Ten Towns Committee
Macroinvertebrate Sampling - 2001
 
The Macroinvertebrate Communities of
the Great Swamp Watershed:
Summer, 2001: Results

A Report to the
Ten Towns Great Swamp Management Committee

Leland W. Pollock, Ph.D.
Department of Biology
Drew University

October, 2001

Executive Summary
The 2001 sampling season shows mixed results in terms of the quality of the macroinvertebrate communities.  There was good news at Primrose Brook, but lower sections of both Passaic River, Loantaka Brook, and Great Brook all saw declines (to lowest levels ever at Great Brook).  The range in environmental conditions and in macroinvertebrate communities is very wide in the Great Swamp watershed, underscoring the challenges before us in improving streams toward the east side and the dangers that threaten those to the west side of the watershed.

Very poor macroinvertebrate communities at both Black Brook sites reflect a unique set of unamenable conditions there (very low gradient and paucity of comparable sampling habitats).  Loantaka and Great Brooks are the special sources of concern.  Frequently sluggish water flow, heavy sedimentation, and lack of suitable macroinvertebrate habitat appear to account for most of the poor macroinvertebrate indicators in both the Loantaka and Great Brooks.  It seems apparent that stream bank erosion coupled with storm drain runoff lead to the sediment burial of desirable, coarser substrate habitat, and that this is, in turn, associated with periodic substantial rain events striking large areas of impermeable cover in these developed watersheds.  Very high levels of TDS are noted throughout Loantaka Brook.  Lower Great Brook sites have declined following sedimentation/turbidity aftermath of Silver Lake dredging.  Passaic River sites (especially PR2) declined this year.  Very high turbidity at PR1 and PR2 may have been involved.

The initiation of a similar survey of macroinvertebrate communities in the nearby Rockaway River watershed provides a larger regional context and a useful source of comparisons for the Great Swamp streams.  I regard the principle difference between Rockaway watershed streams along with the better, western Great Swamp watershed locations vs. the more eastern Black, Loantaka and Great Brooks is that the latter are slow moving, low gradient streams, receiving relatively heavy impact from human sources such as sewage treatment plants, storm drains, and runoff from impervious surfaces.

The June 2002 Great Swamp Watershed Study: Results
An general introduction and descriptions of methods used in this survey are available elsewhere at this site.
Habitats & Environmental Monitoring
To assess habitat conditions, we monitored environmental conditions at all 17 sites between 9 am and 3:30pm on June 11, 2001.  More will be made of the influence of these environmental factors on macroinvertebrates in the analysis to follow.  But it is also useful to view these factors in their own right. 

Figure 01-1 (note that temperatures shown must be doubled and TDS readings multiplied by 50 to yield their true values) and Figure 01-2 show the range and mean values for these factors among the 17 sampling locations.  In Figure 01-1, the broad spread in total dissolved substances (TDS), turbidity, and dissolved oxygen (DO) values is especially noteworthy because these factors are particularly subject to human influence, through added chemicals, sediments/silt, and nutrient/organic loading respectively.  The means and ranges in subjectively evaluated factors (Figure 01-2) are also broad enough to cover the full gambit from near-pristine to significantly disturbed.  Comparable observations from 13 sites within the nearby Rockaway River watershed are available in Figures 01-1 and 01-2.  The distinctly narrower range of conditions in all cases there highlights that comparatively better overall water quality is available within the Rockaway River watershed.  The contrast in human-influenced TDS and turbidity conditions (Figures 01-1) between the two watersheds is especially striking.  Also, note that while ranges in dissolved oxygen are similar between these two watersheds, amounts of dissolved oxygen observed in the Great Swamp streams were generally lower than in the Rockaway River watershed despite somewhat cooler water temperatures.  These observations underscore the continuing cause for concern regarding the impact of human disturbance within the Great Swamp watershed.  Refer to Table 01-1 for site-specific values for these variables.

Patterns can be seen by reviewing some variables individually.  Overall, temperatures were at least a full degree warmer at nearly all sites in summer 2001 compared to summer 2000. Predictably, sites displaying temperatures > 25 C (i.e., BB1, LB2, GB3, GB5, PR1) are all immediately downstream from dammed impoundments.  Strikingly high values for total dissolved substances (TDS > 500 ppm) were observed at all the Loantaka Brook sites (4-5 times higher than for other streams), including sites upstream from both the sewage treatment plant and the Seton Hackney riding stables, suggesting that these are not the primary TDS sources.  The actual source and identity of these high levels of TDS is unknown.  Next highest TDS levels were observed at BB2, the site just downstream from the Chatham Township sewage treatment plant.  Less disturbed sites had TDS values less than 200 ppm.  These patterns precisely match those from summer, 2000.  In general, dissolved oxygen values were considerably higher this summer.  Exceptions toward much lower values occurred at BB1 and LB1.  Values less than 9 mg/L DO were all found in more heavily stressed, eastern watershed streams while higher values were typical of western watershed streams.  Exceptions to this are found in low values at the lower stations on the Passaic River.  Turbidity was very high in the lower sites of the Passaic River and Loantaka Brook, and the upper sites of Great Brook.  As in the past, no discernable pattern was observed in the stream site distribution of pH. 

Most habitat assessment scores match expected trends of increasing quality along an east to west gradient across the watershed.  This is least true for the character estimating the degree to which stream "flow" fills the stream channel a character seemingly more related to recent rainfall conditions (the same for all sites here) than to site-specific features.  The variability in the degree of channelization (the "chann" character in Table 01-1) is also broad throughout the watershed, illustrating the degree to which all of our streams have been tapered with.

Macroinvertebrate Monitoring
To characterize the 17 sites studied here, we examined a total of 3511 individuals representing 104 species of macroinvertebrates.  There was an average of 20.4 species per site, with a range from 9 species at Black Brook site 1 (BB1) to 38 species at Primrose Brook site 2 (PB2).  Actual counts of species for all sites may be viewed in Appendix 01-1.

Beck Index scores are shown in comparison to B-IBI scores in Figure 01-3.  In general, the two indices offered similar views of the 17 sampling sites.  At LB3, PB3, and PR3 sites, the Beck Index was lower than the B-IBI.  Because the Beck Index is focused especially on the impact of organic pollutants, its more pessimistic view makes sense because LB3 is just below a sewage treatment plant.  But the poor Beck rating for the otherwise high quality PB3 and PR3 sites is puzzling.  It continues to be the view of this author that because of its organic stress focus, the Beck Index is a less sensitive evaluator for this region.

Benthic Index of Biological Integrity values from June, 2001 are compared in Figure 01-4 to the range of scores observed in June, 2000.  June, 2001 samples included new low values at BB1, LB4, all Great Brook sites, and PR2.  Although measures at upper sites on Great Brook are not changed from 2000, the decline at lower sites is disturbing.  We note that all sites now lie at the minimal levels observed for the past several years.  At the same time, new high B-IBI values were observed at BB2, all the Primrose Brook sites.  Interestingly, both sites located below sewage treatment facilities, i.e., BB2 and LB3, are included among those whose condition improved substantially.  It is unknown whether any changes in the operation of those facilities may have been involved in this.  A particularly dramatic improvement between 2000 and 2001 can be seen in Primrose Brook site 1 (PB1).  The community last year suffered from inevitable consequences of a major bridge repair just upstream.  It seems clear that at least in this instance, such a disturbance can have short-lived effects on stream conditions.  Improvement at PB3 may reflect the longer recovery necessary downstream from a major siltation incident above this site in 1997. 

A breakdown of B-IBI components is given in Table 01-2 for June, 2000 and June, 2001.  For each site in this table, IBI scores (shown below) are determined by actual values for those components (shown above).  The direction of changes in B-IBI characteristics can be visualized more easily by referring to Table 01-3.  In most cases of declining B-IBI values, the total number of taxa observed has fallen (BB1, LB2, LB4, GB3, GB5). Typically, declining community richness results from habitat simplification, e.g., from siltation covering/clogging spaces among larger sedimentary particles.  Also, negative changes are associated with the increased presence of stress tolerant indicator species (BB1, GB2, PB1, PB2, IG1), which is often associated with major changes in critical environmental variables.  For example, dissolved oxygen was particularly poor at BB1 and turbidity was up at LB2, PB1, PR1 and PR2.  Total dissolved substances were very high at LB4, which also suffered from low water levels during this dry summer. Conversely, substantial increases in B-IBI values between June, 2000 and June, 2001 follow increased number of taxa observed (LB1) and declines in dominance observed (LB3, PB1, PB3).  Changes in species observed of Ephemeroptera (mayflies) (PR3) or of Trichoptera (caddisflies) (BB2, GB2, PB2, PB3) account for some differences between summers 2000 and 2001.

Results from this survey of Great Swamp watershed streams can be placed into a regional perspective, by comparing results B-IBI values from the 13 sites studied within the Rockaway River watershed.  In keeping with historical gradients of development in our region, in both cases, there was a general east-to-west improvement in B-IBI values.  But as with environmental conditions discussed above, a much broader spread in B-IBI values was found in the Great Swamp watershed.  Nearly all of the Rockaway River watershed sites showed results just below the better quality but well above all the poorer quality streams in the Great Swamp watershed. 

Beck Index results from the summer, 2001 survey are added to past results in Table 01-4.

Combining Habitat Assessment and Biotic Metrics
The results of correlation analysis between biotic metrics and stream site conditions are found in Table 01-5.  Correlation coefficients (r) display the degree to which changes in the measured level of each habitat variable at the 17 sites study matches biotic metric value calculated on the basis of the macroinvertebrate community found at these locations.  Correlation coefficients vary from -1.00 (= a perfect negative correlation, i.e., as the value for a variable increases it is matched by a decrease in the corresponding metric) to +1.00 (= a perfect positive correlation, i.e., as habitat values increase, so do community metrics).  Comparing habitat and metrics for 17 sites, and selecting the standard 95% certainty level to determine significance, correlation coefficients > + 0.468 are considered to show significant correlation.  It is important to stress that correlation shows only that the two entities being compared co-vary.  It does not necessarily demonstrate a causal relationship.

In Table 01-5, I have used a box to highlight the signficant correlations between the two biotic indexes calculated here, Beck and B-IBI, and environmental variables monitored on June 11, 2001.  First, it is reassuring that the metrics are highly correlated with one another ( r = 0.972).  But they also both show matching correlations with several variables.  They show significant negative correlation with temperature and total dissolved substances (TDS), i.e., as temperature or TDS increases, the corresponding values for the metrics decrease.  Highest positive correlations are found with stream bank stability (bank), riparian vegetational cover (veget), the level of dissolved oxygen present (DO), the frequency of riffles (riffl), the degree of freedom from sedimentation (sedim), the in-stream habitat cover (cover), and the degree to which stones escape becoming embedded in sediments (embed).  As these important habitat features improve, so does the community index.

Canonical Correspondence Analysis (CCA)
This complex-appearing multivariate statistical procedure offers another way to integrate data on species distribution with environmental measures by arraying species and sampling sites within a matrix formed by the most important surrounding variables.  In Figure 01-5, coded names for species are placed along environmental gradients (indicated by arrows and identified by labels) in "best-fit" locations relative to the combination of conditions in which they were found.  In this plot, for example, it can be seen that species included in Group 1 were found in conditions of high dissolved oxygen (DO) and low total dissolved substances (TDS), while those in Group 4 were found in opposite conditions.  By highlighting all of the species found at a particular sampling site and placing a label for that site at the center of its composite species, the location of the 17 sample sites can be seen within the same environmental gradients in Figure 01-6.

Each stream occupies a partially overlapping but still unique region within the graph, suggesting that stream-specific combinations of conditions are distinctive.  Great Brook, Passaic River, Indian Grave site 1, and Primrose Brook placements are all very similar to the relationships described by the CCA model for summer, 2000.  The Loantaka Brook sites are more coherently clustered in the analysis from summer, 2001.  Oddly, the BB1 and BB2 sites have basically switched places.

All sites on Loantaka and Black Brooks along with GB4 are located toward the poorer dissolved oxygen and higher dissolved chemical substances side of the graph.  Remaining Great Brook sites are found more toward the higher temperature and pH portion of the graph.  The upper right quadrant of the figure reflects high water quality conditions, such as lower temperatures and total dissolved substances along with higher dissolved oxygen.  Primrose Brook and Passaic River sites, including its Indian Grave tributary, are all found in this general portion of the graph. 

Note that sites located just below impoundments (where waters stall and heat and photosynthetically drived pH rises during the daytime) are found lowest in each stream's portion of the figure.  These include BB1 (below a golf course pond), LB2 (below Loantaka Pond), GB5 (below Foote's Pond), GB3 (below Silver Lake), PR1 (below Osbourn Pond), and PB3 (below a small pond on Glen Alpen Road).

Stream Summaries
Indian Grave Brook (B-IBI rating was "good"):
This reference site remained in fine shape producing high B-IBI and Beck values. Environmental conditions here are close to ideal.
Passaic River (lower sites rated "poor" with the upper site from "fair" to "good"):
B-IBI values were lower at all three Passaic River sites this summer.  Declines at the middle site and lower site were greatest.  Relatively small shifts in presence or absence of key species account for most changes.  Turbidity in the lower two sites was higher than at the upper site, while dissolved oxygen was considerably lower.
Primrose Brook (all three sites were "fair" or "good"):
PB1 and PB2 were at their highest values observed in recent years.  The improvement in PB1 following bridge repair work in summer, 2000, is especially reassuring (although it still shows comparatively high turbidity).  It suggests that with care to minimize impact, it should be possible to perform such necessary intrusions in rapidly reversible fashion.  Strong showing at PB2 may signal its recovery from a serious upstream silting incident in 1997.  Conditions within the stream were good, although higher turbidity and lower dissolved oxygen were evident at PB1.
Great Brook ("very poor" B-IBI rating):
Great Brook continues to be a concern.  For the past 2 summers it's stations have yielded poor results.  This is especially true of the downstream stations that may still be showing the effects of the major dredging of Silver Lake in 1998-99.  GB3 showed an alarming 40% decline in taxa richness, suggesting that this site experiences a considerable degree of stress.  GB2 and GB3 have a long way to go to recover to their historical highs.  These sites continue to suffer from high temperatures, fairly high TDS, and problems with habitat quality in upstream sites.
Loantaka Brook ("poor" rating for LB2 and "very poor" for remaining sites):
Substantial gain at just-sub-treatment-plant LB3 come from increased taxa richness and resulting decline in community dominance.  However, this was matched by declines at LB2 (decline in taxa richness especially among predators) and LB4 (modest increase in pollution tolerant indicator types).
Black Brook ("very poor" overall B-IBI rating):
Post treatment plant site, BB2 (high dissolved oxygen, modest TDS, low turbidity), hosted a much improved macroinvertebrate community in 2001.  However, this gain was matched by a steep decline in the community at BB1, located along a different, low dissolved oxygen, high turbidity branch of Black Brook.
The June 2001 Great Swamp Watershed Study:  Recommendations
  1. Continuity of this study, about to enter its second decade, is important.
  2. Especially high levels of TDS are observed in Loantaka Brook.  Highest levels occur farthest upstream.  The source of this high chemical concentration is unknown as is its specific identity, and its impact on macroinvertebrate communities.  This should be investigated.
  3. Lower sites on Great Brook have declined, probably as a result of siltation from the dredging of Silver Lake.  Charting their recovery may help us understand the duration of the impact of impoundment dredging on stream communities.  This bears watching.
Literature Cited

Barbour, M.T., J.Gerritsen, B.D.Snyder, and J.B.Stribling.  1999.  Rapid Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish, Second Edition.  (EPA 841-B-99-002).  U.S.  Environmental Protection Agency; Office of Water; Washington, D.C.

Karr, J.R.  and E.W.  Chu.  1999.  Restoring Life in Running Waters: Better Biological Monitoring.  Island Press, Washington, 206 pp.

Pollock, L.W.  2001.  The Macroinvertebrate Communities of the Rockaway River Watershed, Summer, 2001.  A Report to the Rockaway River Watershed Cabinet.

SCS Water Quality Indicators Guide: Surface Waters.

Strahler, A.N., 1964.  Quantitative geomorphology of drainage basins and channel networks.  IN

Chow, V.T. (Ed). Handbook of applied hydrology.  McGraw-Hill, NY.  Sections 4-11. (Not seen in original)

Acknowledgments
I wish to express my appreciation for the Ten Towns Great Swamp Watershed Management Committee's understanding of the value of using macroinvertebrate communities to document water quality conditions throughout the watershed.  In particular, I am grateful for their funding of this study.  In addition I acknowledge the generous, in-kind support of my home institution, Drew University, and faculty colleagues who have provided encouragement as well as the equipment and facilities necessary to the project's completion.  And finally, I am especially indebted to Audrey Tresham, my indefatigible Drew student assistant during this project, for being so in adopting the techniques used here.
Figures
Figure 01-1.  Mean and range values for various environmental factors observed on June 11, 2001 at the 17 sampling sites within the Great Swamp watershed and the 13 sites within the nearby Rockaway River watershed.  Note: temperatures shown must be doubled and values for total dissolved substances (TDS) must be multiplied by 50 to yield actual values.  Temperature in oC, total dissolved substances (TDS) in ppm, dissolved oxygen (DO) in mg/L, pH in units, and turbidity in FTU (Forel turbidity units).

Figure 01-2.  Mean and range values for various habitat factors observed on June 11, 2001 at the 17 sampling sites within the Great Swamp watershed and the 13 sites within the nearby Rockaway River watershed.  See text for description of characters.

Figure 01-3.  Comparison of Beck Index and B-IBI values for 17 sampling sites within the Great Swamp watershed, June, 2001.  Descriptive terms and limits shown refer to Beck Index values.  For site coding, see Table 1.

Figure 01-4.  B-IBI values for 17 sampling sites within the Great Swamp watershed, June, 2001 vs. June, 2000. 

Figure 01-5.  Results of Canonical Correspondence Analysis (CCA) based on data in Table 2 and Appendix 1.  Species are placed on the graph in best fit (i.e., least squared deviation) locations based on key environmental variables observed at all the sampling sites.  Species data were log-transformed to reduce the unbalanced impact of rare species.

Figure 01-6.  Results of Canonical Correspondence Analysis (CCA), placing each sampling site at the centroid of the species found there.

All Figures.     All Tables.     Appendix.


Printable Version
Please send comments to the webmaster from this or any other page.
Macroinvertebrate Sampling Study - 2001