Great Basin Naturalist

Volume 56 Number 4 Article 1

11-21-1996

Species-environment relationships among filter-feeding (Trichoptera: ) in Rocky Mountain streams

Timothy B. Mihuc Idaho State University, Pocatello, Idaho

G. Wayne Minshall Idaho State University, Pocatello, Idaho

Janet R. Mihuc Louisiana State University, Baton Rouge

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Mihuc, Timothy B.; Minshall, G. Wayne; and Mihuc, Janet R. (1996) "Species-environment relationships among filter-feeding caddisflies (Trichoptera: Hydropsychidae) in Rocky Mountain streams," Great Basin Naturalist: Vol. 56 : No. 4 , Article 1. Available at: https://scholarsarchive.byu.edu/gbn/vol56/iss4/1

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. The Great Basin Naturalist PUBLISHED AT PROVO, UTAH, BY BRIGHAM YOUNG UNIVERSITY

[SSN 0017-3614

VOLUME 56 31 OCTOBER 1996 o. 4

Great Basin Naluralist 56(4), © 1996, pp. 287-293

SPECIES-ENVIRONMENT RELATIONSHIPS AMONG FILTER-FEEDING CADDISFLIES (TRICHOPTERA: HYDROPSYCHIDAE) IN ROCKY MOUNTAIN STREAMS

TImothy B. Mihuc1,2, G.Wayne MinshalJ!, and Janet R. Mihuc3

ABSTRACT.-Species-enviTonmenl relationships were determined fOT filter-feeding macroinvertebrates from 55 Rocky Mountain stream sites to establish species distribution patterns. Species abundance and 20 environmental vari­ ables were measured at each site with species-envkonment relationships determined using canonical correspondence analysis and stepwise multiple regression. Results suggest that the distribution of several taxa was strongly related to upstream-downstream environmental gradients. ArctOPSYCM grandis abundance increased with stream size (width and depth) and decreased with increasing turbulence (Reynolds number). Brachycentru.s abundance also increa~ed with stream sire {depth). Hydropsyche abundance increased with increasing ba.seflow. PMapsyche elsis abundance demon­ strn.ted negative correlation with depth, Froude number, and conductivity. Taxa followed previously reported patteros, partitioning habitat according to stream size. grandis, Brachycent1'W. and Hydropsyche were found in larger (3rd- to 6th·order) streams, while P(Uapsydut elsis was obseT\led in small headwater (lst- and 2nd-order) streams. Other filter-feeding taxa such as Stmulium, Pisiditml, and ostracods exhibited little or no apparent habitat partitioning among stream sites.

Key lJ)cm:ls: species-environment relationships. fJter feeders. Rocky Mountain streams.

Benthic macroinvertebrates adapted for re­ Minshall 1990, Richardson and Mackay 1991). moving particles from suspension (filter feeders) Many studies have determined filterer associa­ are an important component ofstream commu­ tions with food resources and environmental nities. Distribution patterns and habitat associ­ factors such as water velocity or temperature ations among 6lterers have been well docu­ (e.g., Edington 1968, Wallace 1974, Haddock mented, particularly for members of the Tri­ 1977, Wallace and Merritt 1980, A1stad 1982, choptera family Hydropsychidae (e.g., Decamps Hauer and Stanford 1982, Bruns et al. 1987, 1968, Edington and Hildrew 1973, Gordon Osborne and Herricks 1987, Wetmore et. al. and Wallace 1975, Wallace and Merritt 1980, 1990, Voelz and Ward 1992). Few studies have Ross and Wallace 1982, Tachet et al. 1992) and considered the entire filterer component found for lake outlet communities (e.g., Robinson and in natural (unimpounded, unregulated) streams

lStream Ec¢logy Center, Dep,u:tment of Biological Sciences, ld.bo State University, Pocatello. ID 83209-8007. 2PreW(lt addl'e~: Loui.suUla Coopenttive Fhberies and wildHre R~-oo. Unit, School of Fore~tty, Wildlife and FI~he'ies. Lo"i>;iana State Unive""ity. Bat,," Rooge, LA 7Q8OJ. 3.Biology Prog,rllm, 104 Ufe Sc.ienl:e$ Buildins. Louis:illJla Stille Unive,-,ity. Baron Roug¢. LA 70803.

287 2.88 GREAT BASIN NATUJIAUST [Volume 56 and distribution pattems of filterer species as temperature at the time of sampling) and with respect to a wide range of environmental minimum temperature (the freezing point of variahles (Edington and Hildrew 1973, Gordon water). Water chemistry variables included and Wallace 1975, Boon 1978, Ross and Wal­ hardness, alkalinity, pH, and specific conduc­ bee 1982). OUf objective was to assess the dis­ tance. Other biotic variables measured at each tribution patterns of filter feeders in unim­ stream reach were chlorophyll a (n = 5 per pounded Rocky Mountain, USA, streams to site), ash-fi·ee dry mass (AFDM) of periphyton determine relationships with specific environ­ (n = 5 per site), biomass/chlorophyll ratio of menlill variables includin~ flow parameters; periphyton (B/C), and benthic organic matter stream size, depth, and width; benthic organic content (BOM; 11 = 5 per site). This study did matter content; slope; water chemistry; peri­ not address food resources or food acquisition phyton hiomass; and temperature. \,Vhile many among filte.. feeders; therefore sampling of studies have considered current velocity, tem­ transported and henthic nne particulate mater­ perature,

TABLt: L SummUlY ofthe 55 study streams. Sites are arranged by increasing stream order and tncrc:L'iLlig depth witbiu each order. Slrea". Sample dates Order Avg Avg Basellow Slope depth (m) width (m) (m/s) (%)

Cach~. YNP 1988-1992 I 0.06 0.704 0.003 L2 E Blacktail Dt:!er, 'iN P 1988-1992 1 0.13 0.665 0.048 4.7 Twin, YNP 1988-J99'2 1 0.13 0.643 0.06 10.7 W Blacktail Deer, YNP 1988--1992 1 0.17 0.550 0.043 3.S F

Pinneer, ID 1990 2 0.05 0.342 0.13 6 DllltOO, 10 1990,91 2 0.06 0.109 0.07 17 Goat. 1D 1!Jl.X},91 2 0.06 0.089 0.05 18 Cache, y~p 1988--1992 2 0.09 0.764 0.012 10.1 Packhorse, ID 1991 2 0.09 0.4t3 0.04 4

Castle, 10 1992 2 0.09 0.160 0.03 U.5 Yellow, ID 1992 2 0.09 0.220 0.03 8 Rose, 'iN!' 1988-1992 2 0.10 0.416 0.027 7.S Sliver, IV 1991 2 0.10 0.243 0.04 5 EF Whimslick, ID 1991 2 0.10 0.460 0.02 2 Cache. YI\P 1988-1992 2 O.ll 0.832 0.012 8.8 Cliff, to 1988,90,91 2 0.12 00407 0.18 L2 Amphith~,tcr, Y:-.IP 198B---lOO2 2 0.13 1.11 0.146 4.9 POll)'. ID 1992 2 0.13 0.380 008 13 Iron Springs, YNP 1988---1992 2 0.14 0.237 0,1)38 13.1 E McCall. lD 1991 2 0.14 0.190 0.05 2 Blacktail Dt:!t:!r, YNP 1988-]992 2 0.15 0.710 0.151 15.2 H\iry, YN P 1988-1992 2 0.18 0.395 0.083 0.26 WF Cave, ID 1990 3 0.05 0.124 om 6 Doe,ID 1990 3 0.10 0.310 0.02 16

SF Cache, YNP 1988-1992 3 0.16 1.70 0.195 30 Pioneer, JD 1990 3 0.16 0.612 0.31 6 Hellrollring, YNP 1988-1992 3 0.17 1.2:J 0.32 2.5 McCall,ID 1991 .3 0.17 0.196 0.05 2 Pchhle, YNP 1988-1992 3 0.18 1.10 0.592 2.5 CouKar, 10 1990,91 3 0.18 0.297 0.10 12 Cache, YNP 1988-1992 3 0.19 4.60 0.475 1.7 LaV'a, YNP 1988--1992 3 0,24 0.768 0.893 2.1 hun Sprin~s, YN P 1988-1992 3 0.27 0.587 0.520 l.l Beaver. 10 1988 3 0.27 0.800 Ll7 4

(".ache, YNP 1!l88-1992 4 0.18 2.lJ5 0.67 1.2 Ramey. IV L988 4 0.18 0.630 0.74 3.5 Boulder, JD 1992 4 0.19 1.23 0.41 2 He1lroaring. YNP 1988-199'2 4 0.20 2.61 0.4.1 1.8 McCall, to 1991 4 0.22 0.240 0.13 2

Whimstiek Main, ID 1991 4 0.23 0.800 0.10 I WF Rapid. lD 1992 4 0.25 0.930 1.80 3 Lamar, YNP 1988-1992 4 0.34 2.87 2.85 .Ui Socl~ Bulte, Y P 1%8-1992 4 0.35 2.90 3.00 1.3 Indian, IO 1992 5 0.21 1.43 1.31 1.5 Pistol, IO 1992 5 0.33 1.70 1.80 1.8 Rush,10 1988 5 0.35 1.51 1.61 1 Camas. IO 1992 5 0.38 2.10 2.92 1 Chamberlain, 10 1992 6 0.24 1.69 2.43 3.5 Big Ck @Coxey, 10 1988 6 0.31 3.42 5.23 1.5

Rapid, ill 1992 6 0.37 1.48 l.ll 2.5 Loon, ill 1992 6 0.37 2.91 3.29 I Bi~ Ck @Gorge, ID 1988 6 0.37 4.32 8.83 1 Big Ck @Rush, 10 1988 6 0.45 4.3 8.04 1.5 Salmon River. ID 1992 6 0.48 lAO 5.'17 I 290 GREAT BASI~ NATURALIST [Volume 56

CFLOW::> Hy~. <@OTH::> Q>RDEJD A.-grandis BOM EMB SLOPE

'iii ~ t.. ~ . "' ~ Os~COJSa-... :_~ (to)

P. elsis

Axis #1 (X Axis)

Fi~. 1. 13iplot results Hf canonical correspondence annly.~is. Environmental variahles (circled) are listed ill Table 3. Species (lfe plotted USill~ specics names. Positive abundance relationships with a given enviromnental variable arc indj­ cntcd hy species Lhat filII close in the onlination plot to the environmental variable. Species that lall on the opposite end of the plot fronl an environmental variable exhibit a neglltive relationship with that variahle. Species ncar the center of the plot exhibilliltlc relationship with environmental variahles.

did not relate to any ofthe environmental vari­ DISCUSSION ables io the ordination plot and are not consid· ered furIher, Our results support Ule idea that macroin­ Stepwise multiple regression results indi­ vertebrate species in streams respond to envi­ cate species-environment relationships similar ronmental conditions in individualistic ways. to those found in the ordination (Table 3), Arc­ Each taxon was related to a different set of t'opsyche grandis abundance was positively environmental variables. General relationships correlated with stream depth and width and with environmental variables for A. grandis, negatively correlated with turbulent flow Brachycentrus, and Hydropsyche suggest that (Reynolds oumber), Brachycentrus abundance these taxa are adapted for larger river systems was positively correlated with stream depth (3rd--6th order; Fig, 2), BrachycentmR and Hy­ (Table 3), Hydropsyche abundance showed drupsyche are usually found in lower reaches positi.ve correlation with baseflow and negative in river systems (4Il,-6th order; Edington and correlation with water hardness and substrate Hildrew 1973, Boon 1978, Hauer and Stanford size, Pampsyche elsis abundance showed nega­ 1982, Ross and Wallace 1982, Wetmore et aL tive correlation with depth, surface turbulence 1990), A grandiR is most often found in mid­ (Froude number), and specific conductance reaches (3rd-5th order; Alstad 1980, Cuffney (Table 3), and Minshall 1981, Hauer and Stanford 1982). 1996] FII..;l'EH-FEEDING INVERTEBRATES IN ROCKY MOUNTAIN STREAMS 291

TABLE 2. Results ofcanonical correspondence analysis. Eigenvalues give the importance ofan axis on a scale between o and L Total inertia is the total variance in the species data. The species-environment correlations scale the strength of the relationship between species and environment for the axes.

Total Axes 1 2 .1 4 inertia Eigenvalues .444 .360 .174 .085 2.50 Specics-cnviromnent correlations .819 .772 .640 .441 Cumulative percentage ofvariance: ofspecies data 17.8 :32.1 39.1 42.5 ofspecies-environment relationship :37.9 68.6 83.5 90.7 Slim ofall canonical eigenvalues 1.172

Among the taxa adapted for large streams, reach means for all variables in order to iden­ habitat partitioning is apparent in this study as tify factors affecting large-scale (among site) in others (Edington and Hildrew 1973, Boon distribution patterns among ta'\a. Microhabitat 1978, Alstad 1980, Hauer and Stanford 1982, requirements are ultimately responsible for the Ross and Wallace 1982). Taxa exhibited reach­ physical habitat selected by filter feeders (Smith­ scale macrohabitat preferences with Brachy­ Cuffney and Wallace 1987, Wetmore et. a1. centrus distribution related to stream depth, 1990), but reach~scale comparisons allow Hydropsyche related primarily to stream flow, broader scale distribution patterns to be stud­ and A. grandis to a combination of width, ied. The reach-scale comparisons herein indi­ depth, and tmbnlence. cate general conditions at each site in terms of P. elsis was prevalent in headwater stream available macrohabitat. The trends observed in reaches (Fig. 2), a pattern found in several the data indicate preferences for a other studies (Alstad 1980, Hauer and Stanford given reach and its associated habitat condi~ 1982). Distribution pattcrns for P clsis were tions. Differences in reach-scale means among explained by flow and stream-size variables. variables may also renect differences in gen­ Stream temperature may also be an important eral microhabitat conditions available among variable explaining P. elsis distribution patterns sites (e.g., slow- or fast-velocity microhabitats). (Alstad 1980, Hauer and Stanford 1982). Annual Reach-scale means, therefore, can serve as a temperature was measured in this study based useful integrator of microhabitat conditions in only un yearly max/min readings, which may order to facilitate comparisons at larger scales. not adequately reflect diflerences in tempera­ Evolutionary patterns probably have led to ture betw"een headwater sites and downstream habitat partitioning based on current speed locations, resulting in the lack of P. elsis pat­ and filtration rate among filter feeders in Rocky terns explained by temperature in our analysis. Mountain streams with some taxa adapted for Also, previous studies that suggest a down­ larger streams (Brachycentrus, Hydropsyche, stream temperature gradient as the explanation and A. grandis) and some for smaller systems for P elsis distribution (Hauer and Stanford (P elsis; Alstad 1980, 1982). Filter feeders may 1982) did not consider other variables (e.g., be a useful group to address habitat partition­ physical and hydrologic variables) that may con­ ing on large spatial scales in streams because tribute to habitat selection by P elsis. Multiple many filterer taxa appear to have partitioned factors are probably responsible for P elsis high habitat at these scales. In this study, stream abundance in headwater streams, including size (width, depth) and hydraulic parameters temperature patterns and hydraulic conditions. (baseflow, turbulence) were more important in Our results agree with published distribution explaining species-environment relationships patterns for all 4 taxa and provide evidence f<)r than other variables such as water chemistry, physical factors that are important in determin­ periphyton biomass, or benthic organic matter. ing habitat selection for each taxon (Fig. 2). Our results provide support for the idea that Habitat preferences demonstrated in this study evolutionary divergence among benthic maero­ are for distribution patterns among streams at invertebrate filterers has resulted in habitat the reach scale. Data \vere collected within a partitioning according to stream size and hydro­ 250-m reach at each site and expressed as logic parameters in Rocky Mountain streams 292 GREAT BASIN NATURALIST [Volume 56

TABLE 3. Summiu-y of the stepwise multiple regression results of the 4 most abundant species (dependent variable) versm: the 20 t:nvinmmental variables. Partial correlation coefficients and p values (parentheses) are shown for each variahle. Variables included in the regression model for each species are shown (variable included ifP < 0.05). Variable acronyms in Figure 1 are shown in parentheses.

Environmental Arctopsyche Brachycenfrus Hydropsyche ParopsycM vari~d.>le grt1Jldis elsis Stream order Depth 0.25 (0.004) 036 (0.00003) -0.23 (0.009) Width/depth mtio (WiD) Width 0.18 (0.04) Temperclture (TEMP) Slope Emheddedne" (EMB) Baseflnw (FLOW) 0.64 (0.0000) Velocity (VEL) Suhsb..... te size (SUB) -0.21 (0.019) Frollde number (FR) -0.19 (0.031) Reynolds number (RE) -0.30 (0.0008) Hardness (URD) -0.21 (0.018) Alkalinity (ALK) pH Conductivity (COND) -1l.175 (0.049) Periphyton chI. a (CI-IL a) Pedphyton AFDM (AFDM) Periphyton BIC mtio (BIG) Benthic org. matter (BOM)

Multiple R2 0.63 0.22 0.59 0.53

(Gordon and Wallace 1975, Boon 1978, Alstad sured, are necessary before factors affecting 1980, Hauer and Stanford 1982, Ross and Wal­ species distribution and abundauce patterus lace 1982). can be properly understood (Hall et. a1. 1992). While food resources were uot considered in this study, factors such as food type, quality, and particle size are also important in explain­ ing filterer distribution and abundance along stream gradients (Alstad 1980, Cuffney and stream Minshall 1981, Hauer and Stanford 1982, Ross ~ I a.nd Wallace 1982, Richardson and Mackay Depth- 1991, Voelz and Ward 1992). According to the F.oude shredder-collector facilitation hypothesis, lon­ 2 nl.mber- gitudinal distribution patterns among collec­ tor-Hlterers are thought to relate to the genera­ 3 tion of fine particulate organic matter (FPOM) Width .. by upstream shredders (Heard and Richardson • TurblJence - 1995). Whether this distribution pattern relates [Deplh +i primarily to FPOM facilitation by upstream shredders or to physical partitioning of habitat along stream gradients remains to be seeo. 6 Habitat partitioning among taxa in this study indicates that physical babitat requirements, \ apart from those of food, are important in Fig. 2. Summary of major trends in species abundance explaining longitudinal patterns along stream among the 4 most common taxa studied and which envi­ gradients. Multiple variables explained abun­ ronmental variables are most important in explaining dance patterns for t""a studied, supporting the those trends. Downstream gradient is depicted from smaJl streams (1st-2nd order) through large systems (3rd-6t.h idea that comprehensive approaches, where order). Positive and negative relatiQ1lships aloe indicated more than one environmental gradient is mea- by + and -, respectively 1996] FILTER-FEEDING INVERTEBRATES IN ROCKY MOUNTAIN STREAyrS 293

ACKNOWLEDGMENTS larva, Hydropsyche oslmi Banks. Pan-Pacific Ento­ mologist 53: 169-174. HALL, C. A. S., J. A. STANFORD, AND F R. HAUEl\. 1992. Yellowstone National Park streams were The distribution and abundance of organisms as a sampled as part of a study supported by the consequence of energy balances along multiple envi­ University of Wyoming/National Park Service ronmental gradients. Oikos 65: 377-390. Research Center. Idaho stream sampling was HAUER, F R., AND J. A. STA:-JFORD. 1982. Ecological supported by the U.S. Forest Service, Idaho responses ofhydropsychid caddisflies to stream regu­ lation. Canadian Journal ofFisheries and Aquatic Sci­ Division of Environmental Quality, and Idaho ences 39; 1235-1242. State University. Numerous members of the HEARD, S. E., Al\D J. S. RICHARDSOl\. 1995. Shredder-col­ Stream Ecology Center at Idaho State Univer­ lector facilitation in stream detrital food webs: is sity aided in the field sampling, including P. there enough evidence? Oikos 72: 359-366. OSBORNE, L. L., AND E. E. HEl\RlCKS. 1987. Microhabitat Dey, P. Koestier HI, D. E. Lawrence, M. J. characteristics of Hydropsyche and the importance of Mclnuye, J. N. Minshall, G. C. Mladenka, D. C. body size. Journal ofthe North Americ,m Bentholog­ Moser, J. S. Nelson, C. T. Robinson, R. L. Van­ ical Society 6: 115-124. note, and many others. C. T. Robinson did PLATTS, W S., W F. MEGAHAN, AND G, W MINSHALL. 1983. much of the data compilation for the environ­ 1·1ethods for evaluating stream, riparian, and biotic conditions. U.S. Forest Service General Technical mental variables that were used in the analy­ Report INT-138. 70 pp. ses. Data analysis and writing by TBM and RICHARDSO:-J, J. S., AND R. J. MACKAY. 1991. Lake outlets JRM were supported in part by Louisiana and the distribution of filter feeders: an assessment of State University. hypotheses. Oikos 62: 370-380. ROBIKSON, C. T., AND G. W MINSHALL. 1990. Longitudinal development of macroinvertebrate communities be­ LITERATURE CITED low oligotrophic lake outlets. Great Basin Naturalist 50,303-311. ALSTAD, D. N. 1980. Comparative biology of the common Ross, D. H., AND J. B. WALLACE. 1982. Factors influencing Utah Hydropsychidae. American Midland Naturalist the longitudinal distribution of larval Hydropsychi­ 103: 167-174. dae in a southern Appalachian stream system. . 1982. Current speed and filtration rate link cad­ Hydrobiologia 96: 185-199. --d~isfly phylogeny and distributional patterns on a S~1ITH-CUFFNEY, F. L., AND J. B. \VALLACE. 1987. The stream gradient. Science 216: 533-534. influence of microhabitat on availability of drifting BOON, P J. 1978. The pre-impoundment distribution of invertebrate prey to a net-spinning . Fresh­ certain Trichoptera larvae in the North Tyue river water Biology 17: 91-98. system, with particular reference to current speed. TACHET, H., J. P PEIRHOT, C. Roux, AND M. BOVIINAUD. Hydrobiologia 57: 167-174. 1992. Net-building behaviour of six Hydropsyche BRU~S, D. A., A. B. HALE, AND G. W. MINSHALL. 1987. species (Trichoptera) in relation to current velocity and Ecological correlates of species richness in three distribution along the Rhone River. Journal of the guilds of lotic macroinvertebrates. Journal of Fresh­ North American Benthological Society 11: 350-365. water Ecology 4: 163~176. TER BRMK, C. J. F. 1986. Canonical correspondence analy­ CUfFNEY, T F, A:-.lD G. W. MIKSHALL. 1981. Life history sis: a new eigenvector technique for direct gradient and bionomics of Arctopsyche grandis (Trichoptera) analysis. Ecology 67: 1167-1179, in a central Idaho stream. Holarctic Ecology 4: VOELZ, N. J., AND J. V WARD. 1992. Feeding habits and 252-262. food resources offilter-feeding Trichoptera in a regu­ DECAMPS, H. 1968. Vicariances ecologiques chez les Tri­ lated mountain stream. Hydrobiologia 231; 187-196. chopteres de Pyrenees. Annales de Limnologie 4: WALL-\CE, J. B. 1974. Food partitioning in net-spinning Tri­ 1--50. choptera larvae: Hydropsyche venularis, Cheurnato­ EDINGTOK, J. M. 1968. Habitat preferences in net spinning psyche etrona, and Macronema zebmtum. Annals of caddis larvae with special reference to the influence the Entomological Society of America 68: 463-472. of water velocity. Journal of Animal Ecology 37: WALLACE, J. B., AND R. \iV. MERRITT. 1980. Filter-feeding 675-692. ecology of aquatic . Annual Review of Ento­ EDINGTON, J. M., AND A. H. HILDREW. 1973. Experimental mology 25: 103-132. observations relating to the distribution of net-spin­ WETMORE, S. H., R. J. MACKAY, AND R. W. NEWBUIIY. 1990. ning Trichoptera in streams. Verhandlungen der Characterization of the hydraulic habitat of Bmchy­ Internationalen Vereinigung fuel' Limnologie 18: centru.s occidentalis, a filter feeding caddisfly. Journal 1549-1558. of the North American Benthological Society 9: GORDOl\', K, AKD J. B. WALlACE. 1975. Distribution of the 157-169. family Hydropsychidae (Trichoptera) in the Savannah River basin of North Carolina and Georgia. Hydrobi­ Receivecl6 December 1995 ologia 46: 405-423. Accepted 27June 1996 HADDOCK, J. D. 1977. The effect ofstream current velocity on the habitat preference ofa net-spinning caddis fly