Biological Performance Measures: Fish Health as Reflected in the Prevalence of Fish Abnormalities

1
Biological Performance MeasuresFish Health as
Reflected in the Prevalence of Fish Abnormalities
Joan A. Browder, C. Mindy Nelson, Michael
Kandrashoff, Walter Kandrashoff, and Joan W.
Bernstein
2
Research conducted by

• Southeast Fisheries Science Center
• NOAA/National Marine Fisheries Service
• 75 Virginia Beach Drive
• Miami, FL 33149

3
Objectives

• Describe spatial and temporal patterns of
  prevalence of abnormal fish
• Characterize the abnormalities
• Explore potential causal factors
• Canal discharges
• Chemical contaminants
• Resuspended sediments
• Develop performance measures

4
Project Elements

• Field sampling to build a time series of
  prevalence data.
• Statistical analyses to describe sources of
  variability and determine the most probable
  causal factors.
• Collaborative ecotoxicity projects to support
  determination of causal factors.
• Formulation and application of performance
  measures to evaluate local and regional clean-up
  and restoration projects.

5
Field Sampling Design

• Two St. Lucie areas and a reference area
• One day each week in each area
• Hook and line
• Fish commonly caught on hook and line
• Record for each fish caught
• Species, total length, and type and number of
  abnormalities are recorded

6
(No Transcript)
7
Total caught 35,531
8
Major Species Caught

• Irish pompano
• Mangrove snapper
• Spottail pinfish
• Sand drum
• Saucereyed porgy
• White margate
• Black margate
• Lane snapper

• Blue runner
• Porkfish
• Crevalle jack
• Sheepshead
• Hardhead catfish
• Tomtate
• Pinfish
• Pigfish

9
Types of abnormalities

• Fin erosion
• Red spot
• Scale disorientation
• Parasite infestation
• Skeletal or fin anomalies (deformities)

10
Types of abnormalities (cont.)

• Chromatophore clusters
• Hemorrhage
• Ulcers
• Lumps (probable tumors)
• Lateral line anomalies (broken, branched, etc.)
• Eye abnormalities

11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
Statistical analyses of sources of variation in
prevalence of fish with abnormalities
General Linear Modeling
18
Model structure

• Fox-logit(abnormalities) Year Quarter Area
  Species

19
Standardized prevalence of abnormal fish, by
species
Sheepshead
Hardhead catfish
Spottail pinfish
Crevalle jack
Mangrove snapper
Irish pompano
20
Standardized prevalence of abnormal fish, by year
21
Standardized prevalence of abnormal fish, by
quarter
22
Standardized prevalence of abnormal fish, by area
23
(No Transcript)
24
Standardized prevalence of abnormality, by species
Crevalle jack
Hardhead catfish
Pigfish
Schoolmaster snapper
Mangrove snapper
Irish pompano
Sheepshead
Percent of captured
Type of abnormality
25
Standardized prevalence of fish with specific
abnormalities, by year
Year
26
Standardized prevalence of fish with specific
abnormalities, by quarter
27
Standardized prevalence of fish with specific
abnormalities, by area
28
Variation explained uniquely and in common in the
prevalence of specific abnormalities in
individual species
Mang-rove snap-per
Spottail pinfish
Irish pompano
29
Sources of variation

• AREA, YEAR, QUARTER, and SPECIES are all
  significant explanatory variables.
• SPECIES is the major factor explaining variation
  in abnormality prevalence.
• Species differ not only in overall prevalence of
  abnormalities but also in the type of
  abnormalities most prevalent.

30
Sources of variation

• YEAR explains more variation in prevalence when
  the types of abnormalities are examined
  separately.
• The types of abnormalities differ in temporal
  pattern and may relate differently to
  environmental variables.
• The temporal pattern of prevalence differs by
  species.

31
Analyses in relation to environmental variables

• Hydrologic variables

32
Freshwater discharges
33
Models (20)Fox-logit(abnorms)SpeciesAreaC23C2
4C24
Fox-logit(abnorms)SpeciesArea(C23C24C25)

• C23, C24, and C44 separate or summed
• Mean daily discharge
• Proportion of days flow above 75th percentile
• Proportion of days flow above 2000 cfs
• Lags current quarter (CQ), 1-month-(MQ),
  2-month-(MMQ), and 3-month-lagged, or previous,
  (PQ) quarters

34
(No Transcript)
35
(No Transcript)
36
Some Hydrologic Model Results
37
Tests of relationships of hydrologic variables to
specific abnormalities in individual species
38
Probability of hydrologic relationship with fin
erosion in Irish pompano
pp 39
Probability of hydrologic relationship with red
spot in Irish pompano
pp 40
Probability of hydrologic relationship with
chromatophore clusters in spottail pinfish
pp 41
Probability of hydrologic relationship with fin
erosion in spottail pinfish
pp 42
Probability of hydrologic relationship with red
spot in spottail pinfish
pp 43
Probability of hydrologic relationship with fin
erosion in mangrove snapper
pp 44
Variance explained uniquely and in common in fin
erosion in mangrove snapper
45
Conclusions from single-species analyses in
relation to hydrologic variables

• Fin erosion in mangrove snapper is strongly
  related to hydrologic variables.
• Strongest relationships are with the following
  model
• Fox-logit(FE)AREAC23C24C44
• where C23, C24, and C44 are Abov75thtile for
  the quarter lagged by 2 months

46
Summary of Findings

• SPECIES differences were the strongest
  differences in the analyses.
• Year differences were strengthened when
  abnormalities were examined separately.
• Separation by species further strengthened the
  year differences.

47
Findings--Continued

• The strongest hydrologic effects were when
  species and abnormalities were modeled separately.

48
Conclusion

• The prevalence of fin erosion in mangrove snapper
  is most strongly related to all three individual
  parameters of structure flows rather than to the
  sum of flows and most strongly to the parameter
  for the frequency of the highest flows.

49
Hypotheses
or
50
Promising candidate for performance measure to
assess response of St. Lucie system to water
quality aspects of hydrologic changes

• PREVALENCE OF FIN EROSION IN
• MANGROVE SNAPPER

51
We will continue to explore the combined dataset
as well as data for individual species and
abnormalities
52
Biological performance measures fill information
gaps left by chemical-oriented water quality
measures

• Toxicants have interactive effects that cant be
  evaluated only by measuring toxicant
  concentrations.
• There are so many chemicals and chemical analyses
  are so specific and costly that they cant
  measure every biologically harmful material that
  might be present.
• Some contaminants may have detrimental biological
  effects at concentrations too low to be measured.

53
Benefits of Monitoring the Prevalence of Abnormal
fish

• Biological performance measures are important to
  the overall success of the program.
• They provide a biological view of the
  effectiveness of the SLRIT Program.
• A decrease in the prevalence of abnormal fish
  would be a sign of success meaningful to the
  Public and the Legislature.

54
Addendum

55
Models (20)Fox-logit(abnorms)SpeciesAreaC23C2
4C44
Fox-logit(abnorms)SpeciesArea(C23C24C44)

• C23, C24, and C44 separate or summed
• Mean daily discharge
• Proportion of days flow above 75th percentile
• Proportion of days flow above 2000 cfs
• Lags current quarter (CQ), 1-month-(MQ),
  2-month-(MMQ), and 3-month-lagged, or previous,
  (PQ) quarters

56
Hypotheses
or
or