Small ocean currents can make a big difference in population connectivity

Teske, P. R.; Sandoval-Castillo, J.; van Sebille, E.; Waters, J.; Beheregaray, L. B. On-shelf larval retention limits population connectivity in a coastal broadcast spawner. Marine Ecology Progress Series 532: 1-12, 2015. doi: 10.3354/meps11362

A picture is worth a thousand words

Schutte 8.24 1

Why we care

Understanding metapopulation dynamics can (among other things) help predict species vulnerability, response to disturbance, and the spread of invasive species or diseases. It is also critical to designing effective protection for species and ecosystems. For example, between-population connectivity can inform whether it is better to have more small marine protected areas spaced along a coastline or one large protected area.

Population connectivity in marine organisms is often inextricably linked to ocean currents. Many marine species spend some portion of their lifespan pelagically, drifting in the water column. This allows even sessile (non-moving) adult organisms to avoid inbreeding by sending their offspring away to meet and combine with individuals that have novel genes. Organisms may be pelagic when they are just sperm and eggs (corals mass spawn this way) or later in development (as with seeds from marine plants).

Broadcast spawners are those organisms that release sperm and eggs into the water column, allowing fertilization to occur externally. The larval stages of such spawners spend wildly different amounts of time in the water column and may be pelagic for weeks or even months. Time spent in the water column intuitively seems that it should be an indicator of the relative distance an organism is moved. For example, you would expect an organism that drifts for two days to settle closer to its parents than an organism that drifts for two weeks. But instead, topographical features are increasingly being identified as an important influence on how far pelagic stages drift before settling. This paper describes why the typically-studied, large-scale topographical features are not the only ones that influence metapopulation dynamics.


Schutte 8.24 2
A portion of the southern Australia coastline where Nerita atramentosa was sampled. Black arrows show the direction of on-shelf currents while red arrows show those farther away (the “boundary currents”)

The authors collected tissue samples from 870 Nerita atramentosa snails, which live in rocky intertidal habitats in southern Australia. Samples were collected from 30 – 50 snails at each of 21 sites. Microsatellites were genotyped and analyzed using Bayesian modeling.

Dispersal was quantitatively described by modeling genetic spatial autocorrelation. If individuals closer to one another are more closely related than individuals farther from one another, they have positive spatial autocorrelation. Similarly, a more positive spatial autocorrelation value than expected indicates that dispersal is not as great as expected, or is restricted.

Three types of output were created from ocean current models: connectivity between each pair of sites (all sites were paired with all other sites), the percentage of larvae that did not leave the continental shelf, and how many larvae released from a given site returned to the coast to settle.

Multiple regression was used to test for correlation between genetic composition and three types of environmental features: distance, sea surface temperature gradients, and ocean circulation.


Though N. atramentosa larvae spend months in the water column, genetic spatial autocorrelation was higher than expected over short distances. Most larvae settle close to their parents and genetic variation was non-randomly distributed.

Schutte 8.24 3
a) The number of Nerita atramentosa larvae released from each of the 21 sites sampled for this study according to month. b) The percentage of those larvae that remained in on-shelf currents (did not reach the boundary current and did not enter water greater than 100 m deep). Note the correlation between the percentage of larvae retained in on-shelf currents and the number of larvae returning to the coast (the similarity between panels a and b).

Animated simulations showed that during most of the summer and through much of the range analyzed, on-shelf circulation more significantly affected gene flow than did boundary currents. Larvae were much more likely to return to the coast when they did not leave the on-shelf currents.

The correlation between oceanographic features and genetic composition varied temporally. However, models showed that the distance between sites explained genetic composition better than data from the boundary currents did. Ocean temperatures did not successfully predict genetic structure.

This study highlights the necessity of considering the details of ocean currents when describing population connectivity. The major ocean currents (boundary currents) in the area did not explain genetic composition as well as larval geography: whether larvae reached these boundary currents or whether they were retained in on-shelf currents moving in the opposite direction determined how far they moved while in the water column. This dispersal pattern and its implications should apply to many other coastal marine species, making these results relevant to both scientists and policymakers.

Have your say

Do you know which baby animals you’re swimming with in the ocean where you live or visit? Would you like help figuring this out? Let me know in the comments!

Leave a Reply

Your email address will not be published.