What predicts the use by brook trout (Salvelinus fontinalis) of terrestrial invertebrate subsidies in headwater streams?

相似文献   

Energy availability, spatial heterogeneity and ecosystem size predict food-web structure in streams

We used standardized techniques to assemble eighteen food webs in streams. Our aim was to identify the determinants of food-web structure with particular reference to energy availability (related to land use), spatial heterogeneity and ecosystem size (both independent of land use). Forested streams displayed lower algal productivity and higher standing crops of organic matter than the grassland streams. The organic matter in the pine streams was probably of lower quality than that elsewhere. Measures of energy availability and spatial heterogeneity predicted species richness and connectance. A combination of energy availability, spatial heterogeneity and ecosystem size accounted for the representation of particular invertebrate feeding groups in the streams. Algal production and organic matter standing crop were important determinants of invertebrate biomass and overall food-web structure. Grassland sites showed a positive relationship between algal productivity and food chain length whereas forest sites displayed a positive relationship between ecosystem size and food chain length. Therefore, these results provide support for both Pimm's productivity hypothesis and Cohen and Newman's ecosystem size hypothesis.     

Production of benthic macroinvertebrate communities along a southern Appalachian river continuum

1. Annual production was estimated for macroinvertebrate communities of principle habitats along a first- to seventh-order river continuum in the southern Appalachian Mountains (U.S.A.). Annual production was relatively low in depositional habitats, pebble/gravel substrata, and on cobble devoid of plant biomass (mosses and hydrophytes). Production was greater in bedrock habitats and greatest on hydrophyte-covered cobble, with estimates reaching 364 g AFDM (ash-free dry mass)m^–2 yr^–1 in a sixth-order river reach. Annual production in depositional habitats was correlated to standing crops of benthic organic material (BOM) in low-order stream reaches but not in higher-order reaches, indicating differences in BOM availability with stream size. In cobble, pebble/gravel and bedrock habitats production was significantly correlated to standing crops of aquatic plants, which can stabilize substrata and enhance access of collector-filtering invertebrates to entrained food resources. 2. By accounting for proportional availability of habitats along the continuum, estimates of total production ranged from 5 to 154 g AFDM m^–2 yr^–1, and increased significantly with stream size. Annual production estimated for sixth- and seventh-order reaches of the continuum were amongst the highest reported thus far for lotic systems. Organization of the benthic community along the continuum, based on production estimates for individual functional feeding-groups, generally supported predictions of the River Continuum Concept (RCC): shredder contributions were greatest in low-order reaches and declined downstream; scraper percentages were greatest in the middle of the continuum; collector-filterer contributions increased with increasing stream size. Longitudinal trends for collector-gatherers and predators did not support RCC predictions; these groups appeared to be influenced by localized changes in habitat availability and occurrence of vertebrate predators along the continuum.     

Macroinvertebrate diversity in headwater streams: a review

1. Headwater streams are ubiquitous in the landscape and are important sources of water, sediments and biota for downstream reaches. They are critical sites for organic matter processing and nutrient cycling, and may be vital for maintaining the 'health' of whole river networks.
2. Macroinvertebrates are an important component of biodiversity in stream ecosystems and studies of macroinvertebrate diversity in headwater streams have mostly viewed stream systems as linear reaches rather than as networks, although the latter may be more appropriate to the study of diversity patterns in headwater systems.
3. Studies of macroinvertebrate diversity in headwater streams from around the world illustrated that taxonomic richness is highly variable among continents and regions, and studies addressing longitudinal changes in taxonomic richness of macroinvertebrates generally found highest richness in mid-order streams.
4. When stream systems are viewed as networks at the landscape-scale, α-diversity may be low in individual headwater streams but high β-diversity among headwater streams within catchments and among catchments may generate high γ-diversity.
5. Differing ability and opportunity for dispersal of macroinvertebrates, great physical habitat heterogeneity in headwater streams, and a wide range in local environmental conditions may all contribute to high β-diversity among headwater streams both within and among catchments.
6. Moving beyond linear conceptual models of stream ecosystems to consider the role that spatial structure of river networks might play in determining diversity patterns at the landscape scale is a promising avenue for future research.     

Role of habitat in determining macroinvertebrate production in an intermittent-stream system

1. The effect of channel drying on macroinvertebrate production was studied at the habitat and reach scale in a catchment drained by intermittent streams in Maine, U.S.A. The catchment includes two first‐order streams and their second‐order confluence. Six reaches were selected for study based on differences in channel slope and habitat cover (bedrock, riffle/run, debris dam and pool). Stream water in each reach was acidic and oligotrophic. 2. The study reaches had different degrees of channel drying. In the first‐order reaches, surface flow ceased earlier in the season and for longer periods than second‐order reaches. Regardless of reach, pool and debris dam habitats retained water longer than riffle/runs and bedrock. Unlike other habitats, debris dams retained moisture for relatively long periods following cessation of surface flow. 3. Reach‐specific macroinvertebrate production ranged from approximately 1.7 to 2.9 g AFDM m⁻² year⁻¹ which are among the lowest values ever reported. Habitat‐specific production ranged from approximately 0.5 to 5.0 g AFDM m⁻² year⁻¹ (bedrock and debris dams, respectively). 4. At the reach scale, quantities of stored benthic organic matter (range approximately 200–600 g AFDM⁻²) decreased in a downstream direction. 5. A combination of differences in the timing and duration of channel drying, habitat structure and detritus standing stocks appeared to influence levels of invertebrate production among the study reaches. 6. Our interpretation of a canonical correspondence analysis indicates that drying is more important than habitat in affecting macroinvertebrate production in this intermittent stream system.     

Stream community structure in relation to spatial and temporal variation: a habitat templet study of two contrasting New Zealand streams

• 1 The physical characteristics of two contrasting streams, and habitat types within these streams, are described in terms of a two-dimensional physical habitat templet in which disturbance frequency and the availability of spatial refugia are the temporal and spatial axes.
• 2 It is predicted that habitats experiencing a high disturbance frequency and low refuge availability will be characterized by a low invertebrate species diversity, a low biomass of epilithic algae and particulate organic matter and a community made up of mobile, weedy species. Habitats having a low disturbance frequency and high refuge availability will be characterized by a diverse community containing sedentary and specialist species, with high algal and particulate organic matter levels.
• 3 A lower median substrate particle size and higher shear stress regime in Timber Creek were indicative of a higher disturbance frequency than in the Kyeburn. Substrate diversity was lower in Timber Creek than in the Kyeburn and indicated that the availability of refugia was lower in Timber Creek. In both streams, pools were found to have a higher disturbance frequency and lower availability of refugia than riffles.
• 4 Invertebrate species diversity, the biomass of epilithic algae and particulate organic matter and the representation of sedentary species, filter feeders and shredders were higher in the more temporally stable and spatially heterogeneous Kyeburn. The community of Timber Creek, frequently disturbed and having low refuge availability, had a high proportion of mobile and weedy species, with the highly mobile, generalist-feeding Deleatidium spp. (Ephemeroptera; Leptophlebiidae) being the most dominant organisms.
• 5 The predictions made about stream community structure and species characteristics in relation to disturbance frequency and the availability of spatial refugia are generally supported. Now a larger scale investigation is required to test the generality of the predictions. We conclude that the habitat templet approach offers a sound framework within which to pose questions in stream ecology.

     

Effects of forest fire on headwater stream macroinvertebrate communities in eastern Washington,U.S.A.

1. Recent increases in fire frequency in North America have focused interest on potential effects on adjacent ecosystems, including streams. Headwaters could be particularly affected because of their high connectivity to riparian and downstream aquatic ecosystems through aquatic invertebrate drift and emergence. 2. Headwater streams from replicated burned and control catchments were sampled in 2 years following an intense forest fire in northeastern Washington (U.S.A.). We compared differences in benthic, drift and emergent macroinvertebrate density, biomass and community composition between five burned and five unburned catchments (14–135 ha). 3. There were significantly higher macroinvertebrate densities in burned than control sites for all sample types. Macroinvertebrate biomass was greater at burned sites only from emergence samples; in benthic and drift samples there was no significant difference between burn and control sites. 4. For all sample types, diversity was lower in the burned catchments, and the macroinvertebrate community was dominated by chironomid midges. 5. Compared to the effects of fire in less disturbed ecosystems, this study illustrated that forest fire in a managed forest may have greater effects on headwater macroinvertebrate communities, influencing prey flow to adjacent terrestrial and downstream aquatic habitats for at least the first 2 years post‐fire.     

Genetic signatures of translocations and habitat fragmentation for two evolutionarily significant units of a protected fish species

Genetic population structure was evaluated for the White Sands pupfish (Cyprinodon tularosa), a protected fish species comprised of two Evolutionarily Significant Units (ESUs); the Malpais Spring ESU and the Salt Creek ESU. The Malpais Spring ESU is restricted to Malpais Spring, whereas the Salt Creek ESU includes the native Salt Creek population and two Salt Creek-derived populations at Mound Spring and Lost River; all three of these habitats are physically fragmented. We sampled the upper and lower reaches of the four populations, examining 13 DNA microsatellite loci from 40 individuals per population. As expected, significant genetic structure was observed between the two ESUs; Malpais Spring and Salt Creek. Substantial genetic drift was observed for the introduced Lost River population, with modest genetic drift for the introduced Mound Spring population. Taken together with ecological data, neither of the introduced populations successfully replicates the Salt Creek population. We also report significant reductions in genetic diversity for the upper reaches of both Salt Creek and Lost River, indicating that recent habitat changes have altered the genetic structure of these two populations. We consider these findings along with previously reported ecological data to develop guidelines for managing C. tularosa.     

Quantifying the trophic base for benthic secondary production in the Nakdong River estuary of Korea using stable C and N isotopes

The overall dependence of benthic secondary production on the main primary producers at three different habitats in the Nakdong River estuarine system, Korea, was estimated. Inventories of macrobenthic invertebrate biomass were combined with multiple-isotope-mixing models to evaluate the trophic base, comparing Scirpus triqueter-dominated and Phragmites australis-dominated marshes and bare intertidal flat. The feasible contributions of four main food sources, marsh macrophytes, the microphytobenthos, and riverine and marine suspended particulate organic matter (RPOM and MPOM), to the consumer biomasses were calculated using the isotopic mixing model. After weighting the feasible contributions of food sources to each taxon by the consumer biomass, the resultant values were summed for all the consumers at each habitat to quantify the trophic base of the benthic invertebrate community. Dual-isotope-mixing model calculations verified the varying dependence on those potential food sources among the functional feeding groups. In addition, the dependence on each source of the same functional group varied between bare intertidal and salt-marsh habitats, shifting from a dominance of benthic and pelagic microalgal sources on the former habitat to a mixed food source at the latter habitat. The biomasses of the species comprising each functional group differed among habitats and sampling dates, so that each functional group made a different contribution to the whole benthic community and its basal food source. Given the calculation of the overall dependence of macrozoobenthic community on each food source, our results indicate that the microphytobenthic source dominates (nearly half) the trophic base in all the intertidal habitats of different vegetational compositions. Marsh-macrophyte-derived organic matter and RPOM served as considerable subsidies only to salt-marsh food webs, reflecting the use of the mixed food source by salt-marsh-bed consumers. Conversely, the dominance of MPOM in the total food base was equal to that of the microphytobenthos in the bare intertidal ecosystem but increased during spring−summer in the salt-marsh systems. Our results also suggest that the river discharge concentrated during the summer monsoon does not lead to any shift in trophic base for estuarine secondary production.     

Responses in organic matter accumulation and processing to an experimental wood addition in three headwater streams

1. Additions of large wood are being used to restore streams that have been subjected to channelization, wood removal or riparian timber harvest. This added wood potentially increases channel stability, habitat complexity and organic matter retention and improves habitat and productivity of higher trophic levels. However, few stream restorations monitor restoration effectiveness after project completion. 2. We added 25 aspen logs (each 2.5 m length × 0.5 m diameter) to 100‐m reaches of each of three forested headwater streams in the Upper Peninsula of Michigan, U.S.A. These wood‐poor streams drain forests that were completely harvested of timber over a century ago and have been selectively logged for the past 50–60 years. An upstream unmanipulateds 100‐m reach in each stream served as the control. 3. We evaluated responses in organic matter processing by measuring red maple leaf decomposition 1 year before and 2 years after wood addition. We also quantified coarse organic matter standing stocks in the main channel and in debris accumulations associated with large wood. In response to wood addition, we predicted both organic matter standing stocks and leaf decomposition rates would increase, thereby enhancing resource availability to higher trophic levels. 4. Leaf decomposition rates did not change following wood addition. Temporal variation in rates among streams was mostly explained by differences in degree days, water velocity, scour/burial and water column inorganic nitrogen concentrations, but not large wood. Variation within streams across years was explained by differences in degree days, water velocity and shredder biomass. 5. Contrary to our prediction, organic matter standing stocks did not increase significantly at the reach scale. However, the experimentally added wood retained c. 4% of total annual coarse benthic organic matter (CBOM) in the first year and an additional c. 15% in the second year, suggesting accumulation over time in the manipulated reaches. The CBOM held by the new logs may be more biologically available because it is less susceptible to burial and transport than material in the streambed. 6. Some shredding macroinvertebrates responded to changes caused by the wood additions. In particular, the common caddisfly shredder, Lepidostoma sp., increased in abundance in leaf bags following wood addition, whereas the biomass of the winter stoneflies, Capniidae, declined in the first year. 7. Considerable funds are spent to restore in‐stream habitat, but few restorations are monitored, particularly over long periods (>5 years). Our results show that longer‐term monitoring is needed to determine the efficacy of these restorations on ecosystem function; organic matter decomposition in our low‐gradient streams did not respond to a substantial increase in large wood after 2 years.     

Spatial and temporal variability of macroinvertebrates in spawning and non-spawning habitats during a salmon run in Southeast Alaska

Spawning salmon create patches of disturbance through redd digging which can reduce macroinvertebrate abundance and biomass in spawning habitat. We asked whether displaced invertebrates use non-spawning habitats as refugia in streams. Our study explored how the spatial and temporal distribution of macroinvertebrates changed during a pink salmon (Oncorhynchus gorbuscha) spawning run and compared macroinvertebrates in spawning (riffle) and non-spawning (refugia) habitats in an Alaskan stream. Potential refugia included: pools, stream margins and the hyporheic zone, and we also sampled invertebrate drift. We predicted that macroinvertebrates would decline in riffles and increase in drift and refugia habitats during salmon spawning. We observed a reduction in the density, biomass and taxonomic richness of macroinvertebrates in riffles during spawning. There was no change in pool and margin invertebrate communities, except insect biomass declined in pools during the spawning period. Macroinvertebrate density was greater in the hyporheic zone and macroinvertebrate density and richness increased in the drift during spawning. We observed significant invertebrate declines within spawning habitat; however in non-spawning habitat, there were less pronounced changes in invertebrate density and richness. The results observed may be due to spawning-related disturbances, insect phenology, or other variables. We propose that certain in-stream habitats could be important for the persistence of macroinvertebrates during salmon spawning in a Southeast Alaskan stream.     

Respiration and annual fungal production associated with decomposing leaf litter in two streams

1. We compared fungal biomass, production and microbial respiration associated with decomposing leaves in one softwater stream (Payne Creek) and one hardwater stream (Lindsey Spring Branch). 2. Both streams received similar annual leaf litter fall (478–492 g m^?2), but Lindsey Spring Branch had higher average monthly standing crop of leaf litter (69 ± 24 g m^?2; mean ± SE) than Payne Creek (39 ± 9 g m^?2). 3. Leaves sampled from Lindsey Spring Branch contained a higher mean concentration of fungal biomass (71 ± 11 mg g^?1) than those from Payne Creek (54 ± 8 mg g^?1). Maximum spore concentrations in the water of Lindsay Spring Branch were also higher than those in Payne Creek. These results agreed with litterbag studies of red maple (Acer rubrum) leaves, which decomposed faster (decay rate of 0.014 versus 0.004 day^?1), exhibited higher maximum fungal biomass and had higher rates of fungal sporulation in Lindsey Spring Branch than in Payne Creek. 4. Rates of fungal production and respiration per g leaf were similar in the two streams, although rates of fungal production and respiration per square metre were higher in Lindsey Spring Branch than in Payne Creek because of the differences in leaf litter standing crop. 5. Annual fungal production was 16 ± 6 g m^?2 (mean ± 95% CI) in Payne Creek and 46 ± 25 g m^?2 in Lindsey Spring Branch. Measurements were taken through the autumn of 2 years to obtain an indication of inter‐year variability. Fungal production during October to January of the 2 years varied between 3 and 6 g m^?2 in Payne Creek and 7–27 g m^?2 in Lindsey Spring Branch. 6. Partial organic matter budgets constructed for both streams indicated that 3 ± 1% of leaf litter fall went into fungal production and 7 ± 2% was lost as respiration in Payne Creek. In Lindsey Spring Branch, fungal production accounted for 10 ± 5% of leaf litter fall and microbial respiration for 13 ± 9%.     

Effects of fish predation and habitat type on stream benthic communities

The influence of habitat on interactions between a fish predator (brown trout Salmo trutta) and a benthic invertebrate community was studied in nine field enclosures (8 ×3 m) in a creek in southern Sweden. Three habitat treatments were tested, a shallow sandy habitat, a deep habitat containing a mixture of large and small cobbles and a moderately deep habitat with large cobbles. The one month-long experiment showed that there were no major differences in the abundance and biomass of the benthic macroinvertebrate fauna among these habitats as no functional groups of invertebrates and only a few taxa differed between treatments. Invertebrate drift rates decreased over time, which was probably related to seasonal changes in invertebrate life cycles or to effects of predation independent of habitat type, as there was no difference between treatments.     

Correlates of habitat favourability for benthic macroinvertebrates at five stream sites in an Appalachian Mountain drainage basin, U.S.A.

SUMMARY.

• 1 Based on monthly samples taken over a 1-year period, average density (individuals m^-2). average standing biomass and annual production of benthic macroinvertebrates were estimated at five sites within an Appalachian Mountain drainage basin. Two sites were on first order streams and differed from the three second order sites: they were smaller and more shallow and they were depressed in pH and chemical richness.
• 2 Patterns of abundance of individual taxa, of higher taxonomic groups and of functional (feeding) groups differed according to whether abundance was measured as density, as standing biomass or as annual production. Standing biomass was chosen as the measure of macroinvertebrate abundance because available evidence indicates that only standing biomass is consistently, positively correlated with survivorship, and thus with habitat favourability.
• 3 Two non-insect taxa (the crayfish Cambarus and the snail Leptoxis carinata) dominated standing biomass at each site. Consequently, differences among sites in total macroinvertebrate standing biomass and differences within and among sites in standing biomass of functional groups were determined by differences in estimated standing biomass of these two taxa. Differences in estimates of crayfish standing biomass were consistent with an explanation based on the availability of refuges created by large substrate particles. The abundance of L. carinata appeared to be controlled primarily by water chemistry and possibly secondarily by predators. A number of insect taxa exhibited patterns of standing biomass consistent with hypotheses based on effects of annual depth-flow regimes. Hypotheses based on differences in food resource and on competition appeared, in general, to be inconsistent with observed patterns of macroinvertebrate abundance.
• 4 Contrary to predictions of the River Continuum Concept, the shredder functional group in the Guys Run drainage and in other temperate woodland streams was found to be a minor part of total macroinvertebrate standing biomass. Further, in a majority of small forested stream sites studied to date, standing biomass of grazers has been determined to be greater than that of shredders.

     

Relationships between taxonomic and functional components of diversity: implications for conservation of tropical freshwater fishes

相似文献   

Habitat‐specific invertebrate responses to hydrological variability,anthropogenic flow alterations,and hydraulic conditions

相似文献   

Ecological function of constructed perennial stream channels on reclaimed surface coal mines

Mountaintop removal–valley fill mining results in the conversion of steep, forested headwater catchments to low gradient and open canopy channels. We compared the ecological functions of five reference stream channels to five constructed channels (age ranging from 3 to 20 years) on reclaimed mines in southern West Virginia. Variables included stream flow, habitat, water chemistry, riparian vegetation, organic matter (OM) processing, and invertebrate and amphibian communities. Although dissolved metal concentrations remained low, constructed channels produced significantly higher levels of conductivity and total dissolved solids as compared to reference streams. Macroinvertebrate and amphibian richness were comparable between constructed and reference channels; however, there was a distinct shift from sensitive lotic taxa in reference channels to tolerant lentic taxa in constructed channels. Constructed channels also had reduced OM decomposition rates. Nevertheless, constructed channels had significantly higher OM retention than reference channels, and consequently exhibited significantly higher overall OM processing and higher dissolved carbon concentrations. As the time since reclamation increased, we observed slight declines in conductivity and significant increases in total invertebrate richness. Our results provide measures of functional equivalencies between reference and constructed streams, which can serve as a basis for informed permitting and mitigation decisions in mined watersheds.     

Migration patterns and functional groups of spiders in a desert agroecosystem

Abstract 1. Arthropods living in annual crops suffer mortality caused by agricultural practices. Therefore, migration from surrounding habitats is crucial to maintain populations of natural enemies of insect pests in crops. In desert agroecosystems there is a pronounced contrast between managed and unmanaged habitats, where irrigated and fertilised crops are islands of productivity in an arid matrix. This contrast could either enhance or inhibit movement of natural enemies between the landscape components. 2. The importance of the surrounding arid habitats as a source for spiders in crops was examined in the Negev desert of Israel. Spiders were sampled in both arid natural habitat and adjacent wheat fields using pitfall traps and visual searching. In addition, spiders in wheat fields were sampled throughout the winter cropping season using emergence traps at increasing distances from the field edge. Stationary and movable emergence traps were used to distinguish between residents and migrant species. 3. The spider assemblage in the wheat was dominated by three families: Linyphiidae, Theridiidae, and Gnaphosidae. Spider sampling in both natural arid habitat and adjacent wheat fields enabled four functional groups to be recognised that differed in habitat preference, movement patterns, and population dynamics. Thirty‐three per cent of collected individuals were classified as crop residents whereas more than 50% were classified as migrants from the surrounding habitats. These findings suggest that the surrounding habitats influence spider assemblage composition in the fields, in spite of the marked contrast in habitat structure and productivity. 4. Spider assemblages in the wheat fields were dominated by migrant species arriving from the surrounding arid habitats. Migrant spiders inhabited the crop throughout the cropping season. The combined contribution of resident and migrant functional groups may act to prevent insect pest outbreaks in this desert agroecosystem.     

Do invertebrate communities in floodplains change predictably along a river's length?

1. Invertebrate assemblages were described for nine floodplain sites located on a longitudinal gradient of river discharge in the Altamaha River catchment. The Altamaha River and its tributaries constitute one of the few remaining ‘unregulated’ catchments in the southeastern U.S. We predicted that, as the character of lateral flood pulses into backwater swamps changed along the discharge gradient, so would the structure of invertebrate communities. We also examined the relationship between invertebrate assemblages and physicochemical factors (degree of floodplain inundation, pH, conductivity and nutrient concentrations). 2. Cluster analyses of both invertebrate abundance and biomass separated the nine sites into three groups corresponding to their positions in the catchment (upper, mid‐ and lower reach clusters). Non‐metric multidimensional scaling ordinations further corroborated the groupings (with combined axis scores of 92% and 73% for abundance and biomass, respectively) and showed significant correlations with degree of inundation and conductivity (abundance), and conductivity, nitrate and phosphate concentrations (biomass). 3. Floodplains in the upper reaches were dominated by terrestrial taxa, such as earthworms, oribatid mites, collembolans and assorted terrestrial fly larvae, and some rapidly developing aquatics (harpacticoid crustaceans and mosquitoes). In the mid‐reach, the dominant taxa were longer lived aquatic organisms such as mayflies and aquatic oligochaetes, although some terrestrial organisms (elaterid beetles and mites) were still common. Invertebrate families dependent on water flow, such as riffle beetles and some mayflies, were common only in mid‐reach floodplain sites. Lower reaches were dominated by lentic aquatic taxa such as dytiscid beetles and asellid isopods, which commonly persist in wetlands after they dry. 4. Our study indicates that invertebrate community structure varies predictably among floodplains in the Altamaha catchment, with headwater habitats being dominated by terrestrial and rapidly‐developing aquatic invertebrates, mid‐reaches characterised by an influx of invertebrates from the river and lower reaches being dominated by wetland taxa with desiccation‐resistant stages. This spatial variability should be considered when applying the Flood Pulse Concept.     

Invertebrate assemblages in relation to habitat types on a floating mat in Mizorogaike Pond, Kyoto, Japan

Abiotic environmental variables and invertebrate assemblages were compared among four habitat types (bare hollow, sphagnum-rich hollow, pool, and mat edge) on a floating mat in Mizorogaike Pond, Kyoto. We found differences in abiotic environments between two hollows and two inundated habitats (pool and mat edge); pH was significantly lower in hollow habitats than in inundated habitats, and water depths were significantly shallower in sphagnum-rich hollows than in inundated sites. The composition of invertebrate assemblages in the hollow was distinct from that in the inundated habitats. The abundances of some dominant invertebrate taxa or functional feeding groups on the floating mat differed between the hollows and inundated habitats, and were correlated with water temperature, pH and depth. These results indicate that habitat heterogeneity created by the coexistence of hollows and inundated habitats contributes to species diversity on the floating mat in Mizorogaike Pond. A comparison of the pH values in different wetlands revealed that both bog- and fen-specific components coexist within this system. In order to adequately manage and conserve peatland ecosystems, it is necessary to consider the importance and vulnerability of both hollows and inundated habitats in peatlands.