Low mortality rates of juvenile Pacific oysters in the German Wadden Sea are characteristic for invasive species: a reply to Beukema and Dekker

Abstract

Many marine invertebrate species have decreasing mortality rates with age (Gosselin and Qian 1997), which raised a question (Beukema and Decker 2009) on the low mortality we reported for juvenile PaciWc oysters in the East Frisian Wadden Sea (Schmidt et al. 2008). High fecundity is traditionally interpreted as a life history strategy to counteract high mortalities during early life stages, when unfavourable environmental conditions, competition or high predation can lead to a low recruitment success (Olafsson et al. 1994; Hunt and Mullineaux 2002). Beukema and Decker (2009) may therefore not be alone in doubting our values and we welcome the opportunity to clarify that such low juvenile mortalities are indeed a trait observable in the early stages of the invasion of marine invertebrates. Furthermore, as Gosselin and Qian (1997) also highlight, survival rates vary throughout the Wrst year which may necessitate diVerentiation in future studies. Prior to any evaluation of mortality, correct assessment of population parameters has to be assured. This can be confounded by sampling design and sampling errors, in particular where investigations are primarily Weld based, early life stages diYcult to detect, or where high mobility of juveniles can lead to import and export of individuals (Gosselin and Qian 1997). In our study, the establishment of PaciWc oysters (Crassostrea gigas) was studied along a 160 km stretch of intertidal coastline, with a dense net of sampling stations and a high rate of replication, yielding about 1,500 quadrate counts at each annual sampling occasion. All shell material found on the quadrates was visually examined for oyster spat in the Weld, carefully lifting the material oV the sediment, cleaning it from attached sediment and checking all sides of the shells. Even oysters as small as 1 mm could be identiWed, as they diVer in their colouration and structure from other epibionts (e.g. barnacles) or byssus threat attachments. We are thus conWdent that we accurately estimated the small oysters in the Weld. Retrieving material for cross-checks in the laboratory would not have been feasible as the overall goal of the monitoring scheme was to establish a long-term/large-scale data series on the population development of PaciWc oysters in this region. Therefore, Wxed GPS positions marked with bamboo stakes in the Weld were used to relocate the study sites. Beukema and Decker (2009) are concerned that morphological changes in mussel beds could account for variation in settlement substrates between years. Only well-established mussel beds with a stable topography were used for our surveys, documented through mussel monitoring (Herlyn and Millat 2004) and own assessment of the perimeter in 2003 and 2006 (Schmidt 2009; Wehrmann et al. 2009). Beukema and Decker (2009) are further concerned about the allocation to cohorts based on size-frequency distributions, yet the program FISAT II allows for potential overlap, applying standard procedures to diVerentiate cohorts using Bhattacharya’s method (Gayanilo et al. 2005). To corroborate the mortality rate based on the annual monitoring, we calculated mortality for the Wrst year from spatfall collectors deployed in the Weld (Schmidt 2009). Communicated by H.-D. Franke.