The deposition of surficial sediments many centimeters below the sediment-water interface due to the reworking activities of organisms is a potentially important but easily overlooked process in marine sediments. This kind of downward particle transport is difficult to observe in the laboratory or in the field but it has important consequences for bioturbation rates and sediment geochemistry. It is also much more likely to be size dependent than other sediment-mixing mechanisms, such as conveyor-belt feeding, and may also explain some subsurface maxima observed in sediment chemical profiles.We examined the mechanisms behind downward particle transport in Boston Harbor. Laboratory observations indicated that a large cirratulid polychaete, Cirriformia grandis, collected particles (glass beads) near the sediment surface and deposited them at depth. Furthermore, particle collection by this species was size dependent. C. grandis preferred smaller particles in the 16-to 32-m size range relative to larger particles.A mathematical model was developed to simulate the feeding and burrowing mechanisms of C. grandis and to predict the vertical profiles of tracer particles of assorted sizes in the field. The model was tested by comparing predicted profiles with profiles of glass beads measured at the field site. These glass beads were deployed in replicated patches on the bottom of Boston Harbor. Vertical distributions of the beads after 99 d were compared to profiles predicted by the model. Good agreement between predicted and measured profiles indicated that the feeding and burrowing mechanisms of C. grandis were sufficient to determine observed patterns of size-dependent bioturbation rates at this site.

Denitrification is a critically important biogeochemical pathway that removes fixed nitrogen from ecosystems and thus ultimately controls the rate of primary production in nitrogen-limited systems. We examined the community structure of bacteria containing the nirS gene, a signature gene in the denitrification pathway, from estuarine and salt marsh sediments and from the water column of two of the world's largest marine oxygen-deficient zones (ODZs). We generated over 125,000 nirS gene sequences, revealing a large degree of genetic diversity including 1,815 unique taxa, the vast majority of which formed clades that contain no cultured representatives. These results underscore how little we know about the genetic diversity of metabolisms underlying this critical biogeochemical pathway. Marine sediments yielded 1,776 unique taxa when clustered at 95 % sequence identity, and there was no single nirS denitrifier that was a competitive dominant; different samples had different highly abundant taxa. By contrast, there were only 39 unique taxa identified in samples from the two ODZs, and 99 % of the sequences belonged to 5 or fewer taxa. The ODZ samples were often dominated by nirS sequences that shared a 92 % sequence identity to a nirS found in the anaerobic ammonium-oxidizing (anammox) genus Scalindua. This sequence was abundant in both ODZs, accounting for 38 and 59 % of all sequences, but it was virtually absent in marine sediments. Our data indicate that ODZs are remarkably depauperate in nirS genes compared to the remarkable genetic richness found in coastal sediments.

Software, documentation and a complete set of sample data files are available at http://faculty.www.umb.edu/jennifer.bowen/software/FunFrame.zip.

The method for DNA fingerprinting of the 16S-23S rDNA intergenic spacer region was modified to increase resolution of bacterial strains by thermal gradient gel electrophoresis (TGGE) analysis. By utilizing the high melting temperature region of the tRNA gene located in the middle of the 16S-23S rDNA intergenic spacer region as an internal clamp for TGGE, multiple melting domain problems were solved. PCR primers lacking a stretch of GC-rich sequences (GC-clamp) amplified the intergenic spacer region more efficiently than GC-clamped primers. Therefore, PCR artifacts were avoided by using low, 17-cycle, PCR. The method was successfully applied to diverse bacterial species for strain differentiation by TGGE without requiring a special PCR primer set.