Knowledge of the distribution, abundance, and transport of bivalve larvae is limited due to their small size, similar morphologies between species, and lack of an automated approach for identification. The objective of this research is to evaluate and improve the accuracy of ShellBi, a novel supervised image classification method that uses birefringence patterns on the shells of bivalve larvae under polarized light to identify species. The performance of the ShellBi method was tested by rearing Crassostrea virginica (eastern oyster) larvae at different temperatures (21.3 and 27.5°C) and salinities (10.3, 14.1, 14.4, and 20.5). Differences in rearing temperatures resulted in differences in classification accuracy, as did large variations in salinity (≥10 units). Classification accuracies increased from 67-88% to 97-99% when training sets included images of larvae reared in conditions similar to those of the larvae being classified. Additional tests indicate that misclassification rates ranged from 0 to 13% for false positives and from 0 to 22% for false negatives, depending on the proportion of oyster larvae in the sample. Results suggest that this technique could be applied to field samples with high accuracy as long as the images that are used to make classifications include larvae that were reared in conditions that are similar to those in situ. In addition, these findings demonstrate that the ShellBi method can be used to measure and identify bivalve larvae in a different system than the one for which it was developed, suggesting that the method has broad applicability in marine and estuarine systems.

In April 2004, triploid native (Crassostrea virginica) and nonnative (Crassostrea ariakensis) oysters were deployed in cages at four sites along a salinity gradient in Chesapeake Bay. In Maryland, the lowest salinity site was located in the Severn River and two low to mid-salinity sites were located in the Choptank and Patuxent Rivers. The highest salinity site was located in the York River in Virginia. Growth, disease acquisition, and mortality were measured in the deployed oysters through August 2006. Although ANOVA revealed that the nonnative oysters were significantly larger at the end of the experiment than the native oysters at all sites, the differences were much greater at the Virginia site (59 mm) than in Maryland waters (9-23 mm). With the exception of C. ariakensis in the Severn River, Perkinsus marinus infected both species at all sites. Prevalences and weighted prevalences in both species remained relatively low throughout the experiment, but native oysters consistently acquired higher prevalences and weighted prevalences than C. ariakensis by August 2006. With the exception of several mortality-inducing events including winter freezing and hypoxic exposure, mortality was generally low in both species. No disease-related mortality was suspected in either species given the low weighted prevalences observed. In the York River, where a substantial natural spatfall occurred in 2004, more native spat were found on C. ariakensis than on C. virginica. To our knowledge, this is the first comparison of triploid C. ariakensis to triploid C. virginica conducted in the field. Because we did not observe substantial disease-related mortality, it is too soon to draw conclusions regarding the disease tolerance of C. ariakensis in the field or its viability as a replacement for the native species.

No abstract

Bivalve larvae are small (50–400 μm) and difficult to identify using standard microscopy, thus limiting inferences from samples collected in the field. With the advent of ShellBi, an image analysis technique, accurate identification of bivalve larvae is now possible but rapid image acquisition and processing remains a challenge. The objectives of this research were to (1) develop a benchtop automated image acquisition system for use with ShellBi, (2) evaluate the system, and (3) create a protocol that would maintain high classification accuracies for larvae of the eastern oyster, Crassostrea virginica. The automated system decreased image acquisition time from 2–13 h to 46 min per slide and resulted in the highest classification accuracies at the lowest tested magnification (7X) and shortest image acquisition time (46 min). Quality control tests indicated that classification accuracies were sensitive to camera and light source settings and that measuring changes in light source and color channel intensities over time was an important part of quality control during routine operations. Validation experiments indicated that under proper settings, automated image acquisition coupled with ShellBi could rapidly classify C. virginica larvae with high accuracies (80–93%). Results suggest that this automated image acquisition system coupled with ShellBi can be used to rapidly image plankton samples and classify C. virginica larvae allowing for expanded capability to understand bivalve larval ecology in the field. Additionally, the automated system has application for rapidly imaging other planktonic organisms at high magnification.