Population and reproductive dynamics of the polychaete Pygospio elegans in a boreal estuary complex

Abstract

Pygospio elegans is an opportunistic, wide-spread spionid polychaete that reproduces asexually via fragmentation and can produce benthic and pelagic larvae, hence combining different developmental modes in one species. We documented the density, size distribution, and reproductive activity of P. elegans at four sites in the Danish Isefjord-Roskilde Fjord estuary complex, where all modes of reproduction were reported. We compared population dynamics of this species to environmental parameters such as salinity, temperature, and sediment characteristics (grain size, sorting, porosity, water content, organic content, C/N). We observed that new cohorts-resulting either from sexual or asexual reproduction-appeared in spring and fall, and old ones disappeared in late summer and winter. Sexual reproduction occurred from September until May, and although their timing was variable, there were two reproductive peaks at three sites. At those sites, we also observed a switch in larval developmental mode. Asexual reproduction peaked in April. While the seasonal dynamics can be related to temperature to a large extent, the differences in population dynamics among sites also correlated with sediment structure and salinity. Populations from sites with coarse and heterogeneous sediment had high levels of sexual reproduction. At the site with lower salinity, intermediate and benthic larvae were present during winter in contrast to pelagic larvae found at the other sites. However, we could not identify one clear environmental factor determining the mode of development. At present, it remains unclear to what degree genetic background contributes to mode of development. Hence, whether the differences in developmental mode are the result of genetically different cohorts will be further investigated.Additional key words: life history, poecilogony, development, spatiotemporal variation environmental impact Invertebrate Biology 135(4): 370-384.

Seasonal genetic variation associated with population dynamics of a poecilogonous polychaete worm

“…Cohorts based on size of the worms were distinguished previously (Thonig et al., ), and the genetic composition of these cohorts is shown in Figure and Table . Vellerup is not included, as we could not distinguish cohorts based on size at this location (see Thonig et al., for more details). At the other sites, about 50 individuals per site could not be assigned to a distinct size cohort due to overlapping size ranges of the cohorts (cohort not characterized—n.c.).…”

“…Assignment of individuals to different size cohorts (according to Thonig et al., ) and genetic clusters (1, 2, or 3) as determined with Structure when membership coefficient was larger than 60%. Cluster 0 indicates individuals that could not be assigned to a distinct genetic cluster (membership coefficient was less than 60%).…”

“…We investigated the genetic structure of the poecilogonous polychaete P. elegans from four sites in the Isefjord‐Roskilde‐Fjord estuary complex using six temporal samples collected over 1 year. We aimed to evaluate the relationship between the genotype of sampled individuals and previously described differences in population dynamics in these populations (see Thonig et al., ). We observed genetic differences between the sites as well as changes during the year at two of the sites.…”

“…Choosing higher membership thresholds resulted in an increase of unassigned individuals, but did not change the trends. Additionally, when possible, individuals used in this study were assigned to distinct cohorts based on their size (Thonig et al., ). We analyzed whether the different size cohorts are composed of individuals assigned to distinct genetic clusters.…”

“…We compared the observed genetic structure at four time points (March, May, August, and November) with the environmental parameters described previously (Thonig et al., ) using Primer‐E v.6 (Clarke & Gorley, ). For that purpose, genetic differentiation calculated with the R package diveRsity (Keenan et al., ) was input in Primer‐E as a dissimilarity matrix.…”