and anchovy (Engraulis ringens) in the Peruvian Humboldt Current System, determined from 36 acoustic surveys conducted from 1983 through 2003 by the Peruvian Marine Institute (IMARPE). Each index assesses a different aspect of aggregation: the concentration, the percent occupancy of space and the clustering of high‐fish abundance. Both time‐series correlation and tree‐based clustering‐regression method, classification and regression trees (CART), were used to relate each of the indexes to environmental variables (season, temperature anomaly and year). Additionally, a measure of onshore–offshore distribution, the average distance from the coast, and abundance variables (the average acoustic backscatter per occupied sampling unit, and the acoustically estimated total abundance of sardine and anchovy from IMARPE) were related to environmental factors by using CART. We show that the 1983–2003 time series can be divided into three different periods: with shifts in 1992 and in 1997–98. Sardine and anchovy showed large differences in both abundance and aggregation among these periods. Although upwelling ecosystems support dramatic and sudden changes in environmental conditions, fish responses are sometimes smoother than usually suggested and there are transition periods with concomitant high biomasses of anchovy and sardine, but with different spatial aggregation patterns. Observed relationships between environmental proxies and aggregation patterns support the habitat‐based hypothesis that environmentally mediated alterations in range lead to population changes.     

Importance of the Shatsky Rise Area in the Kuroshio Extension as an offshore nursery ground for Japanese anchovy (Engraulis japonicus) and sardine (Sardinops melanostictus)

Recent findings suggest that recruitment of Japanese anchovy (Engraulis japonicus) and sardine (Sardinops melanostictus) depends on survival during not only the first feeding larval stage in the Japanese coastal waters and the Kuroshio front but also during the post‐larval and juvenile stages in the Kuroshio Extension. Spatial distributions of juvenile anchovy and sardine around the Shatsky Rise area in the Kuroshio Extension region and the Kuroshio–Oyashio transition region are described, based on a field survey in the late spring using a newly developed mid‐water trawl for sampling juveniles. All stages of anchovy from post‐larvae to juveniles were obtained in the northern Shatsky Rise area. The Kuroshio Extension bifurcates west of the Shatsky Rise area and eddies are generated, leading to higher chlorophyll concentrations than in the surrounding regions in April and May. When Japanese anchovy and sardine spawn near the Kuroshio front or the coastal waters south‐east of Japan, their larvae are transported by the Kuroshio Extension and are retained in the Shatsky Rise area, which forms an important offshore nursery ground, especially during periods of high stock abundance.     

Modelling potential spawning habitat of sardine (Sardina pilchardus) and anchovy (Engraulis encrasicolus) in the Bay of Biscay

Large amplitude variations in recruitment of small pelagic fish result from interactions between a fluctuating environment and population dynamics processes such as spawning. The spatial extent and location of spawning, which is critical to the fate of eggs and larvae, can vary strongly from year to year, as a result of changing population structure and environmental conditions. Spawning habitat can be divided into ‘potential spawning habitat’, defined as habitat where the hydrographic conditions are suitable for spawning, ‘realized spawning habitat’, defined as habitat where spawning actually occurs, and ‘successful spawning habitat’, defined as habitat from where successful recruitment has resulted. Using biological data collected during the period 2000–2004, as well as hydrographic data, we investigate the role of environmental parameters in controlling the potential spawning habitat of anchovy and sardine in the Bay of Biscay. Anchovy potential spawning habitat appears to be primarily related to bottom temperature followed by surface temperature and mixed‐layer depth, whilst surface and bottom salinity appear to play a lesser role. The possible influence of hydrographic factors on the spawning habitat of sardine seems less clear than for anchovy. Modelled relationships between anchovy and sardine spawning are used to predict potential spawning habitat from hydrodynamical simulations. The results show that the seasonal patterns in spawning are well reproduced by the model, indicating that hydrographic changes may explain a large fraction of spawning spatial dynamics. Such models may prove useful in the context of forecasting potential impacts of future environmental changes on sardine and anchovy reproductive strategy in the north‐east Atlantic.     

Historical fluctuations in spawning location of anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) in the Bay of Biscay during 1967–73 and 2000–2004

The spatial extent of small pelagic fish spawning habitat is influenced by environmental factors and by the state of the adult population. In return, the configuration of spawning habitat affects recruitment and therefore the future structure of the adult population. Interannual changes in spatial patterns of spawning reflect variations in adult population structures and their environment. The present study describes the historical changes in the spatial distribution of spawning of anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) in the Bay of Biscay during two periods: 1967–72 and 2000–2004. Using data from egg surveys conducted in spring, the spatial distributions of anchovy and sardine eggs are characterized by means of geostatistics. For each survey, a map of probability of egg presence is constructed. The maps are then compared to define (1) recurrent spawning areas, (2) occasional spawning areas and (3) unfavourable spawning areas during each period. Sardine spawning habitat is generally fragmented and appears spatially limited by the presence of cold bottom water. It is confined to coastal or shelf break refuge areas in years of restricted spawning extent. For anchovy, recurrent spawning sites are found in Gironde and Adour estuaries whilst spawning can extend further offshore in years of more intense spawning. For both species, the mean pattern of spawning has changed between 1967–72 and 2000–2004. Noticeably, the spatial distribution of anchovy eggs in spring has expanded northward. This trend possibly results from changes in environmental conditions during the last four decades.     

Impact of freshwater input and wind on landings of anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) in shelf waters surrounding the Ebre (Ebro) River delta (north-western Mediterranean)

Time series analyses (Box–Jenkins models) were used to study the influence of river runoff and wind mixing index on the productivity of the two most abundant species of small pelagic fish exploited in waters surrounding the Ebre (Ebro) River continental shelf (north‐western Mediterranean): anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus). River flow and wind were selected because they are known to enhance fertilization and local planktonic production, thus being crucial for the survival of fish larvae. Time series of the two environmental variables and landings of the two species were analysed to extract the trend and seasonality. All series displayed important seasonal and interannual fluctuations. In the long term, landings of anchovy declined while those of sardine increased. At the seasonal scale, landings of anchovy peaked during spring/summer while those of sardine peaked during spring and autumn. Seasonality in landings of anchovy was stronger than in sardine. Concerning the environmental series, monthly average Ebre runoff showed a progressive decline from 1960 until the late 1980s, and the wind mixing index was highest during 1994–96. Within the annual cycle, the minimum river flow occurs from July to October and the wind mixing peaks in winter (December–April, excluding January). The results of the analyses showed a significant correlation between monthly landings of anchovy and freshwater input of the Ebre River during the spawning season of this species (April–August), with a time lag of 12 months. In contrast, monthly landings of sardine were significantly positively correlated with the wind mixing index during the spawning season of this species (November–March), with a lag of 18 months. The results provide evidence of the influence of riverine inputs and wind mixing on the productivity of small pelagic fish in the north‐western Mediterranean. The time lags obtained in the relationships stress the importance of river runoff and wind mixing for the early stages of anchovy and sardine, respectively, and their impact on recruitment.     

Size,permeability and buoyancy of anchovy (Engraulis Encrasicolus) and sardine (Sardina Pilchardus) eggs in relation to their physical environment in the Bay of Biscay

The size and specific gravity of eggs of marine pelagic fish partly determine their dispersal and survival. Using an original dataset of anchovy and sardine eggs, sampled in spring over the last decade in the Bay of Biscay, we provide a parameterization of these properties on ambient water temperature and salinity. We used the density gradient column for measurement of egg specific gravity. The column was also filled with homogeneous water for sinking velocity experiments. For anchovy, these experiments confirm that the effect of egg permeability through the chorion could be neglected when modelling sinking, while it has to be considered for sardine, its perivitelline space representing 78.6% (±6.2%) of the total egg volume, as opposed to 5–10% for most teleosts species. We estimated a coefficient of permeability of the chorion of 0.0038 mm s^?1. However, permeability should not affect the measurement of sardine egg specific gravity in a gradient column, provided a minimum duration before reading is respected for equilibrium to be reached. In relation to their environment, we found that the egg specific gravity is largely determined by sea surface salinity for both species, whereas egg size is weakly but significantly impacted by temperature, for sardine only. On average, the estimated difference in specific gravity between egg and surface water is ?0.92 σ_T for anchovy and ?1.06 σ_T for sardine. The detailed parameterization of the relationship between eggs and water properties should prove useful, in particular to modellers dealing with the dispersal of fish early life stages.     

Environmental variability and growth histories of larval Japanese sardine (Sardinops melanostictus) and Japanese anchovy (Engraulis japonicus) near the frontal area of the Kuroshio

Environmental variability and growth‐rate histories from hatching to capture were investigated for larval Japanese sardine (Sardinops melanostictus) and Japanese anchovy (Engraulis japonicus). Larvae collected around the front of the Kuroshio Current were examined using otolith microstructure analysis, and their movement was estimated from numerical particle‐tracking experiments. Sardine larvae collected inshore of the Kuroshio front originated from a coastal area near the sampling site, while those collected in the offshore area originated from an area 500–800 km west‐southwest of the sampling site. Anchovy larvae collected both inshore and offshore had been transported from widely distributed spawning areas located west of the sampling area. At the age of 13–14 days for sardine and 19–20 days for anchovy, the offshore group exhibited significantly higher mean growth rates than did the inshore group. Although the offshore area was generally warmer than the inshore area, temporal variations in growth rate are not attributable solely to fluctuations in environmental temperature. While previous studies have examined the relationship between larval growth rates and environment based solely on data at capture, the methods used in the present study, combining otolith analysis and numerical particle‐tracking experiments, utilize data up until hatching. Although the relationship between growth rate and environment was not fully confirmed, this approach will greatly advance our understanding of fish population dynamics.     

One-dimensional biophysical modelling of fish egg vertical distributions in shelf seas

Modelling the vertical distribution of fish eggs is important when assessing fish stocks with egg production methods and for monitoring the reproductive potential of fish populations. Fish eggs are passive particles and their vertical distribution is determined by a few parameters such as egg density, egg diameter, wind‐ and tide‐induced turbulence, and vertical hydrographic structure. A one‐dimensional vertical biophysical, numerical model was developed which was adapted to the hydrography of shelf seas under the influence of tidal currents, wind‐induced circulation, and river discharges. The biological part of the model parameterized the ascent velocity of the egg as a function of egg properties (diameter, density) and water properties (density, viscosity, turbulence). The model contains a turbulence closure which makes the model dynamic. The model parameters were surface wind, tidal currents, T‐S profile, and egg diameter and density, which were kept constant in time. The model has the capacity to generate sub‐surface egg maxima in different hydrographic conditions, e.g. in areas under the influence of river plumes, and can also homogenize the egg distribution under wind and tide forcing. Sensitivity tests were carried out to study the response of the model to variations in the model parameters for a variety of hydrographic conditions. The modelled egg vertical distributions were validated by comparison of the model results with egg distributions sampled in the field. The analysis highlighted variability in fish egg density of anchovy, sardine, and sprat across years and areas, with a potential link between egg density and surface sea water density. The validated model is a tool for the analysis of shelf seas fish egg vertical distributions.     

The effect of vertical and horizontal distribution on retention of sardine (Sardinops sagax) larvae in the Northern Benguela – observations and modelling

In the present study, a modelling experiment is conducted to simulate the transport of sardine (Sardinops sagax) eggs and larvae in the Northern Benguela. Based on historical and newly obtained data, different scenarios of vertical and horizontal distribution are applied and the effects on retention are discussed. The simulations showed that vertical and horizontal distribution were important for retention of sardine larvae in the Northern Benguela. By using age‐dependent data on vertical distribution, it was shown that retention of particles in the simulation was substantially enhanced compared with a scenario where particles were distributed in the offshore moving Ekman layer. Retention was lowest during October–December (when upwelling intensity is high) and highest during February–April (when upwelling intensity is somewhat lower). When different spawning areas were considered, highest retention was observed in an area near Walvis Bay. It is concluded that the behaviour of sardine larvae is adapted to the circulation system in the Northern Benguela in a way that promotes retention of the larvae in inshore nursery areas.     

Density‐dependent egg production in chub mackerel in the Kuroshio Current system

A paradigm of proportionality between spawning stock biomass (SSB) and total egg production (TEP) has been largely untested at multidecadal scales mainly because of difficulty in estimating annual TEP. Recently, this paradigm was directly tested for sardine (Sardinops melanostictus) and anchovy (Engraulis japonicus) at a multidecadal scale to reveal that SSB–TEP proportionality was partially distorted by intraspecific (sardine) and interspecific (anchovy) density dependence in total egg production per spawner individual (TEPPS) or unit weight (TEPPSW). In the present study, we demonstrate intraspecific density dependence in TEPPS/TEPPSW for chub mackerel (Scomber japonicus) in the Kuroshio Current system, using a proxy for TEPPS/TEPPSW, calculated from snapshot abundance data based on fishery‐independent egg surveys in combination with fishery‐dependent stock assessment data, at a multidecadal scale (38 years). TEPPS/TEPPSW exponentially declined with SSB, indicating a strong intraspecific density dependence in TEPPS/TEPPSW in chub mackerel. The observed phenomenon for chub mackerel was similar to that for sardine. Hence, intraspecific density dependence in TEPPS/TEPPSW may be a phenomenon that is generally applicable for species with a high maximum biomass and large population fluctuations. Lastly, we recommend the application of a TEP‐based framework to studies on recruitment mechanisms of fish.     

Pacific sardine (Sardinops sagax, Jenyns 1842) landings prediction. A neural network ecosystemic approach

In this study the performances of computational neural networks (CNNs), multiple linear regressions (MLRs) and generalised additive models (GAMs) to predict Pacific sardine (Sardinops sagax) landings and to analyse their relationships with environmental factors in the north area off Chile were studied. For this purpose several local and global environmental variables and indexes (sea surface temperature, sea level and Ekman transport index in the Chilean coast and, sea surface temperature in the area Niño 3 + 4 and Niño 1 + 2, and the south oscillation index) were considered as inputs or independent variables. Additionally, several CNNs were calibrated and validated adding the anchovy (Engraulis ringens) landings in the same area as model inputs. The time lags of the variables considered were selected through analysis of the non-linear cross-correlation functions and an alternative form of sensitivity analysis based on the approach of the missing value problem. The analysis of error measures with validation data set showed that the best results were obtained when local and global variables were used separately and combined with anchovy landings. Globally, the best result was given by a CNN with 18 input variables (model CNN 6(II) which only considered global variables and anchovy landings) and 10 neurons in a hidden layer. For this configuration the explained variance was slightly higher to 86% which supposed a standard error of prediction of 7.66%. These results were significantly better than those obtained with MLRs and GAMs. The strong correlation between predicted and observed sardine landings suggests that CNNs captured the trend of the historical data. Also, the generalisation capacity together the sensitivity analysis allowed us to identify the variables with a high weight in the model and partially to interpret the statistical functional relationships between these environmental variables and sardine landings.     

Disentangling density‐dependent effects on egg production and survival from egg to recruitment in fish

Understanding of density‐dependent effects is key to achieving sustainable management of self‐regulating biological resources such as fish stocks. Traditionally, density‐dependent effects on population abundance in fish have been considered to occur from hatching to recruitment, based on the paradigm of proportionality between spawning stock biomass and total egg production. Here, we demonstrate how the existence of intraspecific and interspecific density dependence in egg production changes the current understanding of density‐dependent processes in the life history of fish, by disentangling density‐dependent effects on egg production and survival from egg to recruitment, using sardine (Sardinops melanostictus, Clupeidae) and anchovy (Engraulis japonicus, Engraulidae) as model species. For sardine, strong intraspecific density‐dependent effects occurred in egg production, but no density‐dependent effects occurred or if any they were weak enough to be masked by environmental factors from hatching to recruitment. In contrast, for anchovy, interspecific density‐dependent effects occurred in egg production. In the survival after hatching, anchovy experienced stronger intraspecific density‐dependent effects than currently recognized. This analysis could overturn the current understanding of density‐dependent effects in the life history, highlighting contrasts between the effects on individual quality and population abundance and between the model species. We propose to reconsider the basis of fisheries management and recruitment studies based on the revised understanding of density‐dependent effects in the life history of the respective species.     

Distribution and settling of Japanese anchovy (Engraulis japonicus eggs at the spawning ground off Changjiang River in the East China Sea

Spatial patterns in the distribution and abundance of Japanese anchovy, Engraulis japonicus, eggs were studied from net surveys in the East China Sea in May 1991. Egg abundance was> 5 × 10³ eggs (100 m³)^?1 in the area off Changjiang River, where a large spawning ground was developed. The vertical distribution of living eggs showed a maximum concentration at the surface (40700 eggs (100 m³)^?1) and rapidly decreased to ～5000 eggs (100 m³)^?1 at 14 m. Newly spawned eggs (stage I) were found throughout the water column but were most abundant near surface. Some eggs were morphologically identified as dead owing to their abnormal development and physical damage. In contrast to living eggs, no dead eggs were found at the surface and the concentrations were low to 21 m depth (40–64 eggs (100 m³)^?1). The concentration increased markedly with increasing depth, reaching a maximum of 634 eggs (100 m³)^?1 near the bottom (35 m). Dead eggs accounted for less than 0.3% of the total caught within 7 m of the surface and increased exponentially to 12% near the bottom. The settling loss of dead anchovy eggs also was studied by employing sediment traps at the spawning ground. The downward flux of settling eggs was low (304–405 eggs m^?2 day^?1) at 15 and 20 m depths, but rapidly increased with increasing depth, reaching a maximum of 1622 eggs m^?2 day^?1 at 35 m (5 m above bottom). By comparing the egg flux at 20 m with the living eggs abundance in the upper 20 m, the settling loss of eggs was calculated as about 0.098 day^?1. This probably represents the natural mortality of anchovy eggs caused by genetic abnormalities and incomplete fertilization.