Springer 2005

Hydrobiologia (2005) 541: 45–54 DOI 10.1007/s10750-004-4666-9

Primary Research Paper

Patterns of macroinvertebrate abundance in inland saline wetlands: a trophic analysis E. Andrew Hart & James R. Lovvorn* Department of Zoology, University of Wyoming, Laramie, WY 82071, USA (*Author for correspondence: Tel.: +1 307 766 6100, Fax: +1 307 766 5625, E-mail: [email protected]) Received 26 April 2004; in revised form 26 September 2004; accepted 11 October 2004

Key words: invertebrate diets, invertebrate community structure, saline wetlands, stable isotopes, wetland food webs

Abstract We used stable isotopes and gut-content analysis to compare trophic relations of macroinvertebrates between two types of saline wetlands in the Laramie Basin, Wyoming, USA. Amphipods (Hyalella azteca), chironomid larvae, and predatory insects (mainly zygopteran larvae) occurred in both wetland types. However, in oligosaline wetlands (0.5–5& total dissolved solids) amphipods were dominant, whereas in mesosaline wetlands (5–18&) amphipods were scarce and chironomid larvae and predatory insects were much more abundant. Salinity alone seemed inadequate to explain these differences, so we examined trophic interactions to address three main questions: (1) why are predatory insects more abundant in mesosaline wetlands, (2) why are chironomid larvae less abundant in oligosaline wetlands, and (3) why are amphipods uncommon in mesosaline wetlands? Zygopteran larvae ate mainly chironomid larvae and zooplankton, and did not eat amphipods; chironomid larvae were an average 93% dry mass of zygopteran intake. Guts of amphipods contained no animal parts and little vascular plant tissue, but rather mainly amorphous detritus. Chironomid larvae ate amorphous detritus as well as diatoms. There is potential for competition between amphipods and chironomids, but the nature of amorphous detritus in these wetlands needs further study. Because amphipods appear unavailable as prey for predatory insects, top-down impacts on chironomid larvae may be greater in oligosaline than in mesosaline wetlands. Our results suggest that (1) predatory zygopteran larvae and hemipterans are more abundant in mesosaline wetlands due to more abundant chironomid prey, (2) chironomids are less common in oligosaline wetlands because of both competition with amphipods and greater per capita consumption by predatory insects, and (3) amphipods are scarce in mesosaline wetlands because, lacking a resting stage and mode of direct dispersal, they do not cope well with extremes created by hydrologic instability in mesosaline wetlands. These mechanisms may apply to wetlands in other regions where similar patterns of invertebrate community structure have been reported. Introduction Most studies of invertebrate communities in inland saline waters have focused on control of species abundance by abiotic factors, mainly salinity (Cannings & Scudder, 1978; Galat et al., 1988; Hammer et al., 1990; Williams et al., 1990; Parker & Knight, 1992; Dexter, 1993; Herbst, 2001). However, biotic factors and interactions among

species can also structure communities, and effects of abiotic factors may be mostly indirect (Wurtsbaugh & Berry, 1990; Lovvorn et al., 1999). For example, salinity may exert bottom–up effects by influencing autotroph communities that provide both habitat structure and organic matter for invertebrate food webs (Herbst & Bradley, 1989; Wollheim & Lovvorn, 1996; Verschuren et al., 2000; Hart & Lovvorn, 2003). Salinity may also

46 directly affect predators or important competitors (via physiological tolerances) and thus exert indirect top–down effects on species occurrence and abundance (Wurtsbaugh & Berry, 1990; Simpson et al., 1998). Understanding the relative importance of various factors in structuring invertebrate communities requires knowledge of trophic pathways. In this paper, we examine patterns of macroinvertebrate abundance in saline wetlands from a trophic perspective. Past research in the Laramie Basin, Wyoming has identified two important wetland types (oligosaline and mesosaline) in which largely the same taxa occur, but with very different abundance (Lovvorn et al., 1999; Lovvorn & Hart, 2004). Hypotheses to explain these patterns could not be evaluated without understanding how feeding relations might differ between wetland types. Thus, we used stable isotopes and gut-content analysis to compare trophic relations between oligosaline and mesosaline wetlands. Our analysis focused on macroinvertebrates that were most abundant as well as most important for waterbirds using these wetlands: chironomid larvae, amphipods, and predatory insects (zygopteran larvae, corixids, and notonectids; Wollheim & Lovvorn, 1995). In mesosaline wetlands (5–18& total dissolved solids), invertebrate biomass is dominated by chironomids and predatory insects (also with abundant zooplankton), while amphipods are scarce (see Lovvorn et al., 1999 for a detailed description). In oligosaline wetlands (0.5–5&), total macroinvertebrate biomass is similar to that in mesosaline wetlands. However, in oligosaline wetlands, amphipods are very abundant if not dominant, while chironomids, predatory insects, and zooplankton occur at much lower levels. A similar dichotomy between wetlands dominated by amphipods vs. chironomids has been reported in a number of other studies, but has not been explained (Bailey & Titman, 1984; Hargeby, 1990; Hargeby et al., 1994; van den Berg et al., 1997). Thus, we organized our study around three main questions: (1) why are predatory insects much more abundant in mesosaline wetlands, (2) why is chironomid biomass low in oligosaline wetlands, and (3) why are amphipods uncommon in mesosaline wetlands?

Predatory insects, especially hemipterans, are common in moderately saline wetlands in North America (Cannings et al., 1980; Hammer, 1986; Hammer et al., 1990; Cox & Kadlec, 1995). Scarcity of fish in more unstable, saline wetlands probably plays a role in the abundance of predatory insects (Hammer, 1986; Lancaster & Scudder, 1987; Batzer et al., 2000; Zimmer et al., 2001). During our research, fish were rare or absent in both oligosaline and mesosaline wetlands we studied. To some extent, large numbers of corixids and nototectids might reflect selection of mesosaline wetlands by immigrating adults. However, for larval zygopterans, which must have originated from eggs, high biomass probably indicates high growth and survival. In addition to the absence of predatory fish, high growth and survival of predatory insects seemed likely to be fueled by more abundant prey in mesosaline wetlands, mainly chironomid larvae and zooplankton (cladocerans and copepods). For chironomid larvae, low biomass in our oligosaline wetlands was unexpected. Chironomids are often one of the most abundant taxa of inland wetlands with moderate to high salinities (Hammer, 1986). In terms of water chemistry, hydrologic stability, and habitat diversity, oligosaline wetlands were expected to be a more favorable habitat than mesosaline wetlands. The oligosaline wetlands were carpeted by dense beds of the submersed macrophyte Chara spp., and chironomid larvae can thrive in this habitat (Horne, 1991). Given these apparently favorable habitat conditions, we hypothesized that chironomid populations might be suppressed in oligosaline wetlands by competition with abundant amphipods. Evaluating this idea required knowing the diets of these taxa in the two wetland types. While Gammarid amphipods are known to mix herbivory with predation on zooplankton and chironomids (Menon, 1966; Anderson & Raasveldt, 1974; Kelly et al., 2002), we wished to know if Hyalella azteca might also consume chironomid larvae. The scarcity of amphipods in mesosaline wetlands was somewhat puzzling. The most common species in the Laramie Basin is Hyalella azteca, which is known to thrive at much higher salinities than those in our study wetlands (Hammer et al., 1990). Ion ratios in these other studies were similar

47 to those in our wetlands, with sulfate the dominant anion at higher salinities (Hammer, 1978; Lovvorn et al., 1999). In general, amphipods (including Hyalella spp.) are considered to be excluded from temporary waters by the lack of a desiccationresistant stage (Wiggins et al., 1980: 123). Although our mesosaline wetlands are not temporary on an annual basis, they can mostly dry up during even mild droughts which occur every 3– 8 years. Vulnerability to drying events seemed a likely factor in amphipod scarcity, but perhaps not a complete explanation. Three alternatives to hydrologic effects were considered. First, emergent vegetation (Scirpus acutus) is abundant around oligosaline wetlands but mostly absent from mesosaline wetlands; if H. azteca feeds as a shredder, its abundance might be reduced by disappearance of this major source of shreddable material (coarse particulate organic matter, CPOM). Second, studies of primary production in these wetlands indicated much lower production of epiphyton in mesosaline wetlands (Hart and Lovvorn, 2000); if amphipods eat mainly epiphyton, they might be food-limited in mesosaline wetlands. Third, higher abundance of predatory insects in mesosaline wetlands might suppress amphipod populations.

Methods In using stable isotopes to identify organic sources and foodweb pathways, the mean of more than one food with different isotope values can erroneously indicate consumption of yet another item that is not eaten. Consequently, we used both gut contents and stable carbon isotopes (d13C) at different times in different wetlands to examine trophic patterns (Mihuc & Toetz, 1994; Hart & Lovvorn, 2002, 2003). We also investigated use of nitrogen and sulfur isotopes (d15N, d34S); however, in our wetlands they were too variable in space, time, and/or trophic-level fractionation to yield useful inference (Hart & Lovvorn, 2002, 2003). We assumed that d13C changed by <1& per trophic level, so that signatures of predator and prey were about the same (Fry, 1991; Neill & Cornwell, 1992). Macroinvertebrates were collected in early June and mid-August 2000 in two oligosaline (George and Nelson) and two mesosaline wetlands

(Creighton and Gibbs) in the Laramie Basin, Wyoming, USA (see Peck & Lovvorn, 2001, for map of study area). Samples were collected at three different sites per wetland; water depth at each site was
0.5 m, and sites were in submersed macrophyte stands with representative density and height for each wetland (Hart & Lovvorn, 2000). In the field, samples for stable isotope analysis were quick-frozen in vials of distilled water placed in an ice-isopropanol bath. Reference samples for gut-content analysis were preserved in 10% formalin. For amphipods and chironomid larvae, the entire foregut was removed and transferred to a depression microscope slide where gut contents were separated from stomach walls. Gut contents were suspended in deionized water, filtered onto 0.45-lm membrane filters, mounted on a slide, and cleared with immersion oil. Each slide included the pooled gut contents of two to five chironomids or five to 10 amphipods. For each wetland and sampling period, three slides were analyzed for each invertebrate taxon. Slides were examined at 400· magnification, and the first 50 particles found along a transect across the slide were classified and measured with an ocular micrometer. Percent volume of a food ingested was assumed equivalent to the area occupied by a given particle type as a percentage of the total area occupied by all 50 particles (Hall et al., 2000; Hart and Lovvorn, 2003). For amphipods and chironomids, gut contents were classified as vascular plant tissue, green algae, diatoms, or amorphous detritus. Both vascular plant tissue and algae have recognizable cellular structures (e.g. rigid cell walls). We defined amorphous detritus according to Hall et al. (2000): particles with no recognizable cellular structure, which appear as discrete aggregations of subcellular-sized particles under phase-contrast microscopy. Examination of many samples indicated that diatoms comprised most of the algal community in the phytoplankton, epiphyton, and epipelon. For zygopterans, we examined the foregut contents of individual larvae. Prey items were identified either as whole organisms or by parts (e.g. chironomid head capsules). Where possible, animal parts were counted to indicate the number of prey items, e.g. two sets of chironomid mouth parts indicated two chironomids in the gut.

48 Percent dry mass of each prey type in the foregut of zygopterans was estimated by counting prey items at 100· magnification, and multiplying the numbers of each prey type by its average mass. For cladocerans and copepods, we determined mean dry mass by weighing field samples. For chironomids, we measured head capsule width with an ocular micrometer and applied Benke et al.’s (1999) regressions of dry body mass vs. capsule width. Samples for d13C analysis were rinsed with 10% HCl to remove CaCO3 precipitates, and rinsed with distilled water before being homogenized. For amphipods and chironomid larvae, gut contents were removed before isotope analysis. Isotopes were measured at the University of Wyoming with a MicroMass Isoprime continuous-flow isotoperatio mass spectrometer (MicroMass, Manchester, UK). Precision for d13C was ±0.1&.

Results Hyalella azteca was by far the dominant amphipod species in these wetlands; only a few Gammarus lacustris were found in oligosaline Lake George. Guts of H. azteca contained mostly amorphous

detritus and little macrophyte material (Fig. 1). Guts of the few G. lacustris collected contained mainly Chara spp. Non-predatory chironomids (mostly Chironomus spp. and Orthocladinae) ate mainly diatoms in oligosaline wetlands and mostly amorphous detritus in mesosaline wetlands. Despite some differences in gut contents, d13C values for chironomids and H. azteca were generally similar within the same wetland (usually within 2&, Fig. 2, Table 1). Values of d13C increased from June to August, with chironomids and amphipods showing similar patterns (except in Creighton Lake where chironomids did not change). Based on 206 prey items in 141 foreguts, zygopteran larvae ate mainly chironomid larvae, copepods, and cladocerans in all wetlands (Fig. 3a). However, because of their much larger size, chironomid larvae represented the bulk of dry mass ingested by zygopterans (Fig. 3b). We detected no snails or amphipods in zygopteran guts. There was nothing to suggest that prey preference changed with zygopteran size. Stable carbon isotopes were not useful in elucidating different prey types of zygopterans: the seasonal shift in zygopteran d13C generally tracked seasonal shifts in d13C of their composite prey (Fig. 4), but without gut-content analysis the d13C values would have wrongly suggested that amphipods were an important food.

George June

August June

Nelson August

Creighton June

Gibbs

August June August

-15

δ13C

-20

-25 Hyalella Chironomids

Figure 1. Percent volume of food items in guts of the amphipod Hyalella azteca (Hy.) and chironomid larvae (Ch., Chironomus spp. and Orthocladinae) from oligosaline wetlands George and Nelson and mesosaline wetlands Creighton and Gibbs in the Laramie Basin, Wyoming. Amorphous detritus was organic particles with no recognizable structures. Percentages are averages for early June and mid-August samples.

-30

Oligosaline

Mesosaline

Figure 2. Stable carbon isotope values (mean ± 1 SD, n ¼ 3) in early June and mid-August for the amphipod Hyalella azteca and chironomid larvae from oligosaline wetlands George and Nelson and mesosaline wetlands Creighton and Gibbs in the Laramie Basin, Wyoming.

49 Table 1. Mean d13C values (&,±1 SD, n = 3) of the amphipod Hyalella azteca, chironomid larvae (Chironomus spp. and Orthocladinae), and zygopteran larvae from oligosaline wetlands George and Nelson and mesosaline wetlands Creighton and Gibbs in the Laramie Basin, Wyoming Wetland

Hyalella

Chironomid

Zygopteran

azteca

larvae

larvae

George June August

)27.3 ± 0.3a* )28.2 ± 1.1ab )27.1 ± 0.3b* )23.4 ± 0.3a )24.7 ± 0.2 )22.9 ± 0.4a

Nelson June

)28.3 ± 1.0a* )26.1 ± 0.5b

)27.4 ± 0.9ab*

August

)26.3 ± 0.4a

)27.4 ± 0.8a

)26.1 ± 0.5a

June

)21.6 ± 0.5a

)17.5 ± 0.5

)21.2 ± 1.8a

August

)17.3 ± 0.6a

)16.7 ± 0.7a

)17.3 ± 0.5a

Creighton

Gibbs June August

)25.5 ± 0.6ab )24.1 ± 0.7ab )25.3 ± 0.2a )22.3 ± 0.3a

)22.2 ± 0.1a

)21.6 ± 0.2

Means for different taxa from the same wetland that are followed by the same letter do not differ significantly within the same month (Bonferroni pairwise comparisions, p > 0.05). Means for the same taxa sharing an asterisk do not differ significantly between wetlands during the same month.

Discussion Despite differing d13C values of invertebrates between wetlands, the similarity of most organic matter sources within the same wetland prevented conclusions about trophic pathways based solely on stable isotopes (Hart & Lovvorn, 2002, 2003). Differences in d13C between wetlands appeared to result mainly from differing availability of dissolved inorganic carbon (DIC) to primary producers, an issue beyond the scope of this paper. Moreover, invertebrate d13C values also differed between seasons within the same wetland. We have interpreted seasonally changing d13C in grazer tissues (Figs. 2 and 4) to mean that they eat mainly products of algal photosynthesis that have high turnover, and therefore track changing availability of DIC to algae (Hart & Lovvorn, 2003). Values of d13C in chironomids and Hyalella azteca were typically very similar within the same wetland (except Creighton Lake in June), even where chironomids ate mostly diatoms and H. azteca ate mostly amorphous detritus. For zygopterans, chironomids were by far the main energy source

Figure 3. (a) Percent occurrence and (b) estimated percent mass of prey items in the guts of zygopteran larvae from oligosaline wetlands George and Nelson and mesosaline wetlands Creighton and Gibbs in the Laramie Basin, Wyoming. Numbers of prey items identified were 20 for George, 41 for Nelson, 86 for Creighton, and 77 for Gibbs. Percentages are averages for early June and mid-August samples.

(Fig. 3b). These data provide insights into our three main questions about communities in these wetlands: (1) why are predatory insects more abundant in mesosaline wetlands, (2) why are chironomid larvae less abundant in oligosaline wetlands, and (3) why are amphipods uncommon in mesosaline wetlands? High biomass of predatory insects in mesosaline wetlands Both zygopteran larvae and hemipterans were more abundant in mesosaline than oligosaline wetlands (Wollheim & Lovvorn, 1995). However, hemipterans have piercing mouthparts used to suck out the insides of prey, so there are no prey

50 Oligosaline -16

George

Mesosaline

N elson

Creighton

Gi bbs

-16

Ch -18

-18

δ13C

Zyg -22

Zyg Am

Zyg Am

-22

Am

-24

Ch

δ13C

-20

-20

-24

Ch -26

-26 Ch

Zyg -28

Zoo

Am Zoo

-30

-30

Zoo

June Aug June Aug Am

-28

Zoo

June Aug June Aug

Amphipods

Zoo Zooplankton

Ch Chironomids

Zyg Zygopterans 13

Figure 4. Stable carbon isotope (d C) values in early June and mid-August for zygopteran larvae and their prey from oligosaline wetlands George and Nelson and mesosaline wetlands Creighton and Gibbs in the Laramie Basin, Wyoming.

head lengths of adult H. azteca in our study (0.8– 1.0 mm) were much larger than in Wellborn’s study (0.46–0.67 mm). Thus, while odonates can be important predators on amphipods, both gut contents and relative sizes of these taxa in our wetlands suggest that such predation was negligible. Although the diets of piercing and sucking hemipterans could not be determined by examining gut contents, it also seems unlikely that corixids and notonectids feed much on amphipods. Behavioral and serological studies indicate that chironomid larvae are a favored and primary prey, while amphipods are not eaten (Reynolds, 1975; Zalom, 1978; Reynolds & Scudder, 1987). The cuticle of amphipods appears too difficult to grasp and/or pierce. Thus, in our wetlands, macroinvertebrate predators probably had little impact on amphipods. Instead, most predation was focused on chironomid larvae, which were far more abundant in mesosaline wetlands. Higher amphipod abundance in mesosaline wetlands

remains in their guts that are visibly recognizable. Consequently, our analysis focused on zygopteran predators. The guts of zygopteran larvae contained many zooplankton (a mixture of copepods and cladocerans, Fig. 3a). However, they also ate chironomids which, being much larger, were far more important energetically (Fig. 3b). Thus, if greater zygopteran abundance in mesosaline wetlands resulted from higher prey abundance, it was probably because of increased availability of chironomids and not zooplankton. Perhaps the most important observation for zygopteran larvae was that they did not appear to feed on amphipods. We found no amphipod parts in the guts of 141 zygopteran larvae examined from both wetland types. Odonate larvae are often reported to prey intensively on amphipods (e.g. Wellborn, 1994), but almost all those examples are for anisopterans (dragonflies) as opposed to zygopterans (damselflies). For anisopteran larvae, Wellborn (1994) stated, ‘there appears to be an odonate size threshold at head width of
3.0– 4.0 mm below which odonates were essentially unable to prey on adult Hyalella’. In our study, the largest predatory odonates (emerging Enallagma spp.) had head widths of only 4.0 mm. Also, the

In the absence of fish, the above discussion precludes predation as a reason for the rarity of amphipods in mesosaline wetlands. Another hypothesis was that amphipod numbers decrease in mesosaline wetlands because of lower food supply. Emergent vegetation is rare or absent in mesosaline wetlands, perhaps resulting in lower supply of coarse particulate organic matter (CPOM) or ‘shreddable’ material. Mesosaline wetlands did contain dense stands of the submersed plant Potamogeton pectinatus, but its biomass was much lower than that of submersed beds of Chara spp. in oligosaline wetlands (Hart & Lovvorn, 2000). Moreover, the latter difference seems inconsequential because even in oligosaline wetlands, where shreddable material from emergent plants was abundant, Hyalella azteca consumed almost no plant tissue. Thus, the low abundance of H. azteca in mesosaline wetlands is probably not related to lower CPOM for food. Rather than being a shredder, H. azteca is generally considered an epiphytic grazer, and epiphyton biomass is typically lower in mesosaline than oligosaline wetlands (Hart & Lovvorn, 2000). Amorphous detritus was by far the dominant item

51 in guts of H. azteca. While this material was probably derived from algae (Hart and Lovvorn, 2003), d13C values of H. azteca were more similar to values of phytoplankton than of epiphyton in both wetland types (Hart and Lovvorn, 2003). Thus, lower amphipod populations did not appear to be linked to lower epiphyton levels in mesosaline wetlands. With no clear effects of food supply, hydrologic instability remains as the most likely reason for low amphipod numbers in mesosaline wetlands. Unlike oligosaline wetlands, mesosaline wetlands are temporary on a semi-annual basis. Because of less consistent surface inflow, their water level fluctuates regularly and usually shrinks to a fraction of maximum volume over the summer. During such events, high water temperatures, much higher salinities, increased bird predation, and perhaps low dissolved oxygen (especially during windless predawn hours) may reduce amphipod populations. Also, under winter ice cover, low oxygen levels in these shallower, saltier wetlands may reduce or prevent overwinter survival of amphipods. For H. azteca at 3 C, death occurred on average at 0.8 ppm O2 (Pennak & Rosine, 1976; see also Nebeker et al., 1992), well above levels often found under the ice in wetlands here and throughout the prairies (Barica & Mathias, 1979; Mathias & Barica, 1980; Rahel, 1990; see also Meding and Jackson, 2003). Anecdotal observations also support the hydrologic instability hypothesis for control of amphipod abundance. After we completed research for this paper, the Laramie Basin experienced a severe drought. During the peak of this drought in summer 2002, even many oligosaline wetlands completely or almost completely dried. In summer 2003, more normal water levels returned. However, sweep-net searches in 2003 revealed no amphipods in oligosaline wetlands where they were formerly abundant, even when those wetlands had not dried completely. We also note that, although amphipods are carried among wetlands on the plumage of waterbirds (Swanson, 1984), amphipods have no means of direct dispersal between isolated wetlands. They also have low production:mean biomass ratios (3.8–5.7) compared to those of chironomids (P30) (Lovvorn et al., 1999), and so multiply

relatively slowly after colonization. Thus, more frequent extreme conditions in mesosaline wetlands probably prevent large amphipod populations. Higher chironomid abundance in mesosaline wetlands An initial hypothesis was that chironomid larvae experienced less competition in mesosaline wetlands owing to decreased abundance of amphipods (cf. Batzer & Resh, 1991; Hill et al., 1992; Duffy & Harvilicz, 2001). There was no clear-cut evidence for this idea; however, in oligosaline wetlands where chironomids were far less abundant, chironomid larvae ate mainly diatoms and amphipods ate mostly amorphous detritus (Fig. 1). In mesosaline wetlands where amphipods were reduced, both taxa consumed mainly amorphous detritus. Microscopic inspection of epiphyton and epipelon samples did not suggest a difference in availability of either diatoms or amorphous detritus between wetland types. While certainly not conclusive, these results suggest the possibility of competition for food if amorphous detritus were preferred by both taxa. Upon initial consideration, it seems unlikely that amorphous detritus would be preferred over diatoms: many studies report low assimilation efficiencies for amorphous detritus (10–27% for amorphous detritus in streams, Benke & Wallace, 1997). However, we argue elsewhere (Hart & Lovvorn, 2003) that amorphous detritus in these wetlands is mainly derived from algae and may be assimilated quite efficiently (i.e. exopolymer secretions, assimilation efficiency >80%, Decho & Moriarty, 1990). In oligosaline wetlands, chironomids and amphipods consumed different foods yet had similar d13C, suggesting that their food was ultimately derived from the same source. Even if they do not compete directly for food, amphipods may still adversely impact chironomids. In these wetlands that usually have few or no fish, Hyalella azteca is highly mobile and can forage over a relatively wide area. When we observed H. azteca in aquaria, its feeding strategy could be characterized as (1) attach to the first surface encountered, (2) attempt to graze, and (3) move if the surface is unsuitable (e.g. if it moves). H. azteca often attempted to feed on the surfaces of zygopteran larvae, trichopteran cases, snail shells, and

52 chironomids before moving on. As occurs in the presence of predators on chironomids, such disturbance by amphipods might restrict chironomid movements and access to feeding sites, and possibly force them to rely more on filter-feeding from within their cases than on deposit-feeding (Ball & Baker, 1995; Koperski, 1998). Higher consumption of diatoms in oligosaline wetlands might indicate increased filter-feeding by chironomids. This evidence is only suggestive, but this potential explanation may warrant further investigation. Another perplexing observation about mesosaline wetlands is the persistence of abundant chironomid larvae in the face of abundant predatory insects that specialize on chironomids (zygopteran larvae and probably hemipterans). However, if we consider amphipods unavailable as prey, the mass ratio of invertebrate predators to chironomid larvae in oligosaline wetlands (
1.8:1) was twice that in mesosaline wetlands (
0.9:1) (Fig. 4 in Wollheim & Lovvorn, 1995). Thus, although predators were a larger fraction of total invertebrate biomass in mesosaline wetlands, they probably consumed a smaller fraction of chironomid production than in oligosaline wetlands. According to most methods of quantifying the impact of predators (Bondavalli & Ulanowicz, 1999), or the interaction strength between predators and prey (Laska & Wootton, 1998), this situation leads to lower impact of zygopterans on chironomids in mesosaline than oligosaline wetlands. Thus, the abundance of amphipods in oligosaline wetlands might have indirectly impacted chironomids by focusing intense predation on chironomid larvae (cf. Batzer et al., 2000). Supporting predation by invertebrates as the dominant reason for low chironomid abundance in oligosaline wetlands, macroinvertebrate predators have been shown to exert stronger control on prey communities than do fish (Prejs et al., 1997). In streams, predatory insects were estimated to consume 50% to >100% of the production of different prey species (Hall et al., 2000; Benke et al., 2001). These results indicate that greater predation pressure on chironomids in oligosaline wetlands could easily depress their numbers. Based on the above evidence, both competition with amphipods for food and increased predation pressure on chironomid larvae remain viable explanations for fewer chironomids in oligosaline wetlands.

Summary We began this study with three main questions about factors that structure wetland food webs in the Laramie Basin: (1) why are predatory insects more abundant in mesosaline wetlands, (2) why are chironomid larvae less abundant in oligosaline wetlands, and (3) why are amphipods uncommon in mesosaline wetlands? Our results and other published information suggest that (1) predatory zygopteran larvae and hemipterans are more abundant in mesosaline wetlands due to more abundant chironomid prey, (2) chironomids are not abundant in oligosaline wetlands because of both competition with amphipods and greater consumption by predatory insects, and (3) amphipods are uncommon in mesosaline wetlands because, lacking a resting stage and direct dispersal mechanism, they do not cope well with extremes created by hydrologic instability. It appears that amphipods play a major structuring role in these food webs, by their population response to physicochemical extremes, and by subsequent direct and indirect effects on chironomid larvae and predation by insects. These mechanisms may apply in other cases where a dichotomy has been noted between wetland types dominated by amphipods vs. chironomids (Bailey & Titman, 1984; Hargeby, 1990; Hargeby et al., 1994; van den Berg et al., 1997). Testing these mechanisms will require experimental manipulation of amphipod, chironomid, and predator populations in different types of wetlands.

Acknowledgments This research was supported by an Edward D. and Sally M. Futch Graduate Fellowship from the Institute for Wetland and Waterfowl Research of Ducks Unlimited, the Delta Waterfowl Foundation, a George E. Menkens Memorial Scholarship, and the Wyoming Water Resources Center. We thank J. M. Welker, R.M. Larson, and M. Otter for assistance with stable isotope analyses, and R.O. Hall, C. Martinez del Rio, J.S. Meyer, and W.A. Reiners for helpful comments on the manuscript.

53 References Anderson, R. S. & L. G. Raasveldt, 1974. Gammarus predation and the possible effects of Gammarus and Chaoborus feeding on the zooplankton composition in some small lakes and ponds in western Canada. Canadian Wildlife Service Occasional Paper 18. Bailey, R. O. & R. D. Titman, 1984. Habitat use and feeding ecology of postbreeding redheads. Journal of Wildlife Management 48: 1144–1155. Ball, S. L. & R. L. Baker, 1995. The non-lethal effects of predators and the influence of food availability on life history of adult Chironomus tentans (Diptera: Chironomidae). Freshwater Biology 34: 1–12. Barica, J. & J. A. Mathias, 1979. Oxygen depletion and winterkill risk in small prairie lakes under extended ice cover. Journal of the Fisheries Research Board of Canada 36: 980– 986. Batzer, D. P. & V. H. Resh, 1991. Trophic interactions among a beetle predator, a chironomid grazer, and periphyton in a seasonal wetland. Oikos 60: 251–257. Batzer, D. P., C. R. Pusateri & R. Vetter, 2000. Impacts of fish predation on marsh invertebrates: direct and indirect effects. Wetlands 20: 307–312. Benke, A. C. & J. B. Wallace, 1997. Trophic basis of production among riverine caddisflies: implications for food web analysis. Ecology 78: 1132–1145. Benke, A. C., A. D. Huryn, L. A. Smock & J. B. Wallace, 1999. Length-mass relationships for freshwater macroinvertebrates in North America with particular reference to the southeastern United States. Journal of the North American Benthological Society 18: 308–343. Benke, A. C., J. B. Wallace, J. W. Harrison & J. W. Koebel, 2001. Food web quantification using secondary production analysis: predaceous invertebrates of the snag habitat in a subtropical river. Freshwater Biology 46: 329–346. Bondavalli, C. & R. E. Ulanowicz, 1999. Unexpected effects of predators upon their prey: the case of the American alligator. Ecosystems 2: 49–63. Cannings, R. A. & G. G. E. Scudder, 1978. The littoral Chironomidae (Diptera) of saline lakes in central British Columbia. Canadian Journal of Zoology 56: 1144–1155. Cannings, R. A., S. G. Cannings & R. J. Cannings, 1980. The distribution of the genus Lestes in a saline lake series in central British Columbia, Canada (Zygoptera: Lestidae). Odonatologica 9: 19–28. Cox, R. R. & J. A. Kadlec, 1995. Dynamics of potential waterfowl foods in Great Salt Lake marshes during summer. Wetlands 15: 1–8. Decho, A. W. & D. J. W. Moriarty, 1990. Bacterial exopolymer utilization by a harpacticoid copepod: a methodology and results. Limnology and Oceanography 35: 1039–1049. Dexter, D. M., 1993. Salinity tolerance of the copepod Apocyclops dengizicus (Lepeschkin, 1900), a key food chain organism in the Salton Sea, California. Hydrobiologia 267: 203–209. Duffy, J. E. & A. M. Harvilicz, 2001. Species-specific impacts of grazing amphipods in an eelgrass-bed community. Marine Ecology Progress Series 223: 201–211.

Fry, B., 1991. Stable isotope diagrams of freshwater food webs. Ecology 72: 2293–2297. Galat, D. L., M. Coleman & R. Robinson, 1988. Experimental effects of elevated salinity on three benthic invertebrates in Pyramid Lake, Nevada. Hydrobiologia 158: 133–144. Hall, R. O., J. B. Wallace & S. L. Eggert, 2000. Organic matter flow in stream food webs with reduced detrital resource base. Ecology 81: 3445–3463. Hammer, U. T., 1978. The saline lakes of Saskatchewan III. Chemical characterization. Internationale Revue der Gesamten Hydrobiologie 63: 311–335. Hammer, U. T., 1986. Saline Lake Ecosystems of the World. Dr W Junk, Dordrecht, The Netherlands. Hammer, U. T., J. S. Sheard & J. Kranabetter, 1990. Distribution and abundance of littoral benthic fauna in Canadian prairie saline lakes. Hydrobiologia 197: 173–192. Hargeby, A., 1990. Macrophyte associated invertebrates and the effect of habitat permanence. Oikos 57: 338–346. Hargeby, A., G. Andersson, I. Blindow & S. Johansson, 1994. Trophic web structure in a shallow eutrophic lake during a dominance shift from phytoplankton to submerged macrophytes. Hydrobiologia 279/280: 83–90. Hart, E. A. & J. R. Lovvorn, 2000. Vegetation dynamics and primary production in saline, lacustrine wetlands of a Rocky Mountain basin. Aquatic Botany 66: 21–39. Hart, E. A. & J. R. Lovvorn, 2002. Interpreting stable isotopes from macroinvertebrate foodwebs in saline wetlands. Limnology and Oceanography 47: 580–584. Hart, E. A. & J. R. Lovvorn, 2003. Algal vs. macrophyte inputs to food webs of inland saline wetlands. Ecology 84: 3317– 3326. Herbst, D. B., 2001. Gradients of salinity stress, environmental stability and water chemistry as a templet for defining habitat types and physiological strategies in inland salt waters. Hydrobiologia 466: 209–219. Herbst, D. B. & T. J. Bradley, 1989. Salinity and nutrient limitations on growth of benthic algae from two alkaline salt lakes of the western Great Basin (USA). Journal of Phycology 25: 673–678. Hill, W. R., S. C. Weber & A. J. Stewart, 1992. Food limitation of two lotic grazers: quantity, quality, and size-specificity. Journal of the North American Benthological Society 11: 420–432. Horne, A. J., 1991. Selenium detoxification in wetlands by permanent flooding: I. Effects on a macroalga, an epiphytic herbivore, and an invertebrate predator in the long-term mesocosm experiment at Kesterson reservoir, California. Water, Air, and Soil Pollution 57/58: 43–52. Kelly, D. W., J. T. A. Dick & W. I. Montgomery, 2002. The functional role of Gammarus (Crustacea, Amphipoda): shredders, predators, or both? Hydrobiologia 485: 199– 203. Koperski, P.,1998. Predator–prey interactions between larval damselflies and mining larvae of Glytotendipes gripekoveni (Chironomidae): reduction in feeding activity as an induced defence. Freshwater Biology 39: 317–324. Laska, M. S. & J. T. Wootton, 1998. Theoretical concepts and empirical approaches to measuring interaction strength. Ecology 79: 461–476.

54 Lancaster, J. & G. G. E. Scudder, 1987. Aquatic Coleoptera and Hemiptera in some Canadian saline lakes: patterns in community structure. Canadian Journal of Zoology 65: 1383–1390. Lovvorn, J. R. & E. A. Hart, 2004. Irrigation, salinity, and landscape patterns of natural palustrine wetlands. In McKinstry, M. C., W. A. Hubert & S. H. Anderson (eds), Wetland and Riparian Areas of the Intermountain West: Ecology and Management. Univ. of Texas Press, Austin: 105–129. Lovvorn, J. R., W. M. Wollheim & E. A. Hart, 1999. High Plains wetlands of southeast Wyoming: salinity, vegetation, and invertebrate communities. In Batzer, D. P., R. B. Rader & S. A. Wissinger (eds), Invertebrates in Freshwater Wetlands of North America: Ecology and Management. Wiley, New York: 603–633. Mathias, J. A. & J. Barica, 1980. Factors controlling oxygen depletion in ice-covered lakes. Canadian Journal of Fisheries and Aquatic Sciences 37: 185–194. Meding, M. E. & L. J. Jackson, 2003. Biotic, chemical, and morphometric factors contributing to winter anoxia in prairie lakes. Limnology and Oceanography 48: 1633–1642. Menon, P. S., 1966. Population ecology of Gammarus lacustris lacustris Sars in Big Island Lake. Ph.D. Thesis, Univ. Alberta, Edmonton, Canada. Mihuc, T. & D. Toetz, 1994. Determination of diets of alpine aquatic insects using stable isotopes and gut contents. American Midland Naturalist 131: 146–155. Nebeker, A. V., S. T. Onjukka, D. G. Stevens, G. A. Chapman & S. E. Dominguez, 1992. Effects of low dissolved oxygen on survival, growth and reproduction of Daphnia, Hyalella and Gammarus. Environmental Toxicology and Chemistry 11: 373–379. Neill, C. & J. C. Cornwell, 1992. Stable carbon, nitrogen, and sulfur isotopes in a prairie marsh food web. Wetlands 12: 217–224. Parker, M. S. & A. W. Knight, 1992. Aquatic invertebrates inhabiting saline evaporation ponds in the southern San Joaquin Valley, California. Bulletin of the Southern California Academy of Science 91: 39–43. Peck, D. E. & J. R. Lovvorn, 2001. The importance of flood irrigation in water supply to wetlands in the Laramie Basin, Wyoming, USA. Wetlands 21: 370–378. Pennak, R. W. & W. N. Rosine, 1976. Distribution and ecology of amphipoda (Crustacea) in Colorado. American Midland Naturalist 96: 324–331. Prejs, A., P. Koperski & K. Prejs, 1997. Food-web manipulation in a small, eutrophic Lake Wirbel, Poland: the effect of replacement of key predators on epiphytic fauna. Hydrobiologia 342/343: 377–381.

Rahel, F. J., 1990. Anomalous temperature and oxygen gradients under the ice of a high-plains lake in Wyoming. Limnology and Oceanography 35: 751–755. Reynolds, J. D., 1975. Feeding in corixids (Heteroptera) of small alkaline lakes in central B.C. Verhandlungen der Internationalen Vereinigung fur Limnologie 19: 3073–3078. Reynolds, J. D. & G. G. E. Scudder, 1987. Serological evidence of realized feeding niche in Cenocorixa species (Hemiptera: Corixidae) in sympatry and allopatry. Canadian Journal of Zoology 65: 974–980. Simpson, E. P., M. R. Gonzalez, C. M. Hart & S. H. Hurlbert, 1998. Salinity and fish effects on Salton Sea microecosystems: benthos. Hydrobiologia 381: 153–177. Swanson, G. A., 1984. Dissemination of amphipods by waterfowl. Journal of Wildlife Management 48: 988–991. van den Berg, M. S., H. Coops, R. Noordhuis, J. van Schie & J. Simons, 1997. Macroinvertebrate communities in relation to submerged vegetation in two Chara-dominated lakes. Hydrobiologia 342/343: 143–150. Verschuren, D., J. Tibby, K. Sabbe & N. Roberts, 2000. Effects of depth, salinity, and substrate on the invertebrate community of a fluctuating tropical lake. Ecology 81: 164–182. Wellborn, G. A., 1994. Size-biased predation and prey life histories: a comparative study of freshwater amphipod populations. Ecology 75: 2104–2117. Wiggins, G. B., R. J. Mackay & I. M. Smith, 1980. Evolutionary and ecological strategies of animals in annual temporary pools. Archiv fur Hydrobiologie Supplement 58: 97–206. Williams, W. D., A. J. Boulton & R. G. Taaffe, 1990. Salinity as a determinant of salt lake fauna: a question of scale. Hydrobiologia 197: 257–266. Wollheim, W. M. & J. R. Lovvorn, 1995. Salinity effects on macroinvertebrate assemblages and waterbird food webs in shallow lakes of the Wyoming High Plains. Hydrobiologia 310: 207–223. Wollheim, W. M. & J. R. Lovvorn, 1996. Effects of macrophyte growth forms on invertebrate communities in saline lakes of the Wyoming High Plains. Hydrobiologia 323: 83–96. Wurtsbaugh, W. A. & T. S. Berry, 1990. Cascading effects of decreased salinity on the plankton, chemistry, and physics of the Great Salt Lake (Utah). Canadian Journal of Fisheries and Aquatic Sciences 47: 100–109. Zalom, F. 1978. Backswimmer prey selection with observations on cannibalism (Hemiptera: Notonectidae). Southwestern Naturalist 23: 617–622. Zimmer, K. D., M. A. Hanson & M. G. Butler, 2001. Effects of fathead minnow colonization and removal on a prairie wetland ecosystem. Ecosystems 4: 346–357.