The Vikramshila Gangetic Dolphin Sanctuary (VGDS) covers a 50 km stretch of the main channel of the Ganga River and is protected for the endangered freshwater dolphin (Platanista gangetica). We studied spatio-temporal patterns of zooplankton community structure during January through December 2009, at three sites. A total of 35 species including rotifers, cladocerans and copepods were identified. Integrating all samples, the top five dominant zooplankton included copepodids, copepod nauplii, B. caudatus, Ceriodaphnia rigaudi and Mesocyclops leuckarti. The zooplankton community was represented by diverse feeding groups including herbivores, carnivores, detritivores and bacterivores. The hierarchical dendrogram revealed distinct spatiotemporal variations. The Sultanganj community was separated at first hierarchical level, whereas species assemblage of August–October at Kahalgaon was distinguished at second hierarchical level. Top three indicative species at site-I (Sultanganj) were C. reticulata, Daphnia pulex and B. quadridentatus throughout the year. During monsoon the top three indicator species were C. rigaudi, M. leuckarti and B. falcatus at Bhagalpur and copepod developmental stages, B. caudatus and Keratella tropica at Kahalgaon. Adult Mesocyclops sp., copepod developmental stages and D. carinata were dominant in winter, whereas, Moina sp., C. rigaudi and M. leuckarti were the top three indicator species in summer at Bhagalpur and Kahalgaon. Of all the sites studied, Kahalgaon was recorded as having higher species richness.
The River Ganga provides the major source of water for drinking, irrigation, navigation, generating electricity, and is a vast reservoir of aquatic biota including the Gangetic Dolphins, the national aquatic animal of India. The Gangetic Dolphin (Platanista gangetica) is listed in Appendix I and II of the Convention on the Conservation of Migratory Species of Wild Animals (CMS). Endemism in the Ganga ecosystem is unusually high (De Silva et al., 2007; Singh and Singh, 2007) including several species of annelids, molluscs, fishes etc. Knowledge about the species richness in this ecosystem is more complete for vertebrates than for invertebrates. Aquatic invertebrates, mainly zooplankton, constitute the food base for larger invertebrates and particularly their developing stages and are an important driver of community structures in aquatic ecosystems (Cohen et al., 2003; Roy, 2007; Nautiyal, 2010). Being bacterivorous and grazers of algae and constituting preferred food for macro-invertebrates and fishes, zooplankton plays a central role in the carbon transfer, biogeochemical cycle and shaping of the overall community structure.
Despite the well-documented role of zooplankton in the transfer of carbon and energy and the ichthyofaunal abundance, few studies have dealt with the zooplankton assemblages in the Ganga ecosystems (Singh and Singh, 2007). Studies on the zooplankton community structure in riverine ecosystems offer opportunities to investigate patterns of responses to recurring variations and episodic disturbances (Hwang et al., 2010). The understanding of plankton dynamics in rivers can also be useful to evaluate the resilience capacity of the ecosystem, which exhibits remarkable variations spatially (Rai et al., 2010; Vass et al., 2010).
Only anecdotal information is available on the zooplankton community of the River Ganga (Bilgrami, 1991; Bilgrami et al., 1993; Battish, 1992; Pandey, 2011). The objectives of the present study are to pinpoint the taxonomic diversity, species richness trends and standing crop of the zooplankton in Vikramshila Gangetic Dolphin Sanctuary (VGDS), the main channel of the River Ganga in eastern Bihar.
Considering the extraordinary resilience of riverine ecosystems and the role of zooplankton as a food base for several organisms, the analyses of zooplankton community assemblages in the Ganga ecosystem in general and in the VGDS in particular is essential to understand its ecosystem health. The middle and lower stretch of Ganges water is affected by higher organic input; consequently, higher bacterial and picoplanktonic production is observed. Therefore through the present time series study of the dynamics of zooplankton communities of the Ganga ecosystem, we tested the hypothesis that (i) the planktonic community in this ecosystem is dominated by the pico- and nanophagous organisms such as rotifers, and (ii) they form a link between the classical (algae-zooplankton-fish) and microbial food web (bacteria-ciliate- zooplankton).
Material and Methods
A 50 km segment of the main channel of the Ganga River between Sultanganj (25.246°N 86.74°E, 1731ft above sea-level) and Kahlgaon (25.27°N; 87.22°E; 52ft above sea-level) in Bihar was declared a Gangetic Dolphin sanctuary in 1991 and named Vikramshila Gangetic Dolphin Sanctuary with an objective of conserving the endangered freshwater dolphin (Platanista gangetica). The average number of dolphins recorded in VGDS was 119 ± 31.8 (SD) with an encounter rate of 1.8 dolphins km−1 (Choudhary et al., 2006). Large groups of ≤13 individuals of dolphins are often sighted in counter-current pools induced by pilings of the Vikramshila Bridge, just downstream from Bhagalpur, and the three monadnocks of Kahalgaon. The Patna High Court (C.J.W.C. No. 5628) directed the governmental agencies to allocate funds for supporting dolphin conservation efforts in Bihar in 2001. In 2004, a management committee for the VGDS was constituted by the State Wildlife Board of Bihar. Though there has been a decline in intentional killing of dolphins and other aquatic wildlife with increased patrolling and people's participation, the population size has not improved so far (Choudhary et al., 2006).
Besides dolphin, this sanctuary is the preferred habitat of a rich diversity of other threatened aquatic wildlife; for instance, the Indian Smooth-Coated Otter, the Gharial, freshwater turtles and water fowl etc. This stretch of the River is also the major source of riverine spawn. The water mass and zooplankton composition in this stretch are mainly influenced by the upstream flow and discharge from various tributaries and sewage fed canals. Additionally dyes of washeries and bleaching industries etc. also affect the water quality in this region.
Water levels rise as much as 10 m and the main channel widens to 5–6 km during the monsoon season. Monadnocks or rock islands induce large countercurrent pools, the primary habitat of river dolphins, at Sultanganj and Kahalgaon respectively. The three sampling sites (Figure 1) are located in the Bhagalpur District of Bihar state and are upstream boundary (western) middle and downstream (eastern) boundary of the core area of the VGDS, the only protected area for the endangered Gangetic dolphins in the world. The three sampling sites are affected by differential anthropogenic pressure and resultantly show different limno-ecological attributes. The upstream stations at Sultanganj (elevation 173 ft above sea-level) are mainly affected by erosion, flood-plain farming, over-fishing, and heavy load of pilgrims (religious bathing) the Bhagalpur stations (elevation 134 ft above sea level) are affected by Champa Nala (a sewage fed canal), dyes of washeries and bleaching industries, and the downstream stations at Kahlgaon (elevation 52 ft above sea level) are mainly affected by the discharge from Super Thermal Power Station and over-fishing. The baseline limnological data for all three stations during summer, rainy and winter seasons are provided in Table S1 (see the online supplementary information).
Collection and processing
Zooplankton samples were collected through a conical plankton net (mouth diameter, 15 cms and 25 cm long) with a mesh size 35 μm. The lower end of the plankton net was tightly fitted with the specimen tube of 30 ml capacity. Known volume of river water (150 to 200 l) was filtered manually through the plankton net and zooplankton samples were preserved in 3% (w/v) buffered formalin on board in a glass bottle (vol 50 ml) immediately after collection. Five samples were taken from 0.5 m deep to surface at each site. Three sampling stations were selected to provide a spatial characterization of the sanctuary (Figure 1). The field work was carried out during 12 monthly expeditions between January and December 2009.
Zooplankton was identified to the lowest possible taxon using standard keys (Sehgal, 1983; Edmondson, 1992; Battish, 1992; Sharma, 1992; Sharma and Michael, 1988; Dodson and Frey, 1991; Dumont, 1994; Fernando, 2002 and other revisional references). Most organisms were counted at species level using sub-samples.
The zooplankton community was analysed on the basis of feeding modes and trophic relationship. Different fractions of zooplankton representing bacterivorous, detritivorous, herbivorous and carnivorous were estimated. Zooplankton fraction capable of feeding at more than one trophic level is considered as omnivorous (Pimm, 1982). Many species historically considered to be strictly herbivorous are reported to feed on heterotrophic and autotrophic food of comparable size range (Rothaupt, 1990; Kumar and Rao, 1999). Therefore, we have taken into account the propensity of a species to particular feeding mode. Feeding mode based assembling of species has been done following Koste (1978), Bogdan and Gilbert (1982), Rothaupt (1990) and Arndt (1993) for rotifers; Dodson and Frey (1991), Frey (1982), Jack and Gilbert (1993a,b), Dumont (1994), Fernando (2002) and Kumar (2004) for cladocerans and copepods.
Spatial and temporal variations in the zooplankton assemblage were analyzed using two factor analysis of variance without replication, with stations and seasons as major factors. In order to reduce higher heteroscedasticity a 4th root transformation was done for abundance data prior to statistical analyses.
To identify the occurrence of particular zooplankton associations and the patterns in community structure, we used cluster analysis, ordinations, and indicator species analysis (ISA). “Species density” –“sample date” matrices composed of samples (36) and species (35) were generated for multivariate analyses and similarity coefficients between samples were computed using the Bray-Curtis similarity coefficient and clustering strategy of flexible links (Clarke and Gorley, 2001). For presentation, the cluster dendrogram is scaled by percentage of information remaining.
The species characterizing each cluster were further identified using the Indicator Value Index (IndVal). The details of indicator species analyses are explained in Kumar et al. (2013). All analyses have been performed using Paleontological Statistics (PAST) software (Hammer et al. 2001).
Integrating all samples from the three stations a total of 35 zooplankton species including rotifers (15), cladocerans (12) and copepods (8) were identified during the study period. The average value of total zooplankton abundance (100.5 ± 35.8 Ind l−1) was lowest at Sultanganj followed by Kahalgaon (187.08 ± 71.52 Ind l−1) and Bhagalpur (231.58 ± 44.04 Ind l−1). A similar trend was recorded for species richness (Table S2 in the SI). Species richness, indices of Shannon-Wiener diversity, and Pielou's evenness, average abundance, relative abundance (RA,%), and occurrence ratio (OR,%) recorded at each sampling location for all samples are listed in Table S2. Time series (monthly) values from January to December 2009 for average abundance, species richness and percent contribution of rotifers, cladocerans and copepods to total zooplankton abundance at three sampling stations are provided in Figures 2, 3 and 4. In terms of abundance and species richness (Figures 2a, 3a, 4a), Rotifera was the largest group contributing 30–60% to the total zooplankton abundance. followed by Cladocera at Sultanganj and Copepoda at Bhagalpur.
Zooplankton community structure revealed differential limnoecologcal attributes of upstream boundary middle and downstream boundary of VGDS. Zooplankton community was represented diverse feeding groups bacterivores, detritivores, herbivores, and carnivores. Various fractions of zooplankton community representing different trophic levels (feeding modes) are shown in Figures 2c, 3c and 4c, respectively, for Sultanganj, Bhagalpur and Kahalgaon. The community is dominated by omnivores followed by bacterivores in rotifers, detritivores in cladocerans and carnivores in copepods. A distinct inverse relationship was observed between copepods and rotifers, at Bahgalpur and Kahalgaon but at Sultanganj rotifers correlated negatively with cladocerans but not with copepods (Figure 2).
The nonmetric multidimensional scaling (NMDS) (Figure S1) segregated the Sultanganj zooplankton community from Bhagalpur and Kahalgaon. The top three indicative species at Sultanganj samples were Ceriodaphnia reticulata, Daphnia pulex and Brachionus quadridentatus (Table S3).
Significantly higher species diversity and density were recorded at Bhagalpur (Figure S1, p< 0.01; Table S2) whereas Sultanganj recorded lesser abundance and diversity of zooplankton (Table S2; Figure S2). The highest richness index values were recorded at Kahalgaon (Figure S1, p < 0.01; Table S2), which was mainly contributed by copepods. In contrast rotifers and cladocerans were significantly higher at Sultanganj (Figure 2). Copepods were captured in significantly higher abundance at Bhagalpur (Figure 3; Table S2). Rotifer abundance did not differ significantly among stations however, spatial variations for copepod abundance were highly significant (p < 0.02; Tukey's test; Figure S1). The lowest copepod abundance was recorded at Sultanganj whereas the highest abundance was at Bhagalpur (middle of the sanctuary). The Bray Curtis similarity based cluster analyses separated samples into five clusters based on stations and seasons. Zooplankton community at Sultanganj (Figure S3; Cluster IB) were entirely segregated from other samples at the first hierarchical level, all groups recorded at Sultanganj were separated registering >∼70% dissimilarity from Bhagalpur and Kahalgaon community. The latter two stations exhibited overlapping (∼40%) community structure and clustered on the basis of seasons (Figure S3). Copepods occurred in higher proportions at Kahalgaon followed by Sultanganj, rotifers were more abundant at Bhagalpur followed by Kahalgaon and Sultanganj. Four highly indicative species of Sultanpur community were Ceriodaphnia reticulata, Daphnia pulex, Brachionus quadridentatus and Filinia longiseta.
Amongst all zooplankton, rotifers were numerically most dominant group, both in terms of diversity and density at all sampling stations except during monsoon and postmonsoon (July–October). The three sampling stations showed differential impacts of seasons on abundance and species dominance pattern. During July–October cladoceran diversity and density were significantly higher at Sultanganj, whereas at Bhagalpur, copepod abundance but not diversity was significantly higher (Figures 2a and 3a). Cladocerans exhibited a decreasing trend towards Kahalgaon during monsoon and postmonsoon (Figure 4). Neither the density nor the diversity of copepods showed any significant temporal variations. The cluster analyses further revealed effects of seasons with high spatial variability (Figure S3) in the occurrence of zooplankton taxa realized during this study. The next cluster level (Cluster II) kept apart the monsoonal (August–September) assemblages (Cluster IIB) recorded at Kahalgaon, displaying higher abundance in autumn (Figure S3). The four highly indicative species for cluster IIB were Copepodid, Copepod nauplius, Brachionus caudatus and Keratella tropica (Table S3). The next hierarchical levels showed clusters coalescing Bhagalpur and Kahalgaon samples (Figure S3) that were separated by season (Figure S3). Cluster IIA further separated the monsoonal (July, August and September) community at Bhagalpur (Cluster IIIB) from the rest. The four highly indicative species were Ceriodaphnia rigaudi, Mesocyclops leuckarti, Brachionus falcatus and Mesocyclops hyalinus (Table S3). The remaining samples were separated on the basis of season (Figure S3).
The results obtained in the present study points to following limnoecological attributes of VGDS (a) zooplankton community is well represented covering diverse feeding modes and trophic levels, (b) rotifers are numerically dominant and diverse group except during monsoon and post monsoon, and they are severely impacted by copepods and cladocerans, (c) there is distinct spatial heterogeneity within the sampled stretch and (d) periodicity of community structure differs spatially.
Zooplankton community has evolved diverse strategies to solve problems like how to survive in changing environmental conditions, where to derive nutrients from, and, how to avoid being eaten. The present results convincingly signpost that the zooplankton community in the Ganga ecosystem shows maximal links in the aquatic food web. The trophic relationship and food web links explicitly elucidate the organization of ecological communities (Havens, 1998). The quantity and quality of food differentially affect the aquatic ecosystems, however in freshwater ecosystems the importance of top down control of zooplankton has been repeatedly emphasized (Rudstam et al., 1994). The phytoplankton and bacteria are well represented in the sampled stretch, the densities are higher during summer than winter (Bilgrami et al., 1993). The inverse relationship between herbivore and carnivore fractions pinpoints to the predation impacts. However, for systematic interpretation of community organization, the body size distribution and trophic structure needed to be analysed (Lewis, 1979; Cohen et al., 2003). In the present study the minimum size range of the plankton included Brachionus angularis, Filinia longiseta, B. caudata, Keratella tropica and nauplii of Mesocyclops. The perennial dominance of pico and nanophagous rotifers proves our first hypothesis. These organisms are primarily herbivorous and/or bacterivorous (Kumar, 2004; Kumar and Rao, 1998, 1999) and approach an absolute lower size limit for metazoans (Lewis, 1979; Hansen, 1994). The size distribution is generally truncated at a size of ∼1 mm (Lewis, 1979). The lower size limit of carnivores is determined by minimum size necessary to subdue, capture and manipulate suitable prey (Hansen, 1994). The smallest carnivore Asplanchna is just large enough to satisfy these criteria. Since the Ganga ecosystem harbours rich ichthyofauna, the diversity and abundance (De Silva et al., 2007) of larger zooplankton are controlled by the vertebrate predation. Cyclopoid abundance also exerts predation pressure on fish larvae. The higher abundance of Mesocyclops may be exerting predation pressure on early stages of fish larvae (Kumar et al., 2012).
Perennial occurrence of copepods suggests that temporal variations in community structure are mainly due to periodicity and occurrence of rotifers and cladocerans. The Copepoda exerts significant pressure on rotifer community at sites II and III, whereas, Cladocera plays a controlling effects on rotifers at Site I. These impacts can be due to interference competition and predation.
Predatory components in zooplankton community are represented by the rotifers genus Asplanchna and the copepods genus Mesocyclops. These predators prefer smaller prey such as heterotrophic protists and rotifers (Dodson and Frey, 1991; Kumar and Rao, 2001; Fernando, 2002) and exert significant impacts on zooplankton community structure (Kumar, 2003). The inverse relationship between copepods and rotifers can be attributed to the combined effects of predation and interference competition (since resources for herbivorous and bacterivorous species are never limiting). In interference competition, one species may limit the access of the competing species to vital resources, which need not necessarily be in short supply. Cladocera and herbivorous copepods captured in the present study have overlapping food requirement and consequently likely to compete. The distinct seasonal shifts in zooplankton community structure observed in the present study are a reflection of strong competitive and predatory interaction. The occurrence of pico and nano particles such as bacteria and organic detritus, which are abundant in the Ganga ecosystem (Bilgrami and Dutta Munshi, 1985; Trivedi, 2010) favors rotifer populations as they represent the most varied and abundant animal group in guild of small filter feeders (Paggi and José de Paggi, 1990). Rotifer dominance is also attributed to high amounts of suspended particles (Jack and Gilbert, 1993a). Suspended particles could have a positive effect on the composition and abundance of ciliates and rotifers by indirectly lessening predation pressure in the meso- and metapotamal stretches of the River Ganga. The amounts of suspended particles are highly perennial in the meso- and metapotamon stretches (Biswas and Konar, 2000) of the Ganga River. Periodicity of rotifer abundance (spring and summer) overlaps with the peak of microbial and algal concentration in Ganga ecosystem (R. Kumar, Centre for Environmental Sciences, Central University of Bihar, Patna, India, unpubl. data). The total bacterial count varied from 4.34–18.68 × 107 l−1 (Prakash, 2013) and phytoplankton were dominated by unicellular diatoms and microalgae which could be attributed to the dominance of rotifers as, non-predatory rotifers are bacterivorous and/or herbivorous preferring unicellular microalgal diet (Cheng et al., 2011). This proves our second hypothesis that pico and nano phagous rotifers play important role in transfer of bacteria- bound carbon to higher trophic levels as they are actively predated by cyclopoid adults and ichthyoplankton (Rao and Kumar, 2002).
The favorable temperature range seems to be (20–30°C) for the rotifer as well as bacterial populations. The prevailing conditions in Ganga ecosystem expected to favor rotifers from exploitative and interference competition with herbivorous crustaceans and visual and sensory predators. The high amount of inorganic suspended sediments (clay and silt) as in the sampled stretch reverses the outcome of competition and predation favoring rotifer community. In a laboratory investigation the rotifer Keratella cochlearis has been observed to be eliminated by Daphnia ambigua and Ceridaphnia dubia however, in suspended clay treated medium the rotifers dominated the cultures (Kirk and Gilbert, 1990).
The present study revealed that in a stretch of 50 km from Sultanganj to Kahalgaon there was a significant change in diversity and abundance of zooplankton. At spatial scale the completely different community structure at Sultanganj, and only ∼40% overlapping between Bhagalpur and Kahalgaon convincingly suggested a strong spatial heterogeneity in the limnoecological attributes. At Sultanganj the diversity and richness of Copepoda were maximum, at Bhagalpur there was moderate representation of all the major taxa of zooplankton while at Kahalgaon the abundance and percentage composition of Rotifera, Cladocera and Copepoda were recorded with maximum values. The heterogeneity observed in the distribution of zooplankton in major river systems is caused by interactions between physical and biological processes (Power and Dietrich, 2002; Nautiyal, 2010). This variability is related to flow-rates, and to the quality and quantity of resources brought into the system by tributaries (Threlkeld and Choinsk, 1985; Betsil and Van Den Avyle, 1994). The impacts of variation can be direct and/or indirect. For instance, during periods of high flux, the structure of the plankton can be strongly influenced by differential loss or addition of species as a consequence of their swimming capacity and reproductive rates, which in this study are reflected by the lower abundance of zooplankton during monsoon and post monsoon (flood period).
Temporal differences in zooplankton community structure in Ganga ecosystem can be attributed to species specific differences in growth and reproduction at different temperature levels, seasonal variations in flow regime, water retention time, differential amount of suspended silt and periodic flooding.
In order to have detailed insight of temporal variation it is important to follow the species-specific periodicity of population peak. During winter the cladoceran community was mainly dominated by the genus Daphnia whereas during summer the dominant cladoceran genera were Ceriodaphnia and Moina. Similar trends have been recorded in several lentic ecosystems in and around Delhi (R. Kumar, Centre for Environmental Sciences, Central University of Bihar, Patna, India, unpubl. data). In winter Daphnia was the main controlling factor of rotifer population at Sultanganj. In a laboratory study Daphnia was suppressed by the brachionid rotifers at 25°C whereas at 18°C rotifers were suppressed by Daphnia (Kak, 1999). Cyclopoid and calanoid copepods exert differential pressure on rotifers and cladoceran as cyclopoid copepods regulate rotifers and smaller cladocerans by predation (Kumar and Rao, 1999), whereas calanoids, mainly diaptomids exert pressure on rotifers and cladocerans through exploitative and interference competition. The contribution of copepods to the total zooplankton is next to the rotifer. Similar observations have also been made in the river Alaknanda (Badola and Singh, 1981).
It is worth mentioning that the varying hydrodynamic forcing factors (e.g. flow rate, turbulence and water depth) are important determinants of overall planktonic community structure in in riverine ecosystems (Allan, 1996; Pahwa and Mehrotra, 1966; Humphries, 2009). Filter feeders, plankton in general increase particle encounter rates through flow regime control (Trevethan et al., 2007; Humphries, 2009). To our knowledge, no study has been undertaken to determine the effects of variable hydrodynamic forcing on the zooplankton community structure. Further study is required aiming to reveal heterogeneity in hydrodynamic forcing and zooplankton community structure.
The Vikramshila Gangetic Dolphin Sanctuary covering a c.50 km segment of the Ganges River between Sultanganj to Kahalgaon is the only protected area established specifically for the protection of the endangered Ganges River dolphins. Extensive population fragmentation has resulted from the widespread construction of barrages (low gated dams). In depth knowledge of abiotic and biotic components of ecosystem is prerequisite for conservation measures of the species intended to protect. We studied zooplankton community structure during January through December 2009, at three sites, the upstream boundary (Sultanganj), middle (Bhagalpur) and downstream (Kahalgaon) boundary of the VGDS. Out of 36 samples from the three stations a total of 35 zooplankton species including rotifers (15), cladocerans (12) and copepods (8) were recorded. The total zooplankton abundance ranged from 100.5 ± 35.8 Ind l−1 to 231.58 ± 44.04 Ind l−1. Rotifera was the largest group contributing 30–60% to the total zooplankton abundance followed by Cladocera at Sultanganj and Copepoda at Bhagalpur. Zooplankton community was dominated by omnivores followed by bacterivores in rotifers, detritivores in cladocerans and carnivores in copepods. A distinct inverse relationship was observed between copepods and rotifers, at Bahgalpur and Kahalgaon however, at Sultanganj rotifers correlated negatively with cladocerans but not with copepods. The top three indicative species in Sultanganj samples were Ceriodaphnia reticulata, Daphnia pulex and Brachionus quadridentatus, whereas, Bhagalpur and Kahalgaon samples clustered on the basis seasons and exhibited differential patterns during different seasons. The prevailing conditions in this segment of the Ganga ecosystem favor rotifers. The high amount of inorganic suspended sediments (clay and silt), heavy bacterial load, and unicellular algae support rotifer community and protect them from competition with cladocerans and predation from copepods. The relative contribution of microbial loop and classical food web in carbon transfer to higher trophic levels needs to be elucidated to understand determinants of community structure in Ganga ecosystem.
We are grateful to three anonymous referees and to Drs. M. Munawar and Dilip Kumar for their valuable comments. We thank Drs. A. P. Sharma and M. Munawar for inviting Ram Kumar to the author's workshop on Major Rivers. M. Tosha is thanked for linguistic improvement. We are also thankful to the University Department of Zoology, Bhagalpur University for facilities required during sampling.
[Supplementary materials are available for this article. Go to the publisher's online edition of Aquatic Ecosystem Health and Management to view the free supplementary files.]