Spatial and seasonal variations in nutrient load and trophic status of Ganga and Yamuna rivers in Uttar Pradesh, India

The quality of the riverine ecosystem is significantly dependent upon external factors (agricultural, industrial, and municipal waste), apart from natural factors. The source of nutrients in the rivers can either be allochthonous (weathering, precipitation, and domestic and agricultural runoff) or autochthonous (instream production and mineralisation of organic matter) (Sharma et al. 2017; Jargal et al. 2021). Rapid population growth has enhanced industrialisation and urbanisation due to changes in land use patterns, fossil fuel combustion, and residential runoff, disturbing the river's nutrient balance. Around 10%–90% of the mean annual flow in the rivers is contributed by treatment plants. Modern agricultural practices with high inputs of fertiliser and other agrochemicals have further aggravated the issue of nutrient pollution in the river systems (Bellos et al. 2004; Haque 2021). The last two decades have witnessed the unfettered use of pesticides and fertilisers, increasing the river's nutrient content (particularly nitrate, potassium, and phosphate). Phosphorus and nitrogen loading in aquatic systems is a major concern, particularly in areas where heavy urban or agricultural land use leads to contamination through diffuse nutrient inputs. Managing nutrients from diffuse sources in riverine systems is more difficult as it requires a combination of mitigation strategies at both the catchment and reach scales (Weigelhofer et al. 2018). This excessive nutrient loading may lead to ecological problems, including nutrient imbalance (C/N/P/Si) and eutrophic conditions and hypoxia zones (Rajmohan & Elango 2005; Ravi et al. 2021).

The Ganga and the Yamuna rivers are northern India's two most significant rivers. Along the banks of these rivers are highly populated cities, and the residents of these cities rely on these freshwater sources for their needs, including for household, agricultural, and industrial uses. Ganga is revered as sacred for its purity and capacity for self-cleansing, and the river's water is part of every Hindu tradition and ceremony. However, it was listed as one of the world's top five most polluted rivers due to anthropogenic activities (Kesari et al. 2022). The river Yamuna is one of the Ganga's most polluted tributaries and receives over 76% of the nation's total pollution load from Delhi-NCR Press Trust of India (2018), effectively converting the river into a ‘sewage drain’ (CPCB 2021a, 2021b). There are more than 100 industries situated along the Ganga River, among which 68 are designated as grossly polluting (NMCG-NEERI 2017). Surface waters receive between 3% and 20% of the phosphorus and 18% of the nitrogen applied to croplands. The global load of anthropogenic nitrogen to freshwater systems from agricultural lands, including leaching and runoff, was projected to be 24.4 × 10⁶ t N per year (Yadav & Pandey 2017). Most freshwater bodies have P-limited conditions that affect the aquatic balance by disturbing the structure and diversity of plants and other micro-organisms vital for maintaining a healthy ecosystem (Wade et al. 2004). Hence, nutrient pollution remains a significant concern as it can create problems with alterations in the food chain, deterioration of water quality, and deleterious effects on the health of living beings (Bende-Michl & Hairsine 2010). To combat this issue of river pollution, the Indian government has started river protection efforts, including the Ganga Action Plan (GAP), Namami Gange, and Yamuna Action Plan (YAP).

Programmes like GAP I and GAP II were introduced between 1985 and 2015, and a total of Rs. 4,000 crores were provided for these action plans over 30 years. However, a new regeneration strategy was needed to maintain the Ganga River's cleanliness. The Namami Gange Programme was started to view the basin as a whole rather than the Ganga as a separate river. An additional Rs. 30,235 crores budget has been allocated to revitalise this river (Balkrishna et al. 2022). Another major river restoration initiative, the YAP, was started to clean the Yamuna River. Action Plans I, II, and III constitute the YAP, a bilateral agreement between the governments of Japan and India. YAP-III is currently being executed with an allocated budget of Rs. 1,656 crores (Srivastava & Prathna 2022) to abate pollution in the river Yamuna. Despite these initiatives, the quality of both these rivers does not seem to have achieved the desired results, as the deterioration of river water quality by nutrient pollution is still a major concern (NMCG-NEERI 2017). It is, therefore, necessary to monitor the nutrient loading of the two major rivers (Ganga and Yamuna) in Uttar Pradesh, has an intensive agricultural region with little information on the trophic condition of river bodies. Most research conducted along the river in Uttar Pradesh has concentrated on the pollution load related to physicochemical parameters, such as DO, BOD, heavy metals, and faecal coliform (NMCG-NEERI 2017; Paul 2017; Singh et al. 2022). There is little research in the literature on the relationship between nutrient load and the risk of eutrophication due to nutrient load and trophic status of aquatic systems. This study was carried out to bridge the knowledge gaps with the objective of (1) determining the spatial as well as seasonal variation in the dissolved nutrient concentration in the Ganga and Yamuna rivers; (2) analysing the nutrient chemistry along with the factors influencing it in the Ganga and Yamuna rivers using multivariate statistical analysis; (3) estimating the dissolved nutrient ratios and dissolved nutrient load in the Ganga and Yamuna rivers; and (4) determining the eutrophication potential of the Ganga and Yamuna rivers using the Indicator for Coastal Eutrophication Potential (ICEP) and Trophic State Index (TSI) values.

The Ganga basin is India's largest river system, covers 26% of the geographical area of India and is home to more than 600 million people, and accounts for more than 40% of the country's GDP (NMCG-NEERI 2017). It originates from the Gangotri glacier as Bhagirathi in Uttarakhand and descends from the mountainous region to the plains at Haridwar. It then turns southeast to Kanpur, the industrial city, and further to Prayagraj, where it is joined by the Yamuna River (Ganga's largest tributary). Here onwards, the river flows in an easterly direction to flow through Varanasi (Garg et al. 2020). Several tributaries join it in its 2,525 km-long course before finally draining into the Bay of Bengal (Purushothaman & Chakrapani 2007; Trivedi 2010). The basin is one of the largest alluvial plains formed mostly over the Quaternary period. The Ganga River, together with the Yamuna, erodes Himalayan sedimentary rocks depositing layers of sediment in the plains, creating a wide alluvial plain about a kilometre in thickness. With an erosion rate three times that of the Amazon and three and a half times the global average, the Ganga River comes in third place for sediment transport behind the Yellow and Amazon rivers (Subramanian et al. 1987; NMCG-NEERI 2017). Most of the region in this basin falls in the tropical climate zone, and around 85% of the rain is received from June to September (monsoon season). The annual average range of precipitation is from 600 to 1,200 mm, while the maximum temperature ranges from 35 to 40 °C (Sinha et al. 2017). The state of Uttar Pradesh contributes 54% of the total wastewater discharged and 76% of the BOD load in the river Ganga (CPCB 2013). Several drains, namely Golaghat Nala, Ranighat Drain, Wazidpur Drain, Sisamau Nala, Permiya Nala, and City Jail Drain join the river Ganga, discharging sewage with high concentrations of nitrate and ammonia into the river (Santy et al. 2020). The major drains in Prayagraj include the Arail Drain, Mahewa Drain, Fort Drain, Rasulabad Drain, Sadananda Ashram Drain, Jhusi Drain, and Chhatnag Drain, while in Varanasi, Ramnagar Drain, Varuna Drain, Assi Drain, Rajghat Drain, and Shivala Drain are the major drains that discharge millions of litres of wastewater into the Ganga River along with the nutrients that degrade river water quality (CPCB 2021a, 2021b; Jamal & Sen 2022). The population density of Kanpur is 1,452/km² (per sq. km.), Varanasi is 2,395/km², and Prayagraj is 1,087/km² (Census of India 2011). The land use of Kanpur mainly includes industrial and public utilities. Agriculture makes up most of the basin's land use, at 65.57%, forest at 16%, and built-up at 4.28% (WRIS 2014). The major portion of Varanasi and Prayagraj was under agricultural use until the last two decades but has recently increased in the built-up area (Jaiswal & Verma 2013; Rousta et al. 2018). The sampling sites along the Ganga River were Kanpur (K), Prayagraj (GPY), and Varanasi (V). The water samples at Prayagraj were collected 2 km before the Ganga and Yamuna confluence points for both rivers. In Prayagraj, the water samples collected from the Ganga and Yamuna rivers are represented as GPY and YPY in this study.

Eighteen water samples were taken from the Ganga and the Yamuna rivers (n = 36) in acid-cleaned polyethylene bottles during the dry (pre-monsoon, April 2022) and wet (monsoon, August 2021) seasons. BOD (biochemical oxygen demand) analysis samples were collected in stoppered glass bottles. Some of the parameters, such as pH, EC (electrical conductivity), TDS (total dissolved solids), and DO (dissolved oxygen), were analysed on-site by a soil and water analysis kit (Labtronics, model- LT68). All the samples were collected, preserved, and carried to the laboratory according to the protocol given in APHA (2005). Water samples were filtered with a nylon filter paper of 0.45 μm and analysed for NH₄-N (phenate method), NO₃-N (brucine method), PO₄-P (ascorbic acid method), and SiO₂-Si (molybdosilicate method) by using a Systronics spectrophotometer. BOD was determined as a difference in DO levels over five days (temperature maintained at 20 °C). All the samples were analysed using the protocols given in APHA (2005), and the values were subsequently compared with their standard limits given by the Bureau of Indian Standards (BIS 2012) and World Health Organisation (WHO 2004). MS Excel 2019 was used for One-Way ANOVA (Analysis of Variance) and Pearson's correlation coefficient analysis. ANOVA assessed the variation in dissolved nutrient concentration in the Ganga and Yamuna rivers. The null hypothesis will be rejected if F_calculated > F_critical and p-value < 0.05. The relationship among the water quality parameters was ascertained using Pearson's correlation coefficient based on the correlation coefficient's value (r). A correlation value greater than 0.5 indicates a significant relationship and points to a similar source of origin or mode of transit within the watershed (Shrestha 2021). The Statistical Package for Social Sciences (SPSS), version 10.0, was used for the factor analysis.

The concentration of measured parameters in the Ganga and Yamuna rivers is given in Table 1. In the dry and wet seasons, the water samples in the Ganga River were mostly alkaline, with observed mean values of 8.2 and 8.72, respectively. In the case of the Yamuna River, it was 8.51 and 7.83 in the dry and wet seasons, respectively (Table 1). Around 42% of the samples in the Ganga and 33% in the Yamuna River exceeded the BIS (2012) limit of 8.5 in Uttar Pradesh. Compared with the solids in surface water bodies, the large water supply does not allow pH to fluctuate drastically, maintaining the pH levels between 7 and 9 (Olalekan et al. 2023). A study by Maji & Chaudhary (2019) reported pH values between 7.9 and 8.7 for the Ganga River in Allahabad, Uttar Pradesh. The EC values varied from 380 μS cm⁻¹ (K2) to 560 μS cm⁻¹ (V5) in the dry period and from 230 μS cm⁻¹ (K2) to 350 μS cm⁻¹ (GPY3) in the wet period in the Ganga River. In the Yamuna River, EC varied from 430 μS cm⁻¹ (YPY2) to 650 μS cm⁻¹ (A4) in the dry season and from 410 μS cm⁻¹ (M1) to 960 μS cm⁻¹ (YPY1) in the wet season (Table 1). The EC value for collected samples was within the recommended limit of BIS (2012). However, two samples (from Agra and Prayagraj each) exceeded this limit in the Yamuna River wet season, indicating contribution from the weathering process and the anthropogenic input. The conductivity of water depends upon several factors, such as number, type, size, degree of hydration, and mobility of ions. It is directly proportional to the TDS as the higher the TDS, the greater the number of ions available for conductance (Jawad et al. 2023). The study assessing the water quality of the Yamuna River in Delhi by Sharma et al. (2017) reported EC values of the Yamuna River in the range of 97–2,846 μS cm⁻¹ due to increased load from natural as well as anthropogenic sources. The mean value of TDS in the Ganga River was 358.33 mg L⁻¹ in the dry and 170.61 mg L⁻¹ in the wet season. In the Yamuna River, the mean TDS value was 519.44 and 373.33 mg L⁻¹ in dry and wet seasons, respectively (Table 1). All the samples of the Ganga River were under the permissible limit of BIS (2012), but around 36% of the samples of the Yamuna River exceeded the limit of 500 mg L⁻¹. Rahman et al. (2021) reported the average value of TDS in the northern floodplains of the Ganga River to be 587 mg L⁻¹ due to the dissolution of cations and anions from the weathering process as well as anthropogenic inputs.

DO is a significant parameter as it supports aquatic life, which removes organic pollution from surface water. It also plays an important role in determining water's corrosiveness and oxidation of inorganic compounds (Wang et al. 2022). The DO concentration was observed to be less than the permissible limit (6 mg L⁻¹) in both seasons for both rivers, indicating the deficiency of oxygen, which subsequently negatively affects aquatic life. It showed the minimum value at Kanpur (K5) and maximum value at Prayagraj (GPY6) in both dry and wet seasons in the Ganga River. In the Yamuna River case, the minimum DO concentration was found at Agra and the maximum at Prayagraj in both seasons (Table 1). Kanpur and Agra are the industrial cities of Uttar Pradesh, where most industries are leather-based. These industries, on average, discharge around 22.1 MLD of wastewater into the rivers, reducing the river water quality (CPCB 2009; MSME 2023). BOD is indicative of the amount of oxygen required by microbes for the degradation of organic matter. The concentration of BOD in the Ganga River was above the permissible limits (2 mg L⁻¹) in all samples taken during the dry season. In the wet season, about 72% of the samples exceeded the BOD limits, indicating that the organic load from the municipal and agricultural sectors was present in the river system. In the case of the Yamuna River, average BOD values exceeded the permissible limit in both dry and wet seasons (Table 1). BOD values decreased by 2.13 times in the dry and 1.14 times in the wet season in the downstream region of the Yamuna River. This decrease in BOD value indicates a dilution of the Yamuna River water as the Chambal River merges with it at Etawah district (CPCB 2006). The COD values exceeded the permissible limit (10 mg L⁻¹) given by WHO in the dry season, and 61% of the samples in the wet exceeded the limits in the Ganga River. A rise of around 1.2 times in the COD values was observed in the downstream region of the Ganga River, indicating the contribution from anthropogenic sources in the catchment area (NMCG-NEERI 2017). The COD values showed high concentrations, particularly in the dry season, reaching a maximum value of 72 mg L⁻¹ at both Mathura (M3) and Agra (A4) (Table 1). Such high COD values indicate high organic load in both rivers from domestic, agricultural, and industrial sectors. Parmar & Singh (2015) reported a BOD value of 58 mg L⁻¹ (maximum) and a DO value of less than 4 mg L⁻¹ in the Delhi-NCR stretch of the Yamuna River, indicating unsuitable conditions for aquatic life.

Nitrogen and phosphorus are the two limiting macronutrients responsible for plant and animal cell growth. The Ganga River receives significant nutrients from both point and nonpoint sources. According to studies, point and nonpoint sources, including atmospheric deposition (either direct deposition on the surface of water or indirect through catchment deposition-coupled surface runoff), significantly regulates the input of nutrients like N and P (Pandey & Yadav 2015; Prajapati et al. 2020). The aquatic system's dissolved inorganic nitrogen (DIN) mainly comprises NH₄-N and NO₃-N. All the samples of the Ganga River had NO₃-N concentrations within the limits of BIS (45 mg L⁻¹) in both the dry and wet seasons. The average concentration of NO₃-N in the Yamuna River was significantly higher than in the Ganga River. However, the average NO₃-N concentration in the Ganga River was 208 and 125 times more than the value reported for the unpolluted river in the dry and wet seasons, respectively (Table 1). Kesari et al. (2022) reported nitrate concentrations between 2.52 and 4.92 mg L⁻¹ in the Prayagraj and Varanasi stretch of the Ganga River. Another study by Chaudhary et al. (2017) reported nitrate content of 1.8–15.8 mg L⁻¹ in the upper Ganga stretch (Table 2). This high nitrate content from the upper stretch of the river, along with instream nitrification and runoff from the catchment area, results in increased nitrogen load in the downstream section of the river runoff (Kumar et al. 2023). For the Yamuna River, the NO₃-N concentration was 486 and 226 times more than that of the unpolluted river in the dry and wet seasons, respectively. The high nitrate concentration in the dry season compared with the wet season indicates the accumulation of inorganic-N from agricultural sources and its subsequent dilution due to rainfall in the wet season. The high nitrate concentration is mainly derived from organic sources (Kurwadkar et al. 2020). Previous studies by Ahmed et al. (2020) and CPCB (2006) reported high nitrate concentrations ranging from 0.01 to 149.32 mg L⁻¹ in Mathura and negligible to 46.20 mg L⁻¹ from Yamunotri to Prayagraj, respectively (Table 2). The nitrate concentration observed for the Yamuna River in this study was comparable to the results reported in previous studies (Tables 1 and 2). The major sources of nitrate in this stretch include horticultural, sewage, agricultural, and other anthropogenic activities. The concentration of NH₄-N was the main concern as all the sites showed significantly high values exceeding the limit given by WHO (0.1 mg L⁻¹). The average NH₄ concentration was 17 and 12 times more than that of the unpolluted river in the dry and wet seasons in the Ganga River. The main reason for such a high concentration was the discharge of untreated sewage from the catchment area (Table 1). The value of NH₄-N obtained in the Ganga River in this study (0.15–0.38 mg L⁻¹) was significantly lower than that obtained by Saxena & Singh (2020) (0.19–1.50 mg L⁻¹) and Santy et al. (2020) (1–6.5 mg L⁻¹) in Varanasi and Kanpur, respectively (Table 2). In the Yamuna River, NH₄ concentration ranged between 0.29 and 0.51 mg L⁻¹, which was much less as compared with the reports of CPCB (2006) from Yamunotri to Prayagraj (trace– 43.34 mg L⁻¹) and CPCB (2021a, 2021b) from Agra to Prayagraj (0.2–39.9 mg L⁻¹). Both these studies included samples from the Delhi region, where NH₄ concentration is relatively higher due to sewage pollution than in other parts of the Yamuna River (CPCB 2021a, 2021b). The relatively lower value observed in this study can be due to the instream nitrification of the NH₄ process, which is also responsible for the high nitrate concentration observed in this study. NH₄-N concentration showed 22- and 18-times higher values than unpolluted rivers in the dry and wet seasons (Table 1). The concentration of PO₄-P ranges from 0.02 to 0.46 mg L⁻¹ in the Ganga River (Table 1). PO₄-P showed the minimum concentration at Varanasi and maximum concentration at Prayagraj in both the dry and wet seasons in the Ganga River. This value was lower than the value reported by Bowes et al. (2020) in the upper Ganga stretch (0.310–0.778 mg L⁻¹) and Kesari et al. (2022) in Varanasi and Prayagraj (0.92–1.82 mg L⁻¹). The major sources of PO₄-P reported in these studies were untreated sewage and agricultural and urban runoff in the Ganga River (Table 2). The concentration of PO₄-P ranges from 0.06 to 1.61 mg L⁻¹ in the Yamuna River (Table 1). PO₄-P showed minimum concentration at Mathura and Agra in the dry and wet seasons and maximum concentration at Agra and Prayagraj in the dry and wet seasons, respectively. Similar phosphate values were reported by Sharma et al. (2017) in their study of the Yamuna River from Barkot to Prayagraj (0.01–2.28 mg L⁻¹) and Shukla et al. (2016) in Kalpi (0.52–1.74 mg L⁻¹). The major contributors of phosphate in the Yamuna River were untreated sewage and agricultural runoff (Table 2). Although all the sites had PO₄-P concentration within the WHO (2004) limits, the average PO₄-P concentration was found to be 13 and 15 times more than that of the unpolluted river in the dry and wet seasons, in the Ganga, and 88 and 27 times more than that of the unpolluted river in the dry and wet season, respectively, in the Yamuna River. Phosphate in the river can be contributed by both natural sources (weathering of rocks) and anthropogenic sources (chemical fertiliser, phosphate mining, and sewage) (Ravi et al. 2021). The main factors contributing to high phosphorus content in the Himalayan rivers are high flow rate, high elevation, and rock weathering. When rocks weather, phosphorus is released as colloidal calcium phosphate and soluble alkali phosphates, most of which are delivered to the river waters. Additionally, phosphorus levels in river water rise because of anthropogenic inputs such as superphosphate fertiliser and alkyl phosphate detergents (Ramesh et al. 2015). Phosphate undergoes similar hydrolysis as carbonate, thereby increasing the alkalinity of water. The high phosphate concentration in both the rivers in this study can be an additional source of alkalinity in the study area (Chowdhary et al. 2020).

The silica concentration mostly ranges from 1 to 30 mg L⁻¹ in surface water. Generally, the high silica content in natural waters is accompanied by high pH, as the alkaline nature of water enhances the release of silica from the rocks. Water with high silica content is generally high in sodium and low in calcium, and the presence of calcium bicarbonate reduces the solubility of silica in water (Milne et al. 2014). SiO₂-Si concentration in the Ganga River was within the WHO (2004) limit of 100 mg L⁻¹ with a minimum value at Varanasi in the dry and Prayagraj in the wet season. It was maximum at Varanasi in both seasons. The values of SiO₂-Si obtained in this study in the Ganga River (0.67–29.9 mg L⁻¹) are significantly higher than in earlier studies published by Siddiqui & Pandey (2019) and Siddiqui et al. (2019). In the Yamuna River, the SiO₂-Si concentration was relatively higher (1.12–41.3 mg L⁻¹) than in the earlier reported values (0.01–2.28 mg L⁻¹) by Sharma et al. (2017). The high silica concentration in this study is mainly due to large floodplain areas in the Uttar Pradesh stretch of the river. In the Yamuna River, the minimum SiO₂-Si concentration was observed at Prayagraj and the maximum concentration at Mathura in both seasons. The average silica concentration observed in both Ganga and Yamuna rivers during this study was higher than the value reported for the unpolluted river (Table 1). Silica in river water is contributed by the weathering of rocks and is essential for the growth of primary producer Diatoms in the aquatic ecosystem (Ravi et al. 2021).

The seasonal variation was significant (p < 0.05) in pH (p = 0.01), EC (p = <0.0001), TDS (p = <0.0001), DO (p = <0.0001), BOD (p = 0.01), COD (p = <0.0001), NO₃-N (p = 0.01), NH₄-N (p = 0.01), and SiO₂-Si (p = <0.0001) with EC, TDS, BOD, COD, NO₃-N, NH₄-N, and SiO₂-Si showing higher mean concentrations in the dry and relatively lower mean concentrations in the wet season. The high concentration in the dry season may be attributed to the reduced volume of water in the rivers, while in the wet season, the pollutants are diluted due to the addition of rainwater (Woldeab et al. 2018). DO showed a higher value in the wet season, indicating the addition of fresh water from rainwater (Figure 2(a)). PO₄-P showed higher values in the wet season, indicating the presence of agricultural areas in the catchment as a contributing source of PO₄-P, along with soil flushing and transport of suspended particles during the wet season (Figure 2(a)).

The results of ANOVA indicated that spatial variation was not significant (p > 0.05) for EC (p = 0.50), TDS (p = 0.35), BOD (p = 0.07), COD (p = 0.06), NH₄-N (p = 0.31), and PO₄-P (p = 0.06) in the Yamuna River (Table 3). Still, the downstream areas showed higher values due to the added load from upstream regions (Figure 2(b)). The BOD and COD values showed higher values in the upstream area of Mathura (Figure 2(b)), which may be due to the addition of sewage waste along with the religious activities (offering flowers and other products in the Yamuna River) that occur regularly on the ghats as Mathura is a popular pilgrim spot attracting a large number of people from across the world (Bhargava 2006). The seasonal changes were significant (p < 0.05) in pH (p = <0.0001), TDS (p = <0.0001), NO₃-N (p = <0.0001), NH₄-N (p = <0.0001), PO₄-P (p = 0.01), and SiO₂-Si (p = 0.01) in the Yamuna River showing lower average values in the wet season indicating the dilution effect due to addition of rainwater in the monsoon period (Figure 2(b)). No significant seasonal change (p > 0.05) was seen in EC (p = 0.89), DO (p = 0.43), BOD (p = 0.18), and COD (p = 0.06) (Table 3), but the values of EC and DO were relatively higher in the wet season. BOD and COD showed higher concentrations in the dry season in the Yamuna River as the low flow in the dry season further enhanced the concentration of organic pollutants in the river (Figure 2(b)).

The factors affecting the nutrient chemistry in the Ganga and Yamuna rivers were determined using Pearson's correlation coefficient and R-mode factor analysis. The results obtained by Pearson's correlation for the dry and wet seasons in the Ganga River are presented in Table 4, and for the dry and wet seasons in the Yamuna River are given in Table 5, respectively. COD showed a strong positive correlation with TDS and BOD in the Ganga River in both the dry and wet seasons, indicating organic waste input in the river system from the catchment area. A negative correlation between pH and DO in the Ganga River showed a decrease in the algal population with increasing pH, thereby affecting the rate of photosynthesis (Dubinsky & Rotem 1974). In the wet season, a positive correlation between NH₄-N and NO₃-N may be attributed to the breakdown of organic matter from domestic and agricultural discharge into the Ganga River. A positive correlation was observed between EC and BOD in the dry season in the Yamuna River system. In the wet season, PO₄-P was positively correlated with EC and TDS, indicating the contribution from agricultural runoff.

The R-mode factor analysis results are presented in Tables 6 and 7 for the Ganga and Yamuna rivers. The analysis showed five factors in the Ganga River for the dry season. The first factor accounts for a 27.51% variation, with positive loading for BOD and COD and negative loading for DO, indicating the addition of organic matter from industries, livestock, agriculture, and municipal wastewater (Mamun & An 2021). Factor 2 explained a 22.31% variance with positive loading for PO₄-P, pointing to the contribution from the agricultural runoff. Factor 3 explained a 14.33% variation with positive loading for EC and TDS, showing the contribution of the weathering process. Factors 4 and 5 showed 12.01% and 11.13% variation, respectively, with positive loading for NH₄-N indicating the sewage contribution from the catchment area. In the wet season, three factors were identified in the Ganga River. Factor 1 explained a 41.60% variation with positive loading for NO₃-N, NH₄-N, and SiO₂-Si, indicating the production and mineralisation of organic matter and sediment transportation from the catchment area. Factor 2 explained a 21.98% variation with positive loading for BOD and COD and negative loading for DO, indicating the addition of organic matter from industries, livestock, agriculture, and municipal wastewater (Mamun & An 2021). Factor 3 explained a 14.52% variation with positive loading for EC, TDS, and PO₄-P, indicating the contribution of the weathering process in addition to agricultural runoff.

In the Yamuna River, four factors were identified in the dry season. Factor 1 explained a 38.84% variation with positive loading for EC and SiO₂-Si, and factor 2 explained a 19.59% variation with positive loading for BOD, COD, and negative loading for DO, indicating the contribution of organic matter from sewage waste. Factors 3 and 4 explained 12.04% and 11.38% variation, respectively, with positive loading for TDS and DO for factor 3 and NO₃-N for factor 4. Three factors were identified for the wet season in the Yamuna River, where factor 1 explained 43.40% variation with positive loading for EC, TDS, and PO₄-P, indicating the contribution from the weathering process along with agricultural and domestic discharge. Factor 2 explained a 22.54% variation with positive loading for pH, and factor 3 explained a 13.70% variation with positive loading for BOD and NH₄-N, indicating the contribution from sewage waste in the Yamuna River system.

Rivers are responsible for transporting dissolved nutrients they carry along their course and the disposing of them into the oceans. The annual flux and specific yield of nutrients can be determined using the discharge, drainage area, and concentration of nutrients. The discharge, drainage area, and specific nutrient yield for the Ganga and Yamuna rivers are given in Table 8. The total drainage area of the Ganga River is 861 × 10³ km^2, and its annual mean discharge is 380 km³ y⁻¹, and for the stretch passing through Uttar Pradesh, it is 294 × 10³ km² and 24.03 km³ y⁻¹, respectively (Rai et al. 2021; SMCG 2021; WRIS 2021). The total drainage area of the Yamuna River is 366 × 10³ km², and the annual average discharge is 131.7 km³ y⁻¹ and for the stretch passing through Uttar Pradesh, it is 70.437 × 10³ km² and 27.76 km³ y⁻¹, respectively (Upadhyay & Rai 2013; Sharma et al. 2017). The Ganga River's annual DIN flux was 8.96 × 10⁴ t y⁻¹, while the Yamuna River's annual flux along the Uttar Pradesh length was 21.96 × 10⁴ t y⁻¹. The Ganga River's annual flux of PO₄-P was 0.07 × 10⁴ t y⁻¹, while the Yamuna River's was 0.37 × 10⁴ t y⁻¹(Table 8). Nitrogen and phosphorus are the major contributors to nutrient pollution in rivers worldwide. A study by Ongley et al. (2010) reported that about 81% of nitrogen and 93% of phosphorus is contributed by nonpoint sources in China, while another study by Bowes et al. (2009) reported that diffuse sources add around 75% of the phosphorus load in the rivers of UK. It was reported by Carpenter et al. (1998) that point sources are responsible for more than 50% of nitrogen and phosphorus discharge, while nonpoint sources contribute more than 90% of the total nitrogen input in the rivers of the USA. In the Ganges River basin, approximately 13.28 Gg (gigagrams) of DIN and 5.29 Gg of DRP were added by point sources annually (Prajapati et al. 2020). According to reports of the National Ganga River Basin Authority (NGRBA), the annual fertiliser consumption of the states along the Ganga River is approximately 10 million tonnes, of which 38% is consumed by Uttar Pradesh alone. Such intensive fertiliser use has produced high nitrogen and phosphorus concentrations in the river through agricultural runoff (NGRBA 2011). The annual specific yield of NO₃-N from the Ganga River was 0.30 t km⁻² y⁻¹_, significantly lower than that of the Yamuna River (3.08 t km⁻² y⁻¹) in Uttar Pradesh (Table 8). The specific yield of NH₄-N was 0.01 and 0.03 t km⁻² y⁻¹ for the Ganga and Yamuna rivers, respectively. The annual yield of PO₄-P was also higher in the Yamuna River (0.05 t km⁻² y⁻¹) than in the Ganga River (0.003 t km⁻² y⁻¹). Uttar Pradesh is one of the most densely populated states with a population of 257,622,800, generating about 3,851.71 MLD of sewage, ultimately discharged into the river, adding to the nutrient level in the river water (CPCB 2009). Along the Ganga River, the maximum amount of nitrogen was contributed by Varanasi due to the discharge of 33% of the untreated sewage into the river (Prajapati et al. 2020), while Mathura contributed the maximum nitrogen load in the Yamuna River from agricultural and domestic sources. Prayagraj contributed the maximum phosphorus load in both rivers due to intensive agricultural activities in the catchment area. The annual flux of SiO₂-Si in the Ganga River was 7.48 × 10⁴ t yr⁻¹; in the Yamuna River, it was 12.71 × 10⁴ t yr⁻¹. The SiO₂-Si annual flux was almost the same in all the sites along the Ganga River, but in the Yamuna River, the maximum annual flux was observed at Mathura, followed by Agra and Prayagraj. The specific yield of SiO₂-Si was 0.25 t km⁻² y⁻¹ in the Ganga and 1.80 t km⁻² y⁻¹ in the Yamuna River. The highest silica load was observed in the Varanasi and Mathura sites for the Ganga and Yamuna rivers, respectively. Sen et al. (2018) studied the nutrient load of the Pandu River, a tributary of the Ganga River, and found that the ammonium, nitrate, phosphate, and silicate yields were 0.248, 0.162, 0.118, and 1.08 t km⁻² y⁻¹. The annual yield of nitrate in their study was lower than the other rivers of the world, such as the Amazon (0.797 t km⁻² y⁻¹), Mississippi (1.302 t km⁻² y⁻¹), and Yangtze (1.736 t km⁻² y⁻¹) River, but the phosphate yield was higher in comparison with the Amazon (0.009 t km⁻² y⁻¹), Yangtze (0.027 t km⁻² y⁻¹), and Ganga River, in this study (0.003 t km⁻² y⁻¹). Agriculture runoff and domestic sewage were the main contributors to nutrients. Being a tributary of the Ganga River, the Pandu River also contributes nutrients, increasing the Ganga River's yield of nutrients. Sharma et al. (2017) reported a relatively higher annual specific yield of PO₄-P (0.17 t km⁻² y⁻¹) than this study because they studied the whole stretch of the River. However, the specific yield of NO₃-N (0.18 t km⁻² y⁻¹) was relatively lower than in the values obtained from the Ganga and Yamuna River samples in this study, given the intensive agricultural practices, use of chemical fertilisers and sewage generation in the Uttar Pradesh stretch of rivers.

N Flx, Si Flx, and P Flx denote the average flux of DIN, DIP, and DSi, respectively. The N-ICEP value for the Ganga River was 0.09 kg C km⁻² day⁻¹, and the P-ICEP value was −0.04 kg C km⁻² day⁻¹. The Yamuna River showed comparatively higher values of N-ICEP, i.e., 0.26 kg C km⁻² day^−1, and the P-ICEP value was the same as that of the Ganga River, i.e., −0.04 kg C km⁻² day⁻¹. The positive N-ICEP values in both rivers indicate an abundance of nitrogen over silica resulting in the growth of phytoplankton (non-diatom species). This abundance might result from fertilisers coming through agricultural runoff from the catchment area. The algal mass is usually flushed down from the eutrophicated zones causing septic conditions in the downstream regions (Pandey et al. 2016). The negative P-ICEP values in both rivers suggest phosphate-limiting conditions for phytoplankton growth. High N will favour minimally silicified diatoms and non-diatom species if P is not a limiting factor.

On the other hand, diatoms that are highly silicified and quickly sinking benefit from high Si:N (Pandey et al. 2016). The N-ICEP values of the Ganga and the Yamuna rivers obtained in this study were significantly higher than that of the Ghaghara River, as reported by Ravi et al. (2021). The high N-ICEP value might be attributed to the large drainage area and high nutrient input from allochthonous sources in the Ganga and the Yamuna rivers.

The water samples of the Ganga and the Yamuna rivers were analysed for nutrient concentrations to determine the elemental nutrient ratio, nutrient load, eutrophication potential, and trophic state of both rivers. The dissolved nutrients showed a significant spatial and seasonal variation. The nutrient concentration was relatively higher in the Yamuna River than in the Ganga River in Uttar Pradesh due to the presence of more industries and tourist footfall in Agra and Mathura along the Yamuna River. It was also observed that the concentration of most of the parameters was higher in the dry season relative to the wet season in both rivers. Such seasonal differences in the concentration indicated a dilution effect due to rainfall during the wet period. The positive correlation between NH₄-N and NO₃-N indicated the input of domestic and municipal sewage from the catchment area. Factor analysis results indicated that nutrient concentration in both rivers is contributed by both natural and anthropogenic sources present in the catchment area. The average DIN/DIP ratio in both rivers indicated the limitation of phosphate for biological productivity. The DSi/DIN ratio indicates nitrogen loading from the catchment area and potential eutrophic conditions in both rivers. Compared with the Ganga River, the annual specific yields of NO₃-N, NH₄-N, PO₄-P, and SiO₂-Si were relatively higher in the Yamuna River, indicating a significant increase in the contribution from anthropogenic sources. Both rivers were at high risk of eutrophication given the abundance of nitrogen, as shown by the positive values of N-ICEP and phosphorus-limiting conditions, as suggested by the negative P-ICEP values.

The estimated TSI value indicated high eutrophication risk, as 44% and 50% of the samples in the Ganga River showed super-eutrophic conditions in the dry and wet seasons, respectively. The condition of the Yamuna River was worse, with all the samples in the dry season and 39% of the samples in the wet season showing hypereutrophic conditions. Both rivers are at high risk, as shown by the ICEP and TSI values, but the health of the Yamuna River requires immediate action. Therefore, proper policy implementation and consistent monitoring are needed to protect this priceless resource. This study generated new information on the trophic status of the Ganga and Yamuna rivers, and this information will be crucial for managing eutrophy and developing nutrient budgets at the regional level. The vision of projects like Namami Gange and YAP needs proper implementation and monitoring along with strict rules as the deterioration of river water quality by nutrient pollution is still a major concern. Other innovative programmes with equal involvement of public and government are the need of the hour by creating more sewage treatment plants to combat the increasing sewage generation with increasing population, restricting runoff from the agriculture areas by creating buffer zones such as riparian corridors and creating awareness among the masses to raise public understanding of the proper way to use water resources without compromising their quality.

The authors thank the Department of Science and Technology, Government of India, for providing an INSPIRE fellowship to K.V.. The authors thank the University of Allahabad for providing the necessary facilities.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.