Weight–length relationships of 33 demersal fish species from Azores archipelago

Weight–length relationships are presented for 33 fish species of Azores archipelago. This work is the first reference on weight–length parameters of 16 species. Samples were collected between the years of 1999 and 2003 using longline at depths ranging from a few meters to 1200 m. The parameters of a and b of the equation W = aL^b were estimated. The b values of the species caught ranged from 2.701 for Lepidopus caudatus to 3.677 for Centroscymnus crepidater. Whenever possible, the b values for the species obtained both in this study and some of the previously reported in the Portuguese waters were compared.     

Otolith–fish size relationship in juvenile gag (Mycteroperca microlepis) of the eastern Gulf of Mexico: a comparison of growth rates between laboratory and field populations

In this study, we conducted experiments on wild-caught juvenile gag Mycteroperca microlepis from the eastern Gulf of Mexico to evaluate the effect of food availability on somatic growth and otolith growth. Juveniles were fed at two different food levels until all fish attained similar sizes. We found that food availability significantly affected growth rates. However, we also found that this manifested itself in differential otolith size. That is, slower-growing gag had larger, heavier otoliths than equal-sized faster-growing gag; an experimental result that has been observed previously among various fish species. We wanted to apply these experimental results to field-caught gag because our initial observations indicated that gag from more southern latitudes along Florida’s west coast were larger than gag from more northern latitudes, at least during the early juvenile period. Applying these relationships to regional field populations, we found that juvenile gag from the more northern latitudes appeared to grow faster than those from southern latitudes, using an otolith–fish size proxy for growth. However, examination of fish length–age relationships revealed that juvenile gag growth rates were not significantly different between regions. These results are contrary to the expectation that larger-sized gag from southern latitudes are growing faster, and suggests that other factors, such as spawning time and habitat quality may explain regional size differences.     

Hydroacoustic in situ target strength of smelt (Osmerus eperlanus (L.))

Dependence of the in situ acoustic target strength (TS) of smelt (Osmerus eperlanus (L.)) on the total fish length (L) was estimated from TS measurements made with a split-beam 120 kHz echo sounder with reference to fish length–frequency distributions in trawl catches. The studies were conducted in the eutrophic and turbid Lake Hiidenvesi in southern Finland. Smelt dominated the catches-in-number with proportions at different sampling stations ranging from 91.8 to 100%. A TS–length equation of the form TS = alog L − b was fitted through the modes in the TS-distributions and in the respective fish length distributions. The estimated model (±parameter standard error) was TS = 23.4 (±2.24)log₁₀ L − 68.7 (±1.82). With a pre-defined slope of 20, the fitted equation was TS = 20 log₁₀ L − 65.9 (±0.23). The estimated day–night difference in TS–length relationship was not significant (t-test, p = 0.37).     

The dynamics of an exploited population of a siluroid catfish (Schilbe intermidius Reupell 1832) in the Cross River, Nigeria

We studied the breeding biology, growth, mortality, recruitment pattern and yield per recruit of a freshwater siluroid catfish (Schilbe intermedius) in the Cross River, Nigeria. Monthly variation in mean gonadosomatic and mean condition indices shows that the species breeds twice in a year (March and September). We used 12 consecutive months length–frequency data together with FiSAT software to study the population dynamics of this catfish which is exploited by artisanal fishermen. Fitting the seasonalized von Bertalanffy growth function to our length–frequency data gives the following growth parameters: L_∞=27.5 cm total length, K=0.29 yr⁻¹, C=0.5, WP=0.46. Using the seasonalized length converted catch curve, we estimated the instantaneous total mortality coefficient Z as 1.85 yr⁻¹. The instantaneous natural mortality coefficient M was 0.81 yr⁻¹ while the instantaneous fishing mortality coefficient F was 1.04 yr⁻¹, giving the current exploitation rate E=0.56. From the analysis of probability of capture of each length class we estimated the length at first capture L_c to be 10.9 cm. The relative yield per recruit analysis predicted the maximum exploitation rate E_max=0.54. The current exploitation rate E is more than this predicted maximum E_max. Thus, the species is on the verge of being overexploited.     

Linear mixed-effects models for weight–length relationships

Length and weight data are often analyzed in fisheries science to derive a parametric weight–length relationship for estimating biomass and to develop indices of condition for comparing the ‘wellness’ of different populations of fish. However, analysis of such data often ignores the inherent spatial and temporal grouping of the observations, and hence, the data hierarchy. This paper proposes the use of linear mixed-effects models as an effective means of analyzing and comparing weight–length relationships and indices of condition when there are many groups. The use of simple linear regression (where grouping is ignored), ANCOVA (where group effects are incorporated as fixed-effects), and linear mixed-effects models (where group effects are random-effects) are compared using data for Atlantic sea scallops (Placopecten magellanicus). The group means of residuals is proposed as a measure of relative weight or index of population condition. Linear mixed-effects models should be used to analyze grouped data because the variability among groups is ignored in simple linear regression and ANCOVA. Also, it is important that explanatory variables be incorporated in analyses of grouped data because their influence may mask the true differences among groups.     

The utility of otolith weight as a predictor of age in the emperor Lethrinus mahsena and other tropical fish species

This study examines the potential use of otolith weight as a proxy for age in the lethrinid Lethrinus mahsena from different sites in the tropical Indian Ocean: the banks of Seychelles, Mauritius and British Indian Ocean Territory (BIOT, Chagos Archipelago). The reliability of age–frequency distributions and individual ages estimated using otolith weight–age relationships was examined through comparison with those estimated through the standard method of ageing using otolith increments. Two other methods for estimating age–frequencies using age-slicing via an estimated growth curve were also examined; these used growth curves estimated by a length-based method (ELEFAN), or by fitting directly to length-at-age data (an ‘age-based’ method). Age-slicing using length-based growth parameters failed to produce reliable age–frequencies, due to inaccuracies in the growth parameter estimates. The use of age-based growth parameter estimates improved the results of age-slicing, however, age–frequencies remained significantly different from those obtained from ageing using otolith increments in two locations. The use of otolith weight–age relationships resulted in estimated age–frequency distributions that in all locations were not significantly different from those assessed through otolith increment counts. In contrast, L. mahsena otolith weight–age relationships could not be used to estimate individual ages accurately, due to the level of overlap in otolith weight between age classes. Where otolith increments are routinely used to age commercial fish species, the fact that otolith weight–age relationships could not be used to age individuals accurately may limit its application. However, where routine ageing of individuals through otolith increments is considered impractical, for instance because of its cost, the use of otolith weight–age relationships to derive catch age–frequencies represents a viable alternative approach. With this in mind, this study has also demonstrated that there is the potential to use otolith weight–age relationships for five other species caught around the Seychelles, following the validation of their otolith increments.     

Age, growth and hatching season of the diamond squid Thysanoteuthis rhombus estimated from statolith analysis and catch data in the western Sea of Japan

Age, growth and hatching season were estimated for the diamond squid (Thysanoteuthis rhombus) migrating into the Sea of Japan based on analyses of statolith growth increments and length–frequency data of the catch in 1999–2004. Growth rates did not differ significantly between the sexes. The oldest squid was 306 days old, which provides further evidence that T. rhombus has a 1-year life span. The hatching season extended from January to September, with a peak in February–March, and the spawning grounds were suggested to extend from the far southwestern Pacific to the East China Sea. Using aging results from statolith analysis, growth was well described by a logistic formula, which also closely corresponded with the sequential progress of the mean mantle length in size–frequency distribution of the catch. Growth rates varied depending on the time of hatching; earlier-hatched squid grew faster than later-hatched ones, suggesting that the former hatched in warmer upper-stream areas of the Tsushima Current or the Kuroshio region and the latter hatched in colder mid- or downstream areas of the Tsushima Current.     

Performance evaluation of rice–fish integration system in rainfed medium land ecosystem

This study was carried out for three successive years during 1999–2001 to evaluate growth and yield performance of fish, prawn and paddy under rice–fish integration system in rainfed medium land ecosystem. Irrespective of stocking density, faster growth rate was recorded for Catla catla followed by Cyprinus carpio, Cirrhinus mrigala, Labeo rohita and Macrobrachium rosenbergii. C. carpio and C. mrigala performed better growth rate against that of L. rohita probably due to the fact that being bottom dwellers, C. carpio and C. mrigala are more tolerant to fluctuation of oxygen concentration. Productivity of fish and prawn was, however, higher (p<0.05) in refuges with 10-cm weir height plots, irrespective of stocking density, while overall yield performance was good at stocking density of 25,000 ha⁻¹. It was observed that, even with supplemental feeding, with increase in stocking density, biomass yield increased up to an optimum and then decreased. An average minimum and maximum yield of 906.6–1282.3 kg ha⁻¹ of fish and prawn has been achieved, which was much higher than the earlier recorded productivity in a season under rice–fish integration system. Highest grain yield was recorded at 15-cm weir height plot (3629 kg ha⁻¹), probably contributed by higher number of panicles per square meter (235.5) and number of filled grains per panicle (121.7). Percentage increase in rice yield under rice–fish integration system was 7.9–8.6% against control, where paddy was cultivated without integration of fish and prawn probably due to better aeration of water, greater tillering effect and additional supply of fertilizer in form of leftover feed and fish excreta. Irrespective of stocking density, the overall rice equivalent yield (REY) of the system was high (4.22–4.55 tons ha⁻¹) at 12.5-cm weir height plots-cum-refuge, without using any pesticide, herbicide, etc.     

Ecomorphological patterns of the sagitta in fish on the continental shelf off Argentine

Morphology and morphometry of the sagittae otolith were studied in fish associated with different substrates. The shape, margins and rostrum of three groups of otoliths from several species were analyzed: group 1 (fish associated with soft substrates, N=10 species), group 2 (fish associated with hard substrates, N=10 species) and group 3 (pelagic fish, not associated with the bottom, N=6 species). E and R indexes were calculated for each species. The value of E=maximum width of the sagitta (WO)/maximum length of the sagitta (LO)%, expresses the relative tendency in the shape otolith (from circular to elongate). The value of R=length of the rostrum (LR)/LO%, expresses the percentage in the total length of the otolith that corresponds to the rostrum. The sagittae of group 1 were circular or polygonal with rounded borders. The rostrum can be absent or poorly developed. The sagittae shape of group 2 was elongated, with ornamented borders and a rostrum. The sagittae of group 3 possessed a prominent rostrum, a deep V-shaped cisure and ornamented borders. Statistical analysis showed no significant differences in the E index of groups 1 and 2, while R values of the three groups were significantly different. These results were compared with 80 other species, belonging to 12 families, from the publisher literature. E and R values could be used to characterize the sagittae of the marine fish and could be considered a useful tool for fish ecology studies.     

基于线性混合模型的蓝点马鲛叉长和体重关系的月间及性别差异

为了研究蓝点马鲛（）生活史特征的异质性,根据2018年10月至2019年3月在东海外海渔场的拖网调查采样数据,对其叉长和体重关系的月间及性别差异进行了研究。依据收集的367尾蓝点马鲛样本,求得其叉长和体重关系（b的估计均值为2.794。本研究构建了广义线性模型和9个线性混合模型,用于研究蓝点马鲛的叉长和体重关系（）在时间及性别上的差异。贝叶斯信息准则（BIC）值和均方根误差值均表明,最复杂的线性混合模型（即月份和性别对两个参数P<0.01）。在最优模型中,a值则与此相反。本研究表明,月份和性别对蓝点马鲛叉长和体重关系具有显著的影响,线性混合模型能把月份和性别的异质性通过随机效应在单个模型中更准确、快速地体现,从而进一步证实了此模型在数据来源异质性研究中的优势。     

Tank culture of yellowfin tuna, Thunnus albacares: developing a spawning population for research purposes

A land-based culture facility for research on yellowfin tuna, Thunnus albacares, was developed at the Achotines Laboratory in the Republic of Panama. Six concrete tanks, and seawater and life support systems were built to maintain a yellowfin broodstock. On average, 50% of the yellowfin caught survived capture and handling, and approximately 30% became broodstock in Tank 1 (17 m diameter, 6 m depth) or Tank 2 (8.5 m diameter, 3 m depth). Each fish was tagged with a microchip implant tag, then weighed, measured, and injected with oxytetracycline (OTC) prior to stocking. Daily rations of primarily market squid, Loligo opalescens, and Pacific thread herring, Opisthonema spp., were regulated based on the feeding activity and energy requirements of the fish. Feeding activity of the broodstock decreased when the water temperatures decreased, and the fish ate decreasing daily rations and increasing calories with increasing size. Spawning occurred in both tanks within 6–8 months of capture. Spawning first occurred in Tank 1 when 24 females ranged in size from 6 to 16 kg and 65 to 93 cm fork length (FL). Spawning was intermittent during the first 2 months and occurred near daily thereafter. Tank size appeared to affect survival rates, the types of mortalities that occurred, and the growth of the fish. Survival rates after 1 year in captivity were higher, and the fish were larger, on average, in Tank 1 than in Tank 2. Most of the mortalities in Tank 1 were the result of wall strikes, which occurred more frequently after the fish reached their highest density of 0.64 kg m⁻³ and sizes greater than 96 cm FL and 19 kg. Non-linear growth models were fitted to the initial stocking sizes and final sizes of fish that died or were removed from Tank 1 during 1996–1999. Estimated growth rates in length (11–48 cm year⁻¹) for fish between 51 and 150 cm FL decreased with increasing length. Estimated growth rates in weight ranged from 9 to 19 kg year⁻¹ for fish less than 19 kg and 20–23 kg year⁻¹ for fish greater than 19 kg. The results of this work demonstrate that the stable environment of a land-based culture facility may be the preferred system for long-term maintenance of a yellowfin broodstock.     

Production of the Australian freshwater silver perch, Bidyanus bidyanus (Mitchell), at two densities in earthen ponds

Silver perch fingerlings (mean weight 15.3 g) were stocked at densities of 21 000 and 7000 fish/ha in six 0.1-ha earthen ponds and cultured for 10 months. There were three replicate ponds for each density. Ponds were aerated for at least 11 h a day and water was added every 4 weeks to replace that lost by evaporation and seepage. Fish were fed a formulated diet containing 35% crude protein at 4% body weight per day for the first 4 weeks and at rates up to 3% thereafter. The mean annual production rate of 9819 kg/ha of fish stocked at 21 000/ha was significantly higher (P < 0.01) than the annual rate of 3699 kg/ha of fish stocked at 7000/ha. The maximum daily production and growth rates achieved in any pond over a 1-month period during summer were 97.7 kg/ha and 5.1 g/fish, respectively. Stocking density did not significantly (P > 0.05) affect survival rate (treatment means for 21 000 and 7000 fish/ha: 92.8 and 94.7%), daily growth rate (0.2–3.3 and 0.3–3.4 g/fish), weight at harvest (434.9 and 473.2 g), food conversion ratio (1.9:1 and 1.8:1) and cost of feeding ($A1.55 and $A1.47/kg), suggesting that higher stocking densities and production rates are possible. Water temperatures ranged from 11.1 to 30.0 °C. Significantly (P < 0.05) slower growth during December was associated with concentrations of NH₃-N up to 0.65 mg/l. The results demonstrate that silver perch is an excellent species for semi-intensive culture in static earthen ponds with the potential to form the basis of a large industry in Australia, based on high-volume, relatively low-cost production.     

Using modified length–weight relationships to assess the condition of fish

This paper develops a model to replace classical fish models used to assess weight and condition. Results from two classical models (M=KL³ and M=aL^b) were compared to results from the proposed model (M=BL²H) to find which one could be most accurately applied over the widest range of genetic, morphological and physiological states. A large database of physical measurements was analyzed. The newly proposed model decreased the variability in calculating weight by between 38 and 44% as compared to the conventional fisheries model M=KL³ and was consistently more accurate than M=aL^b. Condition factor K of the two salmon species studied increased with fish weight, while the B factor was generally invariant. With regards to the salmonids, the B factor indicates a type of bodily streamlining, important for swimming functions. A low B factor indicates a fish has less cross sectional area per unit length than a high B value fish has. As well, the B factor can be used to calculate girth. The new model can, therefore, be used to more accurately predict weight and physical state than other conventional models can.     

Weight-length relationships for selected fish species of the small-scale demersal fisheries of the south and south-west coast of Portugal

Weight-length relationships were estimated for 31 fish species from the south and south-west coasts of Portugal. These species, which are mainly sea breams (Sparidae), dominate the catches of the main fixed gears used in this region and account for more than 50% and 40% of the total fish landings by weight and by value.     

Postlarval growth of the freshwater prawn, Macro-brachium australiense Holthuis, 1950 (Crustacea: Decapoda: Palaemonidae)

Growth of newly metamorphosed Macrobrachium australiense was monitored in individually vidually held and group held laboratory animals over 28 weeks. A breeding population was also established in a farm dam and the growth of young juvenile F₁ progeny was recorded for a similar time. Up to 26 moults were recorded in this period and intermoult periods increased with age. Prawns held in groups grew better than those held individually. Reproductive maturity was attained by some females in group held laboratory and farm dam animals within the observation period. Length—weight regression equations were calculated for laboratory maintained prawns, i.e., log W = 0.44+ 2.44 log L; log W = 0.45 + 2.46 log L for individually held males and females, respectively; log W = 0.27 + 2.70 log L for group held males and females.     

A comparison of the biological characteristics of Sarotherodon niloticus (L.) with those of S. aureus (Steindachner) and other tilapia of the delta and lower Nile

The tilapia species occurring in the lower Nile are Sarotherodon niloticus, S. aureus, S. galiaeus and Tilapia zillii. The distinguishing characteristics between the previously confused S. niloticus and S. aureus are summarised; from these there is no evidence of hybridization of the two species in natural populations in Egypt. The scales are used to estimate the growth rates of tilapia species in two coastal lakes and in both, S. niloticus grows faster than S. aureus after the first year. The possible factors causing the growth checks on the scales are considered.

The spawning season of S. niloticus appears to attain a discrete peak in April–May, whilst the spawning season of S. aureus extends from May to September with at least two actual spawnings within this period. Natural spawning cycles are compared with those observed in fish ponds.

A more extended spawning period of S. aureus may explain the reduced growth rate of the species after the first year. The spawning cycle of all species coincides with the temperature regime of the water bodies. The fecundities of S. aureus and S. niloticus are similar and are described by log F = log 1.33 + 2.23 log L, which suggests that small fishes produce more eggs per g body weight than large.

The salinity tolerance of the Nile tilapia can be ranked as T. zillii > S. galilaeus > S. aureus > S. niloticus. Evidence on chronic and acute effects of salinity are reviewed and upper estimates for salinities giving unimpeded growth are deduced as being, T. zillii 29‰, S. galilaeus 15–20‰, S. areus 10–15‰ and S. niloticus 5–10‰.     

Predicting the age of fish using general and generalized linear models of biometric data: A case study of two estuarine finfish from New South Wales, Australia

Direct ageing of fish can be a laborious and expensive task when age estimates from a large population are required, and often involves a degree of subjectivity. This study examined the application of general and generalized linear models that predict the age of fish from a range of efficiently and objectively measured covariates. The data sampled were from yellowfin bream (Acanthopagrus australis (Sparidae) (Owen, 1853)) and sand whiting (Sillago ciliata (Sillaginidae) Cuvier, 1829) populations from New South Wales, Australia. The covariates evaluated in the models were fish length, otolith weight, sex and location (the estuary from which the fish were sampled). Akaike Information Criteria were used for model selection and residual plots of the final models revealed a satisfactory fit to the observations. The best fitting model for each species included all covariates. An additional investigation considered whether general and generalized linear models that predict age from two different categories of biometric information outperform age-length keys with respect to subsequent estimates of total mortality from catch-curve analysis. The two categories of biometric information differed in the ease and cost with which the information could be collected. The first category only included fish length and location as covariates, whilst the second category also included otolith weight and sex. It was found that traditional age-length keys outperformed the predictive models that estimated age from only fish length and location, because the results from the models were prone to significant bias. However, when otolith weight and sex were added as covariates to the predictive models, some of them, including a generalized linear model with a Poisson-distributed response variable, performed similarly to the age-length key. Given that otolith weight and the sex of fish are cheaper to quantify than age from a sectioned otolith in many situations, general or generalized linear models may represent a cheaper and faster method of estimating mortality compared to age-length keys. Such models can also easily incorporate the influence of spatial, temporal and demographic variation.     

Age, growth, and reproduction of silky sharks, Carcharhinus falciformis, in northeastern Taiwan waters

To examine the age, growth and reproduction of silky sharks, Carcharhinus falciformis, in the waters off northeastern Taiwan, 469 specimens (213 females and 256 males) were collected from August 2000 to January 2002 at the Nanfanao fish market, northeastern Taiwan. The relationship between body weight (W) and total length (TL) for both sexes combined was expressed as: W = 2.92 × 10⁻⁶ TL^3.15 (n = 469, p < 0.01). The relationship between TL and vertebral centrum radius (R) for both sexes combined data was estimated as: TL = 25.979 + 18.197R (n = 250, p < 0.01). Growth bands (including translucent and opaque zones) in precaudal vertebrae formed once a year between December and January and were counted up to 11 and 14 for females and males, respectively. The von Bertalanffy growth function (VBGF) was used to model the observed length at age data. The sexes combined VBGF predicted an asymptotic length (L_∞) = 332.0 cm TL, growth coefficient (k) = 0.0838 year⁻¹, age at zero length (t₀) = −2.761 year (n = 250, p < 0.01). Size at 50% maturity for males was estimated to be 212.5 cm based on the logistic curve, which corresponded to 9.3 years. Females matured at 210–220 cm, which correspond to 9.2–10.2 years. The length at birth was estimated to be 63.5–75.5 cm TL. The number of embryos per litter was 8–10 and sex ratio of embryos was 1:1.     

邛海鱼类资源与空间分布的水声学调查探究

2013年12月和2015年9月,运用科学回声探测仪对邛海的鱼类资源量、空间分布进行了水声学法走航调查评估与分析。野外调查同时采用三层刺网、地笼等网具采样,结合湖岸鱼市及访问渔民等方式,对邛海现有鱼类组成进行分类整理。结果显示,邛海土著鱼类资源稀缺,太湖新银鱼(Neosalanx taihuensis)、鲢(Hypophthalmichthys molitrix)、鳙(Hypophthalmichthys nobilis)、鲤(Cyprinus carpio)、鲫(Carassius auratus)、歺又鱼(Hemiculter leucisculus)、红鳍原鲌(Cultrichthys erythropterus)等外来种为邛海鱼类组成优势种。2013年和2015年,邛海湖区鱼类平均密度分别为(0.1483±0.0715)尾/m~3和(0.1051±0.0279)尾/m~3,鱼类资源量分别为4.67×10~7尾和2.90×10~7尾;体长在13.3cm以上的鱼类资源量分别为9.9×10~6尾和7.6×10~6尾。鱼类水平方向主要集中分布于月亮湾-青龙寺以及青龙寺-邛海湾断面,垂直方向主要分布在中上层,且极显著高于底层(P0.01);调查表明,邛海鱼类的空间分布情况基本类似,4个调查断面及不同水层的鱼类种群分布类型均趋于成群分布(v1)。初步认为邛海鱼类水平分布差异主要是游客休闲娱乐、沿岸居民农业生产等人为干扰以及不同区域水深变化引起的,垂直水层分布差异与优势种鱼类的生态习性紧密相关。     

Habitat variability in young-of-the-year winter flounder, Pseudopleuronectes americanus, in Maine estuaries

The Magnuson-Stevens Fishery Conservation and Management Act defines essential fish habitat (EFH) as “the waters and substrate necessary to fish for spawning, breeding, feeding, and growth to maturity” and the protection of nursery areas has become a key element in US Federal fisheries management. Distribution and abundance by habitat for age-0, young-of-the-year (YOY) winter flounder, Pseudopleuronectes americanus, were compared for 28 Maine estuaries to help define essential fish habitat for this life history stage. The Maine coast was divided into three broad geographic zones based upon geological features and sampled over 5 consecutive years; during April–November of 2000 in the Mid-coast, in 2001 and 2002 along the Southwest coast and in 2003 and 2004 along the Eastern Maine coast. One beam trawl (2.0 m width, 3 mm mesh) sample was collected in one to four habitats in estuaries: eelgrass (Zostera marina), kelp (Laminaria longicruris), drift algae (Phyllophora sp.) and unvegetated sand/mud. Fish were sampled every 2 weeks, April–November 2000–2004. Abundance of YOY winter flounder was greatest in Mid-coast estuaries between Casco and Penobscot Bays and was significantly lower in Southwest and Eastern estuaries. Abundance was similar across all four habitats in Mid-coast estuaries in 2000. In the other years, YOY were found in higher abundance in eelgrass relative to other habitats. A logistic regression model based on nearshore habitat characteristics was developed to predict the distribution of this species along the three broad geological zones of the Maine coast with the physical and biological variables most important in discriminating between habitats with and without individual fish identified. This logistic regression model correctly classified winter flounder 72.4% of the time based on the year, zone, the physical habitat variables (temperature, salinity, depth) and the presence–absence of submerged aquatic vegetation (eelgrass, kelp or algae). These results indicate that the type of habitat most important to YOY winter flounder varies among estuaries and EFH for this species and life stage must be defined with care.