# LINJ 1998 Environment and Ecology

## Science Center Objects

The LINJ study area is in one of the more densely populated areas of the country. There is more than a century of a strong and prosperous industrial economy behind an industrial growth that was centered around the ports and metropolitan areas of New York City and Philadelphia, now considered the older urban portions of the study area. There also is a very productive agricultural heritage around the fringes of these metropolitan areas and in more rural areas of southern and northwestern NJ and eastern LI. These areas, however, are rapidly becoming urban and suburban communities. Below is a basic description of the study setting and resources that contribute to the diversity and complexity upon which the LINJ stratification and water-quality issues are based.

### Population and Land Use

The Long Island-New Jersey Coastal Drainages (LINJ) study unit covers more than 6,000 mi2 in New York and New Jersey. The study unit contains the metropolitan areas of New York City, Philadelphia, and the highly urbanized corridor in between. In 1990, the population of study unit was more than 10 million, concentrated mostly on Long Island and in northeastern New Jersey (figure 1). In 1973, the study area was 35% urban (28 percent urban residential and 7 percent urban nonresidential). We know that a significant increase (+9%) in urban land use and subsequent decrease in forest and agriculture land uses (-6%) has occurred between 1973 and 1990 in the study area. The large extent of urban area and continued growth in NJ and LI is the primary focus of this NAWQA study.

Stratification of the Study Area

Fortunately, there have been several significant efforts in NJ to develop an objective procedure for classifying or differentiating factors controlling both ground- and surface-water systems. Physiography plays a major role. An interagency ecomapping committee headed by the USFS is the most recent effort. Based on the ecoregion concept, they have defined subsections of NJ based on ecosystem differences within physiographic provinces (NJDEP 1994). This ecomapping fits well conceptually and practically with the stratification needs of the NAWQA study. It also works out well with Long Island; we simply treat LI as another subsection of the Coastal Plain. The result is a stratification of the LINJ into 6 ecoregion strata as shown in Figures a and b. We believe there is not enough differentiation between the inner and outer Coastal Plain subregions and so we lumped them as Coastal Plain-NJ. Furthermore, because land use plays such a dominate role in this study area, we believe that both SW and GW stratification can be simplified even further by combining similar ecoregion strata back to 3 primary SW strata and 4 primary GW strata. Land use then becomes the primary differentiating substrata within each of these primary strata.

Description of the Study Area

About two-thirds of the 6,000 square mile study area is Coastal Plain and characterized by flat to gently rolling topography of unconsolidated sedimentary (NJ and LI) and glacial (LI only) deposits. The other one-third is within the Piedmont and New England provinces (north of the Fall Line) and is characterized by rolling to hilly topography of weathered bedrock and glacial deposits in the northern half of each.

Hydrogeologic characteristics of provinces north of the Fall Line differ greatly from those of the Coastal Plain. The New England province is underlain by igneous and metamorphic rocks, where as, the Piedmont province consists of layered shale, siltstone, and sandstone bedrock. Bedrock aquifers of these provinces are associated with water-bearing weathered joint and fracture systems typically within 300 feet of land surface. The top 100-150 feet is largely unconfined. Valleys in the glaciated northern half of each, however, are underlain by stratified drift and glacial till. These stratified drift aquifers are generally 30 to 40 feet thick, but may be 300 feet thick in some valleys.

In contrast, the Coastal Plain is underlain by a thick wedge of unconsolidated sediments that form permeable units of sand and gravel interbedded with poorly permeable units of silt and clay. These sediments differ in areal extent and thickness, but generally dip and thicken southeastward to about 6,500 feet thick at the southern tip of NJ. The Kirkwood-Cohansey aquifer system and the outcrop of the Potomac-Raritan-Magothy aquifer system are the principal surficial aquifers in the NJ Coastal Plain. Groundwater pumpage from confined systems exceeds pumpage from surficial systems by 3 to 1, but since NJ water supply permits for pumpage from confined systems are fully allocated, most of the future population growth in the Coastal Plain will depend directly on surficial sources for supply. Glacial deposits of sand and gravel overlie most of LI and form the surficial aquifer. To large extent these systems are underlain and hydraulically connected to the deeper Magothy aquifer, the principal source of water for LI.

Average annual precipitation ranges from 42 to 52 inches south-to-north, and annual snowfall ranges from 13 in the south to more than 50 inches in the north. Precipitation is evenly distributed throughout the year. Average annual temperature ranges from 56 to 50 oF south to north. Average annual runoff is 21-23 inches in NJ and on LI, but is significantly lower (from <3 to 10 inches) for many LI streams due to the heavy groundwater pumpage on the western part of LI. Annual groundwater recharge varies from 20 to 22 inches for the more permeable aquifers, but is estimated to be less than half that for the bedrock systems of the study area.

The principal river systems within the study unit are the Hackensack (202 mi2), Passaic (950 mi2), Raritan (1,105 mi2), Toms (192 mi2), Mullica (569 mi2), and Great Egg Harbor (347 mi2) Rivers in NJ. Many other smaller rivers and streams drain the Coastal Plain, including the Peconic River (75 mi2) on LI. Middle and downstream reaches of the Hackensack, Passaic, and Raritan Rivers receive urban runoff and treated effluent from point sources. Upstream reaches of these and most other rivers in the study are mixed agricultural, forested, and/or urban land use.

LI and the Coastal Plain of NJ have relatively flat slopes of sandy soils that allow rapid infiltration of rainfall. In fact, 75 to 90 percent of streamflow is derived from groundwater discharge over much of the Coastal Plain (Kirkwood-Cohansey aquifer system). The New England and Piedmont provinces have steeper slopes of thin, clayey soils, which produce runoff more rapidly than Coastal Plain soils. Much of western LI, northeastern NJ, and the corridor between New York City and Philadelphia consists of heavily urbanized areas that yield runoff more rapidly. Groundwater pumpage considerably reduces baseflow in streams on the western portion of LI.

Streams in northern New Jersey generally have upland and lowland portions. The upland portions exhibit steeper gradients, tend to be swift-flowing at higher discharge rates, and have very heterogeneous substrates ranging from bedrock, boulders, and cobbles in riffles to finer particles in depositional areas. Lowland or downstream valley portions have lower gradients, tend to be slower-flowing, and also have very heterogenous substrates perhaps with a tendency towards finer materials and more depositional areas than upland reaches. In the Coastal Plain, streams generally have low gradients, tend to be slow-flowing, and have more homogeneous substrates with bottom materials ranging from sand and gravel to silt and muck.

These habitat variations result in benthic invertebrate communities that exhibit distinct differences in species composition between the northern New Jersey upland/lowland and Coastal Plain systems (NJDEP, 1994). The greatest diversity of pollution-intolerant species has been found in streams of the New England province and areas of the outer Coastal Plain. This is due in part to the varied substrate composition, presence of forest cover, and in part to the generally better surface-water quality conditions in the New England and outer Coastal Plain than in the Piedmont and inner Coastal Plain. The Piedmont and inner Coastal Plain are subject to greater urban, industrial, and agricultural influence, which are generally associated with poorer water and habitat quality and lower invertebrate population density or species diversity. The natural water quality (low pH, alkalinity, and nutrient levels) and habitat (finer grained and unstable bed materials) of the Coastal Plain, in general, results in more depauparate communities and, thereby, reflects lower richness and diversity than other provinces.

Drainage patterns of the study area streams are largely dendritic. Natural lakes are virtually nonexistent in the Coastal Plain, although small man-made impoundments, formed by small earthen and/or rock dams, are common in both New Jersey and on Long Island. Literally hundreds of small impoundments and run-of-the-river dams exist in northern New Jersey. Many of these smaller dams throughout the study area are remnent from their use in providing sources of water power for a multitude of reasons in earlier times. The study area contains 13 major water-supply reservoirs, most north of the Fall Line, and receives about 100 Mgal/d of flow diverted from the Delaware River to the Raritan River basin for water supply. In fact, water-supply facilities in most river systems are highly connected and transfer of water is common. Major water uses in the study unit include domestic, commercial, industrial, mining, power production, and crop irrigation.

Total freshwater withdrawals for the LINJ is 1,617 Mgal/d, serving over 10 million people and related infrastructure. Surface-water withdrawals account for 37% (613 Mgal/d), most of which is in the northern NJ portion. These figures do not include the New York City surface water imports from outside sources to western LI (Queens and Kings Counties). Groundwater withdrawals account for 63% (1,004 Mgal/d) of the LINJ water use; 44% of which is on LI, 37% in southern NJ, and 19% in northern NJ. Only 20 to 40% of the total water use is groundwater from confined aquifers, indicating that much of the used resources in the LINJ are either SW or surficial GW and are, therefore, highly vulnerable to contamination.

Ecological Activities

Numerous natural and anthropogenic factors affect the health of biological communities in the highly urbanized LINJ study unit. Major ecologic issues related to water quality identified by the LINJ liaison committee include (1) the identification of factors (habitat, nutrients, and toxics) that contribute to biologically impaired communities, (2) the relations of land use (non-point sources) and degraded in-stream habitat to biotic condition, and (3) for certain land uses, the impacts of biocides, herbicides, and other organics/toxics on aquatic organisms, especially urban. The synthesis of these issues and integration of biological and water-quality information has been recognized as being of local, regional, and national significance, especially within highly urbanized systems. The LINJ surface water program will concentrate efforts in NE NJ on the largest and most urbanized of these systems, the Passaic and Raritan Rivers, and in the NJ CP near Glassboro in association with GW land use/flow path studies, and in the highly urbanized areas in Nassau and Suffolk Counties on LI. The LINJ study unit has worked hard to select candidate sites where surface-water activities can be directly linked with biological community assessment.
HIP PLANS AS MODIFIED BY CONFERENCE CALLS

As a result of the 1996 conference call and other communications between the NLT and LINJ, our final SW network included:

2 indicator IFS (Bound Brook at Middlesex; Great Egg Harbor River at Sicklerville),
3 indicator BFS (Neshanic R at Reaville; Saddle R at Ridgewood; Stony Brook at Princeton), and
2 integrator BFS (Passaic River at Two Bridges; Raritan River at Queens Bridge).

As a result of the 1997 conference call and other communications between the NLT and LINJ, the following modifications were made to the ecological portion of FY97 workplan:

(1): Two multiple reach sites were to be implemented at Saddle River at Ridgewood (Urban/Suburban indicator) and Neshanic River at Reaville (Agricultural indicator).
(2): Temperature probes were to be installed at all fixed sites.
(3): Bed sediment and tissue assessment was to be accomplished at all fixed sites during the fall of 1997 (FY97-98).

Late winter VOC and early summer pesticide SW synoptics were proposed for FY97 to complement the algae and invertebrate sampling network already established. We proposed expansion of our single reach single year ecological sites to enhance our ability to assess water-quality in NJ and LI streams. Four additional single-reach, single-year intensive ecological sites were to be sampled in FY97 (Beden Brook near Rockey Hill, NJ, Rockaway near Berkshire Valley, NJ, Rahway R. at Washington Park, NJ and Swan River at E. Patchogue, NY). All four of these sites were previously sampled for invertebrates and algae as part of our ecological synoptic completed in FY 1996-97. Therefore, only intensive ecological fish samples and complete habitat Level I assessment were to be done at these sites during FY 1997.

ACCOMPLISHMENTS IN FY 1996-97

All ecological activities proposed in the FY 1996 workplan were accomplished as planned. The following is a summary of those activities.

Ecological sampling at all BFS, IFS, and Synoptic sites began May 25, 1996 and ended October 20, 1996. We implemented the full suite of NAWQA protocols at Bound Brook at Middlesex, Saddle River at Ridgewood, Stony Brook at Princeton, Neshanic River at Reaville, Passaic at Millington and Mulhuckaway at VanSyckle. This included reach assessment, algae and invertebrate RTH, DTH, QMH, Level I habitat characterization and fish sampling. Raritan River at Queens Bridge was sampled for invertebrate and algae RTH, DTH, QMH. The Great Egg Harbor River at Sicklerville was only sampled for fish, and no ecological samples were collected at Passaic River at Two Bridges. No Level II habitat characterization was accomplished in FY96.

Fish community composition was assessed at 7 sites during FY96. We electrofished 9488 fish representing 42 species using the standard two pass method outlined by Meador and others (1993). This was accomplished at all fixed sites except Great Egg Harbor River at Sicklerville where a single pass approach was implemented due to low conductivity and extremely low fish density. Because NAWQA does not attempt to establish density estimates of fish species in streams, a single pass under these conditions is adequate to provide a "representative" sample. More than 3000 fish were captured at Neshanic River. Fish abundances in Stony Brook and Bound Brook were 2537 and 1432, respectively. Species richness ranged from 10 in Great Egg Harbor River to 25 in Stony Brook at Princeton. The most diverse site in 1996 was mixed agricultural/developing urban Stony Brook at Princeton. Our agricultural site at Neshanic River at Reaville was the least diverse. Tesselated darters (Ehteostoma olmsteadi), white suckers (Catostomus commersoni), and spottail shiners (Notropis hudsonius) were the 3 most abundant fish species in NJ streams. Tesselated darters were the only species captured at all fixed sites sampled.

All ecological activities proposed in the FY97 workplan, including changes suggested by the conference call, were accomplished as planned. The following is a summary of those activities.

Temperature monitoring devices (Hobo Temp) were installed at all 7 basic and intensive fixed sites during April 8-10, 1997. In addition, temperature monitors were installed at all synoptic sites where intensive fish assessments were accomplished in FY96-97. This included Swan River at East Patchogue, NY, Passaic River at Millington, Rockaway River at Longwood Valley, Rahway River at Washington Park, Mulhockaway Creek at Van Syckel, and Beden Brook near Rockey Hill. The temperature monitoring device located at Rockaway at Longwood Valley was vandalized, resulting in the loss of 2 months of continuous water temperature data. This device was replaced on 6-10-97.

Ecological assessment at BFS and IFS sites began May 19, 1997 (FY97) and ended October 22, 1997 (FY98). We implemented the full suite of NAWQA protocols at Bound Brook at Middlesex, Saddle River at Ridgewood (3 reaches), Stony Brook at Princeton, and Neshanic River at Reaville (3 reaches). This included reach assessment, algae and invertebrate RTH, DTH, QMH, Level I habitat characterization, and fish sampling. Raritan River at Queens Bridge, Great Egg Harbor River at Sicklerville, and Passaic at Two Bridges were sampled for invertebrate and algae RTH, DTH, QMH and Level I habitat assessment.

No Level II habitat characterization was accomplished in FY97 with the exception of Wolman pebble counts (Wolman 1954) which were done at all 34 sites sampled as part of the special ecological synoptic. Level II habitat characterization at BFS and IFS sites is currently scheduled for May-June of 1998 (FY98). Two multi-reach surveys were completed in FY97 at Saddle River at Ridgewood and Neshanic River at Reaville.

In addition to our fixed site network, four additional single-reach, single-year intensive ecological sites were sampled in FY97 (Beden Brook near Rockey Hill, NJ, Rockaway near Berkshire Valley, NJ, Rahway R. at Washington Park, NJ and Swan River at E. Patchogue, NY). Beden Brook is a 28 mi2 stream located in the Stony Brook-Millstone watershed within the Raritan River basin. Beden Brook is complementary in land-use to Stony Brook at Princeton, composed primarily of agricultural lands that are rapidly converting to residential developments (principally, single-family, town-house, and condominium complexes). Rockaway River near Berkshire Valley is a 25 mi2 stream in a moderately forested basin that is currently showing rapid suburbanization in the upper part of its drainage. This area is located minutes from US Highway 80, a major commuting corridor to New York City. Rahway River at Washington Park is a 20 mi2 urban indicator site located in Hillside, NJ. This area is a densely urban (>80%) with more than 4000 individuals/mi2 . Swan River at E. Patchogue, NY, is an urban indicator site located on LI that was selected to complement our VOC reach synoptic (see SW section - Water Quality Synoptics). All four of these sites were sampled for invertebrates and algae as part of our ecological synoptic completed in FY97. Therefore, only intensive ecological fish samples and habitat Level I assessment were done at these sites in FY97. These sites were also sampled for VOCs and pesticides as part of our surface water synoptics in FY97.

Fish community composition was assessed at a total 12 sites during FY97. We electrofished 21,856 fish representing 43 species  using the standard two pass method outlined by Meador and others (1993). This was accomplished at all single reach single year, single reach multi-year, and multiple reach multiple year sites (Table 4).

More than 3200 fish, representing 26 species were captured at Bound Brook at Middlesex in 1997. Fish abundances in Stony Brook and Neshanic (Reach A) were 2408 and 1994, respectively. Species richness ranged from 11 in Swan River at East Patchogue, NY to 28 in Beden Brook at Rockey Hill. The Shannon-Weiner index was used to assess diversity at all fish sampling sites. This index assumes that all individuals are randomly sampled from an infinitely large population (Pielou 1975, Magurran 1988). The most diverse site was Beden Brook at Rockey Hill. Our urban synoptic site on Long Island (Swan R. at East Patchogue, NY) was the least diverse. White suckers (Catostomus commersoni), tesselated darters (Ehteostoma olmsteadi), and spottail shiners (Notropis hudsonius) were the three most abundant fish species in NJ streams during 1997 and accounted for more that 52% of the total abundance. White suckers and tesselated darters were captured at all fixed sites sampled except Swan River at East Patchogue, NY in FY97. Unfortunately, white suckers (LINJ's target taxa) were not captured at Great Egg Harbor River, our Coastal Plain developing urban IFS. During BS&T sampling this Fall (FY98), a suitable surrogate target taxa will have to be chosen. The most appropriate candidates are either the creek chubsucker (Erimyzon oblongus) or pumpkinseed sunfish (Lepomis gibbosus), because both species overlap with the white sucker at other fixed sites.
VOC and pesticide SW synoptics were accomplished in FY97 to complement the algae and invertebrate network already established (see SW section - Water Quality Synoptics ). We designed these SW synoptics to build multiple lines of evidence' into our urban assessment and supplement the extensive information available on invertebrate and fish populations (from NJDEP).

### FIXED SITE SAMPLING

Proposed Ecological Activity at Fixed Sites FY 1998

Level II habitat characterization at fixed sites is currently proposed for May-June of 1998 (FY98). We intend to complete Level II reach characterization at all indicator fixed sites prior to ecological sampling. A minimum of three cross-sections at each fixed indicator site will be surveyed using a total station or other appropriate surveying approaches. Wolman pebble counts (Wolman, 1954) will be conducted at three transects at each reach corresponding to previous Level I habitat assessment.

Reach and Level I habitat assessments will be made during May & June prior to intensive ecological sampling. Intensive ecological assessments will begin in June and continue through low flow periods in August/September of FY98. It is this SU's intent to establish temporal trends at candidate fixed sites. Therefore, we will continue single-reach, multiple-year intensive ecological assessments at Bound Brook at Middlesex, Neshanic River at Reaville (Reach A only), Saddle River at Ridgewood (Reach A only), Stony Brook at Princeton, and Raritan at Queens Bridge (Table 4). No multiple reach assessments are scheduled for FY 1998.

Currently, Stony Brook at Princeton, Neshanic River at Reaville, and perhaps Saddle River at Ridgewood, represent good candidate indicator sites for continued sampling though the LIP. Stony Brook (agricultural/forested) and Neshanic River (agricultural) are rapidly developing watersheds typical of the urbanization in the mixed agricultural/forested portion of the Raritan watershed. Saddle River is mostly developed with some urbanization yet to occur in the NY portion of the watershed. Raritan River at Queens Bridge is a good candidate integrator site for the LIP. It is the largest LINJ stream; has a good mix of urban, agriculture, and forest; and is undergoing rapid development in the upper portions of the watershed. Our vote would be to choose Stony Brook at Princeton (indicator) and the Raritan River at Queens Bridge (integrator). This provides a good nested design and complements the NJDEP water-quality network.

We intend to install and monitor for 1 year temperature devices (HOBO TEMP) at the four additional sites where intensive fish assessment will be accomplished in FY98 (see above section).

BED SEDIMENT AND TISSUE SAMPLING

Choice of Target Taxa for Tissue Samples (FY98)

A major challenge for the LINJ study unit was the selection of a ubiquitously distributed target taxa that can be used for comparisons of trace elements and synthetic organic compounds within LINJ streams and among other study units. White sucker (Catostomus commersoni) was the target taxa of choice for LSUS, HDSN, and CONN, and recommended regionally for synthesis (Steve Sorenson, pers. comm.). The major drawback of using this species is that it is highly migratory, especially in May and June when adults move from lakes into streams to access spawning areas (Scott and Crossman 1973, Smith 1985). Problems associated with tissue analysis of migratory species (integration of contaminants over wide areas) is somewhat compensated for in the white sucker by its innate homing ability (Raney and Webster 1942, Werner 1979). Thus, tissue concentrations in white suckers are more likely to reflect conditions at a given sampling site than are other catadromous or anadromous species. Additionally, in NJ most are confined to a reduced home range by impassable barriers, impoundments, and presence of salt water.

Sampling for tissue in the spring of 1997 was avoided to prevent the possibility of additional contaminant integration through increasing and decreasing lipid levels in pre- and- post spawning adults, respectively. Another logistical problem associated with using white suckers for tissue assessment is that they are extremely difficult to age, even by the most experienced fishery biologist. We will be taking both scale and spine samples of all individuals for cross referencing of age determination. It is LINJ's intent to use this species to maintain conformity with other tissue analysis in the northeast region and to fulfill the needs of the National Synthesis Team for consistent and broadly applicable tissue data. At sites where the white sucker is not available for tissue assessment (e.g., Great Egg Harbor River at Sicklerville, NJ), a suitable surrogate species will be implemented. Decisions regarding the most appropriate surrogate species for this SU and for the Regional and National database will be made in conjunction with the Regional biological staff and the NST. At present, creek chubsucker (Erimyzon oblongus) represents the most probable candidate because it overlaps with the white sucker at other fixed sites. This species is very similar to the white sucker in feeding and foraging behavior, consuming a variety of benthic organisms and vegetable material (Smith, 1985).

Proposed BS&T Sampling at Fixed Sites (FY98)

The seven LINJ fixed sites (see SW section - Proposed work in FY 1998) were sampled for bed sediment and fish tissue in the fall of 1997 following the protocols of Shelton and Capel (1994) and Crawford and Luoma (1993), respectively. Seven additional sites that correspond to sites sampled as part of the LINJ synoptic network, where bed sediment data currently does not exist, also were sampled for bed sediment only. The seven additional sites were selected from the 34 site algal and invertebrate synoptic network (Table ECO3) and met the following criteria: 1) no bed sediment data collected previously as part of the NJDEP/USGS cooperative water-quality network, 2) NAWQA fish population assessment completed during FY96, FY97, or FY98, and 3) appropriate biological community data comparable to currently selected BS&T sites. Another 6-9 of the 34 sites were considered lower in priority (not having recent (1990s) bed sediment data) but could be sampled for bed sediment in September of FY98 if end-of-year funds are available.

We also sampled two additional single reach, single year synoptic sites (Passaic at Millington and Rahway at Washington Park) for fish tissue. We believe that these two additional tissue (indicator) sites will provide highly valuable information for the northeast region and national synthesis because both have of a high proportion of urban land and both have been intensively sampled for fish (Table 4). It should be noted that because we do not have access to a boat electrofishing unit this fall, sampling at one integrator site (Passaic R at Two Bridges) proved to be unfeasible with backpack electrofishing units. We were unable to locate and capture white suckers at this integrator site during October 1997, so, we will try to solicite the NJDEP's help to sample this site for fish and fish tissues during the low flow period in 1998.

BS&T sampling was conducted in early October 1997 during low base flow conditions when access to depositional zones and fish capture was possible. Tissue collection involved a two to three person team sampling one to two sites per day. Tissue collection was coordinated with the collection of bed sediments samples at the seven fixed sites (+ 3 days between samples). Bed sediment sites were sampled (2-3 sites per day) during the same period by the SW team.

### SYNOPTIC SITE SAMPLING

Accomplishments in FY 1996-1997

In FY97 LINJ completed a 34 site algae and invertebrate synoptic that augments our current fixed site network and integrates it with sites sampled by the NJDEP and USEPA for invertebrates and fish, respectively. The objectives of this survey were to (1) provide LINJ with a more comprehensive understanding of factors affecting aquatic communities in indicator-size basins distributed among our major strata, (2) sample a suite of sites that represents the entire range of the urban land use gradient in NJ streams, and (3) establish multiple lines of evidence by integrating biological community assessment and water-quality data. Currently, little information exists regarding lotic periphyton (attached algae) communities in LINJ watersheds. It was our hope that this synoptic would expand on the current knowledge of algal communities along environmental (land-use, water quality, and habitat) gradients in NJ streams. This is especially important when considering the power of algal communities as indicators of water-chemistry conditions over short time periods. Algal communities are often the first to respond to a water-quality stress and the first to recover following mitigation.

Analysis of existing data determined that 62 sites directly overlap between NJDEP's invertebrate data and USEPA's fish data. Of these 62 sites, approximately 32 were sampled for water quality in the COOP network. Additionally, the NJDEP has sampled 750+ sites as part of their Ambient Biological Sampling Network (AMNET) for aquatic invertebrates and, of those, 43 sites represent what they consider to be regional reference sites based on 152 benchmark (pollution sensitive) organisms showing minimal impairment. Invertebrate RTH samples collected during LINJ's ecological synoptic will be directly compared with samples collected by NJDEP at overlapping sites to assess sampling consistency and invertebrate richness and equitability.

State and federal agencies in NJ are continually looking for ways to expand on the information gained from their efforts (multiple lines of evidence). Synoptic studies like this not only helps their efforts in aquatic community assessment, but at the same time maintains good cooperative working relations and enhances our efforts in assessment of urban impacts within the LINJ. From a national perspective, cooperative efforts like these promote NAWQA's visibility and help bridge the gap between fixed and synoptic site analysis. The NJDEP and USEPA have played a major role in the development of this synoptic investigation by providing coveted data sets of benthic invertebrate and fish communities, respectively, that are currently part of ongoing retrospective analyses (Kennen 1997, in review). Additionally, they have been integral in providing ancillary data on habitat and substrate to complement these data. The NJDEP benefits from this investigation by the enhancement of their AMNET program with an algal component that will undoubtedly provide additional support of trends in water quality in NJ and add one more piece to the multiple lines of evidence' puzzle. The USEPA is currently involved in fine tuning their IBI (Index of Biotic Integrity) approach to fish species composition at 164+ sites in NJ. It is their hope that this synoptic will further expand our knowledge of the health and stability of aquatic communities in NJ and provide complementary algal information in support of known trends. Synoptic sites were also sampled on Long Island as part of our 34 site network. Information about LI aquatic communities is needed to assess and relate levels of urban land-use on Long Island to those in the Coastal Plain of NJ.

We reconned 95 potential surface water sites in NJ and LI during August 16-28, 1996. Of these, 28 NJ and 6 LI locations were chosen for further study based on physiography (NJ), landuse, substrate comparability, existing invert & fish data, and availability of water quality information. Discussions with NAWQA's CR phycologist suggested that the greatest information about algal communities in NJ gained for the least monetary expenditure could be the accomplished with epilithic RTH samples. RTH samples provide estimates of composition, abundance and standing crop from a targeted microhabitat that ideally would be relatively consistent among sites. We realized that the monetary trade-off of not doing QMH samples would subsequently limit extraction of important autecological information, especially for macro-algae and moss species often present at these synoptic sites. However, because macro-and micro-algae are closely associated on stream substrates, many of the RTH samples taken will contain both groups. It was suggested that these samples be taken during the early fall (stable low-flow periods) to correspond with the same sampling period of many other NAWQA Study Units. In addition, it was suggested that we resample a subset of the synoptic sites in the late spring FY97 to assess additional autecological information during the time of year when taxa richness is at a maximum (Stephen Porter, pers. comm.) and to take advantage of timely surface water QW synoptics scheduled this spring. This information would supply a greater `n' for correspondence analysis. Ideally, we would like to have 30 samples in each of our major land-use strata providing us with more sites than species, a caveat associated with most types of correspondence analysis (see Gauch 1982). We realize due to monetary and time constrains that this may not be possible.

In Sep.-Oct. 1997, 34 special ecological synoptic sites (correspond basically to most of the SW pesticide/VOC synoptic sites (Figure SYNOP) were sampled for RTH algae and RTH invertebrates. A modified habitat assessment was implemented at these synoptic sites that included all aspects of the NAWQA Level I habitat protocols except point-quarter methodology. However, Wolman pebble counts (Wolman 1954) were completed at all 34 sites to complement the RTH samples taken in riffle habitats. Substrate size in NJ streams varies from site to site, thus, a quantitative assessment of stream bed material is essential in understanding how algal and invertebrate communities vary relative to substrate size and composition. Minimally, percent substrate composition can be used as a covariable to eliminate potential "noise" in correspondence analysis. Staffing for this synoptic included a minimum of 2 and often 3 SU personnel to accomplish our goal of sampling two sites per day. On most days we achieved this goal, however, inclement weather and extensive travel time did, on occasion, precluded the sampling of two sites in a day.

Proposed Work at Synoptic Sites FY 1998

We will continue to expand ecological sampling in FY98 to enhance our ability to assess water-quality in NJ and LI streams. Most applicable community ordination techniques have an underlying assumption of more sites than explanatory variables (Gauch 1982), therefore, it is in our best interests to increase the number of representative indicator sites within major land-use strata. In addition, the need to assess as much of the urban gradient in NJ streams is paramount to understanding the way in which this type of anthropogenic disturbance affects aquatic communities. It is statistically and ecologically more significant to do comparative analysis on greater that 6 or 7 basic and intensive fixed sites. For this reason, four additional single-reach, single-year intensive ecological sites will be implemented in FY98 (Pequannock River at Riverdale, SB Raritan River at Arch St., Hohokus Brook at Allendale, and Lamington R at Pottersville) to complement our current fixed network.

Pequannock River is an 84 m2 drainage located in the northern part of the Passaic River basin. The majority of the land in this system is still forested (70%) and protected for water-supply purposes and parks. The remaining lands are urban residential and commercial (15%). The South Branch Raritan River near Arch Street (68.8 mi2) is in a moderately forested basin (51%) that is showing rapid urban and suburbanization (currently 30%) in much of its drainage. This site is located upstream of Spruce Run, a NJ drinking water reservoir, however, many nonpoint sources of pollution currently contribute to the water-quality problems in this river. Hohokus Brook at Allendale is a 10 m2 urban indicator site located in Bergen County, NJ. This area is a densely urban (76%) with more than 1700 individuals/mi2. Land in this drainage is extensively developed and contains many older communities and industrial centers. Aquatic invertebrate communities at all sites sampled by the NJDEP in the Hohokus system were found to be severely impaired (NJDEP 1992). Lamington River at Pottersville (33mi2) is a mixed land use basin (33% urban, 15% agric, and 42% forest) that drains into the North Branch of the Raritan River. This stream runs parallel to Rt. 206 and under Highway 78, two major NJ commuting corridors. All four of these sites were sampled for invertebrates and algae as part of our ecological synoptic completed in FY96-97. Therefore, only intensive ecological fish samples and habitat Level I assessment will be done at these sites during FY98. These sites have also been previously sampled for VOCs and pesticides as part of our surface water synoptics accomplished in FY97. With the inclusion of these four additional fish sampling sites, the LINJ SU may have a large enough sample size to assess potential underlying urban gradients relative to physical, chemical and environmental parameters in these streams.

Proposed Coastal Plain Synoptic in FY 1998

There is currently a need to establish 8-12 ecological synoptic sampling sites in the Coastal Plain of NJ to complement the currently established 34 site algae and invertebrate synoptic network (Table ECO3) which concentrated efforts in the northern part of the SU. These samples will be taken in concert with a SW pesticide synoptic study in the Mullica, Tom's, and Metedeconk basins proposed for June of 1998. The Tom's River-Metedeconk basins are some of the most rapidly developing areas in NJ and although numerous water quality studies have been conducted in these basins in recent years, pesticide and biological data are currently lacking (see SW section - Water-Quality Synoptics section, Proposed work in FY 1998). It is our intention to work in close cooperation with the New Jersey Pinelands Commission to choose representative candidate sites appropriate for these types of samples. The sampling protocols applied previously to our initial 34 site synoptic network in northern NJ and LI, will be used at these Coastal Plain synoptic sites (see previous section on accomplishments in FY97).