This study reveals the relationship of biotic assemblages to their environments in the Central Highland Ecoregion rivers (Ken, Paisuni and Tons). Such knowledge will enhance our predictive abilities in the ecological studies and resource management in the concerned region. Rivers of an ecoregion are expected to harbour similar flora, fauna, and communities. This hypothesis was examined by tracking the spatial variation in richness, density, and taxonomic composition of benthic diatom and macroinvertebrate assemblages in three Plateau rivers whose course lay within 3°N latitude. Taxonomic richness and density decreased with the reduction of substrate size from the headwaters to the lowlands. Cluster and ordination analyses indicate prevalence of intrabasin rather than interbasin gradients of abundance. In both assemblages, the most abundant taxon is rarely similar longitudinally or across the river sections in the ecoregion. Hence, proximate factors govern the assemblages rather than ecoregional properties (latitude, altitude). This, and the high richness of diatom flora, points to diverse assemblages from the headwater to the lowland section in each river and among the rivers of the ecoregion.

Introduction

Correspondence between ecoregional classification and the biota should depend on the strength of linkages. Ecoregions may poorly classify components of the biota that are strongly regulated by local factors. Algae, macroinvertebrate and fish assemblages each respond to environmental changes at different temporal and spatial scales because of their different life histories, physiologies, and mobilities. Thus, changes in each assemblage may reflect local conditions of stream habitats including microhabitats, channel unit, reach, and watersheds (Frissel et al., 1986). Diatom assemblages often respond to land use, especially agricultural activities (Pan et al., 2000). Thus, diatom assemblages not specific to bioregions may be ideal as unbiased indicators of stream water quality (Charles et al., 2006). The distribution of benthic macroinvertebrates also varies within and among the rivers (Hawkins and Sedell, 1981; Milesi et al., 2009; Mesa, 2010; Ezekiel et al., 2011). Macroinvertebrate compositions are known to exhibit much stronger relationships to local environmental conditions (at reach scale) than to catchment variables (Carter et al., 1996). The factors influencing freshwater biodiversity at local and regional scales in India (Rodgers et al., 2002) remains poorly understood.

Our study area is within the Central Highlands subdivision of the major physiographic region the Peninsular Plateau, synonymous with the 6th biogeographic region—Deccan Peninsula. The Central Highlands alongside Chota Nagpur are the northernmost biotic provinces of this biogeographic region (Source: wiienvis.nic.in/database/htmlpages/bioprovincemap.htm WII, 2000). The Central Highlands province has an east–west orientation and continental climate, with rainfall decreasing westwards. The Gangetic drainage primarily descends from the Himalayas as the Yamuna and Ganga Rivers. The Plateau Rivers, descending from the western and central Vindhya Ranges, join the Yamuna, while those arising from the eastern ranges join the Ganga. The Central Highlands is an important region for the freshwater biodiversity of hill streams of the Indian subcontinent. This ecoregion represents a vital junction of mountain streams and rivers.

Agenda 21 (1992) emphasizes the role of biodiversity in the sustainable development of the uplands. Only recently has biodiversity been documented for the Central Highlands ecoregion (Nautiyal and Verma, 2009a; Verma and Nautiyal, 2010; Nautiyal and Mishra, 2012, 2013; Kulshrestha, 2009; Lakra et al., 2010; Dwivedi and Nautiyal, 2012; Nautiyal et al., 2011). Stable native food webs are the key to health of the ecosystem. Lower trophic levels in wild rivers, especially grazers, require pristine conditions. It is imperative to document the baseline conditions of benthic biota and assemblages (diatom, macroinvertebrates). Recent studies of the benthic communities of the Ken, Paisuni, and Tons Rivers during the last decade help predict ecoregional features of the Central Highlands (Nautiyal and Mishra, 2012, 2013; Mishra and Nautiyal, 2013a; 2013b).

The benthic diatoms and macroinvertebrates (usually insect larvae) account for the majority of species among the producers and consumers. A profitable fishery depends on healthy food webs that involve lower trophic levels within benthic communities. Fishing is a commercial activity that generates considerable employment along the rivers, from those manufacturing fishing nets, to fishing vessels, fishermen, and traders. This study primarily examines the richness and composition of benthic diatom and macroinvertebrate communities and the factors governing them in the Central Highlands ecoregion. The focus of this study was to generate information on the spatial distribution of benthic biota along the length of the rivers.

Materials and methods

Study area

The Ken, Paisuni, and Tons drainages in the northeastern Central Highlands are the subject of this study. The Central Highlands consist of the Malwa, Bundelkhand and Chota Nagpur Plateaus that form the northern subdivision of the ancient (Gondwana) triangular-shaped tableland—the Peninsular Plateau. This geographical region is bound by the Gangetic Plains to the north and east, and the Deccan Peninisula to the south. The Vindhya Ranges give rise to northward flowing tributaries of the Yamuna (Chambal, Kali, Sindh, Parbati, Betwa, Dhasan, Ken, Paisuni) and Ganga (Tons, Sone) Rivers. These rivers are important for irrigation, domestic supply, and power generation including proposed river linking projects on the Indian Peninsula (NWDA, 2006). The rivers also have high religious significance. These drainages lie within 24 to 26°N latitude, 79 to 82°E longitude, and altitude 360 to 72 m above sea level from source to confluence. The examined length of rivers Ken, Paisuni, and Tons are ca. 340 km, 100 km, and 305 km long, respectively. Locations sampled on the Ken are labelled K1 to K4, Paisuni P1 to P4, and Tons T1 to T4 (Figure 1), dividing the river into four gross sections: upper (K1, P1, T1), middle (K2, P2, T2), and lower plateau (K3, P3, T3), and a lowland section (K4, P4, T4). The course of these rivers from upper to lower sections moves through the Plateau, which constitutes the highland sections, while the mouths fall in the lowland section, which stretches for approximately 80 km in the Ken and Tons, and 30 km in the Paisuni. The highland stretch of the Ken and Tons Rivers consists of short headwater sections up to 50 km, and 100–120 km long middle and lower sections. In contrast, the Paisuni headwater section is approximately 10 km and the middle and lower sections are 30–40 km. The physiography, climate, and vegetation have been described previously for these rivers (Mishra and Nautiyal, 2011; Nautiyal and Mishra, 2012).

Sampling

Annual intensive sampling is considered suitable to analyze the distributional patterns of benthic macroinvertebrate assemblages (Corkum, 1989). Thus, intensive sampling in the dry period (October to June) was considered appropriate. The river's flora and fauna diversify after monsoon floods (July to September). The dry period extends from October to June but sampling was restricted to December 2003 to March 2004 because long stretches of the rivers tend to dry up during the summer season (March to June), forming pools of various sizes that disrupt the continuum.

Water temperature and pH were analyzed with portable meters (WT-Mextech, multimeter, pH-Hanna). Current velocity (CV) was measured by an EMCON current meter. Substrate predominance at each sampling station was estimated visually. If the river bed was extensively littered with cobble substrate followed by boulder and gravel, then substrate type was recorded as C;B:G. Substrate names and their categories were determined through Wentworth classification of substratum particle size (http://fwcb.cfans.umn.edu/courses/fw8459/Private/SubstrateandHyporheos08-2.pdf).

Diatom samples were obtained by scraping the surface of 3×3 cm samples of cobbles collected from the river bed Samples were collected from various habitats at each station. Each sample was preserved in 4% formaldehyde solution. Samples were treated with HCl and cleaned by boiling in H2O2. The treated samples were washed repeatedly to remove traces of acid and peroxide. Permanent Naphrax mounts were prepared and examined under bright field using a BX-40 Trinocular Olympus microscope (x10 and x15 wide-field eyepiece) fitted with a PLANAPO x100 oil immersion objective. Diatoms were identified according to Hustedt (1931–1959), Krammer and Bertalot (1986–1991), Sarode and Kamat (1984), and Gandhi (1998). The slides are archived at the Aquatic Biodiversity Unit. The relative abundances (in percent) were determined on the basis of 300 valves from each sample at each station.

Benthic macroinvertebrate fauna was sampled intensively (20 quadrants per station). Sampling procedures at each station involved lifting stones (boulder, cobble, pebble, gravel) and sieving clay and silt from a 0.09 m2 area in different flows (turbulent, swift, slow, placid). The substrate was washed to dislodge the fauna, which was preserved in 5% formalin for further analysis. As broad taxonomic classifications are acceptable (Corkum, 1989), the fauna was identified at a family level (Edmondson, 1959; Edington and Hildrew, 1995). Counts were made to obtain density (indiv. m−2) and relative abundance. The functional groups were defined according to Cotta Ramusino et al. (1995) and Dudgeon (1984).

The Kruskal–Wallis test (Henderson, 2003) was applied to determine significant differences between stations in total density of the community and higher taxa (order, families). Principal Component Analysis (PCA) determined the characteristic taxa at each station, and Canonical Correspondence Analysis (CCA; ter Braak and Smilauer, 2002) helped to identify the environmental variable(s) causing longitudinal variation in the taxonomic composition of the benthic macroinvertebrate fauna. The cluster analysis to classify river sites and the multivariate analysis (CCA and PCA) were based on the raw data, i.e. counts of diatom species and invertebrate fauna from each quadrant.

Results

The range of water temperature (WT) and pH differs in the rivers Ken (16–27°C; 7.0–7.5), Paisuni (15.5–33°C; 7.0–7.7) and Tons (17–31.4°C; 7.0–7.8). The range value of current velocity (CV) was higher in the Ken (1.0–60 cms−1) and Tons (0.1–50 cms−1) compared with the Paisuni (0.0–30.9 cms−1). The WT and pH increase downstream of the source, while current velocity decreases (Figure 2). Substrate particle size decreases gradually from the upper to lower plateau sections, but abruptly changes to small sediments in the lowland section, which lies in the Gangetic Plains (Table 1).

Benthic diatoms

The diatom flora is represented by 205, 202 and 211, taxa, respectively, in the rivers Ken, Paisuni and Tons, with a total overall 293 taxa (species, variety, form) from 50 genera (see Appendix A in the online supplemental material). In the Ken, richness declines downstream of the headwater section, while in the case of the Paisuni and Tons, richness increases from the headwater to middle sections but decreases from the lower to lowland sections (Figure 3). Total density decreases downstream of the sources in the Ken and Tons, while in the Paisuni density increases from the headwater to middle section and declines in the lowland section (Figure 3). Richness as well as density vary among the headwater (168–182; 1198–1404 cell. mm−1), middle (178–185; 1098–1343 cell. mm−1), lower (157–169; 803–1003 cell. mm−1), and lowland (151–156; 843–934 cell. mm−1) sections of these rivers, most noticeably in the lowland sections compared with the headwater to lower sections. The raphid elements, especially biraphid, constitute a very large share of diatom flora throughout the ecoregion and in each basin. The biraphid Navicula, Cymbella, and Nitzschia are species-rich genera in each river (see Appendix B in the online supplemental material).

In the ecoregion, 18 diatom taxa attain 10% or more relative abundance in one or more sections. Amphora twentiana (9%) is also listed, as it is probably unique to the rivers of Central India. The middle section of the Tons contains eight taxa that attained >10% abundance, compared with four and two taxa, respectively, in the lower and lowland sections (Table 2). Notably, all Achnanthidium species occur in 9–14% abundance at all the locations of these rivers. Another 11 taxa exhibit >10% abundance at either the headwater or lowland location/section. Aulacoseira granulata, occurs in >10% relative abundance in all the sections of the Ken only. Based on abundance patterns, Achnanthidium chitrakootense was the representative taxa for Paisuni River, while Achnanthidium minutissimum var. scotica and A. minutissimum were representative taxa for the Ken and Tons, respectively. These taxa are, therefore, the abundance signatures of the respective rivers (Table 2).

Examination of the abundance pattern among the sections of the Ken, Paisuni, and Tons reveals the similarity between the headwater section of the Ken (K1), and headwater as well as middle section of the Paisuni (P1, P2). Similarly, the middle to lowland sections of the Ken (K2 to K4) resemble the lower section of the Paisuni (P3). The Tons forms a separate cluster of its own sections (Figure 4).

Benthic macroinvertebrates

Twenty-eight invertebrate taxa (class/families), including those from seven insect classes, occur at the ecoregion scale, and 21, 24 and 27 taxa are seen at the basin scale in the Ken, Paisuni and Tons, respectively. The richness varies among the headwater, middle, and lower sections of these rivers. Richness is high in the middle and lower sections of the Ken, the middle and lowland sections of the Paisuni, and the headwater and lower sections of the Tons (Figure 3). Longitudinally, the mean density decreased from the headwater to the lowland section in the Ken and Tons, while it increased in the Paisuni (Figure 3). The density differed significantly among the sections (Observed H(0.5,) for headwater, middle, lower and lowland sections was 16.49, 13.76, 39.88 and 41.77, respectively).

Insect larvae are the major component of the macroinvertebrate fauna followed by molluscs and annelids (Table 2). The order Ephemeroptera is a family-rich taxon in all of the Plateau rivers (Appendix C). The cluster analysis revealed that the invertebrate fauna reflect the close similarity among the middle, lower, and lowland (P2, P3, P4) sections of the Paisuni, as well as the headwater, middle, and lowland (T1, T2, T4) sections of the Tons. Only the lowland sections of the Ken and Tons (K4-T4) are similar (Figure 5). The headwater and middle sections of the Ken (K1 and K2) exhibit similarity with the lower section of the Tons (T3). Similarly, the headwater section of the Paisuni (P1) resembles the headwater and middle sections of the Tons (T1 and T2). The lower and lowland sections of the Paisuni (P3 and P4) are similar to the lower section of the Ken (K3). Functionally, the collectors (suspension and deposit feeders that filter fine particles from the water column or gather loose particles in depositional areas) are dominant longitudinally throughout the rivers, from the headwaters to the lowland sections. The percentage of scrapers (feeders of photosynthetic algae growing on any substrate) increases substantially downstream in the Ken and Tons, while it decreases in the Paisuni (Figure 6).

Multivariate analysis

PCA ordination for diatoms reveals that Axes 1 and 2 represent 41.4% and 29.1% of the variance in the lowland sections of the Ken and Tons (K4, T4). A. minutissimum, Gomphonema parvulum, Navicula viridula and Navicula cryptotenella are characteristic taxa in these sections. A. chitrakootense, A. petersenii, Navicula seminulum and Navicula rostellata are characteristic in the entire Paisuni and middle section of the Tons (Figure 7). PCA ordination for macroinvertebrate fauna indicates that Axes 1 and 2 represent 36% and 17.6% of the variance in taxonomic composition. The characteristic taxa are similar in the lowland sections of the Ken and Tons. The headwater of the Paisuni resembles the middle section of the Ken (Tabanidae, Sialidae) and faintly resembles the headwaters of the Ken and middle Tons, characterised by Glossoscolecidae, crustacean and Caenidae. The remaining sections of the Paisuni slightly resemble the lower section of the Ken, as Chironomidae lies intermediate to them. The rivers Ken and Tons are similar compared with the Paisuni, with the exception of PI (Figure 8).

CCA indicated that land use, current velocity, and substratum are important variables for the distribution of benthic macroinvertebrate fauna in the rivers Ken, Paisuni, and Tons, respectively. However, depth (53.8%) and water temperature (46.2%) emerge as the most important factors with respect to the ecoregion (Figure 9).

Discussion

Benthic diatoms

The richness of the diatom flora for the ecoregion is considerably higher than that found in a lesser Himalayan stream (Verma and Nautiyal, 2009) and the Plateau River Damodar (Nautiyal and Nautiyal, 1999a), but is comparable with what is seen in the Mandakini basin (Nautiyal et al., 2004a) and the Alaknanda-Ganga (Nautiyal and Nautiyal, 1999b). When the 153 diatom taxa (38 genera) recorded from the Betwa River (Figure 1, Appendix D) are added, the total number of taxa in the Central Highlands increases to 322 representing 50 genera (Appendix A). Kawecka and Olech (2004) recorded 270 taxa from Finnish Lapland, which was lower than the numbers seen in the Central Highland ecoregion. Thus, the richness of the Central Highland rivers is higher than that in the rivers of more temperate regions: 165 taxa found in the Estonian River (Vilbaste, 2001) 72 taxa in streams of Coastal Oregon (Naymik et al., 2005), and 98 and 94 taxa, respectively, from the Bulgarian rivers Vit and Osum (Stancheva et al., 2007).

Richness is relatively low in the Paisuni and Ken compared to the Tons. However, the shorter Paisuni is proportionately richer than the longer Ken and Tons Rivers. The island biogeography concept suggests that the larger area of the longer rivers should have higher richness than the Paisuni, provided the number of habitats increase with area (Begon et al., 1990). However, diatom richness did not increase with area, probably due to the low richness in the lowland sections compared with the upper plateau. High richness of the Paisuni in proportion to its length supports the view that freshwater biodiversity can be highly localized, since freshwater habitats are relatively discontinuous and many freshwater species do not disperse easily across the land barriers (WRI, 1992). Nautiyal et al., (2004a) also observed five centers of high diversity in different altitudinal zones of the Mandakini basin.

Species richness of the diatom community exhibits a general decrease from the headwater to the lowland sections in all of the rivers, despite a slight increase in the middle sections of the Tons and Paisuni, attributed to the longitudinal gradients of decreasing substrate size and heterogeneity from source to mouth (Nautiyal and Verma, 2009a). The richness of diatom flora and density in the headwater and lower sections decreases towards the east from the Ken to the Tons basin, but increases in the middle and lowland sections from the Ken to the Tons. The diatom richness in the Paisuni is higher than in the river basins on either side, while density is slightly less.

The Central Highlands ecoregion supports high Shannon species diversity () and low evenness: Ken (5.01; 0.30), Tons (5.02; 0.33), and Paisuni (5.07; 0.23; Nautiyal and Verma, 2009b). The Shannon species diversity and evenness were observed to be 3.03 and 0.50, respectively, in the Betwa of the same ecoregion. The species diversity is higher in the Central Highlands ecoregion ( = 5.56) compared with the Himalayan ( = 4.40) and other Plateau rivers (Nautiyal and Nautiyal, 1999a). Biodiversity maximizes at intermediate levels of disturbance (Townsend et al., 1987; Minshall, 1988), which can operate on a local scale. Factors like predation, herbivory, fluctuations in physical factors, and catastrophes can be lumped together as disturbance (Krebs, 1994). Human activities also act as disturbances, indirectly in the form of agriculture and directly in the form of impoundments across the Betwa.

The taxonomic composition of diatom flora from the Central Highlands is dominated by biraphids. In Himalayan rivers, biraphid content is also high (Nautiyal and Nautiyal, 1999b; Nautiyal et al., 2004b). The Peninsular area or other parts of India, including the mountainous zones like the Western Ghats (Krishnamurthy, 1954), are equally rich in biraphid flora. Most species-rich genera are biraphid; Navicula, Cymbella and Nitzschia in the Paisuni and Ken, and Navicula, Nitzschia, and Cymbella in the Tons, contribute to this trend (Appendix B).

PCA and cluster analysis on diatom abundance indicated longitudinal patterns in contrast to the spatial patterns (among the river sections). The floral composition (species abundance) is similar in the lowland sections of the Ken and Tons due to similar ecological conditions (soft sediments and flooding). However, other sections of the Ken are similar to the Paisuni, attributable to similar riparian conditions (Saal forest and prevalence of agriculture) in their basins.

Benthic macroinvertebrates

The macroinvertebrate fauna identified in our study is comparable with other Peninsular and Himalayan rivers/basins. No definite pattern of invertebrate richness is evident for the tropical or temperate conditions. The number of benthic fauna is high at the regional scale rather than in individual rivers because the species differ between one site and the next (Allan, 1996; Vinson and Hawkins, 1998).

In the case of benthic macroinvertebrate fauna, the taxonomic richness in the Ken and Paisuni is relatively the same, though the Ken is longer than the Paisuni. The Tons, which is relatively shorter than the Ken, has the highest richness among all three rivers. In these rivers, the richness does not increase with river length (or area). Instead, the higher richness in the Tons may be attributed to the possibility of more habitat availability caused by substrate heterogeneity in the Tons, a concept accepted in island biogeography (Begon et al., 1990). However, few studies are available indicating that richness increases with respect to sampling site (Milesi et al., 2009; Nautiyal and Mishra, 2012). At a spatial scale, macroinvertebrate richness increases slightly from the Ken to the Tons in the lowland sections, while a reverse trend of decrease is seen with regard to species density. The lowland section of the Paisuni supports high richness and density compared with the lowland sections of the other two rivers. The richness does not differ in the lowland sections of the Ken and Tons, where substrate heterogeneity is greatly reduced by annual flooding that deposits smaller and softer sediments on the river beds.

At latitudes comparable with the Central Highland rivers, 58 taxa were recorded from the Brazilian Plateau rivers (27°N) and 25 taxa were found in the Lules River basin (26 °N; Mesa, 2010). At the tropical latitude (12°N) of India, macroinvertebrate richness (43 taxa) was relatively higher in the River Kaveri (Sivaramakrishnan et al., 1995). Corkum (1992) observed that density was not consistent among the rivers of different biomes.

The diversity of benthic macroinvertebrates is known to increase with increasing substrate heterogeneity (Singh et al., 1994; Nautiyal and Mishra, 2012) represented by the middle and lower sections of the Ken and Paisuni Rivers. Diversity is high in the agricultural lands (Richards et al. 1997) and lower in urbanized areas (Walsh et al. 2001). The proximate factors appear to be more responsible for the assemblages of the benthic macroinvertebrate fauna than large-scale factors (Strayer et al., 2003). The density of benthic macroinvertebrate fauna declines generally from the headwaters to the lowland sections of each river, except the Paisuni (Mishra, and Nautiyal, 2011; Nautiyal and Mishra, 2012; Mishra and Nautiyal, 2013a). The macroinvertebrate fauna of the Central Highlands consists largely of insects. The mollusc and annelid elements are very few in the fauna. This is comparable with insect-dominated fauna from rivers of the Western Ghats (Sivaramakrishnan et al., 1995; Subramanian and Sivaramkrishnan, 2005), as well as the mountain part of the Gangetic drainage in Uttarakhand. In contrast, mollusc-dominated fauna has been reported from the rivers of the Malwa Plateau (Kulshrestha, 2009) in the western part of the Central Highlands.

The most abundant macroinvertebrate taxon is longitudinally similar in various sections of the rivers Ken, Paisuni, and Tons. Longitudinal variations in abundance are attributed primarily to substrate preferences of the fauna and land-use practices in the rivers of this ecoregion (Mishra and Nautiyal, 2011). It can be argued that similarity of substrate and riparian conditions along the river lead to longitudinal similarity of the fauna, as it appears in the Paisuni and Tons but not in the Ken, as was also observed by Nautiyal and Mishra (2012). Vannote et al. (1980) state that gradients of physical and chemical characteristics from headwater to mouth result in different community and functional feeding groups (FFG). Longitudinally, collectors are dominant in all the rivers. Therefore, the rivers are functionally heterotrophic (Mishra and Nautiyal, 2013b). This study is in agreement with the River Continuum Concept, as heterotrophic conditions prevail in the headwater sections of the rivers (i.e. third order), though in the present study these rivers represent a third order condition from headwater to mouth.

The faunal abundance (shown as percent) also exhibits longitudinal similarity within each river (intra-basin), as the dominant taxa are identical, especially in the headwater and middle sections of the Ken and Tons. However, in the lowland sections of the Ken and Tons in the Gangetic Plains, the fauna lacks inter-basin similarity. This is attributed to the deposition of soft sediments by the Yamuna and the Ganga around the confluence of these rivers during annual flooding. Thus, the rivers are not similar section by section with respect to the macroinvertebrate fauna. Cluster analysis for the river sections clearly indicates that the faunal abundance in the headwater or middle sections of the Ken and Tons are not similar, though the rivers originate in close proximity. Only the lowland section of the Ken and Tons resemble one another. The headwater section of the Paisuni is similar to the headwater, middle, and lower sections of the Tons, while the lower and lowland sections of the Paisuni are similar to the lower section of the Ken. Pan et al. (1999) showed that the spatial patterns of diatom assemblages in the Mid-Atlantic Highlands were best discriminated by both land cover/land use in catchments and by site-specific factors such as riparian conditions, as observed in the Central Highlands ecoregion.

Factors governing distributional patterns of macroinvertebrates

CCA indicated that land use, current velocity, and substratum are important variables for distribution, while depth and water temperature emerge as the most important factors with respect to the ecoregion. These factors have figured in other studies too: current velocity in the Paisuni (Mishra and Nautiyal, 2011), land use in the Ken (Nautiyal and Mishra, 2012); substratum in the Tons (Mishra and Nautiyal, 2013a), and current velocity, substrate size, conductivity, and abundance of aquatic plants in the Patagonian Plateau rivers (Miserendino and Pizzolon, 2003).

The central highland rivers: Are they diverse?

Rivers of an ecoregion are expected to harbour similar flora, fauna, and communities. This hypothesis was examined in respect to the richness, density, and taxonomic composition of benthic diatom and macroinvertebrate assemblages in three Plateau rivers. Observations reveal that the three rivers are similar in species richness of the diatom community, but the rivers differ in respect to species richness of benthic macroinvertebrate fauna. With some exceptions, the pattern of general decrease in the species richness and density from headwater to lowland sections is also similar for both communities among these rivers. All rivers support high diatom species diversity. The species-rich genera (Navicula, Cymbella and Nitzschia) were also similar. The insect fauna prevails in stretches of river with hard substrate, while mollusc and annelid species are found in lowland sections having soft sediments. The faunal elements lack inter-basin similarity. Thus, the rivers are not similar section by section with respect to the macroinvertebrate fauna. The diatom taxa A. chitrakootense is specifically abundant in the Paisuni. Navicula seminulum and Navicula rostellata are characteristic of the Paisuni and the middle section of Tons. The Paisuni, which is three times shorter than the Ken and Tons, is exceptional in some respects, being unique both in respect to benthic diatom and macroinveretebrate species, while the Ken and Tons resemble each other relatively more. Though the longitudinal trends for species richness and density of benthic flora and fauna are similar, the examined communities differ in elements and abundance. Thus, generalisations regarding similar flora, fauna, and communities in the rivers of the ecoregion cannot be made for the Central Highland rivers. In other words, these rivers are diverse because the species differ between one site and the next. In view of the Ken–Betwa river link, this ecoregion needs to be examined for diversity of different trophic levels (NWDA, 2006).

Species richness of diatoms is high in the Central Highland rivers compared to temperate rivers (Kawecka and Olech, 2004). The floral composition and abundance at subtropical latitudes in the Himalayas (Nautiyal et al., 2013) is different from the Central Highland rivers. Species richness of macroinvertebrates in the Central Highland rivers somewhat resembles that of the Plateau rivers of South America (Mesa, 2010), but is certainly lower than that found in the tropical latitudes of India itself (Sivaramakrishnan et al., 1995). The faunal elements differ considerably from the Deccan Plateau at tropical latitudes, as well as subtropical Plateau rivers and basins of South America (Nautiyal and Mishra, 2012).

Science and management

Study of the benthic community is important because it reflects changes in the aquatic environment either due to natural or anthropogenic effects. The Ken–Betwa river link has been approved to sustain and enhance agricultural production in the Bundelkhand region. The link involves building another dam on the Ken and diverting the water to the Betwa through a linkage canal (Figure 1). Presently, the fish fauna of the Betwa are threatened by habitat fragmentation (Lakra et al., 2010). In addition to social impacts (Krueger et al., University of Michigan Master of Science Degree, 2007), the link will reduce spatial heterogeneity, constrain the species represented at sites (Pan et al., 2000), and hinder its role as a vital junction for the mountain biotic elements of the subcontinent, especially in the donor river, Ken (NAAS, 2004, Policy Paper 28). The present study has demonstrated that the Plateau rivers are diverse with certain fish species being specific to some rivers (Appendix E), as are some diatom species. Maintaining scientifically recommended flows with regular monitoring is thus imperative for the Plateau river ecosystems.

Priority gaps and needs

The present study explains only the Bundelkhand part of the Central Highlands ecoregion, while the rest remain unknown. It is important to generate information on this ecoregion because of the native elements and the region's value as a vital junction of mountain biota.

Speculation on future trends and potential for fish species, fisheries, and habitats

The Betwa has more fish species (63) than the Ken (56), the Paisuni (52), and the Tons (48). Many (43) species are common to all three rivers. The species richness of fish declines eastwards in the Central Highlands ecoregion. The food web comprising allochthonous + autocthonous inputs, along with the producer community of which diatoms constitute a very significant part, and the invertebrate consumer community that utilises detritus and algal biomass, are the major drivers of fishery. Although, the species richness of benthic communities declines in the rivers from source to mouth, the species richness of fish fauna increases in the lower part of the rivers as they widen, providing a variety of habitats that various fish species can utilise. Homogeneous softer sediments, decreased current velocity, and a greater volume of water creates river depths that allow the plankton community to flourish. Because of this, there is a decline in benthic species richness.

Sperata spp. is highly prized followed by L. rohita, T. tor, while L. calbasu is the least prized fish for commercial fisheries. The high exploitation rate for Tor tor in the Paisuni and Tons Rivers, and L. rohita in the Ken has threatened these species (Dwivedi and Nautiyal, 2012). These rivers are stressed habitats because shallow river sections dry up during summer. The river linking will reduce desired flows, reduce fish diversity, and the more tolerant common carp will dominate the fish landings. Presently, Cyprinus carpio, Labeo calbasu, and miscellaneous groups constitute the major portion of capture fishery in the lower sections of the Ken, Paisuni and Tons.

Conclusions

In the present study, taxonomic richness and density of benthic flora and fauna decrease longitudinally from the headwater to lowland sections in all the rivers (except for the macroinvertebrate species in the Paisuni). This is due to the gradual reduction of substrate particle size from boulder (stony) to silt-clay (soft), which provide low surface area for production of producer and consumer communities. The abundant benthic flora varies longitudinally in all the rivers, but in the case of benthic fauna variation occurs only in the Ken, but there is similarity from middle to lowland in the Paisuni as well as to the headwater, middle and lowland in the Tons. The Central Highlands ecoregion supports high richness and diversity of diatoms. Spatially, there is no definite trend in the flora and fauna (except in the lowland sections) in all measured aspects (richness, density, and abundant taxa). Therefore, each river and section has its own characteristic flora and fauna, which suggests that the proximate factors govern the various features of assemblages, rather than regional factors such as latitude, altitude, and climate. This study demonstrates high diversity across these river basins in the Central Highlands ecoregion.

Acknowledgements

Co-authors acknowledge the financial assistance for the Doctor of Philosophy program at the University of Allahabad. We thank H.N.B Garhwal University, Srinagar for academic support.

Supplemental material

Supplemental data for this article can be accessed on the publisher's website.

ORCID

Prakash Nautiyal http://orcid.org/0000-0002-7958-5755

The text of this article is only available as a PDF.

References

Agenda 21
,
1992
.
Programmed or Action for Sustainable Development. Rio Declaration on Environment and Development
.
The final text of agreements negotiated by Governments at the United Nations Conference on Environment and Development (UNCED)
3–14 June 1992
Rio de Janerio Brazil
.
Chapter 13 Sustainable Mountain Development
.
Allan, J. D.,
1996
.
Stream ecology - structure and function of running waters
.
Chapman and Hall Press
,
London
.
Begon, M., Harper, J. L., Townsend, C. R.
1990
.
Ecology individual, populations and communities
.
Cambridge, Massachusetts, USA
.
Carter, J. L., Fend, S. V., Kennelly, S. S.,
1996
.
The relationships among three habitat scales and stream benthic invertebrate community structure
.
Freshwat. Biol.
35
,
109
124
.
Charles, D. F., Acker, F. W., Hart, D. D., Reimer, C. W., Cotter, P. B.,
2006
.
Large-scale regional variation in diatom-water chemistry relationships: rivers of the eastern United States
.
Hydrobiologia
,
561
,
27
57
.
Corkum, L. D.,
1989
.
Patterns of benthic invertebrate assemblages in rivers of northwestern North America
.
Freshwat. Biol.
21
,
191
205
.
Corkum, L. D.,
1992
.
Relationship between density of macroinvertebrate and detritus in the rivers
.
Arch. fur Hydrobiol.
239
,
149
166
.
Cotta Ramusino, M., Villa, S., Calamari, D.,
1995
.
River continuum concept and correspondence analysis to study Alpine stream macroinvertebrate assemblage
.
Mem. lst. Ital. Idrobiol.
53
,
101
114
.
Dudgeon, D.,
1984
.
Longitudinal and temporal changes in functional organisation of macroinvertebrate communities in the Lam Tsuen River, Hong Kong
.
Hydrobiologia
,
111
,
207
217
.
Dwivedi, A. C., Nautiyal, P.,
2010
.
Population dynamics of important fishes in the Indian region. Status and structure of commercial fishes in the Vindhyan river
.
Lap Lambert Academic Publishing, Amazon Distribution GmbH
,
Leipzig
.
Dwivedi, A. C., Nautiyal, P.,
2012
.
Stock assessment of fish species Labeo rohita, Tor tor, and Labeo calbasu in the rivers of Vindhyan region, India
.
J. Environ. Biol.
33
,
261
264
.
Edington, J. M., Hildrew A. G.,
1995
.
Caseless caddis larvae of the British Isles
.
Freshwat. Biol. Asso. Sci Publi.
53
,
1
133
.
Edmondson, W. T.,
1959
.
Freshwater Biology
. 2nd edition.
John Wiley and Sons, INC
,
New York
.
Ezekiel, E. N., Hart, A. I., Abowei, J. F.N.,
2011
.
The distribution and seasonality of benthic macroinvertebrates in Sombreiro River, Niger Delta, Nigeria
.
Res. J. Appl. Sci., Eng. Tech.
,
3
(
4
),
264
271
.
Frissel, C. A., Liss, W. J., Warren, C. E., Hureley, M. D.,
1986
.
A hierarchical framework for stream habitat classification: viewing streams in a watershed context
.
Environ. Manag.
10
,
199
214
.
Gandhi, H. P.,
1998
.
Freshwater Diatoms of Central Gujarat (with a Review and Some Others)
. Bishen Pal Singh, Mahendra Pal Singh,
Dehradun
.
Hawkins, C. P., Sedell, J. R.,
1981
.
Longitudinal and seasonal changes in functional organization of macroinvertebrate communities in four Oregon streams
.
Ecology
62
(
2
),
387
397
.
Henderson, P. A.
2003
.
Practical Methods in Ecology
.
Blackwell Scientific Publication
,
Oxford, London
.
Hustedt, F. C.
1931–1959
.
Die Kieselagen Deutschlands Oesterriches und der Schweiz Bd.7, Teil 2. (Translated by N.G. Jensen as The Pennet Diatoms 1985)
, pp.
918
.
Koeltz Scientific Books: Koenigstein
.
Kawecka, B., Olech, M.
2004
. Diatom diversity of streams in Finnish Lapland and Maritime Antarctica. In: M. Poulin (Ed.),
Proceedings of 17th International Diatom Symposium
, pp.
161
186
.
Ottawa, Canada
,
2002
August
25–31
.
Biopress Limited, Bristol
.
Krammer, K.., Lange-Bertalot, H.,
1986–1991
.
Bacillariophyceae. Die Süsswasserflora von Mitteleuropa (The Freshwater Flora of Central Europe. In German). Vol. 2/1 Naviculaceae
, p.
1
876
mit 206 pl
.
Vol. 2/2 Bacillariaceae, Epithemiaceae, Surirellaceae
, p.
1
596
(
1988
).
Vol 2/3 Centrales, Fragilariaceae, Eunotiaceae
, p.
1
576
(
1991
).
Vol. 2/4 Achnanthaceae, Kritische Erganzungen zu Navicula (Lineolatae) und Gomphonema
. p.
1
437
(
1991
).
Vol. 2/5 English and French translations of the keys and supplements
(
2000
).
Stuttgart and Heidelberg
.
Krebs, C. J.,
1994
.
Ecology: The Experimental Analysis of Distribution and Abundance. 4th Edition
.
Harper Collins
,
New York
.
Krishnamurthy, V.,
1954
.
A contribution to the diatom flora of south India
.
J. Indian Bot. Soc.
33
,
354
381
.
Kulshrestha, S. K.,
2009
. Biodiversity of some aquatic environments in Central India. In: Singh, H. R., Nautiyal P. (Eds.)
Biodiversity and Ecology of Aquatic Environments
, pp.
1
28
.
Narendra Publishing House
,
Delhi
.
Lakra, W. S., Sarkar, U. K., Kumar, R. S., Pandey, A., Dubey, V. K., Gusain, O. P.,
2010
.
Fish diversity, habitat ecology and their conservation and management issues of a tropical river in Ganga basin in India
.
Environmentalist
DOI .
Mesa, L. M.,
2010
.
Hydraulic parameters and longitudinal distribution of macroinvertebrates in a subtropical Andean basin
.
Interciencia
,
35
,
759
764
.
Milesi, S. V., Biasi, C., Restello, R. M., Hepp, L. U.,
2009
.
Distribution of benthic macroinvertebrates in subtropical streams (Rio Grande do Sul, Brazil)
.
Acta. Limnol. Bras.
21
(
4
),
419
429
.
Minshall, G. W.,
1988
.
Stream ecosystem theory: a global perspective
.
J. North Am. Benthol. Soc.
,
7
,
263
288
.
Miserendino, M. L., Pizzolon, L. A.,
2003
.
Distribution of macroinvertebrate assemblages in the Azul- Quemquemtreu river basin, Patagonia, Argentina, New Zealand
.
J. Mar. Freshwat. Res.
37
,
25
39
.
Mishra, A. S., Nautiyal, P.,
2011
.
Factors governing longitudinal variation in benthic macroinvertebrate fauna of a small Vindhyan river in Central Highlands ecoregion (Central India)
.
Trop. Ecol.
,
52
(
1
),
103
112
.
Mishra, A, S., Nautiyal P.,
2013a
.
Longitudinal distribution of benthic macroinvertebrate assemblages in a Central Highlands River, the Tons (India)
.
Proceedings of National Academy of Sciences (Biological Sciences), Allahabad, Sci. (January–March 2013)
83
(
1
):
47
51
,
U.P. DOI
.
Mishra, A. S., Nautiyal, P.,
2013b
.
Functional composition of benthic macroinvertebrate fauna in the north flowing Vindhyan rivers (Bundelkhand, Central India)
.
J. Threat. Taxa,
5
(
13
):
4752
4758
.
National Academy of Agricultural Sciences (NAAS)
,
2004
.
Impact of Inter River Basin Linkages on Fisheries
.
Policy Paper No. 28
, National Academy of Agricultural Sciences,
India
.
National Water Development Agency (NWDA)
,
2006
.
Terms of Reference for Preparation of the Detailed Project Report: Interlinking of Rivers
. http://nwda.gov.in/writereaddata/linkimages/9[
2007
March
15
].
Nautiyal, P., Mishra, A. S.,
2012
.
Longitudinal distribution of benthic macroinvertebrate fauna in a Vindhyan river
.
Internat. J. Environ. Sci.
1
(
3
),
150
158
.
Nautiyal P., Mishra, A. S.
2013
.
Variations in benthic macroinvertebrate fauna as indicator of land use in the river Ken (Central India)
.
J. Threat. Taxa
, India.
5
(
7
),
4096
4105
.
Nautiyal, P., Verma, J.,
2009a
.
Taxonomic richness and diversity of the epilithic diatom flora of the two biogeographic regions of the Indian Subcontinent
.
Bull. Natn. Inst. Ecol.
20
,
15
21
.
Nautiyal, P., Verma J.,
2009b
.
Longitudinal variation in the composition of Diatom flora in a Central Highlands River of Vindhya region, the Ken
.
Bull. Natn. Inst. Ecol.
20
,
22
28
.
Nautiyal, P., Kala, K., Nautiyal, R.,
2004a
. A preliminary study of the diversity of diatoms in streams of the Mandakini basin (Garhwal Himalaya). In: M. Poulin, (Ed.)
Proceedings 17th International Diatom Symposium
, pp.
235
269
.
Ottawa, Biopress Limited
,
Bristol
.
Nautiyal, P., Nautiyal R., Kala, K., Verma, J.,
2004b
.
Taxonomic richness in the diatom flora of the Himalayan streams (Garhwal, India)
.
Diatom
.
20
,
123
132
.
Nautiyal, P., Dwivedi, A. C., Joshi, K D.,
2011
.
Sex ratio and structure of certain cyprinids of Vindhyan region in Central India
.
J Inland Fish Soc, India
43
(
1
),
77
82
.
Nautiyal, P., Nautiyal, R., Semwal, V. P., Mishra, A. S., Verma, J., Uniyal, D. P., Uniyal, M., Singh, K. R.,
2013
.
Ecosystem health indicators in the Ganga Basin (Uttarakhand, India): Biodiversity, spatial patterns in structure and distribution of benthic diatoms, macroinvertebrates and ichthyofauna
.
Aquat. Ecosyst. Health Mgmt.
16
(
4
),
362
373
.
Nautiyal, R., Nautiyal, P.
1999a
. Spatial distribution of diatom flora in Damodar river system of Chhota Nagpur. In: M. Joseph (Ed.),
The Fourth Indian Fisheries Forum
. Kochi p.
17
22
.
Asian Fisheries Society (Indian Branch)
,
Mangalore
.
Nautiyal, R., Nautiyal, P.,
1999b
. Altitudinal variations in the pennate diatom flora of the Alaknanda-Ganga river system in the Himalayan stretch of Garhwal region. In: Mayama, S., Idei, M., Koizumi, I. (Eds.),
Proceeding 14th International Diatom Symposium
, pp.
85
100
.
Tokyo, Koeltz Scientific Books
,
Koenigstein
.
Naymik, J., Pan, Y., Ford, J.,
2005
.
Diatom assemblages as indicators of timber harvest effects in coastal Oregon streams
.
J. North Am. Benthol. Soc.
,
24
,
569
584
.
Pan, Y. D., Stevenson, R. J., Hill, B. H., Kaufmann, P. R., Herlihy, A. T.,
1999
.
Spatial patterns and ecological determinants of benthic algal assemblages in Mid-Atlantic streams, USA
.
J. Phycol.
35
,
460
68
.
Pan, Y. D., Stevenson, R. J., Hill, B. H., Herlihy, A. T.,
2000
.
Eco-regions and benthic diatom assemblages in mid Atlantic High land stream, USA
.
J. North Am. Benthol. Soc.
,
19
,
518
540
.
Richards, C., Haro, R. J., Johnson, L. B., Host, G. E.,
1997
.
Catchment and reach-scale properties as indicators of macroinvertebrate species traits
.
Freshwat. Biol.
,
37
,
219
230
.
Rodgers, W. A., Panwar, H. S., Mathur, V. B.,
2002
.
Wildlife Protected Area Network in India: A Review (Executive Summary), Wildlife Institute of India, Dehradun
.
Sarode, P. T., Kamat, N. D.,
1984
.
Freshwater diatoms of Maharashtra
.
Saikripa Prakashan
,
Aurangabad
.
Singh, H. R., Nautiyal, P., Dobriyal, A. K., Pokhriyal, R. C., Negi, M., Baduni, V., Nautiyal, R., Agarwal, N. K., Nautiyal, P., Gautam, A.,
1994
.
Water quality of the river Ganga (Garhwal-Himalaya)
.
Acta. Hydrobiol.
34
(
1
),
3
15
.
Sivaramakrishnan, K. G., Venkataraman, K., Sridhar, S., Marimuthu, U.,
1995
.
Spatial patterns of benthc macroinvertebrate distribution along river Kaveri and its tributaries (India)
.
Intern. J. Ecol. Environ. Sci.
,
21
,
141
161
.
Stancheva, R., Mancheva, A., Ivanov P.,
2007
Taxonomic composition of the epilithic diatom flora from rivers Vit and Osum, Bulgaria
.
Phytol. Balcan.
13
(
1
),
53
64
.
Strayer, D. L., Beighley, R. E., Thompson, L. C., Brooks, S., Nilsson, C.,
2003
.
Effects of land cover on stream ecosystems: roles of empirical models and scaling issues
.
Ecosystems
,
6
,
407
423
.
Subramanian, K. A., Sivaramakrishnan, K. G.,
2005
.
Habitat and microhabitat distribution of stream insect communities of the Western Ghats
.
Curr. Sci.
89
(
6
),
976
987
.
Ter Braak, C. J. F., Smilauer, P.,
2002
.
CANOCO reference manual and CanoDraw for Windows User's guide: Software for Canonical Assemblage Ordiantion (version 4.5)
.
Microcomputer Power
(
Ithaca, NY, USA
).
Townsend, C. R., Hildrew, A. G., Schofield, K.,
1987
.
Persistence of stream invertebrate communities in relation to environmental variability
.
J. Anim. Ecol.
56
,
597
616
.
Vannote, R. L., Minshall, G. W., Cummins, K. W., Sedell, J. R., Cushing, C. E.,
1980
.
The River continuum concept
.
Can. J. Fish. Aquat. Sci.
37
,
130
137
.
Verma, J., Nautiyal, P.,
2009
.
Longitudinal patterns of distribution of epilithic diatoms in a lesser Himalayan Stream
.
J. Hill Res.
,
22
(
2
),
105
109
.
Verma, J., Nautiyal P.,
2010
.
Floristic Composition of the epilithic diatoms Central highland region of Indian Subcontinent: Thalassiosiraceae, Fragilariaceae, Eunotiaceae and Achnanthaceaea
.
J. Indian Bot. Soc.
89
(
3/4
),
397
400
.
Vilbaste, S.
2001
.
Benthic diatom communities in Estonian rivers
.
Boreal Env. Res.
6
,
191
203
.
Vinson, M. R., Hawkins, C. P.,
1998
.
Biodiversity of stream insects: Variation at local, basin and regional scales
.
Ann. Rev. Entomol.
43
,
271
193
.
Walsh, C. J., Sharpe, A. K., Breen, P. F., Sonneman, J. A.,
2001
.
Effects of urbanization on streams of the Melbourne region, Victoria, Australia. I. Benthic macroinvertebrate communities
.
Freshwat. Biol.
46
,
535
551
.
World Resources Institute (WRI)
,
1992
. World Conservation Union (IUCN), and United Nations Environment Programme (UNEP).
1992
.
Global Biodiversity Strategy: Guidelines for Action Save, Study, and Use Earth's Biotic Wealth Sustainably and Equitably
.
Washington, D.C.
:
World Resources Institute
. wiienvis.nic.in/database/htmlpages/bioprovincemap.htm
WII, 2000
.

Supplementary data