Differences in benthic macroinvertebrate assemblages associated with a bloom of Didymosphenia geminata in the Lower American River, California.
Many studies indicate that didymo is likely to have a significant effect on ecosystems due to its ability to alter abundance and distribution of organisms at the base of the aquatic food web (e.g., Gillis and Chalifour, 2010; C. Kilroy et al., in litt.; S. Larned et al., in litt.). In waters where didymo is abundant, macroinvertebrate taxonomic composition tends to shift from a highly diverse assemblage of large-bodied taxa (e.g., Ephemeroptera, Plecoptera, and Trichoptera; also referred to as EPT) to a less-diverse assemblage of smaller-bodied taxa such as Diptera, especially Chironomidae (Mundie and Crabtree, 1997; Blanco and Ector, 2009; Gillis and Chalifour, 2010; James et al., 2010). Furthermore, total macroinvertebrate abundance often increases in areas dominated by didymo (Blanco and Ector, 2009; Kilroy et al., 2009; Gillis and Chalifour, 2010).
In early 2012, fisheries researchers monitoring the effects of instream gravel supplementation on reproduction of Chinook salmon (Oncorhynchus tshawytscha) and steelhead (Oncorhynchus mykiss) in the Lower American River, Sacramento County, California, observed for the first time that didymo was locally abundant in the main channel immediately downstream from Nimbus Dam. Due to its ability to carpet the stream bottom in thick mats, resource managers were concerned that excessive growth of didymo could interfere with on-going efforts to restore spawning and rearing habitats for anadromous salmonids. Because salmonids bury their embryos in redds under several centimeters of gravel, resource managers postulated that didymo mats might inhibit adults from spawning or hatchlings from emerging through the gravel. In addition, shifts in macroinvertebrate taxonomic assemblages could potentially affect juvenile salmonids by influencing the quality and quantity of their forage supply (Shearer et al., 2007). Although there is currently no published scientific evidence to suggest this diatom has adversely affected salmonids in California or elsewhere, we initiated our study to provide preliminary information on the effects of didymo under bloom conditions to benthic macroinvertebrates that serve as potential forage for juvenile Chinook salmon and steelhead. Specifically, our objective was to compare the taxonomic composition and abundance of benthic macroinvertebrates in the main channel of the Lower American River, where didymo was abundant, with those in a nearby side channel where didymo was seemingly rare or absent.
Study Area and Methods--Our two study reaches were located in the Lower American River downstream from Nimbus Dam (crest elevation, 40.2 m), a 333-m-long concrete gravity structure located 10.94 km downstream from Folsom Dam that serves to reregulate water released from Folsom Lake. The first reach (38[degrees]38.087'N, 121[degrees]13.719'W) was located 0.77 km below Nimbus Dam in a gravel-bottomed main channel heavily overgrown with didymo. The second reach (38[degrees]38.244'N, 121[degrees]15.469'W) was located 3.31 km below Nimbus Dam in a gravel-bottomed side channel connected at both ends to the main channel. Didymo was not grossly visible in the side channel. During our study, the two reaches exhibited similar water quality (temperature, 13.6-17.1[degrees]C; dissolved oxygen, 9.311.6 mg/L; turbidity, 1.1-1.6 nephelometric turbidity units) and, although not recorded, similar water depths and current velocities (K. Sellheim, pers. comm.).
Sampling Methods--Benthic macroinvertebrates were sampled concurrently from three randomly selected sites within each reach during May and June 2012 using a Wildco[R] Hess sampler (Wildlife Supply Company, Yulee, Florida; 33-cm diameter, 40-cm high, 363-pm mesh with attached 368-[micro]m dolphin bucket). We obtained macroinvertebrates within the 0.086-[m.sup.2] area by scrubbing the bottom substrate to a depth of ~15 cm with a brush. Samples were stored in 500-mL Nalgene bottles containing 95% ethyl alcohol and transported to a laboratory in West Sacramento, California, for sorting and analysis. Macroinvertebrates were identified to family level with standard taxonomic keys (e.g., Merritt and Cummins, 2008; Thorp and Covich, 2010) and counted using a 60x stereoscopic microscope.
Statistical Analysis--The taxonomic characteristics of macroinvertebrate samples were summarized using PC-ORD v. 6 (MjM Software, Gleneden Beach, Oregon) and SAS v 9.1 (SAS Institute Inc., Cary, North Carolina). Prior to analysis, rare taxa (taxa occurring in two or fewer samples) were omitted and counts of macroinvertebrates were relativized by conversion to ranks (McCune and Grace, 2002). We used a hierarchical clustering procedure (Ward's method, also known as minimum variance clustering) to determine if patterns existed among macroinvertebrate taxa from the various sampling sites. The number of "significant" clusters was determined by visually examining plots of the semi-partial [R.sup.2] statistic, the pseudo F statistic, and the pseudo [t.sup.2] statistic (SAS Institute Inc., 1989). We corroborated the groupings of sites by cluster analysis by plotting results from nonmetric multidimensional scaling. The nonmetric multidimensional scaling procedure was performed by using the relative Sorensen distance measure, and included 500 iterations per run (beginning with a random configuration), with the best solution (lowest stress) chosen from 250 runs of real data. To determine if macroinvertebrate taxonomic composition differed between the groupings of sites, a multi-response permutation procedure analysis was conducted using Euclidean distances. If we found that macroinvertebrate taxonomic composition differed significantly, the taxa responsible for the difference were identified by using Indicator Species Analysis (ISA; Dufrene and Legendre, 1997). Two-way analysis of variance (ANOVA) was used to compare variations in counts of macroinvertebrates for each major taxon identified by ISA and within arbitrary classifications such as total macroinvertebrates (all taxa combined), the percentage of EPT (%EPT; a measure of the total contribution of large-bodied macroinvertebrates), and selected diversity indices (Margalef richness, Pielou's evenness, and Shannon-Wiener diversity). Brown and Guy (2007:15) recommended that nonparametric statistical procedures be employed "... if the sample contains fewer than a dozen or so observations." Thus, we used rank-transformed data when computing F statistics in a two-way ANOVA (SAS Institute Inc., 1988). Unless indicated otherwise, the significance level for all statistical comparisons was [alpha] = 0.05.
Results--Cluster analysis of macroinvertebrate taxa indicated the existence of two major groups of sample sites corresponding to either the main channel where didymo was abundant or the side channel where didymo was seemingly rare or absent (Fig. 1). These two groups (clusters) of sample sites were also observed in the nonmetric multidimensional scaling ordination plot (3D solution, final stress = 0.035; Fig. 2).
According to multi-response permutation procedure, macroinvertebrate taxa from two groups of sampling sites comprising the main channel and the side channel exhibited a highly significant difference (A = 0.18, T = -5.69, P < 0.001). ISA identified seven macroinvertebrate taxa that distinguished the main channel from the side channel (Table 1). Four taxa (Hydroptilidae, Hirudinidae, Chironomidae, and Gammaridae) were indicators of the main channel (i.e., they were most constant and abundant within the main channel), whereas three taxa (Baetidae, Glossosomatidae, and Tipulidae) were indicators of the side channel.
According to a two-way ANOVA, Reach*Date interactions were not significant for counts of the seven indicator taxa and total macroinvertebrates, %EPT, and measures of the three diversity indices (Table 2). Thus, we focused our attention on the main effects (Reach or Date). Counts of the seven indicator taxa exhibited significant Reach effects (main channel versus side channel; Table 2). Hydroptilidae, Hirudinidae, Chironomidae, and Gammaridae were more numerous in the main channel than in the side channel, whereas Baetidae, Glossosomatidae, and Tipulidae were more numerous in the side channel than in the main channel (Fig. 3). Reach effects were also significant for counts of total macroinvertebrates, %EPT, and the three diversity indices (Table 2). Counts of total macroinvertebrates were significantly higher in the main channel than in the side channel (Fig. 3). By comparison, %EPT was higher in the side channel (median = 35.3%) than in the main channel (median = 3.1%). The three diversity indices were also higher in the side channel than in the main channel (Fig. 4).
The only variables exhibiting significant differences for Date effects were Chironomidae, total macroinvertebrates, and %EPT (Table 2). Chironomidae and total macroinvertebrates were more numerous in May than in June (for Chironomidae in May, median = 1,916.5 and in June, median = 822.5; for total macroinvertebrates in May, median = 2,653.5 and in June, median = 1,199.0). On the other hand, %EPT was lower in May (median = 11.0%) than in June (median = 21.3%).
Discussion--Judging from our results, the didymo bloom influenced the composition and abundance of benthic macroinvertebrates in the main channel of the Lower American River. Small-bodied taxa (e.g., Hydroptilidae, Chironomidae) and taxa that are strongly thigmotactic (react negatively to light; e.g., Hirudinidae, Gammaridae) characterized the main channel where didymo was abundant. Conversely, larger-bodied taxa (e.g., Baetidae, Glossosomatidae, Tipulidae) characterized the side channel where didymo was seemingly rare or absent, a finding consistent with observations made by other investigators in North America, New Zealand, and elsewhere (e.g., Shearer et al., 2007; Kilroy et al., 2009; Gillis and Chalifour, 2010; James et al., 2010). Moreover, we documented a higher count of total macroinvertebrates and a lower taxonomic diversity in the main channel than in the side channel, which also corroborated observations made by other investigators (e.g., Mundie and Crabtree, 1997; Kilroy et al., 2009; Gillis and Chalifour, 2010).
The ecological basis for a shift in macroinvertebrate composition from large-bodied taxa to smaller-bodied taxa in the presence of a didymo bloom is still poorly understood. According to Rice et al. (2008), excessive growth of didymo may create a microhabitat exhibiting reduced flow (water velocity) and little or no predation that favors proliferation of smaller macroinvertebrates. Rapid expansion of didymo mats can smother the original habitat (e.g., gravel or cobble) while simultaneously creating new habitat from the extensive stalk matrix, leading to changes in algal community structure and the taxonomic composition of aquatic invertebrates that graze on algae (Kilroy et al., 2009). The dense stalks secreted by didymo cells also provide a habitat preferred by clinging and burrowing invertebrate taxa such as chironomids and some baetids (Whitton et al., 2009; James et al., 2010). A. M. Larson (in litt., cited by Whitton et al., 2009) reported that relative abundances of Oligochaeta and Hirudinea were positively correlated with percent cover of didymo. Dense mats of didymo might also lead to reduced intergravel water circulation in the streambed, creating impaired hyporheic water quality conditions (especially low dissolved oxygen concentrations) that favor hardier and less-diverse macroinvertebrate taxa generally associated with polluted waters (Blanco and Ector, 2009).
Drift-feeding salmonids, including juvenile Chinook salmon and steelhead, are known to prey selectively upon larger macroinvertebrates because they offer the greatest energy reward for effort (Healey, 1991; Shearer et al., 2007). In addition, large invertebrates are the most prone to drifting (Shearer et al., 2007), presumably because swift river currents are more likely to dislodge large rather than small organisms. Thus, if blooms of didymo result in decreased abundance of larger-bodied macroinvertebrates, coupled with an increase in smaller macroinvertebrates, the quality of forage organisms available to juvenile Chinook salmon and steelhead could be adversely affected.
Resource managers are also concerned that nuisance blooms of didymo might inhibit adult Chinook salmon and steelhead from constructing redds or that didymo could restrict water flow through redds and impede delivery of oxygen to, and removal of metabolic wastes from, embryos and yolk-sac larvae. However, a recent experiment by Bickel and Closs (2008) in the upper Clutha River system, South Island, New Zealand found no evidence of detrimental effects from didymo cover on selection of redd sites by trout or intragravel water flow in redds.
Little empirical evidence exists that didymo infestations have negatively affected salmonid populations. In North America, the first documented major bloom of didymo occurred in 1989 in the Heber River on Vancouver Island, British Columbia (Bothwell et al., 2008). Assessment of records for chum salmon (Oncorhynchus keta) and coho salmon (Oncorhynchus kisutch) escapement and productivity suggested that the didymo infestation had either no detectable impact or, in some cases, a positive impact on productivity. Similarly, steelhead productivity was the same in rivers with and without didymo both before and during the period of mass growths. In Norway, the most productive rivers for Atlantic salmon (Salmo salar) are also the ones with the most-extensive growths of didymo (Lindstrom and Skulberg, 2008). The situation in Iceland is similar, and there have been no negative impacts on Atlantic salmon populations, despite concerns about excessive growth by didymo (Jonsson et al., 2008).
Despite evidence to the contrary, the possibility exists that didymo blooms in the Lower American River could compromise ongoing restoration efforts aimed at increasing the spawning and rearing habitat for Chinook salmon and steelhead. In addition to the Lower American River, didymo has been observed upstream in the South Fork of the American River (Spaulding and Elwell, 2007; D. Tarbell, in litt., cited by Blanco and Ector, 2009) and in several other rivers within the Central Valley. These include the Feather and Bear rivers (C. Love, http:// www.fieldandstream.com/blogs/fishing/2010/08/ rock-snot-california-bane-anglers-and-mystery-sciecne [sic]), the Middle and South forks of the Yuba River (Rost etal., 2011), the Mokelumne River (Rost et al., 2011), the Stanislaus River (S. D. Wilcox et al., in litt., cited by Blanco and Ector, 2009), and the Tuolumne River (National Park Service, 2010). It is likely that additional blooms of didymo will be discovered in the near future. Clearly, more study is needed to further assess the threat from didymo proliferation and to identify risks to anadromous salmonid populations in the Lower American River and elsewhere in the Central Valley where habitat restoration programs are planned or underway.
We thank K. Dove, K. Martens, J. Siegel, J. Slingsby, and J. Sweeney for assistance in the field and laboratory and D. Richards and two anonymous reviewers for commenting on early drafts of this report. Our work was funded by the U.S. Bureau of Reclamation, the U.S. Fish and Wildlife Service, and the Sacramento Water Forum under Contract Number 20101049, and by Cramer Fish Sciences.
Bergey, E. A., J. T. Cooper, and B. C. Phillips. 2010. Substrate characteristics affect colonization by the bloom-forming diatom Didymosphenia geminata. Aquatic Ecology 44:33-40.
Bickel, T. O., and G. P. Closs. 2008. Impact of Didymosphenia geminata on hyporheic conditions in trout redds: reason for concern? New Zealand Journal of Marine and Freshwater Research 59:1028-1033.
Blanco, S., and L. Ector. 2009. Distribution, ecology and nuisance effects of the freshwater invasive diatom Didymosphenia geminata (Lyngbye) M. Schmidt: a literature review. Nova Hedwigia 88(3-4):347-422.
Bothwell, M., B. Holtby, D. R. Lynch, H. Wright, and K. Pellett. 2008. Did blooms of Didymosphenia geminata affect runs of anadromous salmonids on Vancouver Island? Pages 50-53 in Proceedings of the 2007 international workshop on Didymosphenia geminata (M. L. Bothwell and S. A. Spaulding, coeditors). Canadian Technical Report of Fisheries and Aquatic Sciences 2795.
Brown, M. L., and C.S. Guy. 2007. Science and statistics in fisheries research. Pages 1-29 in Analysis and interpretation of freshwater fisheries data (C. S. Guy and M. L. Brown, editors). American Fisheries Society, Bethesda, Maryland.
Dufrene, M., and P. Legendre. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs 67:345-366.
Gillis, C. A., and M. Chalifour. 2010. Changes in the macrobenthic community structure following the introduction of the invasive algae Didymosphenia geminata in the Matapedia River (Quebec, Canada). Hydrobiologia 647:63-70.
Healey, M. C. 1991. Life history of Chinook salmon (Oncorhynchus tshawytscha). Pages 311-393 in Pacific salmon life histories (C. Groot and L. Margolis, editors). UBC Press, Vancouver, British Columbia.
James, D. A., S. H. Ranney, S. R. Chipps, and B. D. Spindler. 2010. invertebrate composition and abundance associated with Didymosphenia geminata in a montane stream. Journal of Freshwater Ecology 25:235-241.
Jonsson, I. R., G. S. Jonsson, J. S. Olafsson, S. M. Einarsson, and T. Antonsson. 2008. Occurrence and colonization patterns of Didymosphenia geminata in icelandic streams. Pages 41-44 in Proceedings of the 2007 international workshop on Didymosphenia geminata (M. L. Bothwell and S. A. Spaulding, coeditors). Canadian Technical Report of Fisheries and Aquatic Sciences 2795.
Kilroy, C., S. T. Larned, and B. J. F. Biggs. 2009. The nonindigenous diatom Didymosphenia geminata alters benthic communities in New Zealand rivers. Freshwater Biology 54:1990-2002.
Kirkwood, A. E., L. J. Jackson, and E. McCauley. 2009. Are dams hotspots for Didymosphenia geminata blooms? Freshwater Biology 54(9):1856-1863.
Lindstrom, E.-A., and O. M. Skulberg. 2008. Didymosphenia geminata--a native diatom species of Norwegian rivers coexisting with the Atlantic salmon. Pages 35-40 in Proceedings of the 2007 international workshop on Didymosphenia geminata (M. L. Bothwell and S. A. Spaulding, co-editors). Canadian Technical Report of Fisheries and Aquatic Sciences 2795.
McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, oregon.
Merritt, R. W., and K. W. Cummins. 2008. An introduction to aquatic insects of North America. Second edition. Kendall/ Hunt Publishing Company, Dubuque, Iowa.
Mundie, J. H., and D. G. Crabtree. 1997. Effects on sediments and biota of cleaning a salmonid spawning channel. Fisheries Management and Ecology 4:111-126.
National Park Service. 2010. Invasive species: didymo a.k.a. "rock snot." Yosemite National Park, National Park Service, El Portal, California.
Patrick, R., and C. W. Reimer. 1975. The diatoms of the United States, exclusive of Alaska and Hawaii, Volume 2, Part 1. Entomoneidaceae, Cymbellaceae, Gomphonemaceae, Epithemaceae. Monographs of the Academy of Natural Sciences of Philadelphia 13:1-213.
Rice, S. P., T. Buffin-Belanger, J. Lancaster, and I. Reid. 2008. Chapter 24. Movements of a macroinvertebrate (Potamophylax latipennis) across a gravel-bed substrate: effects of local hydraulics and micro-topography under increasing discharge. Pages 637-659 in Gravel-bed rivers VI: from processes understanding to river restoration (H. Habersack, H. Piegay, and M. Rinaldi, editors). Elsevier, Waltham, Massachusetts.
Rost, A. L., C. H. Fritsen, and C. J. Davis. 2011. Distribution of freshwater diatom Didymosphenia geminata in streams in the Sierra Nevada, USA, in relation to water chemistry and bedrock geology. Hydrobiologia 665:157-167.
SAS Institute Inc. 1988. SAS[R] Procedures guide, release 6.03 edition, Cary, North Carolina.
SAS Institute Inc. 1989. SAS/STAT[R] User's guide, version 6, fourth edition, volume 1. Cary, North Carolina.
Shearer, K. A., J. Hay, and J. W. Hayes. 2007. Invertebrate drift and trout growth potential in didymo (Didymosphenia geminata) affected reaches of the Mararoa and Oreti rivers: April and August 2006. Prepared for MAF Biosecurity New Zealand. Cawthron Report No. 1214.
Spaulding, S. A., and L. Elwell. 2007. Increase in nuisance blooms and geographic expansion of the freshwater diatom Didymosphenia geminata. United States Geological Survey Open-File Report 2007-1425.
Thorp, J. H., and A. P. Covich. 2010. Ecology and classification of North American freshwater invertebrates. Third edition. Elsevier, San Diego, California.
Whitton, B. A., N. T. W. Ellwood, and B. Kawecka. 2009. Biology of the freshwater diatom Didymosphenia: a review. Hydrobiologia 630:1-37.
Submitted 4 November 2013.
Acceptance recommended by Associate Editor, Jerry L. Cook, 15 July 2014.
Ian J. Anderson, Michael K. Saiki, * Kirsten Sellheim, and Joseph E. Merz
Cramer Fish Sciences, 3300 Industrial Boulevard, Suite 100, West Sacramento, CA 95691
* Correspondent: firstname.lastname@example.org
Table 1--Results of Indicator Species Analysis (ISA) for seven macroinvertebrate taxa determined to be significant contributors of dissimilarity between the main channel (MC) where didymo was abundant and a side channel (SC) where didymo was seemingly rare or absent. IV from randomized groups Taxa Group Observed Mean SD P-value indicator value (IV) Trichoptera: MC 73.1 56.5 4.79 0.004 Hydroptilidae Hirudinea: Hirudinidae MC 71.8 56.5 4.81 0.006 Diptera: Chironomidae MC 70.5 56.5 4.82 0.011 Amphipoda: Gammaridae MC 69.2 56.5 4.77 0.016 Ephemeroptera: Baetidae SC 73.1 56.4 4.79 0.004 Trichoptera: SC 69.2 55.8 4.22 0.014 Glossosomatidae Diptera: Tipulidae SC 69.2 55.8 4.25 0.014 Table 2--Results of two-way analysis of variance (ANOVA), as Fvalues and significance levels, for seven macroinvertebrate taxa; total macroinvertebrates (all taxa combined); the combined percentage of Ephemeroptera, Plecoptera, and Trichoptera (%EPT); and three diversity indices (Margalef richness, Pielou's evenness, and Shannon-Wiener diversity). We based the ANOVA computations on rank-transformed data from two reaches (the main channel where didymo was abundant and a side channel where didymo was seemingly rare or absent) on two sampling dates (May and June 2012). Matrix Source Two-way ANOVA df MS F P Baetidae Reach 1 108 26.72 0.0009 * Date 1 2.08 0.52 0.4932 Interaction 1 0.08 0.02 0.8894 Error 8 4.04 Glossosomatidae Reach 1 75.00 19.15 0.0024 * Date 1 4.08 1.04 0.3371 Interaction 1 4.08 1.04 0.3371 Error 8 3.91 Hydroptilidae Reach 1 108.00 31.61 0.0005 * Date 1 2.08 0.61 0.4574 Interaction 1 4.08 1.20 0.3061 Error 8 3.42 Tipulidae Reach 1 75.00 16.51 0.0036 * Date 1 1.33 0.29 0.6027 Interaction 1 1.33 0.29 0.6027 Error 8 4.54 Chironomidae Reach 1 85.33 28.44 0.0007 * Date 1 33.33 11.11 0.0103 * Interaction 1 0.33 0.11 0.7475 Error 8 3.00 Gammaridae Reach 1 75.00 11.08 0.0104 * Date 1 12.00 1.77 0.2198 Interaction 1 0.33 0.05 0.8300 Error 8 6.77 Hirudinidae Reach 1 96.33 33.51 0.0004 * Date 1 21.33 7.42 0.0261 * Interaction 1 1.33 0.46 0.5151 Error 8 2.88 Total Reach 1 75.00 21.95 0.0016 * macroinver- Date 1 40.33 11.80 0.0089 * tebrates Interaction 1 0.33 0.10 0.7628 Error 8 3.42 %ept Reach 1 108.00 108.00 < 0.0001 * Date 1 27.00 27.00 0.0008 * Interaction 1 0.00 0.00 1.0000 Error 8 1.00 Margalef Reach 1 56.33 8.78 0.0181 * richness Date 1 27.00 4.21 0.0744 Interaction 1 8.33 1.30 0.2874 Error 8 6.42 Pielou's Reach 1 108.00 25.92 0.0009 * evenness Date 1 0.33 0.08 0.7845 Interaction 1 1.33 0.32 0.5871 Error 8 4.17 Shannon-Wiener Reach 1 108.00 40.50 0.0002 * diversity Date 1 5.33 2.00 0.1950 Interaction 1 8.33 3.12 0.1151 Error 8 2.67 * P-values that are statistically significant.
|Printer friendly Cite/link Email Feedback|
|Author:||Anderson, Ian J.; Saiki, Michael K.; Sellheim, Kirsten; Merz, Joseph E.|
|Date:||Sep 1, 2014|
|Previous Article:||Occupancy patterns of western yellow bats (Lasiurus xanthinus) in palm oases in the lower Colorado Desert.|
|Next Article:||Elevational range expansion in a neotropical live-bearing fish from Panama: implications for adaptive evolution.|