Differences In Atlantic Sprattus Sprattus Growth Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

A project involving Otolith microstructure was undertaken to examine if Sprat Sprattus sprattus from three locations throughout the Atlantic have different otolith structure possibly due to geological distribution and changes in diet or spawning date. The daily nature of otolith growth increments provides a reliable source of data about the life history of each sprat. This data was corroborated using image analysis and statistical techniques to provide information on these sprat populations. It was determined that there is a significant difference in otolith growth between these populations.

The sprat Sprattus sprattus (Linnacus 1758) is a small shoaling fish with multiple batch spawning species that is pelagic throughout its life. This paper aims to find if Sprat from three different areas have otolith growth and morphology differences and whether these findings may warrant studies on the potential possibility of different sub-populations within the species and implications of these studies.

This project will also aim to establish that if there is significant growth and differences in Sprat, are these differences in response to different food availability and seasonal variations across the Atlantic.

Fisheries science is a field where managing and understanding the fishery population requires an understanding of the whole food web and its trophic levels. Sprat aren't an important fishery for human consumption in most of Europe but have been shown to be important prey species for Cod and Salmon, two very important commercial fisheries and are also a important constituent in many fish feeds and baits. So as a result Sprat should be studied to learn more about how they fit into their habitats and whether various populations of Sprat have different growth and breeding cycles to help give a fisheries manager better understanding of when and how Sprat populations affect their fisheries.

Otoliths are the ear bones of fish made by successive layers of calcium carbonate laid on the otolith as the fish grows and are generally held to be one of the most effective dating tools in identifying a fish's age. Otolith microstructure examination is now a preferred tool for the study of most fish species and can provide a wealth of information to a biologist such as age determination, daily growth rates estimations, mortality, migration and environmental history, competition, abundance and taxonomy.

Methods and Materials

Sprat were collected from 2010 to 2011 by several groups such as the Marine Institute and Acoustic survey team. Sprats were then measures to the nearest 0.5 cm, and were weighed to the nearest 1 g. Sex and maturity stages were assigned by visual inspection of the gonads, and otoliths were removed. Otoliths were viewed in water under a Olympus Stereoscopic microscope and the opaque (summer) and translucent (winter) bands were enumerated to derive estimates of age. After age determination, subsamples of the sprat were selected for otolith microstructure analysis. In this case two hauls by the Marine Institute and one from the acoustic survey were used. The hauls chosen are indicated in Figure 1. along with approximate positions of the fish hauls.

Figure 1. Haul locations from around Ireland.

Over 70 random otoliths from the three hauls were selected and then cleaned of any tissue adhering to them, and stored in labbeled grid boxs until they are prepared for mounting onto a microscope slide by fixing them in place with an crystal bond heated to over 160oc using a fine forceps to ensure the saggital curve remains facing upwards. Once the otolith is mounted wait for the slide to cool down and check the slides to ensure a relatively even covering.To begin grinding and polishing the otoliths, a marble slab was utilised with 30 micron calcinated aluminium oxide dust mixed with deionized water. Creating figures of 8 by lightly gliding the slides through the solution gradually grinds away both crystal bond and the otolith. After about 30 seconds of grinding the otolith slide would be cleaned and polished with a chalk block to remove scratches and examined under a Lieca GME light microscope to determine how many microns of resin and otolith remained until the daily growth increments and larval core are uncovered. As the otolith nears completion of the grinding process progressively smaller grades of aluminium oxide dust are used such as 15 or 3 micon dust to ensure a polished finish, Once againa chalk block was used to remove any scratches to ensure unimpaired visibility to the microstructure of the otolith the otolith is then given a final check at 1000X to ensure good quaility data can be obtained from it and then it is placedback into the labbeled grid box.

The finished Otolith slides are then examined under a Olympus BX-51 light microscope at 1000X. Otolith measurements are made by using a Retina 2000r Qimagine camera which is then linked to a desktop running Image proplus 6.3 program. Once all the measurements and lens are calibrated images of Otoliths (Figure2.) are captured and adjusted to provide the best possible visibitility. Analysis can begin by measuring increment width from spawning date to until up to one hundred rings are covered along the longest growth axis visible.

(Figure 2.) Otolith no. 11 from haul 197 at 1000X. The red line is 108 microns long.

The data gathered is then transferred to an excel sheet. From this data a further subsample of 10 otoliths from each of the three haul is used for statistical analysis. Mean increment widths from increments 20-30 were chosen for use in these tests. Once the data is checked for homogeneity of variances and normality using Minitab 15 it is placed into a one-way analysis of variance (ANOVA) that tests the hypothesis that the means of several populations are equal. This method is an extension of the two-sample t-test, specifically for the case where the population variances are assumed to be equal. A one-way analysis of variance requires the following a response or measurement taken from the units sampled in this case mean otolith width and a factor, or discrete variable that is altered systematically in this case the Haul site. The different values chosen for the factor variable are called levels of the factor. Each level of the factor in the analysis corresponds to a larger population with its own mean. The sample mean is an estimate of the level mean for the whole population. A one-way ANOVA can be used to tell you if there are statistically significant differences among the level means. The null hypothesis for the test is that all population means (level means) are the same. The alternative hypothesis is that one or more population means differ from the others and could provide the statistical evidence that Sprat from different areas have different feeding strategies and breeding cycles and as a result have different otolith increments.


Comparing the mean otolith width for rings 20 to 30.Otolith width is the quantitative response variable

Haul is the qualitative predicting factor. Factor has three groups/levels.

Ho: Var1=Var2=Var3

No difference of variance between the 3 populations so variances are homogenous

Ha: Oposite of Var1=Var2=Var3

No difference of variance between the 3 populations so variances are NOT homogenous

Rule: Accept Ho of no difference, if B-test < Chi^2(0.05,K-1), K = 3

Alternatively, accept Ho if p-value > 0.05

Data is normal based on Ryan-joiner tests.

Test for Equal Variances: Mean otolith width (um) versus Haul

95% Bonferroni confidence intervals for standard deviations

Haul N Lower StDev Upper

H-156 10 1.91727 3.00966 6.40481

H-197 10 0.27741 0.43547 0.92671

H-9 10 1.41917 2.22776 4.74085

Bartlett's Test (Normal Distribution)

Test statistic = 21.64, p-value = 0.000

Levine's Test (Any Continuous Distribution)

Test statistic = 9.16, p-value = 0.001

Accept Ho that samples are taken from populations with equal variances

B-test =21.64.00 < Chi^2 (0.05, 3-1) = 5.08

p - value = 5.08 >> 0.001

Accept Ho that samples are taken from populations with equal variances


Data assumptions are met can proceed with parametric ANOVA.

One-way ANOVA

Data assumptions:

Samples from populations with normal pdf and equal variance are met.

Mean otolith width is the quantitative response variable

Haul type is the qualitative predicting factor

Factor has three groups/levels.

Running the test.

Fixed factor Anova

k=3; N= 10 x 3 = 30

Ho: No difference in response variable in the three hauls.

Ha: There is a difference in the response variable in the three hauls.

Rule: Accept Ho IF

F-test < F-critical at (0.05, k-1, N-k) df, Or if p-value >0.05

One-way ANOVA: Mean otolith width (um) versus Haul

Source DF SS MS F P

Haul 2 352.48 176.24 37.21 0.000

Error 27 127.90 4.74

Total 29 480.38

S = 2.176 R-Sq = 73.38% R-Sq(adj) = 71.40%

Individual 95% CIs For Mean Based on Pooled StDev

Level N Mean StDev +---------+---------+---------+---------

H-156 10 5.476 3.010 (---*----)

H-197 10 1.505 0.435 (----*----)

H-9 10 9.897 2.228 (----*----)


0.0 3.0 6.0 9.0

Pooled StDev = 2.176

S is the estimated measure of the within sample standard deviation.

Note that S^2 = MS Error.

This is equivalent to the pooled standard deviation used n calculation the individuals confidence interval levels.

R-sq the coefficient of determination indicates how much variation in the response is explained by the model.

The higher the R^2, the better the model fits your data.

Statistical interpretation: Reject Ho and Accept Ha that there are significant differences in otolith increment width between hauls.

Tukey 95% Simultaneous Confidence Intervals

All Pairwise Comparisons among Levels of Haul

Individual confidence level = 98.04%

Haul = H-156 subtracted from:

Haul Lower Center Upper --------+---------+---------+---------+-

H-197 -6.387 -3.971 -1.555 (---*---)

H-9 2.006 4.421 6.837 (---*---)


-6.0 0.0 6.0 12.0

Haul = H-197 subtracted from:

Haul Lower Center Upper --------+---------+---------+---------+-

H-9 5.976 8.392 10.808 (---*---)


-6.0 0.0 6.0 12.0

Individual Value Plot of Mean otolith width (um) vs Haul

Boxplot of Mean otolith width (um)

Residual Plots for Mean otolith width (um)

Trellis diagram

H-156 H-197 H-9

Level N Mean StDev 5.476±3.010 1.505±0.435 9.897±2.228

H-156 10 5.476 3.010 0 -1.505* 4.421*

H-197 10 1.505 0.435 3.971* 0 8.392*

H-9 10 9.897 2.228 -4.421* -8.392* 0

Significant differences denoted by *


Sprats in the Northeast Atlantic are currently assessed as a single stock and very little is known about how many are discarded or used as bait leading to unknown fishing mortality and the potential for a loss of a sub-population that fish such as Cod or Salmon may rely upon for energy causing further losses in the fishery. As of February 2011 The Celtic Sea Herring Management Advisory Committee (CSHMAC) has entered its herring, sprat and sardine fisheries for MSC assessment. The herring, sprat and sardine fisheries join a growing number of northern Europe pelagic fisheries seeking MSC certification and, if successful, the sprat fishery will be the first certified sprat fishery. Failure to account for structure in the stock could create bias or poor modelling in its assessment and leading to inaccurate estimates foe quotas and biomass for Sprat, and the possible extinction of a subpopulation such as what the herring fishery of the Celtic sea, Which is only currently in the later stages of a recovery plan following a collapse of the herring stock during the late 1990's through overfishing and poor understanding of the population dynamics in Herring. Following the introduction of spawning area closures, a ban on roe fishing and severe quota restrictions, the Celtic Sea herring stock is expected to have reached full recovery by 2012. The CSHMAC vessel group catches 90 per cent of Ireland's quotas for the species with total landings in 2010 around 7,900MT. Most of the landings go to markets in Germany, Poland, Lithuania and Russia with products  that include frozen fillets, dressed herring, whole frozen herring, pickled fillets and 'roll mops' providing a valuable market and jobs for Irish fishermen.

The factors responsible for the variability observed in otolith growth increments are not known at this time. One factor that may account for such distinct groups to appear in the results would be despite the surveys large amount of hauls they only caught fish aged at 0-2 years old and as a result of the highly variable nature in the density of Sprats main prey of copepods and nauplii may result in different groups of Sprat having "good" or "bad" seasons for feeding affecting their otolith growth positively or negatively. It is unlikely that the spatial variability in otolith growth was greatly influenced by measurement error along a restricted increment sequence. Although fish from different areas show some overlap in increment width it would seem the three populations are distinct. My own recommendation would be for a much larger sample size for analysis across many year classes to provide better quality data.