Measuring Larval Dispersal

1
Measuring Larval Dispersal

• Modeling Approaches

2

• Options models, genetics, microchemistry
• Genetics
• poor interannual and interdecadal resolution
  (Largier 2003).
• 10 migrants/gen. enough to prevent all but minor
  gene frequency differences (Palumbi 2003)
• Physical models and Biological balance
• large spatial scales (10s 100s of km) for most
  physical models.
• Finer spatial scales for biological processes.
  Many organisms capable of sophisticated
  behavioral responses to sensory cues, and are not
  passive drifters.

3
Dispersal and Oceanography

• Dispersal heavily influenced by oceanography
• Biological models constant oceanographic
  environment
• Phys models specific local oceanographic
  patterns, simplistic biological behavior
• (trade-off)

Bhaud 2000
4
The eastern Straits

• Processes Drift of debris illustrates passive
  transport and accumulation sites
• Spatial extent of dispersal, accumulation away
  from shores, follow specific transport
  trajectories through water

Ebbesmeyer et al. 1991
5
(No Transcript)
6
Klinger and Ebbesmeyer 2001

• Larval linkage potential in the SJA drifters
  model
• Q exploring the pattern of larval transport
• Larval export hypothesis
• MPA network design
• Approach drifters to infer larval transport by
  surface currents (general)
• Over 6 months
• Results
• EB as larval pool (95 recovered there)
• Selectivity in accumulation (70 recovered on 15
  of shoreline)
• Also as to specific areas of accumulation

7
GNOME

• General NOAA Oil Spill Modeling Environment
  (GNOME)
• Strengths broad temporal scale and specific
  current patterns
• Weaknesses no wind on surface mixed layer,
  reference pt (reliable timescale) , passive
  transport (simple biology)
• What it has been used for larval dispersal in
  the SJA and around the Strait of Juan de Fuca
• Copper rockfish in the San Juans
• -- Spatial extent and transport patterns in the
  Strait of Juan de Fuca

8
GNOME to GIS

• Study area .05 x .05 dec. degrees 626 grid
  cells
• 4 days
• 16 release sites
• 2 ref. tide stations
• Temporal Variation 2001-2003, April June
• Level of accumulation calculate in each grid
  cell and radius
• Caveats same!

G
F
G
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
G
F
9
Connectivity
10
Conclusions Engie

• Is there a larval pool? Yes
• Does the level of accumulation correspond to
  drifter studies? Yes
• What is the spatial extent of dispersal?
• Basin-wide, but localized
• Management Implications the design of MPAs
• Spacing (dispersal distance)
• Temporal and Spatial Variability 3 yrs, 3 mo.s
  (constant)

11
Kendall et al. 2003 copper rockfish

• Q What are the effects of MPAs on larval output?
• Examines assumptions from larval-export
  hypothesis
• Can we observe an effect through ichthyoplankton
  sampling?
• If anticipated increase in larvae from MPAs.
• Simulate a larval sampling program
• One of most abundant fished species
• Stationary adults
• Knowledge of dispersal
• Virtually nothing on the small scale
  (distribution, behavior)
• 1-2 mo. larval duration, juv.s settle on kelp or
  soft-bottom habitats
• Move to rocky areas as they grow

12
Sampling area 2 km x 2 km 359 grid
cells. Sampling station in middle of each cell.
Compute larval distribution in and outside of
reserves Outside 90 of land and fish. Total
fish 2.141 x 106 (Palsson), 5050 sex ratio,
avg. fecundity (2 sources), surface area 3.906
x 107 larvae/km2 Inside 55-fold increase (lower
adult mortality and larger fish) 2.900 x 1011
16 sites in SJ Skagit Co. Active transport
the potential is recognized
13
Trajectory Analysis Planner (TAPS)

• Prediction after 3 days
• 72 hrs after spawning from 16 proposed MPAs
• N 100
• 1000 particles from each MPA
• Each particle 181,279 larvae
• Over 24 hrs
• Total
• of larvae passing through each cell

Kendall et al. 2003
14

• Conclusions
• Is there potential for MPAs to have an effect?
  Yes (long timescales)
• Can differences in larval abundance be detected?
  Yes

15
English sole