Hydropower development in the Himalaya: identifying critical river stretches

Two hydroelectric projects are operational in the upper and lower stretches of the river Alaknanda near Lambagarh and Srinagar that are respectively ~ 140 km apart. The health of the river ecosystem in the flowing stretches between and below these projects was bio-assessed, based on widely used diatom indices and van Dam ecologic values computed by OMNIDIA software. Samples were collected from Birahi (S3), the semi-natural stretch ~ 53 km downstream of the powerhouse, Narkota (S5), the impacted stretch ~ 27 km upstream Srinagar dam and Bagwan (S8), the highly regulated stretch ~ 22 km downstream Srinagar. IPS and IBD revealed mild pollution in the upper and lower stretch compared to relatively elevated pollution at S5. However, TDI indicates moderate organic pollution at S3 and S5, whereas the IDS/E indicates low degradation at all stations. The prevalence of tolerant N-autotroph, β-mesosprobe, mesotrophic, and aquatic to aerophilic categories of van Dam environmental variables indicate that a certain level of impact already persists due to organic-cum-nutrient load and perturbed flow regime. The prevalence of eutrophic and moderate O2 categories shows that the ecosystem is in a critical state at S5, as also observed for species diversity. Based on the dominant categories, the river stretch was classified as “polyoxybiontic-cum-mesotrophic” at S3 and S8 while “moderate O2-cum-eutrophic” at S5, implying higher organic and nutrient load in the river section at S5.


Introduction
The mountains are a favorable terrain for hydropower development. The Indus, Ganga, and Brahamaputra river systems are visualized as a vast potential in the Indian Himalayan region (IHR). The government of India aims to construct 292 dams throughout IHR, which is projected to meet ~ 6% national energy needs by 2030 (Grumbine and Pandit 2013). The Ganga river system constituted by the tributaries Alaknanda and Bhagirathi, each ~ 200 km long, accounts for 24% of the hydropower in IHR.
The 400 MW and 330 MW hydroelectric project (HEP) at Vishnuprayag (VHEP) and Alaknanda hydropower corporation Ltd. (SHEP) near Lambagarh and Srinagar respectively have modified the Alaknanda almost along its whole length except for ~ 30 km stretch between the holy pilgrimage shrine of Badrinath at ~ 3100 m asl and VHEP. Based on a steep fall of ~ 2100 m in the upper 100 km length of the river compared with a gentle fall of ~ 500 m in the lower 100 km, they were considered as upper and lower stretch, alias ecosystems. The upper stretch is characterized by shooting currents that cause considerable scouring and hence support lower densities of benthic producer and consumer communities (Mishra et al. 2013;Nautiyal unpublished 1 ). The declining gradient in the lower stretch allows the development of relatively higher densities and diversity of benthic algae (Nautiyal 2005). Consequently, snow trout fishery becomes an economic activity in the lower stretch, characterized by wider valleys that allow the operation of the indigenous fishing traps and nets. Besides a small quantity of detritus, adequate primary production by diatom-dominated benthic communities ) is essential to sustain this fishery , as snow trouts are herbivores (primary consumers), a grazing-chain element.
Thus, the lower stretch is a subject of concern and deeper scrutiny. Past studies on biotic communities in relation to HEP focused only on river sections just upstream (u/s) and downstream (d\s) of the dam and barrages (Nautiyal et al. 1988;Badoni et al. 1997;Rajvanshi et al. 2012). Studies in the flowing sections of the river between two or more HEP's are scarce (Smolar and Mikos 2014;Kumar and Nautiyal 2019). The HEP impacts get mingled with those from the habitations and human activities along the river banks and basin. Hence, this study examines the impact of hydropower and anthropogenic activities (largely tourism, pilgrimage, habitations) so as to identify and designate current critical stretches in the flowing sections of the river. This would serve as a measure of river health impacted by HEP and pollution from the anthropogenic sources.
Hence, the assessment was restricted to the use of ecologic criteria (van Dam et al. 1994) and some frequently used diatom indices for ecological state and organic pollution respectively in the serially impounded river Alaknanda. Selected indices were used; Pollution Sensitivity Index IPS (CEMAGERF 1982), Biological Diatom Index IBD (Coste et al. 2009), Trophic Diatom Index TDI (Kelly and Whitton 1995), Index of Saprobity Eutrophication IDS/E (Leclercq and Rosengarten 2008) with ample share (as %) of species in the examined samples; IPS (89.5 to 93.2), IBD (64.9 to 77.3), TDI (73.7 to 79.5), and IDS/E (36.8 to 40.9). The indices for species diversity, richness, and evenness were calculated to understand the impact on biodiversity. The van Dam criteria have been successfully used in the Himalayan rivers and streams (Cantonati et al. 2001;Nautiyal and Nautiyal 2018;Singh et al. 2018;Bartwal et al. 2020) as the large number of speies reported from the Himalayan rivers (Nautiyal et al. 2004a, b) are similar to those present in the database of the OMNIDIA software, indicating its suitability for present investigations.

Material and methods
The area of influence (AoI) for SHEP extends for ~ 27 km consisting of ~ 20 km long impounded section u/s of the Srinagar dam site and a short ~ 7 km F-D (flow-deficient) d/s section from the dam to PH. Beyond PH, the remaining ~ 30 km stretch till Devprayag is the F-R (flow-regulated) section. Since the interest lies in the flowing sections of the lower-half stretch, that are distantly located from the modifying structures (as long distances allow the river to regain the ecosystem integrity), S3, S5 above SHEP impoundment and S8 in the F-R section below SHEP were considered for this study (Fig. 1), which is part of a larger study funded by National Mission for Himalayan Studies, GBPNIHESD, Kosi-Katarmal, Almora. i) Birahi (S3): This station outside VHEP-AoI lies in the beginning of the lower Alaknanda river ecosystem. It carries F-R discharges from VHEP but is considered here as a reference location and a surrogate for unhindered but impacted flow representing a semi-natural zone as it is ~ 53 km d/s of the VHEP-PH and is far u/s of the SHEP impoundment. ii) Narkota (S5): This location is ~ 1.5 km u/s of the tail of SHEP impoundment and 73 km d/s of S3. iii) Bagwan (S8): This represents the F-R stretch and lies in the middle of the Srinagar-Devprayag section.
The OMNIDIA ver 6.0.8 (Lecointe et al. 1993) software requires the valve count of each diatom species to compute the richness species diversity, evenness, diatom indices, and van Dam ecologic criteria to assess the river health based on benthic diatom communities in the selected sections. The acid-peroxide-treated diatom Naphrax mounts from each sampling station were examined using three replicates collected from 3 × 3 cm 2 surface area of cobble substrate in varying flows from the riffle-pool-riffle sequence at monthly intervals. Species count was generated from each replicate and pooled for the year 2016-17. Inventories were prepared and imported in the software. The following indices serving as a measure of organic pollution and eutrophication were computed. Since these indices are used widely, they were selected so as to make the results comparable. Also, the highest number of species (exceeding 70%) in our samples is being used by these indices, while other indices in OMNIDIA use less than 50% species from the study area except ROTT-TI, EPDI, and CEE that have around 60% species in their list. Only IDSE which uses a very less share of species in the sample needs to be used judiciously. Others appear to be robust.
The ecologic values for the categories with higher shares of each characteristic were recorded to determine the ecological state vis-a-vis river health. Since each environmental variable, excluding unknown, has 4 to 7 categories, only those with the highest share of the variable were considered for the present analysis. The order of all categories extends from excellent to good and while bad to worse for O 2 requirements, saprobity, trophic state, moisture-aerophily, and N 2 uptake. pH is one instance where acidic or highly alkaline conditions do not represent a healthy river.

Diatom indices-based assessment
IPS shows almost low pollution at S3, which increases further at S5 and S8. IBD also indicates an increase at S5 compared to S3 and S8. IDS/E shows low degradation at all stations. TDI indicates that organic pollution is likely to contribute significantly to eutrophication at S3 and S5 and some evidence of organic pollution at S8. The low value of each index at S5 indicates the stressed state of this river section (Fig. 2).

van Dam-based assessment
The dominant categories (those with relatively higher shares) of van Dam environmental variables were found to indicate river health (Table 2). Tolerant N-autotroph and β-mesosprobe categories consistently dominate at all stations. The alkaliphilic and occasionally aerophilic categories dominate only at S5 and S8, while neutrophilic and aquatic-aerophilic are restricted to S3. The polyoxybiontic and mesotrophic state prevail in discontinuity (S3, S8), while moderate O 2 and eutrophic categories prevail at S5. Based on the dominant categories, the river stretch was classified as "polyoxybiontic-cum-mesotrophic" at S3 and S8 while "moderate O 2 -cum-eutrophic" at S5, implying higher organic and nutrient load in the river section at S5.   River regulation is a practice to maintain a constant discharge for generating the required hydropower. Since the discharge is inherent to a river, its modification to suit the generation of hydropower to full potential, impairs the ecosystem. Each hydroelectric project differs from the other by its design and hence the impacts cannot be generalized. Below the PH, the HEP either terminates into another project immediately or after flowing for some distance. In some instances, the river may appear to be free-flowing but actually carries regulated discharges. The F-R river sections with modified flows are crucial for the health of serially impounded river ecosystem like Alaknanda. In these sections, the flows maintain the O 2 requirements of the biota and replenish the necessary nutrients to sustain the community of primary producers. The O 2 profile of a river is primarily governed by water temperature, followed by discharge and current velocities that facilitate mechanical mixing required for diffusion through air-water interface. Despite the difference in the flows, the O 2 levels varied barely among the sections. O 2 maxima are almost similar in both F-R and impounded sections, but the minima are lower in the impounded and subsequent F-R section (S3, 7.3-8.4; S5, 6.8-8.8; S8, 6.6-8.6 mg l −1 ). This regime is attributed to increasing water temperature (S3, 5-12.5; S5, 8.12-16.2; S8, 9.7-19 °C) and decreasing current velocity profiles in months other than monsoon (S3, 0.3-0.8; S5, 0.4-0.7; S8, 0.3-0.6 m s −1 ). In contrast, the van Dam criteria show moderate O 2 condition (50% O 2 saturation) at S5 compared to F-R sections of VHEP and SHEP that show polyoxybiontic category (100% O 2 saturation) at S3 and S8. These physicochemical characteristics show no apparent impact, but examination of their temporal rhythms reveals an intermittent, though minor increase and decrease from month to month, especially in current velocity and O 2 profile (Supplementary Table 1). These are attributable to daily flow modulations for power generation. In contrast, earlier, a gradual increase or decrease was noted for most of the characteristics in the Alaknanda (Nautiyal 1984).
The flows have other important functions too. In presence of constant discharge, the effluent carrying organic and nutrient load develops a fertilizer potential which modifies the community structure. However, some wastes are either flushed or take more time to decay and hence accumulate over a period of time, which can lower oxygen levels as evidenced at S5 and influence the various properties of a community, such as biodiversity. Observations on biodiversity reveal a decrease in the community features (species diversity, evenness, richness) and an overwhelming abundance of C. placentula var. euglypta indicating an adverse impact at S5. The increase of these features at S8 possibly explains the role of intermediate levels of disturbance (Connel 1978). It augurs that the water current velocities and O 2 profile are impacted temporally in the form of interrupted gradient caused by mild increase or decrease. The bio-assessment by van Dam criteria registers a detrimental change in the categories of O 2 requirements and trophic state. Thus, it is a sensitive tool that should be used along with physico-chemical characteristics, as also opined by Borja et al. (2008).
This bio-assessment uses diatom indices and ecological criteria (van Dam et al. 1994) to measure the impact vis-a-vis response of diatom (producer) communities to HEP operations. All the indices show that the river is more stressed at S5, which is also indicated by the ecological values. However, IDS/E reveal low degradation at all stations possibly attributed to a low share of species having a preference for degraded conditions. In this study, the mesotrophic state prevails in F-R sections despite low and high flows, as the river keeps flowing and prevents accumulation, in contrast to the tail of the impounded section causes prolonged water retention, resulting in a eutrophic state. Nogueira et al. (2010) opined that rising water temperatures, prolonged water retention time, and nutrient enrichment associated with dam operation can induce phytoplankton blooms that lead to the eutrophication of the dam reservoir. The other van Dam variables viz. tolerant N-autotroph, β-mesosaprobe (BOD-2-4 mg l −1 ) persist in the whole stretch of the Alaknanda under observation, irrespective of the flow regimes. Contrarily, neutrophilic and aquatic to aerophilic occurred at S3, while the alkaliphilic and occasionally aerophilic categories persist at S5 and S8. Contrasting features were recorded for mesotrophic and polyoxybiontic categories as they persist at both ends of the stretch (S3 and S8) but turn eutrophic and moderate O 2 (50% saturation) at S5, indicating a perturbed river stretch. Alkaliphilous, tolerant N-autotroph, high O 2 requirements, and β-mesosprobe state have been observed in many rivers and streams of the Himalaya (Nautiyal unpublished 2 ).
The van Dam ecological criteria, diatom indices, and diversity show that the continuum is disrupted near the tail of SHEP impoundment as the polyoxybiontic and mesotrophic category occur in a discontinuity at S3 and S8, while moderate O 2 (50% O 2 saturation) and eutrophic categories prevail at the tail of impoundment (S5) and hence identified to be "critical." Among the species contributing to this category, C.p.var. euglypta occurred in overwhelming abundance. This study identifies the ailing river section that requires restoration of ecosystem health and suggests periodic monitoring of this and other such river sections.

Conclusions
The flow modifications seem to have homogenized the river. Considerable share (not mere presence) of these categories in merely ~ 200 km length of the river from its source shows that the river is in a state of low-level impact, as a pristine river should contain a greater share of sensitive N-autotroph, neutrophilic, oligotrophic, polyoxybiontic, oligosaprobic, and aquatic categories of respective variables. The river gets perturbed at S5 as the tail of the impoundment swells and shrinks according to the discharge requirements for power generation, thus impeding the flow and creating eutrophic conditions. Evidently, the static hydraulic and hydrological regime around the tail of the impoundment aggravates the impact of organic and nutrient load due to the lack of flushing. Moderate O 2 category was attributed to a decrease in mechanical mixing. The river stretch was hence classified as "polyoxybiontic-cum-mesotrophic" at the ends of the selected stretch (S3 and S8) while "moderate O 2 -cum-eutrophic" at S5, implying higher organic and nutrient load in the river section at the tail of SHEP impoundment, as also reflected by the pollution and diversity indices. Thus the river stretch around S5 was considered critical.