Results and Discussion

Station #2:    Grain-size Analysis Results
                        Organic-Mineral Aggregates

Station #3:    Grain-size Analysis Results
                         Organic-Mineral Aggregates

Station #2

Grain-size Analysis Results

Results of grain-size analysis of the monitoring period at station #2 from February 1990 to May 2001 are shown on Table 1 and Figure 4. The overall sediment texture of the interval from 0 to 0.5 cm has exhibited great consistency over time, with mean percent sand values exceeding 95% for the majority of the monitoring period (Table 1). Error bars of duplicate and triplicate samples indicate a within-station variability of sand typically less than 8% (Table 1 and Figure 4). On the basis of the size analysis results over the monitoring period, this station is classified as a moderately well-sorted fine sand. Percent mud (silt plus clay) typically ranged from 0.5% to 5% (Figure 5).

A sediment texture map of the outfall area based on sediment profile imaging camera surveys identified major modal grain size and inferred kinetic energy on the benthic environment (Shea and others, 1991). On the map by Shea and others (1991), the USGS station #2 falls within the boundaries of an area designated as a coarse sand to gravel. The fact that the USGS analyses and Hilbig and others (1997) indicate fine sand at station #2 probably reflects a different methodology and the patchiness of sediment in this region.

A map of the major modal grain size of the surface sediments based on grab sampling of a 2-square-mile area west of the diffuser site further demonstrates and refines the spatial variability of the sedimentary environment regimes in this area (Hilbig and others, 1997). From this map, the USGS station #2 falls within the fine sand mode environment defined by that study, which also encompasses the diffuser site.

An apparent increase of 100% to 200% mud (silt plus clay) was evident from samples recovered from cruises W7 and W8 (October 1991 and February 1992, respectively) and represents higher mud concentrations than at any other sampling times between 1990 and 2000. However, only one of the duplicates from W8 contained an elevated mud concentration. There were no apparent errors associated with the collection, processing, or laboratory analyses of these samples to account for the slightly higher mud concentrations measured in W7 and W8 samples.

Fluctuations in the relative amounts of surficial sands and muds over time have also been observed comparing baseline data from 1992. Yearly sampling studies ensued through 1997 using sediment profile imaging and sediment grain size analyses of grab samples from monitoring sites in Massachusetts Bay (Hilbig and others, 1997: Blake and others, 1998). Surficial grain size fluctuations have been ascribed to the variety of sediment sources, their response during high runoff or storm related events, and the seasonal variability and activity of benthic infauna.

Organic-Mineral Aggregates

Organic-mineral aggregates were separated from the sediment with wet sieving techniques. Identifiable from microscopic observations, they were primarily composed of mineral grains with variable amounts of the following components: shell fragments, slag, plant fragments, rock fragments, Foraminifera tests, diatoms, copepods, unidentifiable larvae, sponge spicules, worm tube fragments, and fecal pellets.

These particles were bound in an amorphous elastic mucilaginous exudate. This mucus- like exudate is produced by the action of microbes, diatoms, and other benthic organisms living within the upper few centimeters of the sediment (Rhoads and Boyer, 1982). The mucus binds the individual particles together and may increase the bottom stress required for resuspension. The mucus matrix appeared colorless under both transmitted and reflected light, and the aggregate grains varied in size from 100 to 2000 micrometers as measured along the longest axis.

The aggregate fraction of the sediments was found to be consistently less than 2% by weight of the total sample throughout the 11-year monitoring period (Figure 6). The low percentage of aggregates may be related to the low percentage of silt and clay and the low abundance of benthic infauna (Rhoads and Boyer, 1982; Wheatcroft and others, 1994). There appears to be no seasonal trend for the occurrence of aggregates in the sediments. There was no difference between aggregates measured before and after the October 2001 storm (cruise W8). However, the aggregates were much lower on cruise W11 after the major storm with a longer duration, in December 1992.

The primary component of the aggregates were mineral grains, which commonly composed upwards of 95% of the total aggregate fraction. The predominant mineral grains were quartz (20% to 95%), which varied in coloration from a clear translucent to a brownish coloration (due to iron staining). The grains were angular to subrounded, and some of the quartz grains appeared to be encrusted with organic matter. Other mineral grains present were micas, heavy minerals, feldspars, and unidentified silt- and clay-sized particles. The remaining components of the aggregates (shell and rock fragments, slag, and biogenic skeletal components) occurred in amounts typically less than 1% of the total.

Fecal pellets were the second most numerous component of the aggregate fraction. They are encapsulated waste products produced either by organisms living and feeding at the sediment-water interface or by burrowing organisms such as tubiculous worms or other sediment-dwelling organisms (such as bivalves), which excrete the pellets at the sediment surface. These pellets generally totaled less than a few percent of the total aggregate components. The pellets were ellipsoidal and were retained primarily on the 2-phi (0.25-mm, medium sand) sieve. Measurements of the long axis of the pellets ranged from 300 to 700 micrometers.

Results of long-term biological monitoring of western Massachusetts Bay has shown that a mixed benthic infauna assemblage is present near USGS station #2 that is composed of polychaetes ( such as cirratulids, syllids, and spionids), amphipods and isopods. The infaunal density is low, and the species types are quite variable from year to year (Kim, written communication, 1992: Hilbig and others, 1997; Blake and others, 1998). Worm tube mats of polychaetes were not observed at the sediment-water interface in cores collected at this site by USGS investigators during the length of the monitoring period.

The mucus-bound aggregates are somewhat delicate and are easily disaggregated by typical treatments during laboratory grain-size analysis such as the use of dispersants, sonification, and drying. In nature, the aggregates are easily destroyed by near-bottom turbulence during storms and thus, they are not readily transported nor found intact at depth in the sediment (Rhoads and Germano, 1986).

Station #3

Grain-size Analysis Results

Results of grain-size analysis for the monitoring period from February 1990 (cruise W2) through May 2001 (cruise MH36) at USGS station #3 are shown on Table 1 and Figure 7. The overall sediment texture of the surficial interval from 0 to 0.5 cm is finer grained than that of station #2 and somewhat more variable in distribution. The overall concentration of sand for station #3 varied from 10% to 52%, silt from 40% to 70%, and clay from 5% to 47%. Within-station variability ranged from approximately 2%-15% for sand, 1%-13% for silt, and from 1.5% to almost 10% for clay (Table 2). No seasonal differences were detected.

Results from the monitoring period indicate that the sediment at station #3 is classified as coarse to medium silt based on the method of moments using mean phi values. The sediment contained greater than 50% mud (silts plus clays) over the entire monitoring period (figure 8).

No apparent effects to the overall grain-size distribution were evident from samples collected on W8 (February 1992) after the passage of the intense Halloween Nor'easter storm which occurred in October 1991. This storm had peak waves of 9 m for a few hours and waves greater than 5 m for 0.9 day. In contrast, the biggest change in grain size was found in samples collected on cruise W11 (February 1993) when the clay concentration had increased from about 10% to 45% (Figure 7). The W11 cruise followed a storm in December 1992, which had peak waves of 7.3 m and waves greater than 5 m for 2.6 days. The correlation of major storms and collection rates in sediment traps indicates that surface waves generated during Nor'east storms are the major cause of sediment resuspension in western Massachusetts Bay (Butman and others, 1992). We speculate that the 1992 storm, considerably longer in duration than the more intense 1991 storm, was responsible for resuspending and winnowing fine-grained material from surrounding areas, which accumulated, at least temporarily, in depositional areas like station #3. Higher concentrations of silver and Clostridium perfringens in the W11 sample suggest that the storm transport included material originally deposited from the sewage outfall in Boston Harbor. The absence of a dramatic change in the grain size or sediment chemistry following the October 1991 cruise may indicate that the duration of resuspension processes was insufficient to move a significant mass of sediment. This increase in clay concentration may reflect the variability of sediment response to storms of different strength.

For station #3, the grain-size distribution after sampling cruise W11 (February 1993) appeared to return to a somewhat siltier mode much like that as observed during the W2-W9 sampling cruises (Figure 5). However, cores collected during sampling cruises MH 35 and MH 36 (February 2001 and May 2001) also indicate an apparent increase in the concentration of clay-sized particles of 30% and 40%, respectively (Figure 8). Examination of bottom current data collected at a nearby USGS mooring site indicates that only current events of low strength and short duration (not much longer than tidal cycles) occurred during these periods. No intense storms, such as the Halloween Nor'easter of 1991 or the December 1992 storm, were observed over these time frames. Changes within a sedimentary environment over time may simply reflect the natural variability in bottom sediment response to near-bed flow conditions (Snelgrove and Butman, 1994). We will be assessing the impact of the suspended material discharged from the Massachusetts Bay Outfall at station #3. The concentrations of silver and Clostridium perfringens are not elevated in bottom sediment samples collected 6 months after the start of the outfall on September 6, 2000, compared with background values unrelated to storms (Bothner and others, 2001).

The depth profile of sand, silt, and clay percentages in two cores (3-1 and 3-3) collected on cruise W15 (June 1994) are in the same range but show maximums and minimums at different depth (Figures 9, 10, 11, 12). The variability between these cores from the same location may be due to factors such as non-uniform bioturbation by sediment dwelling organisms, variability in net deposition on a small spatial scale, or by local disturbance of the bottom possibly by commercial fishing activity.

An additional core of over 50 cm in length was collected during a summer cruise (W24) in 1997. Figure 13 shows the textural variability with depth from core 3-4. Trace gravel concentrations of slag, quartz, and shell fragments occur in the upper third and near the bottom of the core. Sand concentrations are highly variable in the upper 4-5 cm and near the bottom of the core, whereas from 5 to 23 cm the concentration of sand is fairly uniform, varying less than 15%.

Concentrations of mud throughout the core appear widely variable as well, ranging from 55% at the near surface to over 95% slightly below the midpoint of the core (Figure 14).

Organic-Mineral Aggregates

Components of the aggregates from station #3 appeared visually similar to those collected at station #2 as observed under the light microscope. The aggregates were bound within a clear to translucent mucus-like coating and consisted of mineral grains, rock fragments, diatoms, worm tube fragments, unidentifiable biogenic components, silt and clay size particles, shell fragments, slag, heavy minerals, and fecal pellets.

With the exception of two sampling cruises (W12 and W13), the weight percentage of the aggregate fraction was quite uniform with an average value of 7.8% ± 4.6% that typically ranged from about 5% to 15% throughout the monitoring period. Significantly higher values of 27% and 53% occurred during the summer and fall of 1993 (Figure 15, sampling cruises W12 and W13). Of these aggregates 75% were fecal pellets.

Fecal pellets were by far the most abundant non-mineral particle observed within the aggregates ranging from 25% to 90% of the total particles. The fecal pellets from station #3 were smaller than those from station #2. They were typically retained on the 4-phi (0.0625-mm, very fine sand) sieve instead of the 2-phi (0.25-mm, medium sand) sieve as at station #2. The fecal pellets were ellipsoidal with the long axis between 60 to 150 micrometers in length. Because fecal pellet/aggregate formation might be expected to vary seasonally, the percent aggregates for each season were averaged (excluding W12 and W13). Mean values of weight percentages for samples collected in September (7.8 ± 4.3 %), February ( 6.7 ± 3.4 %), and May (9.1 ± 6.0 %) are not statistically different.

The observed increases in the percentage of fecal pellets during the summer and fall of 1993 (cruises W12 and W13) correlate with similar increases of infauna abundance at station #3 as well as from other nearby sites in Massachusetts Bay (Hilbig and others, 1997; Blake and others, 1998). The benthic community in Massachusetts Bay and at station #3 is dominated by various species spionid polychaete worms that feed on surface deposits (Wheatcroft and others, 1994; Gallagher and Keay, 1998). This assemblage might experience episodic changes in numbers, distribution, and diversity due to the life cycles of the organisms, to sediment transport events, and to changes in the flux of organic matter from both natural and anthropogenic sources.

The increase in fecal pellet occurrence produced from the benthic infauna at station #3 during the spring and fall of 1993 might also be a result of a "pulse" of fine-grained particles rich in organic matter that was deposited by the December 1992 storm. This new material could have been winnowed from the coarse sediment that exists landward of station #3, or it could have come from the harbor itself.

The elevated concentrations of silver and Clostridium perfringens suggest that the source of contamination is from sewage (Bothner and others, 2001). However, these results do not distinguish whether the source of these sewage related components originated from Boston Harbor or from sources outside it on the inner continental shelf areas.

Organic-mineral-aggregate concentrations at the near surface of the sediments is about 21% and 24% for cores 3-1 and 3-3, respectively, collected in June of 1994 on cruise W15 (Figures 16 and 17). Both these concentrations decrease rapidly with depth in the core to less than 5% near the bottom of the core. The decrease of aggregates and fecal pellets with increasing core depth is probably due to the inherently fragile nature of the pellets. Also, these aggregates may be incorporated as a food source by other organisms within the sediment.

A plot of the aggregates from core 3-4 collected on cruise W27 in the summer of 1997 shows a maximum of slightly over 4% of aggregates at the surface with a more gradual decrease than cores 3-1 and 3-3, to less than 2% near the bottom of the core (Figure 18).