Cod and fmounder (seasonal) habitat mapping Berthe Vastenhoud - - PowerPoint PPT Presentation

cod and fmounder seasonal habitat mapping
SMART_READER_LITE
LIVE PREVIEW

Cod and fmounder (seasonal) habitat mapping Berthe Vastenhoud - - PowerPoint PPT Presentation

Cod and fmounder (seasonal) habitat mapping Berthe Vastenhoud Alexandros Kokkalis Sofja Ferreira Elliot Brown Josianne G. Stturup 12 December 2018 Map fjsh habitats to contribute to the FishHab II project knowledge basis for Maritjme


slide-1
SLIDE 1

Cod and fmounder (seasonal) habitat mapping

Berthe Vastenhoud Alexandros Kokkalis Sofja Ferreira Elliot Brown Josianne G. Støturup

12 December 2018

slide-2
SLIDE 2

FishHab II project

Map fjsh habitats to contribute to the knowledge basis for Maritjme Spatjal Planning 10 commercially important species in the inner Danish waters

slide-3
SLIDE 3

Objectives

Mapping of habitats Explore seasonality of the spatjal distributjons, relatjve to life stage (juvenile, adult) habitats

slide-4
SLIDE 4

KASU dataset

Sampled twice a year Q1 : February/March Q4 : October/November/December

  • 1994-2017

Danish part of Baltjc Internatjonal Trawl Survey (BITS)

SEASONAL DISTRIBUTIONS

Fitued maturatjon ogive Additjonal sampling campaign in 2016 (Q3) 2005 – 2015: checked to validate 2016

  • Number of individuals per length

6 maturity stages

1. Virgin - Immature 2. Maturing - Mature 3. Spawning - Mature 4. Spent - Mature 5. Restjng- Mature 6. Abnormal - Mature Separate analysis juveniles/adults

slide-5
SLIDE 5

Distributions

  • Seasonal
  • Physical characteristjcs
  • Temperature
  • Oxygen
  • Salinity
  • Current speed
  • Sediment type
  • Biological characteristjcs
  • Food
  • Predatjon

Habitat distributjon models:

  • identjfy factors infmuencing species'
  • ccurrence
  • visualize the spatjal arrangement of

habitats

slide-6
SLIDE 6

Modeling

Main model including all possible (environmental) variables, reduce stepwise removing the least signifjcant variable untjl all are signifjcant

Generalised Additjve Models (GAMs)

(Separate models for juveniles and adults) Environmental factors

  • Temperature
  • Oxygen
  • Salinity
  • Current speed
  • Depth
  • Sediment type

Spatjotemporal factors

  • Quarter
  • Seasonal component
  • Sampling positjon

Ofgset efgort Number of individuals

slide-7
SLIDE 7

Adults

slide-8
SLIDE 8

Results - Adults

Quantjles: cut points dividing the range of observatjons into intervals with equal probabilitjes

Signifjcant variables:

  • Salinity
  • Oxygen
  • Temperature
  • Quarter
slide-9
SLIDE 9

Uncertainty - Adults

slide-10
SLIDE 10

Declining stock: many of the earlier spawning sites are no longer observed & changed to:

  • ofg the coast of Falkenberg
  • very close to the entrance to the

Sound

(Vitale et al., 2008; Börjesson et al., 2013)

Spawning: Q1 (& Q2)

in the Katuegat and the Sound

  • north of Læsø
  • southern Katuegat including

Skälderviken and Laholmsbugten

  • Great Belt

Western Baltjc stock: Kiel Bight

?

slide-11
SLIDE 11

Can we use 2016 as an example for other years?

Model for 2005-2015 Similar selectjon process to 2016 model Predict numbers at statjons in 2016

Compare

  • Center of mass

Average of weighted positjon coordinates of the distributed mass

  • standard deviatjon ellipse (SDE)

Dispersion of weighted point observatjons along two

  • rthogonal axes
slide-12
SLIDE 12

Juveniles

slide-13
SLIDE 13

Signifjcant environmental variables:

  • Salinity
  • Oxygen
  • Temperature
  • Depth
  • Quarter

Results - Juveniles

slide-14
SLIDE 14

Uncertainty - Juveniles

slide-15
SLIDE 15

Results - Juveniles

Associated to habitats that provide refuge (greater structural complexity than sand/mud)

  • Sediment type not signifjcant
  • Too few observatjons in coarse

substrate/mixed sediment

  • Quality sediment data could be

improved

Other studies:

  • Surface salinity
  • Depth
  • Temperature < 16°C
slide-16
SLIDE 16

Can we use 2016 as an example for other years?

Compare

  • Center of mass

Average of weighted positjon coordinates of the distributed mass

  • standard deviatjon ellipse (SDE)

Dispersion of weighted point observatjons along two orthogonal axes

slide-17
SLIDE 17

Adults vs. Juveniles ADULTS JUVENILES

slide-18
SLIDE 18

Conclusions

What has been done?

  • Mapping habitats and exploring seasonal distributjons
  • Difgerences between seasons and between juveniles and adults
  • Modeling based only on one year (2016)
  • Fittjng a model for the years 2005-2015 predicts 2016 quite well

Next steps

  • Extend analysis spatjally and temporally
  • Improve modeling of spatjal auto-correlatjon (e.g. LGCP model)
  • Length-based model
slide-19
SLIDE 19

Flounder - Adults

slide-20
SLIDE 20

Results - Adults

Quantjles: cut points dividing the range of observatjons into intervals with equal probabilitjes

Signifjcant variables:

  • Salinity
  • Depth
  • Temperature
  • Quarter
slide-21
SLIDE 21

Uncertainty - Adults

slide-22
SLIDE 22

Flounder - Juveniles

slide-23
SLIDE 23

Signifjcant variables:

  • Sediment type
  • Oxygen
  • Temperature
  • Depth
  • Quarter

Results - Juveniles

slide-24
SLIDE 24

Uncertainty - Juveniles