November 18, 1998

Dear Colonel Pratt:

The Department of Environmental Management has completed its review of the Draft Environmental Impact Statement for the Providence River and Harbor Maintenance Dredging Project.

The attached comments were generated after a thorough review of the DEIS by scientists and engineers throughout the agency with expertise in marine fisheries, water quality, freshwater wetlands, groundwater quality, agriculture, waste management, forest environment, and air quality. Due to the volume of information and time constraints imposed by seasonal field work, the comments that pertain to marine fisheries issues are limited to impacts associated with the proposed dredging and preferred disposal option (CAD cells in the Providence River and Bay disposal at site 3 off Hog Island).

As the comments indicate at greater length, the Department of Environmental Management has serious questions and concerns both with the Army Corps of Engineers preferred disposal alternative and with the dredging methodology of the Providence River channel. Specifically, DEM recommends that the disposal of dredge material at Site 3 -- the seventy foot-deep hole in Narragansett Bay between Hog and Prudence Islands -- be rejected as environmentally unacceptable, and that dredging activity be suspended during the months of March through July. DEM's recommendations are based on its analysis of data contained in the DEIS and other data referenced and included in the attached comments.

The Department of Environmental Management recommends that the ACOE take the actions below to respond to these concerns. (The supporting documentation is attached.)

• It is imperative that dredging activity be suspended during the months of March - July to protect the larvae of already stressed commercially and recreationally valuable fish and shellfish species from mortality associated with suspended solids and BOD generated by dredging activity.
• Disposal of dredge material at Site 3 should be rejected as environmentally unacceptable for the following reasons:
• The area serves as a seasonal cold water refuge vital to the migration habits and life cycle of recreationally and commercially valuable species such as lobsters and winter flounder, a function that would be compromised or destroyed by the reduction of depth associated with the disposal of dredge material.
• Survey data shows that it and the Bay's other deep "holes" support four times the density of finfish as do shallower waters and that the reduction of depth associated with dredge material disposal is predicted to result in a 48% reduction in finfish abundance.
• Both the area of the dredge project and the preferred disposal area qualify as Essential Fish Habitat (EFH) under the Magnuson Fishery Conservation and Management Act of 1976 as amended by the Sustainable Fisheries Act of 1996 and the Atlantic Coastal Fisheries Cooperative Management Act of 1993. Pursuant to those acts Essential Fish Habitats must be protected from avoidable adverse impacts including specifically those caused by channel dredging and the disposal of dredge material.
• Although the Department recognizes that there are specific local issues associated with the creation of the confined aquatic disposal (CAD) site proposed in the Draft EIS, DEM 's environmental evaluation of the data provided in the Draft Environmental Impact Statement does not indicate that the use of this site for the CAD would be environmentally unacceptable.

Additional recommendations are made at the conclusion of the attached report.

We have appreciated the efforts the Corps has made to address our agency's concerns throughout the development of this DEIS. We believe that this process has been very helpful, and the Department will continue to offer our assistance and cooperation to the Corps throughout this process. If you have questions about the Department's comments that you would like to discuss further, please call Roger Greene at 222-4700 extension 2402.

Sincerely,

Andrew McLeod, Director

Department of Environmental Management

Benthic Habitat Impacts (Including Shellfish and Lobster) and Finfish Habitat Impacts

In terms of its fisheries resources, Narragansett Bay is a system under stress.

Narragansett Bay has undergone dramatic changes in the composition of its fishery resources, which are indicative of stress (sensu Fausch et. al. 1990). These dynamics should be kept in mind when evaluating potential impacts from the project. Jeffries and Terceiro (1985) and Jeffries (1994) correlated species shifts in the Bay with long term patterns in water temperature while Gibson (1996) associated a multi-species collapse in Mt. Hope Bay with power plant degradation of the local habitat. Lee et al. (1991) examined winter flounder liver pathology and found a gradient consistent with pollutant sources but researchers have been unable to generally correlate declining water quality with reduced fishery production (Desbonnet and Lee 1991). Oviatt and Nixon (1973) showed that fish distribution in Narragansett Bay was correlated with depth, sediment type and water temperature while Meng and Powell (1998 in press) have emphasized the importance of habitat by linking juvenile fish abundance to bay physical attributes. Overfishing has occurred on several important species, confounding efforts to identify natural and anthropogenic impacts (Gibson 1998a, 1998b).

Restructuring of the fisheries resource community has continued to present day. In Figure 1 (attached), results from Division of Fish & Wildlife (DFW) trawl surveys and stock assessments are plotted as abundance indices by species groups. It is clear that the pelagic fish/squid and crustacean resources have increased in abundance while demersal finfish and shellfish have declined. Crabs and lobsters have replaced bottom dwelling fish such as winter flounder, windowpane flounder, tautog, and sea robins. Despite a recent increase in oyster populations, overall shellfish biomass, which is dominated by quahaugs, has declined. There has been a steep increase in the biomass of pelagic fish and squid, particularly the Bay anchovy. Despite substantial variability in the underlying components, aggregate biomass levels have been stable suggesting that primary production has been redirected into alternate pathways. With shifts in the species guilds has come a reduction in the average size of fish caught in surveys (Figure 2, attached). These changes likely result from the interaction of several factors such as selective over exploitation, habitat degradation, and long term environmental forcing factors.

The shifting resource complex has implications for fishers exploiting the assemblage. Rhode Island commercial landings, like the survey indices, have been stable in the aggregate. There has been, however, a strong shift in the species harvested so that landings of pelagic species and lobsters have increased while landings of demersal fish and shellfish have declined (Figure 3, attached). Recreational landings show much the same pattern with catches of demersal fishes near historic lows but striped bass landings at historic highs. If mean body size is considered a proxy for trophic status, the available survey and landings data for the Rhode Island area support the worldwide observation by Pauly et al. (1998) that a progressive fishing down of marine food webs is occurring (Figure 4, attached).

Major efforts are underway to reverse these changes and rebuild depressed fish stocks. The Sustainable Fisheries Act (SFA) and Atlantic Coastal Fishery Cooperative Management Act (ACFCMA) govern management activities in federal and state waters respectively and mandate the rebuilding of overfished stocks. Existing plans for species of primary importance to the state include; winter flounder, tautog, summer flounder, lobster, scup, striped bass, bluefish, menhaden, and weakfish. There is little room for state discretion in these plans and failure to meet stock rebuilding benchmarks can have serious consequences including penalties. All of these management plans contain elements and requirements related and to the protection and preservation of fish habitat in recognition of its essential nature in fishery production. Given the rigorous nature of the plans, it is important that States make strong efforts to protect fish habitats in their jurisdictions.

1. Projected Impacts of, and Concerns Related to, Dredge Windows

The ACOE categorically rejects the need for imposition of dredge windows during the project period although sequencing of dredging is offered in limited circumstances. Their general justification is that dredging will cause minimal impact to living resources in the area. We believe this conclusion to be scientifically indefensible because of the well-known effects of suspended sediments on fish and fish larvae (Newcombe and MacDonald 1991). Contrary to the position adopted by the ACOE, the rigor with which dredge windows are applied should increase with the size of the project. From a fisheries management standpoint the period of greatest sensitivity is February through August when biological activity by spawning adult and larval fish is most vulnerable to the adverse effects of dredging. The ACOE has inappropriately devalued the Providence River reach by referring to the Keller and Klein-MacPhee (1992) study, which suggests that the epicenter of fish larvae in the Bay has shifted to a more mid-bay distribution. We note that these results are skewed as a result of recent changes in the distribution and increased abundance of bay anchovy and do not reflect accurately the distribution of the principal species of concern such as winter flounder and tautog. The ACOE also projects minimal impact on fisheries resources since adult spawning habitat is generally located outside of the federal channel. However, this analysis ignores the well-known phenomenon that it is dispersed larvae, not spawning adults, which suffer the most serious and pervasive impacts of dredging.

With this in mind, abundance of larval fish remains high in the Providence River as evidenced by monitoring data for 1992-1997 collected by Marine Research Inc. near Manchester Street Station (MRI 1997a). The RIDEM examined 1972-1973 icthyoplankton data in the Bourne and Govoni (1988) comprehensive baywide survey and compared it to modern MRI monitoring in the Providence River. The Bourne and Govoni study identified the Providence River reach (sector 2 in Figure 5, attached) as having the highest densities of fish eggs and larvae in the Bay (Figure 6, attached). They noted that concentrations of eggs and larvae from estuarine spawners in the upper bay were consistent with bay circulation patterns and larval retention mechanisms. Their results were reported as arithmetic mean densities whereas the recent MRI results are geometric means. Because of the skewed statistical distribution characteristic of icthyoplankton catches, a geometric mean is considerably smaller than the arithmetic, so a conversion function is needed to make comparisons. The RIDEM employed summary results in the MRI appendix document (MRI 1997b) to calculate a regression equation relating the two estimators: AM=1.625*GM^1.018. When this conversion is applied, there is little evidence that densities of eggs and larvae are lower in the Providence River now than in the 1972-1973 study. For most species and life stages, the current and historic densities have overlapping ranges (Table 1, attached). Menhaden, silversides, and possibly tautog show some evidence of a decline. Still, the importance of the Providence River reach is not diminished by reduced larval abundance, which may simply relate to stock declines.

Fisheries demographic theory holds that the numerical size of fish stocks is ultimately controlled by the size of the larval retention areas (Sinclair 1988). Habitats, which by virtue of their unique physics retain and nurture larvae and juveniles, give rise to adult stocks in proportion to the size of those areas. Larvae have evolved a suite of behaviors adapted to estuarine physics, which enhance retention in such optimum habitats (Epifanio 1988). Biological synchrony across geographic areas can be examined by correlating abundance at various life stages over spatial scales (Drinkwater et al. 1997). Because the Providence River reach is an important larval retention area contributing to fish recruitment in Narragansett Bay, we would expect larval abundance in this reach to be correlated with bay wide abundance in later stages. This is, in fact demonstrably the case. RI DEM extracted 1992-1997 egg and larval abundance indices for several species in the Providence River from the MRI (1997) report. They were related to young of the year (YOY) abundance indices from the DFW seine or trawl surveys in the bay. Demersal and littoral species such as winter flounder, tautog, and silversides are well sampled in the YOY stage by beach seining whereas the pelagic species such as anchovy, menhaden, and herring are better sampled by trawl. Regression equations were fit using log- transformed data to stabilize the variance. The results of the analysis are plotted in Figures 7a and 7b (attached). For winter flounder, tautog, and silversides there was a significant (P<0.05) correlation between Providence River larval abundance and DFW YOY abundance later in the season. For these species, there is a clear pattern of higher YOY production in years where larval abundance was higher. The same pattern holds for pelagic species, although the grouping of species makes interpretation of the regression model more difficult. These results clearly demonstrate that the Providence River reach is among the Bay's most important larval retention areas, the production from which is linked statistically to baywide production of juvenile demersal and pelagic fishes of considerable recreational and commercial value in the aggregate. Although causality is not conclusively established, the relationships suggest clearly that larval losses during dredging will be manifested in turn as lost juvenile production and ultimately as reduced recruitment to adult stocks. Several of these species are at very low abundance already and can ill support juvenile production losses without stock collapse (Gibson 1998a, 1998b).

RIDEM dredge surveys also show that the Providence River reach contains a dense population of adult quahaugs (Table 2) not identified by the ACOE. Size composition data from the surveys indicate that spawner biomass in the Providence River contributes to recruitment in conditional areas further down the Bay. In Figures 8a-8c, recruiting small clams are more abundant in conditional areas downstream of the Providence River. The Providence River reach had the highest larval quahaug density along a north-south gradient sampled by Butell and Rice (1996). Adult stock density also declines in a north-south gradient but larval quahaug density in the Providence River reach is proportionately higher than local spawner biomass (Figure 9, attached). These observations are consistent with a larval retention area in the Providence River reach. While settlement of larvae in the Providence River may be impaired by high adult density, periodic harvesting facilitates settlement in the nearby conditional areas. As noted by the ACOE, quahaugs are a commercially important species in Narragansett Bay. Stock assessment studies using fishery CPUE and survey data indicate declining biomass on a baywide scale (Gibson 1997). This is likely related to overfishing, pollution closures, and possibly an increase in benthic predators. The stock can ill afford additional larval mortality likely to be associated with dredging in this sensitive area during periods of high larval density.

Table 3 contains the life history timetable for several recreationally and commercially important species in Narragansett Bay. Much spawning activity takes place in the spring and summer with larval abundance persisting into August. Brief summaries for these species and their seasonal sensitivity to larval mortality are included below.

Winter flounder - This species spawns in the Upper Bay/Providence River from February through April. Eggs are demersal and hatch in 15-28 days. Larvae are present in the water column as late as the end of June. Data collected by Marine Research Inc. during 1993-1997 at and downstream of the Manchester Street Power Station show high abundances of winter flounder larvae in the water column from March through May. During 1998 larvae did not metamorphosis and settle to the bottom until late June. Winter flounder abundance in Narragansett Bay is at low levels (Figure 10, attached). Surveys in 1998 showed the lowest juvenile abundance ever which bodes poorly for adult abundance 2-3 years from now. This stock can not absorb any additional stress at this time (Gibson 1998a).

Windowpane flounder- This flatfish spawns mostly in inner continental shelf waters but also uses estuaries as nursery areas. The species exhibits two spawning periods with peaks in June and September as evidenced by MRI egg sampling in the Providence River. Larvae are generally present from July to October. Eggs are buoyant and hatch in 8 days at 10.6-13.3 C. Although not as valuable as winter flounder, they have been harvested in commercial quantities since 1974 with Rhode Island landings reaching 1 million pounds in 1988-89. Like most of the demersal fishes in Narragansett Bay, they have declined to very low levels in recent years (Figure 11, attached).

Tautog - Spawning by tautog occurs in the northern reaches of Narragansett Bay (Ohio Ledge and area south of Conimicut Point) during the months of May and June, often extending into July. Eggs are buoyant and hatch in 42-45 hrs. Larvae are planktonic and become juveniles in approximately 25 days. Ichthyoplankton studies done by Marine Research Inc. in 1972 and 1973 found the upper Bay (north of Prudence Island) to have the highest abundance of tautog larvae during the summer. Tautog are also at low abundance in the Bay (Gibson 1998b) and need to be protected (Figure 12, attached).

Weakfish - Spawning by this species in Narragansett Bay occurs from June to August. Eggs are buoyant with hatching occurring in 36-40 hours. MRI has sampled eggs or larvae in the Providence River since 1992. The coastwide stock of Atlantic weakfish collapsed to low levels in the early 1990's. It has recently showed local recovery with increased spawning in Narragansett Bay as evidenced by YOY collections in DFW and MRI surveys. It is a popular fish, which once yielded both commercial and recreational harvests. The few fish that continue to spawn in the Bay should have as near pristine conditions as is possible.

Menhaden - Atlantic menhaden spawn along the inner continental shelf and in estuaries. Menhaden eggs appear in the MRI surveys in May and larvae persist through July. The Atlantic menhaden stock is believed to be reasonably healthy but the spawning component in Narragansett Bay collapsed in the mid 1980's as evidenced by egg and larval surveys in Mt. Hope Bay by NEP/MRI. Anecdotal evidence suggests renewed spawning in 1998 as large numbers of YOY fish have been observed. In view of the controversy between commercial menhaden seiners and recreational fishers, every effort should be made to protect the Bay-spawning component and facilitate recolonization.

River Herring (alewives and blueback herring) - River herring migrate through the project area from the end of March into June. There are self-sustaining populations in the Providence/Seekonk Rivers and in the Warren, Barrington and Palmer Rivers. Smaller tributaries along the Providence River also have self-sustaining populations of anadromous species. Data from the DFWs Anadromous Fish Restoration Program and The Fisheries Resources of the Seekonk and Providence Rivers, Pawtucket, Providence, East Providence, Cranston, Warwick, and Barrington, Rhode Island, 1997 study indicate the presence of anadromous fish runs in the project area. There is ample data to support a dredge window for anadromous fish to avoid impacts on migration.

Quahaugs- Bay quahaugs spawn in May and June. Data in Butell and Rice (1996) indicate that larval abundance peaks in mid to late June and persists through July. Their sampling data also show a strong gradient in larval density along a transect running from the Providence River to the West Passage of the Bay. The high density of quahaug larvae in the Providence River reach is likely related to the dense beds of spawning adults in the upper bay closure area which was recently surveyed by the DFW (Gibson 1997). This abundant spawner biomass and larval production is likely very important to fishery production in the upper Bay conditional areas. It is believed that the Narragansett Bay quahaug stock exists as a metapopulation with patches in closed, conditional, and open areas linked by larval dispersal. Until the DFW completes its study of the Providence River-Conditional Area system, disturbances, which might reduce larval production, should be avoided. Estimated quahaug stock abundance in Narragansett Bay is given in Figure 13 (attached).

It is RI DEM's opinion after a review of Sections 7.2.2, 7.3.2, 7.3.3, and 7.4 of the DEIS that the impacts of suspended sediment on the marine resources of the Bay have been seriously underestimated. Because of this, the scope of the project, and the documented adverse impact of suspended sediment on sensitive life history stages of various living marine resources (Newcombe & MacDonald 1991), it is RI DEM's position that dredge windows are critical during this project. We base this contention on our model runs using the Newcombe & Jensen (1996) model which indicate that there will be lethal impacts to larval fish in the vicinity of the dredge plume. We used the Group 5 model parameterization for adult estuarine nonsalmonids rather than the Group 4 larval salmonid parameters used by ACOE. Group 4 parameters estimate lower severity impacts for a given sediment dose than do Group 5 parameters because of a significant difference in the exposure parameter. This likely results from the fact that larval and juvenile salmonids are naturally resistant to high total suspended solids (TSS) because of the lotic nature of their riverine nursery areas. The ACOE used group 5 parameters for site 3 disposal and we believe that estuarine fish larvae will be no more tolerant than adults of dredge related suspended sediments. We did make one modification to the model. Examination of regression diagnostics indicate that the Newcombe-Jensen model as originally published tends to overestimate observed severity impacts (SEV) at the low end of the scale so that lethal effects would begin to occur at background TSS levels in just a few days. Although the Providence River reach is not pristine habitat and elevated larval mortality rates are likely, we felt an adjustment to the model was needed. This was accomplished by setting the Group 5- intercept term to zero and forcing the regression through the origin. In effect, the model is corrected for background levels of TSS which average 5.99 mg/l in the Providence River reach (Pilson and Hunt 1989).

In our model runs, a range of concentration-duration doses from Table 7.2.2-7 of the DEIS was used to estimate a range of severity impacts. We don't agree that the Table 7.2.2-7 exposure durations are maximums because of tidal action or experiences at the Conley terminal. The target organism is fish larvae, which move with tidal currents as opposed to sessile bottom organisms. Indeed, larvae in estuaries migrate vertically in the water column so as to ride flood tides upstream while drifting back with ebb tides thereby maintaining position in systems with net outflow (Epifanio 1988). This means that contact durations with the dredge plume will not be strictly intermittent with respect to tidal cycle as the ACOE contends. The ACOE reference to the Conley project is also misleading. In Table 7.2.2-4, they provide TSS concentrations at 500 and 1000 feet from point source but compare these on page 7-94 to Table 7.2.2-2 TSS data for other projects. We note that Table 7.2.2-2 data is for 100 to 400 foot distances and, given the exponential decay process characteristic of TSS in dredge plumes, the two tables are consistent. TSS concentrations inside of 500 feet will be substantially higher than the Conley project data.

We have arranged our model results in a matrix of severity indices for various doses (Table 4, attached). Severity indices of 10 are associated with the onset of lethal effects. At the lower end of the TSS range in Table 7.2.2-2 for enclosed bucket dredging (25-50 mg/l), lethal effects will occur at exposures of 4-5 days. At the high range (100-300 mg/l), lethal effects begin at 3-4 days. Given a 200 day continuous dredging format and the aforementioned larval distribution properties, this level of lethal exposure will certainly occur for significant portions of the larval population. As a consequence, significant larval mortalities will result. How much of a given years production will be lost, however, cannot be conclusively established based on existing data.

Further independent review of the ACOE application of the Newcombe & Jensen (1996) model by the developer of the model concurred that the DEIS under-estimates the impact of the suspended sediment plume from dredging and dredged material disposal on finfish. Dr. Charles Newcombe completed the review at the request of RI DEM. The following is extracted directly from the comments sent to DFW by Dr. Newcombe:

US ARMY CORPS OF ENGINEERS IMPLEMENTATION OF SUSPENDED SEDIMENT DOSE RESPONSE MODELS (NEWCOMBE AND JENSEN 1996).

Two of the dose-response models of Newcombe and Jensen (1996) are used in the report: Eggs and Larvae of Salmonids and Nonsalmonids (Group 4) page 99-100 and Adult Estuarine Nonsalmonids (Group 5) pages 112-113. The US ACE implementation of the models is based on scenarios (concentrations and durations of exposure) that seem to be much less than the worst case scenarios that I developed earlier in this memo (my data choices are referenced to page numbers and table numbers in the US ACE report; so are those of the US ACE. So, the main arena for discussion is the actual exposure (concentration and duration) of any species at any time in their life-history. In my opinion based on the information I reference from the US ACE report, fish mortalities potentially can occur during any hour of dredge operation, day or night, over the 200 day span of the project; and, fish mortalities can occur during any hour when spoil dumping occurs, and for some interval of time (an hour or a number of hours) thereafter as the concentration of the suspended sediment plume decreases. The US ACE report concludes (page 7-100) that "some mortality and reduced growth rates of winter flounder larvae entrained within the dredge induced TSS plume may occur." We agree that there will be fish mortality, but I believe the zone of impacts (including lethal concentrations and durations) is larger in spatial dimension, and greater in duration than the US ACE report acknowledges.

Considering RI DEM's analysis and Dr. Newcombe's comments, it is our opinion that adverse impacts to living marine resources in the dredging area will be considerably greater than concluded in the DEIS. Given the status of the Providence River reach as a larval retention area and the critically low abundance of key species, imposition of dredge windows is needed to minimize these impacts. The dynamic nature of larval fish and shellfish distribution in time and space in this reach does not allow for small scale sequencing of the dredge operation. Examination of the station abundances in sector 2 of the Bourne and Govoni (1988) study show that the distribution of larvae is fairly uniform within the reach so that sequencing would be ineffective in protecting impacted larvae.

The ACOE states that there are no self-sustaining populations of anadromous fish in the Providence River or associated tributaries. Field studies conducted by the DFW from January through December 1996 showed the presence of river herring (alewives and blueback herring), hickory shad, and American shad in the Providence River from Conimicut Point north to Bold Point at the confluence of the Seekonk River. Sufficient numbers of these species were collected during this study to indicate that there are self-sustaining small populations of alewives, blueback herring, and American shad in the Providence and/or Seekonk River systems or smaller tributaries emptying into these rivers. We know that there is a population of river herring in the Ten Mile River, a tributary of the Seekonk River. This population has been partially sustained by fisherman netting the ripe and running adults over the Omega Pond dam. The low population numbers of all these species indicate that any adverse impact caused by the dredging or disposal would be particularly damaging.

There are also self-sustaining populations of river herring and American shad in the Warren, Barrington, and Palmer River system as indicated by studies done for Mobil Oil by Stonybrook Laboratories. These populations migrate through the project area and would be impacted by suspended sediment resulting from the dredging and disposal operations. They were not addressed in the DEIS. Newcombe & Jensen (1996) indicate that increased suspended sediment can inhibit migration of anadromous species.

We also note that no catadromous fish resources or concerns were identified in the DEIS, although the American eel is found throughout the area.

The Narragansett Bay ecosystem is a mosaic of different habitats, which include varied bathymetric features. The species inhabiting the estuary have evolved to exploit different habitats at various times during their life. Juvenile winter flounder have higher temperature tolerances than adult stages (ASMFC 1992) and adult lobsters become freer ranging than cryptic juveniles which remain in the cobble habitat (Krouse 1973). This diversity of habitats supports a diversity of living resources. Deep water habitats such as Site 3 are rare in Narragansett Bay (Chinman and Nixon 1985). Habitat greater than 20 meters in depth makes up less than 6% of the total Bay area. Mean summer water temperatures in the deep areas of the Bay are considerably cooler than shallow areas based on measurements made during DFW and URI trawl surveys (Figure 14, attached). Based upon tagging studies, we know that winter flounder use these areas during migrations into and out of the Bay. We, therefore, disagree with the ACOE assertion that site 3 is not used by winter flounder. To the contrary, winter flounder tagged in the Upper Bay, and in Mt. Hope Bay, by RI DEM have been recovered in this area. Moreover, Oviatt and Nixon (1973) sampled at site 3 among others in the Bay and found that winter flounder were the most abundant species overall with a distribution correlated with water depth. The fitted temperature regression in Figure 14 predicts a mean June-September temperature of 17.2 C at a depth of 70 feet. If the depth at the Hog Island site is reduced to 40 feet as a result of dredged material disposal, mean summer temperature is predicted to rise to 19.3 C. Temperature preferences for winter flounder were recently summarized by Pereira et al. (1998) in reference to essential fish habitat. Adult and juvenile winter flounder have different preferred temperatures. Adults prefer temperatures of 13.5 C and leave for cooler areas when temperature reaches 15 C while juveniles prefer 18.5 C. The predicted rise in temperature at site 3 will certainly exclude adult winter flounder and may limit the residence of juveniles. Also, the ACOE has ignored tautog migration patterns. Analysis of tag returns by Lynch (1994) show that site 3 is along a migration route for tautog.

There is a clear relationship between tow depth and catch per unit effort of fish in the DFW trawl survey from 1990-1997 (Figure 15, attached) as was the case in the 1971-1972 survey of Oviatt and Nixon (1973). Deep water stations like site 3 generally produce about four times as much finfish catch as shallow-water stations. Lobster utilization of this area is also well documented. Data from RI DEM Lobster Research and Management Project collected from Narragansett Bay from 1990 to 1998 clearly show that catch per trap haul is highest in deep water sites in the Bay (Figure 16, attached). Lobster CPUE in deep sites is generally twice that of the shallow areas. The survey and bathymetric data are combined and summarized in Figures 17 and 18 (attached).

In summary, then, a disproportionate share of the survey catch comes from the 5.4% of the Bay with depths over 20 meters. We calculated, in fact, that these areas support 25 % of the Bay's standing biomass. Given these clear relationships, deepwater areas such as site 3 are considered uniquely valuable habitats, which must be protected. We dispute the ACOE contention that this dumpsite would return to "normal" after a period of time since, by their own projections, its depth would be permanently reduced from 20 to 14 meters. The above cited relationships between increased depth and increased biomass predict a permanent 48 % loss in fish abundance in the 500-acre area impacted by dredged material disposal as a direct consequence of depth reduction.

The fidelity of winter flounder to their natal spawning areas and established migratory routes is well documented. Tagging studies by Saila (1962) and Powell (1989) showed that adult winter flounder return to the same spawning area in Narragansett Bay where they were born, and use the same migratory routes to reach the spawning areas. Saila showed that 25% of fish captured in Mount Hope Bay and released in the Sakonnet River were recaptured in Mount Hope Bay, and 36% of the fish caught in the Sakonnet River and released in Mount Hope Bay were recaptured in the Sakonnet River. The remainder of the fish tagged in these areas was recaptured outside the tag and release area. In studies of spawning winter flounder in Rhode Island coastal ponds Perlmutter (1947) and Saila (1962) both found, with few exceptions, that winter flounder return to the same pond in which they were spawned. Crawford (1990) not only showed that winter flounder return to the same pond but also return to the same area of the pond. A tagging study conducted by Howes and Coates (1975) showed little interchange between winter flounder tagged both north and south of Cape Cod; and recent tagging studies by Lawton (pers. com. 1998) indicate that winter flounder tagged in Plymouth Harbor, MA area have a strong fidelity to that estuary.

These studies show that spawning and migration behavior of winter flounder has evolved over thousands of years and demonstrates that man-induced alteration of any habitat upon which the population depends will have severe adverse impacts on the population. The elimination of the deep water "hole" at Hog Island (Site 3) which is used by winter flounder will impact migration of this important species generally and may be expected to have a particularly severe impact on those discrete populations which have evolved to incorporate Site 3, as distinct from other deep areas of the Bay, into their annual migration and spawning patterns.

We are concerned about the impacts to shellfish adjacent to the channel that may be destroyed during dredging of the channel or by burial. The DEIS states that quahogs and soft-shell clams are abundant in the lower river below Fields Point (pages 6-50 and 6-75). The DEIS further states that at areas directly adjacent to the dredge, siltation may cause additional mortalities to the benthic communities in the project area (page 7-60), but does not evaluate the impact to the shellfish in these areas from burial. It is our opinion that the shellfish resources in the areas to be dredged have not been adequately addressed or accurately identified. Five, one square meter quadrats sampled per reach in the navigational channel are not adequate to estimate the spatial distribution and abundance of this resource. We agree that few quahaugs are located at the bottom of the channel but believe based upon our own surveys that shellfish have colonized the existing channel slopes over the years. In this regard, RI DEM dredge surveys during the spring and fall of 1998 in the area adjacent to the navigation channel between Field's Point and Ohio Ledge show abundances of quahaugs up to 30.9 per m². Mean quahog density was 15.0 per m² with a 95 % confidence bound of 11.3-18.7. The DFW survey dredge cannot fish directly in the channel but recent transplant operations by the URI RV Captain Bert under contract to DFW have found substantial quantities of quahaugs in deeper water adjacent to and in the channel near Conimicut Point. Although density estimates have not yet been computed from the URI tows, catches are in the range of 400-500 lbs per tow, which will be in the high range of densities reported above. Recent survey results are plotted in Figure 19 relative to the channel. These resources will likely be destroyed during dredging.

Magnuson-Stevens Act

In 1976, the Magnuson Fishery Conservation and Management Act established a management system for fisheries in the U.S. exclusive economic zone (EEZ) which extends from 3 to 200 miles offshore. Under this system, eight Regional Fishery Management Councils develop fishery conservation and management measures. After approval by the Secretary of Commerce, these management measures become Federal regulations enforced by the National Marine Fisheries Service (NMFS). In October 1996 the Sustainable Fisheries Act (SFA), among other things, amended the habitat provisions of the Magnuson Act. The renamed Magnuson-Stevens Act calls for direct attention to stop or reverse the continued losses of fish habitats. Congress mandated the identification of habitats essential to manage species and measures to conserve and enhance that habitat. Essential fish habitat (EFH) is broadly defined to include "those waters and substrate necessary to fish for spawning, breeding, feeding, or growth to maturity."

To improve fish habitat protection, the SFA requires or authorizes the Regional Fishery Management Councils, NMFS, and other federal agencies to:

1. describe and identify the essential habitat for species managed by the Council;
2. minimize to the extent practicable adverse effects on EFH caused by fishing; and
3. identify other actions to encourage the conservation of EFH.

The Councils must describe EFH and identify adverse impacts and conservation measures for the managed species. Councils, with the assistance of NMFS, must amend all EEZ Fishery Management Plans to include these components. Habitat management elements are also incorporated into plans produced for inshore state waters by the Atlantic States Marine Fisheries Commission (ASMFC) and the State of Rhode Island.

The Act also requires EFH that are judged to be particularly important to the long-term productivity of a species or be particularly vulnerable to degradation to be identified as "habitat areas of particular concern" (HAPC). These areas must meet one or more of the following criteria:

1. the importance of the ecological function provided by the habitat;
2. the extent to which the habitat is sensitive to human-induced environmental degradation;

3. whether, and to what extent, development activities are or will be stressing the habitat type; and
4. the rarity of the habitat.

The intent of the HAPC designation is to identify those areas that are known to be important to species that are in need of additional levels of protection from adverse impacts.

Under the New England Fishery Management Council's Essential Fish Habitat Amendment (Draft dated September 16, 1998) the Council is required to identify and characterize activities other than fishing that potentially reduce the quantity and/or quality of essential fish habitat. Dredging and dredged material disposal are listed as anthropogenic non-fishing threats to EFH. Clearly, the Providence River reach as a larval retention area qualifies as EFH-HAPC under criterion 1. The deep water holes in the Bay qualify under criterion 4. In Section V. Conservation and Enhancement Measures, all fishery management plans (FMPs) are required to identify actions to promote the conservation and management of fishery resources. The Council also has the discretion to provide comments on non-fishing activities authorized by federal and state agencies that impact EFH. To address anthropogenic non-fishing threats to EFH the council has made the following recommendations regarding physical threats to EFH caused by dredging:

Congress has also provided for cooperative management in state waters with the Atlantic Coastal Fisheries Cooperative Management Act of 1993 (ACFCMA). Under this act, the Atlantic States Marine Fisheries Commission (ASMFC) is delegated responsibility to coordinate inshore (< 3 miles) management of migratory fish stocks. The ASMFC program charter identifies standards for interstate fishery management plans:

Under the management program elements, the Commission's charter lists supporting information and analyses of which item B is;

As an example of specific FMC requirements we refer to the ASMFC winter flounder management plan. Relating to habitat, the ASMFC (1992) plan requires that:

These habitat protection activities are listed in the FMP and include:

Similar habitat elements are found in the ASMFC tautog and weakfish management plans. Section 4.5.2 of the April (1996) tautog plan, Avoidance of Incompatible Activities, states:

Comparable language in the May (1996) weakfish plan, section 3.1.2 Avoidance of Incompatible Activities, states:

Based on RI DEM's above described analysis of the project impacts and the regulatory provisions of the Magnesun-Stevens Act and the Atlantic Coastal Fisheries Cooperative Management Act, RI DEM believes that the Providence River reach and deep water areas of the Bay qualify as EFH with HAPC designation.

The issue of essential fish habitat (EFH) has not been addressed in the DEIS. Considering the above information and the importance of this issue to Narragansett Bay fisheries, the issue of EFH needs to be addressed by the ACOE.

The State of Rhode Island CCMP was approved by the Narragansett Bay Project executive committee on August 4, 1992 and became part of the State Guide Plan in December 1992, making the implementation of the Plan a state policy. The Governor and the EPA Administrator signed the document in January 1993. The CCMP sets forth goals and implementation strategies to improve and sustain the health of Narragansett Bay. One of the objectives in the CCMP is to protect critical habitat and resources. The CCMP states that "the State of Rhode Island and the Commonwealth of Massachusetts should protect all critical natural resources; manage all designated critical resource areas for the benefit of the public and the ecological protection of the Bay and its tributaries; protect these critical resource areas from any irreversible degradation; and where necessary, restore impaired critical resources". Based on the DFW determination that disposal site 3 is EFH and a HAPC, there is clearly a need to recognize that deep holes of the upper Narragansett Bay are rare critical habitat, and are critical to migratory staging behavior of several species, especially winter flounder. The loss of any of these by filling in deep holes in the Bay will likely have permanent impact and repercussions on the ability of species like winter flounder to eventually return to normal sustainable levels. The CCMP mandates that we protect such areas (the deep holes are apparently critical staging areas and refugia for temperature-driven behaviors like migration for at least winter flounder and lobsters and very likely for other species as well).

The DEIS states that research conducted by Truitt has shown that disposal losses typically range from 1-5% of the amount disposed. We identified a 1989 report prepared by Gentile and Pesch (EPA) and Scott and Munns (SAIC) entitled "Applicability and Field Verification of Predictive Methodologies for Aquatic Dredged Material Disposal", which assessed the fate of disposing dredged material from Black Rock Harbor in Long Island Sound. Approximately 55,000 cubic yards of material was dredged from the channel and disposed in 66 feet of water. The report concluded that based on mass balance estimates of the dredging and disposal operations 20-30% of the material was unaccounted for, and only 30-60% of selected contaminants reached the sea bed at the disposal site. The report further stated that "the high organic content (6%), high water content (69%), and fine grained nature of this maintenance dredged material creates special problems for containment during both dredging and disposal operations". This is similar to the material from the Providence River channel. The report provides no further information, so it is unclear how much of the material that is unaccounted for was related to the dredging operation and how much was related to the disposal operation. The RIDEM would like to discuss the estimates that the ACOE used in the DEIS to determine whether these are reasonable.

The RIDEM remains concerned about long- term impacts from the disposal of unsuitable material caused by losses of sediment during the disposal process and erosion of uncapped sediments, if capping of unsuitable material is selected.

The volume of unsuitable material that may be disposed at an open water site is 1.2 million cubic yards. Assuming a 5 % loss of sediments during disposal (as indicated on page 7-16 of the DEIS), this means that 60,000 cubic yards of unsuitable material will be released to the water column.

Sediment dispersion will also occur prior to capping. The DEIS states that if an environmental bucket is used and interim capping is done, the duration that unsuitable material would be exposed to erosive currents would be 90 days. If it is assumed that the material will erode at a constant rate extrapolation of 16 peak velocity occurrences/month multiplied by the .03 feet presented in table 7.2.1-1, total erosion depth would be .48 feet of material per month. This assumption seems reasonable because an armored or more consolidated layer will not occur until after disposal has ceased. Using site 3 as an example, the DEIS states that a majority of the material will be contained in a 134 acre area. The total volume of unsuitable material that will be eroded until an interim cap is completed will be about 310,000 cubic yards (.48 feet/month x 134 acres x 3 months).

Based on this analysis, the dispersion of unsuitable material caused during disposal and erosion will be about 370,000 cubic yards, which represents about one third of the total volume of unsuitable material. If it is assumed that these sediments all deposited within one half mile of the disposal site, it would result in a 7.5 inch layer of unsuitable material. Further analyses are needed prior to placement of unsuitable material in these open water sites (if such disposal is identified as part of the preferred alternative or recommended for implementation) to demonstrate that no long term impacts to the aquatic environment will occur.

The RIDEM also remains concerned about long-term impacts from the disposal of clean material caused by losses of sediment during the disposal process and erosion of unconsolidated sediments. The volume of clean material that may be disposed at an open water site is 4.4 million cubic yards (2.8 million cubic yards from the channel and 1.6 million cubic yards from the CAD site). Again assuming a 5 % loss of sediments during disposal, this means that 220,000 cubic yards of clean material will be released to the water column.

The erosion analysis that was performed only presents erosion that will occur after the material has consolidated. If it is assumed that the material will erode at a constant rate extrapolation of 16 peak velocity occurrences/month multiplied by the .03 feet presented in table 7.2.1-1, total erosion depth would be .48 feet of material per month until disposal is complete and the material has consolidated. This assumption seems reasonable because during disposal and after disposal until the sediments have consolidated surface fine material will be exposed to these currents and winnowed away. Using site 3 as an example, the DEIS states that a majority of the material will be contained in a 134 acre area. The total volume of clean material that will be eroded after disposal has ceased and prior to consolidation will be about 415,000 cubic yards (.48 feet/month x 134 acres x 4 months). If it is assumed that disposal will occur over 18 months and 7.5 acres are used for disposal each month, which represents one eighteenth of the total disposal area, the total volume of material that will be eroded during disposal will be 105,000 cubic yards (.48 feet/month x 7.5 acres x 1 month).

The DEIS concludes that because the overall amounts of material that may be eroded are expected to be small, placement of clean dredged material would be acceptable. Based on this analysis, the dispersion of clean material caused by losses of dredged material during disposal and erosion during disposal and after disposal until the material consolidates will be about 740,000 cubic yards, which represents about one sixth of the total volume of clean material. Further analyses are needed prior to placement of clean material in these open water sites to demonstrate that no long-term impacts to the aquatic environment will occur.

Appendix A, page 25 discusses water- related recreation, but does not specifically mention kayaking. Sites located in shallow waters are searched out and used by kayakers. The ACOE should evaluate whether any of the near shore sites are used by kayakers.

Appendix B, page B-3 states that consolidation of the dredge material will occur after disposal and provides an estimated consolidation factor of 0 to 50 percent. Are the final depths that are described in the DEIS for the open water disposal sites the depths that will be present after consolidation has occurred? If not, the ACOE needs to explain the final depths that can be expected at the open water sites after the material has consolidated and the time it will take for this to occur.

The quantity of dredged material, depths of dredging, and/or area of dredging that is indicated for some of the applicants are not consistent with the applications that were submitted. Simply comparing the quantities of dredged material indicated in the applications with the quantities indicated in the DEIS and Appendix F shows that the quantity of dredged material is underrepresented by 81,800 cubic yards. The ACOE should review each application to determine the correct depths of dredging and area of dredging and calculate the correct quantity of dredged material that will be generated for each applicant.

We concur with the recommendation for a dredging window of November 1 to January 1 for the associated dredging areas, as stated on page 7-97 of the DEIS. Site specific impacts and requested extensions to the above dredging window will be dealt with on an individual and species specific basis.

It is unclear what area of benthic habitat will be impacted by disposal. The figures shown in Section 4.12.1 for the open water sites do not accurately represent the area of benthic habitat that will be impacted, as indicated in the text. For example, for site 3 the DEIS states that the disposal mound would cover about 430 acres of bottom at this site, with a majority of the material contained in a 134 acre area. The area shown on Figure 4.12.1-2 represents only about 100 acres. The areas indicated in Section 4.12.1 are also inconsistent with the information presented in Section 7.3 regarding impact to benthic organisms. The ACOE needs to provide an accurate representation (on a map) of the total area of benthic habitat that will be impacted for each open water site. The map should also clearly delineate the area within which each benthic organism type will be destroyed by burial. The ACOE also needs to explain how these areas of impact were determined.

The DEIS does not assess quantitatively the number of finfish, crustaceans, and shellfish (by weight and/or number) that will be destroyed by the disposal operation. The EIS Scope and Workplan indicated that the ACOE would evaluate compensatory measures against the significance of the resource. The ACOE needs to include this information so that it can be determined whether compensatory measures should be considered.

The DEIS states that the grain size distribution of the disposal sites is not known, although sediment profile images indicate that they are primarily silty. The DEIS further states that permanent alteration of the substrate would occur if the uppermost material disposed at an open water site is of substantially different grain size distribution than the existing sediments (page 7-81). The ACOE needs to perform further testing to verify the grain size distribution of the disposal sites and re-evaluate the conclusions about the potential for permanent alterations, if an open water site is selected.

Short-term water quality monitoring at any open water site needs to be considered during disposal to verify compliance with the State's water quality standards outside the mixing zone. A third party independent observer should do this activity. The monitoring should include measurements taken before, during and after dredging operations. Spatial and temporal extent and concentration of suspended sediment should be monitored for threshold levels established by the Newcombe (1996) model. Remobilization and fate of contaminants should also be monitored. Threshold tissue levels should be established for the consumption of finfish and shellfish meats. Prior to this monitoring background levels of these contaminants in these organisms should be established. The ACOE also needs to consider measures that can be taken to comply with water quality standards if monitoring indicates water quality standards violations outside the mixing zone.

Short and possibly long term monitoring at any open water site needs to be considered after disposal if clean material is disposed to determine whether the benthic habitat quality has restored to its existing condition and whether the material at the site is remaining in place. If unsuitable material will be capped at an open water site, short and long term monitoring needs to be considered to determine whether the unsuitable material has been properly capped and the cap is stable and whether the benthic habitat quality has restored to its existing condition. The ACOE also needs to consider measures that can be taken if monitoring indicates that the benthic habitat has not restored to its existing condition, unsuitable material has not been properly capped, the cap is not stable, or clean material disposed at the site is migrating off-site

We agree that further modeling is needed to define the mixing zone requirements for CAD cell disposal and open water disposal, if these sites are selected. Results indicate that dilution is needed for several parameters to reach the acceptable criteria. These parameters are toxicity, coliform bacteria, cyanide, ammonia, silver, and copper. For CAD cell disposal, toxicity requires the greatest dilution to reach the acceptable criteria, therefore, this parameter and not copper (as stated in the DEIS) would dictate the size of the mixing zone. For open water disposal, the RIDEM is not convinced that toxicity requires the greatest dilution. The DEIS states that the required dilution for coliform bacteria is 620 and the required dilution for toxicity is about 1700. It is unclear from the discussion in the DEIS how a dilution of 1700 was determined for toxicity. Also, as the background water results indicate that the ambient water exceeds the acute water quality criteria for copper, the ACOE needs to evaluate how much over these ambient levels can be expected for CAD cell disposal and open water disposal.

The RIDEM assessed the coliform bacteria concentrations that can be expected over background for the maximum plume concentrations that will occur in the Providence River channel during dredging using worst case assumptions. This analysis indicated that the coliform bacteria concentrations that can be expected over background are 3 coliform bacteria/100 ml. This increase will not result in a violation of the water quality criteria for the Upper Narragansett Bay conditional shellfish area. Based on this analysis, we do not anticipate the need for additional shellfish closures except in the area around an open water disposal site during disposal and for some period after disposal has ceased, if an open water site is recommended. The RIDEM would need to close this area to comply with the National Shellfish Sanitation Program (NSSP) requirements for all waters that are approved for shellfish harvesting that are within Narragansett Bay and offshore waters within 3 miles of the shoreline. The extent and duration of the closed area still needs to be determined based on further modeling that will be done to define the mixing zone.

The ACOE needs to perform further engineering studies to determine whether the proposed CAD sites have the capacity to accommodate the volume of unsuitable material that will be generated by this project. It is unclear whether the CAD sites have the capacity to accommodate all the unsuitable material in the Federal channel, the material overlaying the CAD site, and the material from the non-Federal applicants. The ACOE prepared suitability determinations for Brewer's Cove Haven Marina, Brewer's Yacht Yard, Brewer's Sakonnet Marina, and Brewer's Wickford Cove Marina, which were reviewed by EPA and NMFS. It was determined from this review that this material is unsuitable for unconfined open water disposal and would need to be disposed in a confined aquatic site, unless the applicants wanted to perform further biological testing to demonstrate that it was acceptable for unconfined open water disposal. It is possible that the other 7 applicants outside the Providence River may need to dispose of their dredged material in a confined aquatic site as well.

The ACOE needs to clarify the depth of material overlaying the CAD site at Watchemoket Cove that will be removed for later burial in the CAD site. The DEIS indicates that the top 1 foot of material will be removed, however, previous discussions with the ACOE indicated that the top 2 feet of material would be removed. The RIDEM is not convinced that removal of only the top 1 foot will be sufficient.

The ACOE needs to discuss the depth of material below 40 feet MLW for the in-channel CAD site that is considered suitable for open water disposal and how this conclusion was determined. There is no discussion of this in the DEIS.

The ACOE needs to clarify whether the material overlaying the CAD site at Watchemoket Cove and the material in the in-channel CAD site below 40 feet MLW that is considered unsuitable for open water disposal will be stored in scows for the entire period while the CAD site is being constructed. For the 48- acre CAD site (assuming 1 foot of material is unsuitable), about 80,000 cubic yards of material would need to be stored in scows, which would require twenty-six 3,000 cubic yard scows or thirteen 6,000 cubic yard scows. Storing these scows for the period of time it will take to construct the CAD site does not seem practical and may result in nuisance conditions to the surrounding areas.

We have reservations regarding the methodology and conclusions drawn from the assessment of the economic impact to the commercial fisheries. In particular the assumption of a uniform distribution of fish landings in Narragansett Bay is not supportable. Not all habitat is equal and this needs to be taken into account. Also, the economic impact to recreational fisheries was not evaluated. In some cases the recreational fishery is equal to or larger than the commercial fishery and generates considerable economic activity. The DEIS also does not indicate what multiplier, if any, was used for the in-Bay sites to account for further spending.

Section 5.2 of the DEIS summarizes the various reasons that were used by the ACOE to determine that open water disposal at site 3 was the preferred alternative. One of these reasons is that the other alternatives that were considered had much higher costs for disposal. The RIDEM has a number of questions regarding the cost estimates that were presented in Appendix B for the various disposal options.

Table 1 includes a cost of $19/CY to dredge and dispose of material into the CAD site. How was this cost figure determined? Does the cost include the additional volume that will be generated by bulking? If not, the cost for the 4 options shown should be revised for this volume. For the Watchemoket Cove CAD site, does this cost include the removal of the top layer of contaminated sediment from the CAD site and the disposal of this material in the CAD site? If not, the cost for the 3 options shown should be revised for this volume.

Table 1-A includes cost figures of $13CY to $22/CY to dredge and dispose of material at an open water site. How were these cost figures determined? For the Watchemoket Cove CAD sites, do these costs include the additional volume of clean material that must be dredged from the CAD site to accommodate the top layer of contaminated sediment from the CAD site and the additional volume that will be generated by bulking? If not, the costs for the 3 options shown should be revised for these volumes. Also, for the Watchemoket Cove CAD sites there is a discrepancy between the capacity figures shown in this table and the capacities indicated on page B-6.

Tables 2 and 3 include cost figures to dredge and dispose of material at the Bay sites. Do these costs include the additional volume that will be generated by bulking? If not, the costs need to be revised for these volumes.

Table 4 includes cost figures for dredge disposal, dike/tube construction, bulkhead construction, and landscaping. How were these cost figures determined? For the Spar Island sites, do the costs for dredge disposal for the 225 acre, 280 acre, 740 acre, and 930 acre sites include the additional volume that will be generated by bulking? If not, the costs need to be revised for this volume. Do the costs for dike/tube construction include the cost for construction of the entire capacity of the 225 acre, 280 acre, 740 acre, and 930 acre sites? If not, the costs need to be revised to include the entire capacity of the sites, which were indicated in the text to be 6,000,000 CY for the 225 acre and 740 acre sites and 7,500,000 CY for the 280 acre and 930 acre sites.

Table 5 includes cost figures for dredge/disposal, de-watering, hauling, and upland construction. How were these cost figures determined?

The RIDEM remains concerned about the dissolved oxygen impacts during dredging and disposal. The DEIS references only one study that was done that monitored dissolved oxygen concentrations during dredging, which was during dredging of parent materials for the Conley Terminal in Boston Harbor. The DEIS indicates that parent material is more cohesive than surface silts and less parent material is suspended in the water column during dredging (page 7-13). Also, parent material will have a lower oxygen demand than surface silts because of the lower organic content of the sediment. Based on this it appears reasonable to assume that greater reduction than .6 mg/l would occur from dredging silt from the Providence River. The only study that the RIDEM could find on this topic was one published by Mark LaSalle, ACOE, Waterways Experiment Station, entitled "Physical and Chemical Alterations Associated with Dredging: An Overview." The author summarized the results of four previous studies that had been done where direct measurements of dissolved oxygen concentrations around dredges had been taken. These studies were a bucket dredge operation in a highly industrialized channel in New York (1968), a cutterhead dredge operation in Gray's Harbor, Washington (1976), a hopper dredge operation in a tidal slough in Oregon (1982), and a bucket dredging operation in a widened portion of the lower Hudson River, New York (1988). The results from these studies indicated that dissolved oxygen reductions were observed at all four sites. Reductions in New York channel were 16%-83% in the mid to upper water column and by as much as 100% in near bottom layers. Periodic reductions of bottom water dissolved oxygen of up to 2.9 mg/l were observed in Gray's Harbor. Reductions in dissolved oxygen (1.5-3.5 mg/l) at the Oregon site were observed during slack water periods in the lower third of the water column, lasting until tidal flow resumed. Reductions in dissolved oxygen at the Hudson River site were generally <.2 mg/l and restricted to the lower water column. The reports by Doering, Pilson, and Oviatt (1988), Doering, Oviatt, and Welsh (1990), and the most recent data collected by Turner in June 1995 and October 1996 all indicate that hypoxic (less than 3 mg/l) conditions occur in bottom waters in the Providence River during the summer months.

The two model approaches used in the DEIS were initially developed to assess dissolved oxygen depletion around a bucket dredging operation. It is unclear whether these models can be used to accurately predict dissolved oxygen depletion at an open water disposal site. As we indicated in our March 6, 1998 letter to the ACOE, the RIDEM has several questions regarding field verification of these model approaches that we need to discuss further with the ACOE. We also have questions about the values that were used in the models for benthic oxygen demand and ferrous iron and sulfide concentrations. Also, there is some evidence that dissolved oxygen water quality violations (hypoxia) may exist at disposal site 3. The Narragansett Bay Estuary Project (NBEP) reported on oxygen depletion occurring in Bay waters as far south as Ohio Ledge during late July and early August. Data collected by the Narragansett Bay National Estaurine Research Reserve (NBNERR) at the South Prudence T-Wharf showed 8 days between July 12 and September 22, 1997 with dissolved oxygen levels at 6 mg/l or less. There were 26 measured occurrences of dissolved oxygen levels below 6 mg/l during summer months in 1994 through 1998, which were measured once per week. A water quality monitoring instrument was located one meter above the bottom at a depth of 12 feet mlw throughout the summer of 1997. Readings were taken continuously at 30-minute intervals. The hypoxia was usually observed in early morning before first light and again in early evening after sunset. There were a number of short-term events with levels of dissolved oxygen below 3 mg/l. Daytime dissolved oxygen levels were usually between the range of 5 to 7 mg/l. The NBEP observations and the NBNERR data indicate the potential for July and August oxygen depletion, which bracket disposal site 3. We can only extrapolate dissolved oxygen depletion observed at other locations to disposal site 3. In addition, the data NBNERR has indicates hypoxia but does not rigorously define the temporal and spatial magnitude of the problem. A copy of this data has been included for the ACOE information.

The ACOE needs to provide further justification for dredging the channel to maintain the authorized 40-foot project depth. The DEIS discusses two primary reasons for maintaining the authorized 40-foot depth, which are economic cost savings and environmental harm posed by lightering and increased truck traffic (if companies decide to off-load at another port and transport petroleum by truck to Rhode Island). The information on pages 7-131 through 7-134 provides little reason to dredge beyond depth 37. The cost savings are only $50,000 per year and there is only 1 less lightering.

The ACOE should also do the economic benefit analysis of dredging to -40 feet, similar to the economic benefit analysis on page 7-133 on 2 way traffic. If the result of this analysis changes the overall project as far as the area that needs to be dredged or the depth of dredging, this needs to be explained.

We agree that further modeling is needed to define the mixing zone requirements for de-watering site effluent, if upland disposal is selected. As with open water disposal, toxicity requires the greatest dilution to reach the acceptable criteria, therefore, this parameter would dictate the size of the mixing zone.

It is unclear what minimum acreage is needed for a de-watering site or whether all of the available sites have been considered. Appendix B, page B-8 states that 25 and 40-acre sites were considered. The ACOE used 54 acres in Appendix A. Also in Appendix A, the ACOE included a letter from the Economic Development Corporation that indicated that a 36-acre site was needed. The criterion of a minimum of 54 acres assumed a rate of production of 250,000 cubic yards per month of dredged material, which is only correct if all the material from the Providence River is disposed upland. Disposal of smaller volumes of dredged material or sequencing of dredging would lower the acreage that is needed. The DEIS states that 2 sites were considered at Fields Point (Appendix A, page 10), however, it does not state whether either of these sites are currently used by RIDOT to de-water dredged material. If not, this site may be available. The ACOE would need to clarify the acreage that is needed for a de-watering site, if upland disposal is selected, and explain whether all of the available sites at Fields Point and Quonset Point for that acreage were considered.

The RIDEM has comments regarding acceptable truck routes and impacts to freshwater wetlands and agriculture.

The DEIS states that an acceptable route to sites 27, 28, and 152 was not available because the route to the south involves a one-way street, busy streets with schools, and impacts on a residential area. The RIDEM noted that the routes for these sites (until they connect to major highways) are the same as site 74, so it is unclear why only site 74 had an acceptable route from Fields Point.

Freshwater Wetlands

Site 28 would likely require an application due to the wetlands located immediately southwest of the deposition area. Site 27 will definitely require an application and it may be significant. Wetlands will be eliminated and adjacent wetlands are contiguous to Buckeye Brook. Site 82 has a forested wetland that fringes the site along the entire eastern and southeastern border, which has not been noted in the DEIS. A flowing waterbody is present in this wetland. The filling of the wetlands in their entirety may be significant as opposed to what is stated in the DEIS on page 7-119. There will be a permanent loss of wildlife habitat due to filling into existing "retention ponds". These retention ponds contain macrophytic vegetation. Site 152 has many wetlands and watercourses that surround and pass through the site, which is not noted in the DEIS. A very involved permitting process with edge verification may be helpful to assess overall potential impacts of the landfill area.

The sites do not appear to pose any significant impacts to agricultural land or activities if they were utilized (assuming lining of all disposal sites to prevent groundwater contamination). It is evident that the screening process has eliminated lands that the Division of Agriculture has expressed previous concerns regarding disposal of dredged material and impacts to agriculture.

One concern the Division of Agriculture has regarding the upland disposal sites involves the trucking and transport of dredged material. The disposal sites themselves are not in proximity to agricultural land, but the potential routes of travel from the de-watering site to the disposal site may involve travel past roadside farmstand operations and fields in active agricultural usage. Dredged material blown off of transport vehicles has the potential to contaminate fruits and vegetables for sale at roadside stands as well as fields bordering roads, which are in active usage. Adequate safeguards must be in place on transport vehicles to prevent this from occurring.

Assuming that some of the dredge materials will be disposed of on shore in some type of CAD cell, there will be shellfish present in the dredged materials. These will likely be unfit for human consumption due to the presence of either fecal coliform bacteria or heavy metals. There has to be a level of security present at the dredge discard site to insure that members of the public do not remove these shellfish either for personal consumption or sale.

We agree that further modeling is needed to define the mixing zone requirements for effluent from salt marsh creation or island expansion, if a beneficial use site is selected. As with the de-watering site and open water disposal, toxicity requires the greatest dilution to reach the acceptable criteria, therefore, this parameter would dictate the size of the mixing zone.

The use of geotextile bags filled with dredged materials was evaluated under reef creation. One of the disadvantages cited on page 4-134 was that using geotextile bags is a new and still unproven technology. In 1994, personnel from the ACOE Waterways Experiment Station presented information to members of the Dredging Contractors of America (DCA). A scientific paper titled Dredged Materials Filled Geotextile Containers by Jack Fowler and C. Joel Sprague (WES personnel) was provided to the DCA. The conclusion of their research and a demonstration project in Mobile, Alabama harbor were that "filling of geotextile containers with fine-grained contaminated dredged material is feasible, gaining popularity, is cost effective, and minimizes impact on the environment".

The ACOE needs to reconcile the statement in the DEIS with the information that was provided to the DCA. The ACOE should seriously consider the use of geotextile bags, if this is a feasible technology.

The RIDEM performed an investigation at site 149 relating to a hypochlorite spill in the area, which identified an abundance of soft shelled clams, as well as razor clams, mussels, oysters, mud snails, and 3 species of crabs. A copy of this report has been included for the ACOE information.

The conceptual upland dike section shown in Appendix B, plate U-2 will generate leachate that must be disposed if a liner is installed. Proper disposal of leachate is an important factor that needs to be discussed in the DEIS and this Appendix, if upland disposal is selected. Disposal of the leachate should also be included in the costs for the upland disposal sites.

The analysis that was performed on the cap thickness appears complete, and the RIDEM concurs with the recommended cap thickness of 40 inches. We also agree that further analysis would need to be considered as part of the design of temporary and permanent cap thickness, as described on page 7-7 of the DEIS.

The DEIS mentions in several sections that site 16 is outside of the State of Rhode Island's jurisdiction. The ACOE should clarify, as explained in the ACOE 19 September 1989 Dredging Guidance Letter, that within the territorial sea the ACOE policy is to apply for state 401 water quality certification and CZMA consistency for ocean dumping activities as a matter of practice.

1. In view of the EFH-HAPC status of the Providence River reach, its significance as a larval retention area, poor stock status of several species, and the Newcombe model results and comments, RIDEM recommends that no dredging activity take place from March through July. (This still allows 210 days per year for dredging.)
2. We concur with the recommendation for the use of a closed bucket.
3. In view of the EFH-HAPC status of deep water habitats in Narragansett Bay, their significance as cold water refugia for several valuable species, and their disproportionately high biomass, we recommend that Site 3 be rejected as environmentally unacceptable for the purpose of dredge material disposal.
4. The ACOE should re-survey the shellfish resources in the dredge path, especially the animals residing on the slopes of the navigational channel and the contiguous channel. These non-motile resources will be lost through removal or burial during dredging. An alternative to this loss would be to transplant the shellfish to down-Bay sites for eventual harvest or to a spawner sanctuary as part of mitigation.
5. The economic impact to commercial fisheries should be reviewed by a marine resource economist in light of the ACOE under-calculation of fisheries impacts both in regard to direct impacts on fish, shellfish, and lobster resources and in regard to areas such as Site 3 where commercial and recreational fishing resources are particularly concentrated and, hence, disproportionately valuable.
6. The impact of the project on essential fish habitat (EFH) has not been addressed in the DEIS. This shortcoming should be remedied in the Final EIS.

Atlantic States Marine Fisheries Commission (ASMFC. 1992). Fishery management plan for inshore stocks of winter flounder. Atlantic States Marine Fisheries Commission Fishery Management Report No. 21.

Bourne, D.W., and J.J. Govoni. 1988. Distribution of fish eggs and larvae and patterns of water circulation in Narragnasett Bay, 1972-1973. American Fisheries Society Symposium 3: 132-148.

Butelle and Rice. 1996. Larval quahaug abundance in Narragansett Bay. URI Fisheries Report

Chinman, R.A., and S.W. Nixon. 1985. Depth-Area-Volumen relationships in Narragansett Bay. University of Rhode Island Graduate School of Oceanography. NOAA/Sea Grant Marine Technical Report 87.

Crawford, R. 1990. Winter flounder in Rhode Island coastal ponds. R.I. Sea Grant Pub. #RIU-G-909-001. 24p.

Desbonnet, A., and V. Lee. 1991. Historic trends: water quality and fisheries, Narragansett Bay. The University of Rhode Island Coastal Resources Center Contribution No. 100 and National Sea Grant Publication #RIU-T-91-001. Graduate School of Oceanography, Narragansett, Rhode Island 101 pp.

Drinkwater, K.F., G.C. Harding, K.H. Mann, and N. Tanner. 1996. Temperature as a possible factor in the increased abundance of American lobster, Homarus americanus, during the 1980's and early 1990's. Fish. Oceanogr. 5:3/4 176-193.

Epifanio, C.E. 1988. Transport of invertebrate larvae between estuaries and the continental shelf. American Fisheries Society Symposium 3: 104-114.

Fausch, K.D., J. Lyons, J.R. Karr, and P.L. Angermeirer. Fish communities as indicators of environmental degradation. American Fisheries Society Symposium 8: 123-144.

Gibson, M.R. 1995. Comparison of trends in the finfish assemblage of Mt. Hope Bay and Narragansett Bay in relation to operations at the New England Power Brayton Point Station. Report to the Brayton Point Technical Advisory Committee. RI Division of Fish and Wildlife. Res. Ref. Doc. 95/1.

Gibson, M.R. 1997. RI Quahaug assessment: technical analyses in support of a baywide quahaug management plan. RI Division of Fish and Wildlife.

Gibson, M.R. 1998a. Recent trends in abundance, recruitment, and fishing mortality for winter flounder in Narragansett Bay. RI Division of Fish and Wildlife.

Gibson, M.R. 1998b. Recent trends in abundance, recruitment, and fishing mortality for tautog in Narragansett Bay, RI Division of Fish and Wildlife.

Howe, A.B. and P.G. Coates. 1975. Winter flounder movements, growth, and mortality off Massachusetts. Trans. Am. Fish. Soc. 104(1):13-29.

Jeffries, H.P., and M. Terceiro. 1985. Cycle of changing abundance in the fishes of the Narragansett Bay area. Marine Ecology Series 25:239-244.

Jeffries, H.P. 1994. The impacts of warming climate on fish populations. Maritimes 37(1): 12-15.

Keller, A. and G. Klein-MacPhee. 1992. Finfish, Chapter 6. In D. French and others, Final Report: Habitat Inventory/Resource Mapping for Narragansett Bay and Associated Coastline. Volumne II, Chapter 6, pp. 228-318. Report to Narragansett Bay Project, by Applied Sciences Ass. Inc.

Krouse, J.S. 1973. Maturity, sex ratio, and size composition of the natural population of American lobster, Homarus americanus, catch along the Marine coast. Fish. Bull. 73: 862-871.

Lawton, R.P. 1998. Per Comm. MA Div. Marine Fish., Pocasset, MA.

Lee, T.C., S.B. Saila, and R.E. Wolke. 1991. Winter flounder contaminant and pathological survey: Narragansett Bay and vicinity. Narragansett Bay Project Report NBP 91-51. Narragansett Bay Project, Prov. RI 65 pp. and appendices.

Lynch, T.R. 1994. Tautog tagging studies-Narragansett Bay and Rhode Island Coastal Waters 1987-1993 Completion Report. Rhode Island Division of Fish and Wildlife. Project No. F-54-R-5. Ref. Doc. TT-396.

Marine Research Incorporated (MRI). 1997a. Aquatic ecology studies in the Providence and Seekonk Rivers in the vicinity of the Manchester Street repowering project, February-November 1996. Report for the New England Power Company by Marine Research, April 1997.

Marine Research Incorporated (MRI). 1997b. Aquatic ecology studies in the Providence and Seekonk Rivers in the vicinity of the Manchester Street repowering project, February-November 1996. Appendix Document. April 1997.

Newcombe, C.P., and D.D. MacDonald. 1991. Effects of suspended sediments on aquatic ecosystems. North American J. Fish. Management 11: 72-82.

Newcombe, C.P., and J.O. Jensen. 1996. Channel suspended sediment and fisheries: a synthesis for quantitative assessment of risk and impact. North American J. Fish Management 16: 693-727.

Odum, W.E. 1970. Insidious alteration of the estuarine environment. Trans. Am. Fish. Soc. 99: 836-847.

Oviatt, C.A., and S.W. Nixon. 1973. The demersal fish of Narragansett Bay: an analysis of community structure, distribution and abundance. Estuarine and Coastal Marine Science (1973) I: 361-378.

Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. Torres Jr. 1998. Fishing down marine food webs. Science 279: 860-863.

Pereira, J.J., R. Goldberg, and J.J. Ziskowski. 1998. Winter flounder life history and habitat requirements essential fish habitat source document. Northeast Fisheries Science Center, National Marine Fisheries Service, Milford Laboratory.

Perlmutter, A. 1947. The blackback flounder and its fishery in New England and New York, Bull. Bingham. Oceanogr. Collect. Yale Univ. 11(2):1-92.

Pilson, M.E.Q. and C.D. Hunt. 1989. Water quality survey of Narragansett Bay: a summary of results from the SINBADD 1985-1986. Narragansett Bay Project. Report #NBP-89-22.

Powell, J.C. 1986. Winter flounder tagging study, 1986-1988 with comments on movements. RI Division Fish & Wildlife. 19 pp.

Powell, J.C. 1989. Winder flounder tagging study, 1986-1988 with comments on movements. Rhode Island Division of Fish and Wildlife. Res. Ref. Doc. 89/3.

Saila, S.b. 1962. Proposed hurricane barriers related to winter flounder movements in Narragansett Bay. Trans. Am. Fish. Soc. 91:189-195.

Sinclair, M. 1988. Marine populations: as essay on population regulation and specification. Washington Sea Grant. University of Washington Press, Seattle.