2013 NORTHEASTERN NATURALIST 20(1):91–102 An Impoverished Benthic Community Shows Regional Distinctions Erica L. Smith1,*, David Coté2, and Murray H. Colbo3 Abstract - Monitoring programs using benthic macroinvertebrates are well-used and expanding to areas where communities are species-poor. The sensitivity of these depauperate communities to environmental conditions, however, is not well known. In this study, impoverished benthic invertebrate communities were compared from three climatically and geologically distinct regions of Newfoundland. Differences in community structure were evident among regions at both the genus and family level. These results indicate that widely dispersing and depauperate macroinvertebrate communities can be sufficiently diverse to respond to regional variation in environmental conditions and therefore remain promising for detecting anthropogenic-induced changes. Introduction Benthic invertebrate communities have long been used for environmental monitoring because of such characteristics as ease of qualitative sampling (Hellawell 1978), widespread distribution and niche differentiation (Barbour et al. 1999), and the sensitivity of numerous taxa to environmental perturbation (Griffith et al. 2005). However, individuals of species-poor communities often show increased niche breadth (Preston 1980), which may mask environmental signals that would be evident in more diverse assemblages. Understanding the environmental sensitivity of communities, particularly in regional analyses, is critical for biomonitoring studies (Poff 1997, Vinson and Hawkins 1998). Consistent with the island biogeography theory (MacArthur 1967), the fauna of the Island of Newfoundland is impoverished relative to adjacent mainland regions of comparable land area. For example, species of Ephemeroptera number 35 in Newfoundland (106,000 km2) versus 160 species in Maine (91,650 km2) (Burian and Gibbs 1991, Larson and Colbo 1983), while Odonata number 38 species in Newfoundland and 128 species in the Canadian Maritimes (133,852 km2) (Brunelle 1997, Larson and Colbo 1983). The fauna present on the Island are for the most part widely dispersed throughout North America (Larson and Colbo 1983, Merritt et al. 2008) and show strong dispersal capacity, establishing on the Island since the last glaciation (ca. 10,000 years) (South 1983). In contrast to the limited faunal diversity, Newfoundland has notable habitat diversity (Damman 1983), as the geology and climate differ considerably 1Box One, Portage la Prairie, MB, Canada R1N 3P1. 2Glovertown, NL, Canada A0G 2L0. 3Lovett Road, Coldbrook, NS, Canada B4P 2R6. *Corresponding author - SmithEricaL@ gmail.com. 92 Northeastern Naturalist Vol. 20, No. 1 across the landmass. Under such conditions, regional differentiation in macroinvertebrate community structure would be expected. However, it is unknown whether these impoverished communities can be responsive to this environmental variation. This question was addressed by comparing benthic communities across riffle habitats from three climatically and geologically distinct regions of Newfoundland: the Avalon Peninsula, Terra Nova, and Gros Morne. The absence of such differentiation would suggest that the effectiveness of using impoverished stream invertebrate communities for tracking anthropogenic impacts is also limited. Methods Study location and design The Island of Newfoundland (51°38'–46°37'N, 59°24'–52°37'W) falls in the boreal forest biome (Roberts 1983) and was completely glaciated in the most recent ice age (Rogerson 1983, Shaw et al. 2006). As a result, most of the flora and fauna was obtained post-glaciation via colonization and human-mediated introductions (South 1983). In general, the climate of Newfoundland is cool and wet, with a shorter growing season than is observed on the mainland (Table 1). The west and central parts of the Island tend to be colder and have more snow in the winter and experience an earlier spring and warmer summer than the eastern Avalon Peninsula (Banfield 1983). The western portion of the Island is the most diverse geologically and topographically; the area encompassing Gros Morne National Park is an extension of the Appalachian Mountains. The central portion of the Island is a plateau of a different geological origin, largely sedimentary in nature but with metamorphic presence, while the eastern portion is sedimentary bedrock (Rogerson 1983). Approximately half the Island is forested, the other half largely made up of barrens, peatlands, and lakes (Roberts 1983). Generally, the percentage of bog and peatland increases from west to east. On the western side of the Island, the southern part is densely forested and grades toward increased bog and barrens Table 1. Mean seasonal and yearly temperature and precipitation and average number of degree days for locations in Newfoundland (NL), Nova Scotia (NS), New Brunswick (NB), and Quebec (QC), calculated over a minimum of 15 years between the years 1971 and 2000. (Environment Canada 2000). Mean temperature (°C) [mean precipitation (mm)] Degree days January May July Year (<10 °C) St. John’s, NL -4.5 [170.7] 6.5 [103.9] 15.7 [84.1] 5.0 [1571.9] 543.1 Terra Nova, NL -6.8 [105.7] 7.0 [87.6] 16.1 [88.3] 4.4 [1183.2] 581.2 Rocky Harbour, NL -7.5 [145.5] 6.4 [73.7] 15.4 [99.6] 3.6 [1316.5] 483.8 Halifax, NS -6.0 [100.6] 9.8 [109.7] 18.6 [102.2] 6.3 [1238.9] 883.8 Sydney, NS -5.7 [151.5] 7.8 [102.9] 17.7 [86.8] 5.5 [1504.9] 748.9 Moncton, NB -8.3 [108.8] 10.7 [99.1] 19.4 [99.8] 5.8 [1143.5] 966.2 Fredericton, NB -9.5 [104.4] 11.2 [100.1] 19.3 [89.7] 5.6 [1124.2] 966.2 Quebec City, QC -11.8 [89.8] 11.2 [106.1] 19.0 [127.8] 4.0 [1230.3] 908.6 2013 E.L. Smith, D. Coté, and M.H. Colbo 93 moving north. On the Avalon Peninsula, the northern portion is largely forest, while the southern portion of the peninsula is completely barren (Roberts 1983). The Avalon Peninsula is also the most densely populated region of the Island (South 1983). Streams were selected from three geographic areas (Fig. 1). On the west coast, 29 sites were selected in the vicinity of Gros Morne National Park. In central Newfoundland, 18 sites in and near Terra Nova National Park were chosen. An additional 10 sites were located on the Avalon Peninsula, the easternmost portion of the Island (Fig. 1). All sites were at least 500 m from a lake outlet to avoid potential influences on stream macroinvertebrate communities from lentic habitats (Lomond and Colbo 2000). Riffle-dominated reaches were chosen for sampling to maximize species richness (Angradi 1999, Wohl et al. 1995). Methods for collecting measures of wetted width, bankfull width, and velocity were taken directly from Reynoldson et al. (2003), while ArcGIS was used to gather information on watershed size and the extent of forest in the whole watershed. These five variables have been shown to affect macroinvertebrate composition (Bronmark et al. 1984, Hawkins et al. 1982, Kilgour and Barton 1999, Malmqvist and Hoffsten 2000) and were compared across regions to Figure 1. Study regions on the Island of Newfoundland (bottom left), with insets depicting sample locations (black dots) within each region. Note some cities (white dots) are included for geographic context. 94 Northeastern Naturalist Vol. 20, No. 1 evaluate potential confounding factors associated with region. Samples were collected from 4 October–27 October 2004, as this season was concluded to be the best in which to identify insects in Newfoundland in terms of specimen maturity (Smith 2009). Macroinvertebrate sampling Macroinvertebrates were collected using a 5-minute kick-net protocol with a 250-micrometer D-frame net. The sampling entailed traversing a reach in a zigzag pattern while disturbing the stream substrate with kicking and rubbing. The sample was then poured through a 250-micrometer screen and preserved with 90% ethanol. In the laboratory, samples were divided among 100 cells using a Marchant box (Marchant 1989), whereupon cells were randomly selected and invertebrates removed. Cells were selected and processed until the cumulative number of invertebrates reached or exceeded 300 individuals (Reynoldson et al. 2003). All invertebrates recovered were placed in a vial with 75% ethanol. Invertebrates were identified to the lowest feasible taxonomic level (species for mature insect specimens, genus for immature or damaged insect specimens, family for Chironomidae and most non-insects) using numerous taxonomic keys (Adler et al. 2004, Merritt and Cummins 1996, Morihara and McCafferty 1979, Peckarsky et al. 1990, Wiggins 1996). Statistical analyses We evaluated uniformity amongst regions in velocity, bankfull width, and watershed size (the latter two log transformed to meet assumptions of normality) using analysis of variance (ANOVA; R Core Team 2010). The macroinvertebrate data were reduced to the genus level and square root transformed prior to analysis to reflect our interest in the diversity of these communities as opposed to a few dominant taxa (Clarke and Gorley 2006). Taxonomic richness across regions was assessed in terms of taxa per sample and total taxa across all samples. Richness per sample was compared using ANOVA after the model residuals were visually inspected for assumptions of normality and heteroscedasticity (Zuur et al. 2009). However, since the taxonomic richness across all sites in a region would be expected to be higher in areas with greater sampling effort, a randomization approach (Manly 1991) was used to determine if richness across regions differed beyond what would be expected due to unequal sample sizes. The randomization approach permuted site labels randomly across regions without replacement. In each of 10,000 permuted datasets, differences in regional taxonomic richness were assessed for each regional comparison. Observed field values were then compared to the ranked permuted distributions to determine if real regional differences in taxonomic richness were greater than those caused entirely by disparities in sample size. The following tests were performed on both genus-level and family-level data. Using a Bray-Curtis resemblance matrix (Bray and Curtis 1957), an analysis of 2013 E.L. Smith, D. Coté, and M.H. Colbo 95 similarity (ANOSIM) (Clarke and Gorley 2006) test was run to test for statistical differences amongst regions. A non-metric multi-dimensional scaling (MDS) ordination was used to visualize the similarity in invertebrate communities amongst sites. One thousand iterations were conducted for the MDS ordination to ensure that the resulting plot was the best conformation (lowest stress). Region-specific differences in multivariate dispersion were evaluated using the PERMDISP routine (9999 permutations) in PERMANOVA+ (Anderson et al. 2008), a permutation-based test using the ANOVA F statistic to compare amongtreatment distances from their group centroid. Results Among measured habitat variables, only watershed size differed significantly between regions (F = 6.45, P = 0.003; Table 2). This result was largely due to the inclusion of four watersheds in the Terra Nova region which were considerably larger than other sampled watersheds. While more expansive habitats are expected to have greater species richness, there were no differences detected in richness between the four large watersheds and the remaining samples (ANOVA F = 0.727, P = 0.407). All measured features tended to have broad ranges in each region. Standard deviations around means were similar among regions for each variable. The permutation test showed that at least some differences in species richness among regions were not due to unequal sample sizes. The Avalon Peninsula had the fewest taxa across all samples at the genus level. However, once sample size was accounted for with the permutation approach, only the Terra Nova region Table 2. Means, standard deviations, and ranges of common stream-scale habitat variables within regions. Std. dev. = standard deviation. Significant P-values in bold. F-statistic Habitat variable Region Mean (std. dev.) Range (P-value) Bankfull width (m) 0.8993 Avalon 13.11 (2.12) 5.70–23.00 (0.4129) Terra Nova 15.29 (3.32) 3.00–53.70 Gros Morne 17.49 (2.19) 5.20–55.50 Velocity (m/s) 0.1205 Avalon 1.10 (0.25) 0.68–2.92 (0.8867) Terra Nova 0.85 (0.06) 0.42–1.20 Gros Morne 1.22 (0.66) 0.18–19.36 Watershed size (km2) 6.4544 Avalon 2.26 x 107 (6.22 x 106) 6.34 x 106–5.68 x 107 (0.0031) Terra Nova 1.05 x 108 (4.55 x 107) 1.51 x 106–7.63 x 108 Gros Morne 1.60 x 107 (3.67 x 106) 2.66 x 105–9.26 x 107 Forested watershed (%) 2.1833 Avalon 24.04 (7.94) 0.00–57.98 (0.1200) Terra Nova 41.73 (4.96) 8.36–76.69 Gros Morne 27.87 (5.46) 0.00–85.02 96 Northeastern Naturalist Vol. 20, No. 1 significantly exceeded the Gros Morne region’s taxonomic richness (P < 0.001). Mean taxonomic richness per site differed significantly across regions (ANOVA F = 7.644, P = 0.001); Terra Nova had the highest mean taxonomic richness, while Gros Morne had the lowest (Fig. 2). Sites on the Avalon Peninsula had the least amount of site-to-site variation, while Gros Morne sites varied substantially in richness (Fig. 2). In general, Avalon Peninsula sites and Terra Nova sites tended to have distinct clusters, but both were interspersed amongst Gros Morne sites (Fig. 3). Community variability, as measured by differences in distance to regional centroid, was not significantly different at the genus level (F = 0.392, P = 0.080) and marginally significant at the family level (F = 3.38, P = 0.044). Pairwise tests showed that the dispersion detected using PERMDISP was only significant between Terra Nova and Gros Morne (t = 2.77, P < 0.008; Bonferroni- adjusted α = 0.016). Figure 2. Number of taxa per site in each region. Mid-quartiles are represented by the box, horizontal lines represent the median, and the whiskers represent adjacent data falling within 1.5 times the inter-quartile range. 2013 E.L. Smith, D. Coté, and M.H. Colbo 97 Figure 3. Multidimensional scaling (MDS) plot of macroinvertebrate communities sampled in the Avalon, Terra Nova, and Gros Morne regions. Top panel represents communities identified to the genus level, and bottom panel represents communities identified to the family level. 98 Northeastern Naturalist Vol. 20, No. 1 The benthic community structure at both the genus and family level of identification was significantly different across regions (genus: ANOSIM R = 0.274, P = 0.001; family: ANOSIM R = 0.175, P = 0.002). Furthermore, pairwise tests between the Avalon Peninsula and Terra Nova, and the Avalon Peninsula and Gros Morne indicated that the regions were significantly different from one another at both levels of identification (all ANOSIM R = >0.300, all P = 0.001). However, Terra Nova and Gros Morne communities were only statistically different from one another at the genus level (genus: ANOSIM R = 0.169, P = 0.003; family: ANOSIM R = 0.016, P = 0.298). These differences remain significant even when alpha levels are adjusted using the Bonferroni method (α = 0.016) to account for multiple comparisons. Discussion Anthropogenic induced changes to invertebrate communities are manifested through, among other things, changes in water chemistry and changing thermal and hydrological regimes (e.g., possibly related to climate change). On the Island of Newfoundland, the extent of anthropogenic impact (e.g., urbanization) is relatively mild, particularly beyond the Avalon Peninsula, due to the relatively low population density and the tendency for anthropogenic development to be located along the coast, where freshwater impacts are minimized. While this study did not directly evaluate the utility of stream invertebrates for assessing anthropogenic impacts, we have determined that Newfoundland’s depauperate invertebrate communities are sufficiently diverse to respond to regional environmental differences related to climate and geology (and resulting water chemistry). As a result, these communities are also expected to respond to anthropogenic impacts within regions, assuming that such impacts would result in environmental differences at least as large as those observed in this study. Supporting this contention is that regional comparisons of community structure involving the more urbanized Avalon Peninsula region showed the greatest divergence from other areas of the Island. These findings are consistent with studies of taxonomically poor tundra streams where family-level benthic invertebrate data were found to be suitable for detecting anthropogenic impacts (Medeiros et al. 2010). Regional taxonomic richness was expected to be higher in areas of high habitat diversity (Thienemann 1954, as outlined by Vinson and Hawkins 1998). Therefore, the Gros Morne region of the Island was expected to have the greatest diversity and richness due to the high geological and topographical diversity in the area (Lomond and Colbo 2000, Vinson and Hawkins 1998). As expected, the community structure associated with the Gros Morne sites was more variable relative to other regions and virtually encompassed the range of community structure observed in other regions. Surprisingly, taxonomic richness of the region as a whole was significantly less than Terra Nova, 2013 E.L. Smith, D. Coté, and M.H. Colbo 99 a region with substantially less habitat diversity. Among the sites of Gros Morne are those characterized by extreme environmental conditions (extreme spate amplitudes, low pH) and low invertebrate density and richness. It appears that the communities inhabiting these sites are comprised of subsets of taxa found in richer sites and therefore do not contribute new taxa to the richness of the region. This lack of niche specialization could be reflective of the low taxonomic richness of the Island’s fauna. Taking the relatively low explanatory power of the multivariate models together with the low richness of Gros Morne indicates that region is insufficient to properly define Newfoundland communities. Similar conclusions were derived for diverse assemblages of benthic invertebrates in Swedish ecoregions, where habitats were too diverse at such spatial scales to properly classify communities (Sandin and Johnson 2000). Physical and/or chemical habitat attributes, collected at the local scale, are needed to improve predictions of community structure, richness, and diversity of Newfoundland riffle communities. Conclusions The macroinvertebrate fauna of Newfoundland are sufficiently diverse to respond to regional-scale habitat attributes of the Island. Such responsiveness suggests that impoverished communities can remain useful for detecting environmental change. However, the absence of qualitative relationships between within-region habitat diversity and taxonomic richness suggests that the species compliment of the Island of Newfoundland generally lacks taxa that specialize in occupying extreme habitat niches. Acknowledgments Many thanks are extended to Parks Canada, the Heritage Foundation, and Young Canada Works for supplying field workers. NSERC, Parks Canada, and Memorial University of Newfoundland are most gratefully remembered for their provision of funds. Literature Cited Adler, P.H., D.C. Currie, and D.M. Wood. 2004. The Black Flies (Simuliidae) of North America. Cornell University Press, Ithaca, NY. 941 pp. Anderson, M.J., R.N. Gorley, and K.R. Clarke. 2008. PERMANOVA+ for PRIMER: Guide to Statistical Methods. PRIMER-E, Plymouth, UK. Angradi, T.R. 1999. Fine sediment and macroinvertebrate assemblages in Appalachian streams: A field experiment with biomonitoring applications. Journal of the North American Benthological Society 18:49–66. Banfield, C.E. 1983. Climate. Pp. 37–106, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. 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. 2[EPA 841-B-99-002]. US Envrionmental Protection Agency, Office of Water, Washington, DC. 339 pp. 100 Northeastern Naturalist Vol. 20, No. 1 Bray, J.R., and J.T. Curtis. 1957. An ordination of the upland forest communities of Southern Wisconsin. Ecological Monographs 27:325–349. Bronmark, C., J. Herrmann, B. Malmqvist, C. Otto, and P. Sjostrom. 1984. Animal community structure as a function of stream size. Hydrobiologia 112:73–79. Brunelle, P.M. 1997. Distribution of dragonflies and damselflies (Odonata) of the Atlantic Provinces, Canada. Northeastern Naturalist 4:61–82. Burian, S.K., and K.E. Gibbs. 1991. Mayflies of Maine: An annotated faunal list. Maine Agricultural Experiment Station Technical Bulletin. No. 142. 115 pp. Clarke, K.R., and R.N. Gorley. 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E Ltd., Plymouth, UK. 190 pp. Damman, A.W.H. 1983. An ecological subdivision of the Island of Newfoundland. Pp. 163–206, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. Environment Canada. 2000. National climate data and information archive: Canadian climate normals or averages 1971–2000. Available online at http://www.climate. weatheroffice.gc.ca/climate_normals/index_e.html. Accessed 1 April 2012. Griffith, M.B., B.H. Hill, F.H. McCormick, P.R. Kaufmann, A.T. Herlihy, and A.R. Selle. 2005. Comparative application of indices of biotic integrity based on periphyton, macroinvertebrates, and fish to southern Rocky Mountain streams. Ecological Indicators 5:117–136. Hawkins, C.P., M.L. Murphy, and N.H. Anderson. 1982. Effects of canopy, substrate composition, and gradient on the structure of macroinvertebrate communities in Cascade Range streams of Oregon. Ecology 63:1840–1856. Hellawell, J.M. 1978. Biological surveillance of rivers: A biological monitoring handbook. Water Research Centre, Stevenage, UK. 331 pp. Kilgour, B.W., and D.R. Barton. 1999. Associations between stream fish and benthos across environmental gradients in southern Ontario, Canada. Freshwater Biology 41:553–566. Larson, D.J., and M.H. Colbo. 1983. The aquatic insects: Biogeographic considerations. Pp. 593–677, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. Lomond, T.M., and M.H. Colbo. 2000. Variations in lake-outlet Ephemeroptera, Plecoptera, and Trichoptera communities amongst regions of eastern Newfoundland, Canada. Canadian Journal of Zoology 78:1536–1543. MacArthur, R.H. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, NJ. 218 pp. Malmqvist, B., and P.-O. Hoffsten. 2000. Macroinvertebrate taxonomic richness, community structure, and nestedness in Swedish streams. Archiv für Hydrobiologie 150:29–54. Manly, B.F.J. 1991. Randomization and Monte Carlo Methods in Biology. Chapman and Hall, London, UK. 281 pp. Marchant, R. 1989. A subsampler for samples of benthic invertebrates. Australian Society for Limnology Bulletin 12:49–52. Medeiros, A.S., C.E. Luszczek, J. Shirley, and R. Quinlan. 2010. Benthic biomonitoring in arctic tundra streams: A community-based approach in Iqaluit, Nunavut, Canada. Arctic 64:59–72. 2013 E.L. Smith, D. Coté, and M.H. Colbo 101 Merritt, R.W., and K.W. Cummins. 1996. An Introduction to the Aquatic Insects of North America. 3rd Edition. Kendall/Hunt Publishing Company, Dubuque, IA. 862 pp. Merritt, R.W., K.W. Cummins, and M.B. Berg. 2008. An Introduction to the Aquatic Insects of North America. 4th Edition. Kendall/Hunt Publishing Company, Dubuque, IA. 1158 pp. Morihara, D.K., and W.P. McCafferty. 1979. The Baetis larvae of North America (Ephemeroptera: Baetidae). Transactions of the American Entomological Society 105:139–221. Peckarsky, B.L., P.R. Fraissinet, M.A. Penton, and D.J. Conklin, Jr. 1990. Freshwater Macroinvertebrates of Northeastern North America. Cornell University Press, Ithaca, NY. 442 pp. Poff, N.L. 1997. Landscape filters and species traits: Towards mechanistic understanding and prediction in stream ecology. Journal of the North American Benthological Society 16:391–409. Preston, F.W. 1980. Noncanonical distributions of commonness and rarity. Ecology 61:88–97. R Core Team. 2010. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at http://www.Rproject. org. Accessed 21 March 2012. Reynoldson, T.B., C. Logan, T. Pascoe, and S.P. Thompson. 2003. CABIN (Canadian Aquatic Biomonitoring Network) invertebrate biomonitoring field and laboratory manual. National Water Research Institute, Environment Canada, Dartmouth, NS, Canada. 59 pp. Roberts, B.A. 1983. Soils. Pp. 107–162, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. Rogerson, R.J. 1983. Geological evolution. Pp. 5–35, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. Sandin, L., and R.K. Johnson. 2000. Ecoregions and benthic macroinvertebrate assemblages of Swedish streams. Journal of the North American Benthological Society 19:462–474. Shaw, J., D.J.W. Piper, G.B.J. Fader, E.L. King, B.J. Todd, T. Bell, M.J. Batterson, and D.G.E. Liverman. 2006. A conceptual model of the deglaciation of Atlantic Canada. Quaternary Science Reviews 25:2059–2081. Smith, E.L.S. 2009. Newfoundland macroinvertebrate communities and their potential for use in bioassessment. M.Sc Thesis. Memorial University of Newfoundland, St. John’s, NL, Canada. 225 pp. South, G.R. 1983. Introduction. Pp. 1–4, In G.R.South (Ed.). Biogeography and Ecology of the Island of Newfoundland. Dr. W. Junk Publishers, The Hague, The Netherlands. 723 pp. Thienemann, A. 1954. Ein drittes biozonotisches Grundprinzip. Archiv für Hydrobiologie 49:421–422. Vinson, M.R., and C.P. Hawkins. 1998. Biodiversity of stream insects: Variation at local, basin, and regional scales. Annual Review of Entomology 43:271–293. 102 Northeastern Naturalist Vol. 20, No. 1 Wiggins, G.B. 1996. Larvae of the North American Caddisfly Genera (Trichoptera). University of Toronto Press, Toronto, ON, Canada. 457 pp. Wohl, D.L., J.B. Wallace, and J.L. Meyer. 1995. Benthic macroinvertebrate community structure, function, and production with respect to habitat type, reach, and drainage basin in the southern Appalachians (USA). Freshwater Biology 34:447–464. Zuur, A.F., E.N. Ieno, N.J. Walker, A.A. Saveliev, and G.M. Smith. 2009. Mixed Effects Models and Extensions in Ecology with R. Springer Science and Business Media, New York, NY. 574 pp.