EXPERIMENTAL DESIGN AND METHODS:
Macrofauna were collected at a variety of stations throughout the Plum Island Sound Estuary. Stations 1-9 represent samples collected using a 15.2 meter length beach seine and 4.8 mm (3/16 inch) mesh. Stations 10-21 represent samples collected from a small boat with a sixteen foot shrimp trawl (5/8" net with 3/16" cod end mesh). Replicate or triplicate dates represent replicate or triplicate collections for that species at that site.

Seining Method
The seines were 15.2 m length and either 1.2 or 1.8 m in height with a 4.8 mm (3/16 inch) square ace mesh. The middle of the seines were fitted with a bag (1.8 x 1.8 m) of the same mesh size as the wings of the net. A lead line insured that the bottom of the net remained in contact with the substrate. At each sampling station the actually sampling loci were carefully chosen after an inspection for rocks, cobbles, and any other potential impediments. Once a locus was selected, one person held one end of the net and stood at a fixed position at the edge of the water. A second person took the other end of the net and walked out in the water perpendicular to the shoreline at the point where the first person was standing. When the net was straight, it would be checked along its length to insure that it was deployed correctly. The person at the deep water end would then move toward shore describing an arc while the person at the water's edge remained fixed. Thus each seine tow covered a quarter of a circle with a radius of 15.2 m and an area of about 181 m2.  When the far end of the net was almost at the shoreline, the two people holding the ends of the net would walk toward each other and then move slowly up the beach until the net was completely out of the water. Fish and selected macroinvertebrates (decapod crustaceans) were then placed in plastic bags and stored in a cooler for later identification and enumeration back at the laboratory. Triplicate seine hauls were taken at all stations, with care taken in the second and third haul to avoid areas that had been impacted by the previous haul.  A seine team typically consisted of three or more people. In addition to the two people holding the ends of the seine, a third person checked along its length while it was being deployed to insure that it was sampling properly, i.e. lead line down, no tangles, etc. Additional people, often volunteers, helped with collecting the samples from the net after it was hauled ashore. At each seine station, temperature and salinity (by refractometer) were measured, and the depth at the deep end of the net was estimated. Almost all depths were between 0.4 and 1.0 meters.

Shoreline stations sampled by beach seine. Stations 1-6 were selected to be as close as possible to shoreline stations sampled by DMF in 1966. Station 1. Great Neck, Ipswich. This station was located several hundred meters north of DMF's 1967 Little Neck station near Pavilion Beach, a narrow strip of land that separates Great Neck from Little Neck on the western shore of Plum Island Sound. The old DMF station was too cobbly at this point to reliable seine. Station 1 was located on a flat, very gently sloping sandy beach just north of an area of small rocks interspersed with salt marsh vegetation that separates the site from the northern end of Pavilion Beach. Sampling was always carried out within two hours of low tide.
Station 2. Bluffs, Ipswich. This station was located on the eastern side of Plum Island Sound several hundred meters south of the private residence on Stage Island. It was in a relatively sandy area at the seaward edge of a tidal flat that extends out from Stage Island. We sampled at low tide. The gradient was gentle.
Station 3. Knobs. This station was a small gently sloping sandy beach bordered by salt marsh on three sides. The station is at the end of the dirt road that extends west past the southern border of the Bill Forward pool on the Parker River National Wildlife Refuge (PRNWR). It was seined at mid to low tides. New salt marsh vegetation covered the upper edge of the beach, thus unlike DMF who reported that the beach was devoid of vegetation, this area could no longer be seined at higher tides.
Station 4. Nelson's Island. This station, also on the PRNWR, was located at the edge of a tidal flat at the end of the dirt road that runs from the refuge parking lot at the end of Stackyard Road across the salt marsh to Nelson's Island. The substrate was a muddy sand with occasional disturbance by commercial clammers. Since the upper edge of the station was bordered by a steeply-banked salt marsh, it was seined at mid to low tides (2-3 hours on either side of a low tide) when the marsh was not flooded. We avoided extreme low water, because the flat sloped steeply into a channel that was too deep to allow us to extend the net to its proper length.
Station 5. Subheadquarters. This station was located on the extensive mud flat just west of the maintenance buildings of the PRNWR. In sampling this station we were faced with the greatest time constraint since it was too dry during much of low tide, but could also not be seined from mid to high tide because of the steeply banked salt marsh at the upper edge of the mud flats. There was generally a 45 minute "window" between two to three hours before or after dead low where this station could be seined. This mud flat was frequented by commercial clammers.
Station 6. Newbury Town Landing. This station was located on mud flats on the north shore of the Parker River several hundred meters downstream from the Newbury's old town landing at the end of Cottage Street. The old DMF sampling station was at the landing itself. We moved downstream slightly to be out of the influence of the mooring and launching area for boats. The mud flats bordered a steep salt marsh bank on the PRNWR. The soft sediments were interspersed with occasionally "clumps" of salt marsh peat that had probably slumped from the edge of the bank, thus care had to be taken to find a free area for seining. Sampling was done within three hours of maximum low water. This station had strong tidal currents, particularly as the tide approached mid tide.
Station 7. Rowley River flat. This station was located on a mud flat that bordered the salt marsh on the south side of the Rowley River about 0.75 km downstream from the Ecosystems Center's Rowley Field Station at the end of Railroad Avenue. The substrate was muddy. The gradient of the mud flat was gradual across much of its breadth, however the lower end of the flat sloped off more steeply to the channel of the river. The station was seined between 1-2 hours after or before maximum low water.

Trawling Method
Trawling was carried out with a 4.9 m otter trawl (mesh size of 19.1 mm in the wings and 4.8 mm in the cod end) towed behind a 16 foot Boston Whaler. Trawls in Plum Island Sound (stations 10-15) were generally carried out for 2 minutes at speeds of about 1.5 m s-1. Trawls in the Parker River (stations 16-21) were carried out for only 1 minute, due to the obstructions within the river that precluded longer trawls. Shorter tows were noted on the data sheets and factored into our calculations. Each station was trawled in triplicate at each sample date, however the narrow confines of some Parker River stations (see below) often forced us to trawl over the same area.

The location of stations in Plum Island Sound was based largely on those sampled by DMF in their 1967 study. We did not sample their offshore station OS1 (Camp Sea Haven) in Ipswich Bay, which they sampled three times over the course of their study, because we were limited to our smaller boats. We also did not sample off Great Neck (DMF's OS4) because the area now contains a large number of moorings. We added one station at the entrance to the Parker River.  Trawling in the Parker River itself presented a challenge because of the large number of underwater obstacles and meanders. Rather than attempting to resample the same stations as in the DMF study, we selected six stations that ran the length of the tidal part of the river that we felt could be reliably seined for at least one minute. Trawl Stations sampled by shrimp trawl. Stations in Plum Island Sound were selected to be as close as possible to the DMF offshore stations except where noted. In the Parker River upstream from Route 1A, we made no attempt to overlap exactly with DMF sites because of the existence of obstacles, such as underwater snags, that limited what could be trawled now.. Nonetheless, our six Parker River stations should represent similar habitats to theirs.
Plum Island Sound Stations:
Station 10. Castle Neck. This station was at the mouth of Plum Island Sound and could be sampled at all tides. It was similar to DMF's OS2, however we trawled in shallower water. The substrate was sand. Depth was about 1.5 m at low water.
Station 12. Middle Ground. This station was located just north of the Middle Ground, a salt marsh island in the southern section of Plum Island Sound. Equivalent to DMF's OS3. The substrate was sand. Depth was about 1.5 m at maximum low tide, and the station was sampled at all tides except for maximum low.
Station 13. Nelson's Island. This station was located in the channel offshore of Nelson's Island and shoreline station 4 on the west side of Plum Island Sound. It was equivalent to DMF's OS5. Depth was about 1.6 m at low tide, and the station was sampled at all tides except for maximum low.
Station 14. Cape Merrill. This station was located in the Plum Island River just beyond its confluence with the Parker River at the north end of Plum Island Sound. It is analogous to DMF's OS6. The substrate was sand and its minimum depth was about 1.35 m, and the station was sampled at all tides.
Station 15. Entrance to Parker River. This station was located north of Dole Island where the Parker River enters Plum Island Sound. The substrate was sand, and it had a depth at low water of about 1.45 m. The station was sampled at all tides. DMF did not sample in this area.  Parker River Stations:
Station 16. Parker River. This station was on the Parker River between the Little and Mill Rivers, several hundred meters east of the railroad bridge. The substrate was mud. No equivalent DMF station was within 2 km in either direction. Depth at low water was about 1.5 m.
Station 17. Parker River. This station was two hundred meters to the west of the railroad bridge. The substrate was mud. The nearest DMF stations are OS8 (Mill River) and OS9 (South Shore). Depth at low water was about 1.5 meters. Sampling was carried out at mid to high tides.
Station 18. Parker River. This station was just east of the first meander in the river east of U.S. Route 1, approximately 0.8 km east of this road. The substrate was mud. This station is just west of DMF's OS9 (South Shore). Depth at low water was about 1.5 meters. Sampling was carried out at mid to high tides.
Station 19. Parker River. This station was several hundred meters west of Thurlow's Bridge (Middle Road). The substrate was mud. The nearest DMF station was their OS10 (Thurlow's Bridge). Depth at low water was about 1.6 m, and this station was trawled at all tides.
Station 20. Parker River. This stations was about 2 km west of Thurlow's Bridge (Middle Road). The substrate was mud. No DMF station was located in the immediate vicinity. Depth at low water was about 1.8 m, and this station was trawled at all tides.
Station 21. Parker River. This station was about 1 km east of the falls at Orchard and Central Street in Byfield. The substrate was mud. Depth at low water was about 1.2 m, and this station was trawled at all tides. This station is several hundred meters east of DMF's OS 11 (Woolen Mill). We attempted to sample there as well but could not due to obstacles in the water.

Identification, enumeration, and biomass of fish and decapods
After field collection, the fish samples were brought back to the Ecosystems Center's Rowley River Laboratory for identification. Fish and decapod crustaceans from each replicate were sorted by species and each fish individually counted. Fish identifications were based on Bigelow and Schroeder (1953), Robins et al. 1986), Scott and Scott (1988) and an estuarine fish key developed by the Fish Ecology Laboratory of University of Massachusetts at Amherst (Basher 1989). As described above, we routinely collected decapod crustaceans, including Crangon septemspinosa, Palaeomonetes spp. and crabs. These were identified using Smith (1964) and Gosner (1971, 1978). Fish numbers were expressed as fish per 100m2.