4. Discussion

4.1 NH4+recycling rates and contribution to the N budget in Yangtze River

REGs and Upots from Yangtze River were 0.26 (0.05~1.19) μmol N L−1h−1 and 0.34 (− 0.22~1.99) μmol N L−1 h−1, respectively. Generally, the REGs and Upots in this study were within the range of values reported elsewhere (Table 2 ). In detail, they were comparable to several major polluted inflow rivers of Lake Taihu, China (Jiang et al. , 2019), but were higher than that from the low nutrients and low primary production Old Woman River (Mccarthy et al. , 2007b), Aransas River and Mission River (Bruesewitz et al. , 2015). Compared to other eutrophic lakes, such as Lake Taihu, Lake Petit saut and Lake Maracaibo (Gardner et al. , 1998; Collos et al. , 2001; Jiang et al. , 2019), insignificantly lower REGs and Upots were observed in the river-estuary continuum of Yangtze River. In addition, it is noteworthy that compared with lake ecosystems, river ecosystems tend to be found the greater REGs than Upots, suggesting a potential release risk of NH4+ in the river water column.
Sources of NH4+ to Yangtze River mainly include external nutrient inputs (river input, atmospheric deposition, and sediment release) and undocumented regenerated NH4+. Although it has been confirmed that the internal recycling processes play a vital important role in the NH4+ budget and ecological effect in eutrophic lakes (Wu et al. , 2017; Paerl et al. , 2011), the contribution of regenerated NH4+ to river-estuary continuum of Yangtze River is still unclear. Here we provide the first estimation of contribution from regenerated NH4+ to the N budget. Previous study has reported that river N input from the upper and middle reaches of Yangtze River was 57.14 × 108 kg N yr−1, and N input in the lower reaches of Yangtze River (study area of this study) was estimated as 4.87 × 108 kg N yr−1 (Liu et al. , 2018). Atmospheric deposition in our study area was estimated as 2.76 × 108 kg N yr−1, which was calculated as the deposition to emission ratio (Chen et al. , 2016). N input from sediments was estimated as 2.19 × 108 kg N yr−1(estimated using the total mineralized N in the sediments minus the sum of denitrification, anammox and microbial N assimilation) (Lin et al. , 2016). Based on The Changjiang Bulletin issued by the Changjiang Water Resources Commission of the Ministry of Water Resources in 2018, the annual runoff of Datong station (Fig.1) in 2018 was 8028 × 108 m3. Together with average REGs in the river-estuary continuum, we can estimate riverine regenerated NH4+ was approximately 21.8 × 108 kg N yr−1. Compared with other N sources in the Yangtze River (Fig. 5) , the regenerated NH4+ is lower than riverine input from upper and middle reaches, but much higher than N input from lower reaches, atmospheric deposition, and sediments release. The regenerated NH4+ accounts for 25% of total N inputs in the study area, suggesting that regenerated NH4+ is an important internal N source for microbes and may influence nutrient dynamics in lower coasts. Although the summer rates of REGs can be significantly higher than in other seasons (Bruesewitz et al. , 2015; Jiang et al. , 2019), which made the annual regenerated NH4+ be overestimated, this ratio is high enough and could not be neglected.

4.2 Effects of SS on NH4+ recycling rates

No significant correlation between REGs and Chl-a was found in this study (r =0.220, p > 0.05) (Table 3) , indicating that algae was not the major source of regenerated NH4+. Unlike in eutrophic lakes (e.g., Lake Taihu, Jiang et al. , 2019), algae abundance in Yangtze River was low (2.8 ± 3.2 μg L−1) due to high turbidity and washout from the high velocity of water flow in the Yangtze River, which leads to low concentrations of algae-derived organic N. However, significant correlations between REGs and SS instead of between REGs and DOC (r =0.197, p > 0.05) or DON (r =0.111, p> 0.05) suggested that REGs were mainly influenced by allochthonous particulate matters. SS is a complex mixture of organic detritus, microorganisms, and other organisms (Turner and Millward, 2002; Odman et al. , 1999), and SS in Yangtze River mainly come from the input of basin and soil erosion (Yang et al. , 2007). In this study, SS were positively correlated with COD (r =0.535, p< 0.01), and TN (r = 0.304, p < 0.05), and PN (r =0.478, p < 0.01) (Fig. S2) , suggesting that SS can act as a vector of nutrient to promote N cycling processes (Bilotta and Brazier, 2008; Zhang et al. , 2019). Moreover, significant correlations between REGs and COD (roughly represent the content of particulate and dissolved organic matter) (r =0.608, p< 0.01) and PN (r =0.455, p < 0.01) provide further evidence that heterotrophic bacterial (e,g., ammonifying bacteria) degradation of allochthonous particulate organic matters influences regeneration rates of NH4+. Ammonifying bacteria, a major participant for NH4+ regeneration activities, is abundant and widespread in river systems and tended to attach on SS (Xiaet al. , 2013). Ammonifying bacteria population in culture with 5 g L−1 SS of river systems were shown two orders of magnitude higher than that without SS (Xia et al. , 2013). Thus, high regeneration rates of NH4+ in Yangtze River were deduced to be mainly contributed by heterotrophic bacteria attached on high levels of SS and PN in SS provides important substrate for bacterial metabolism.
NH4+ uptake is primarily composed of nitrification as well as assimilation by phytoplankton (Hampel et al. , 2017). However, the low biomass of phytoplankton in the turbidity river may contribute less to NH4+uptake. This is supported by the insignificant correlation between Upots and Chl-a (r =0.257, p> 0.05). Study reported that nitrification rates increased with SS as a power function in the Yellow River, China, which was characterized by high SS concentrations (Xia et al. , 2009). Combined with high SS concentrations and low algae biomass in the Yangtze River, we can infer that NH4+uptake is mainly due to nitrification on SS. In this study, the SS concentrations were found to be significantly correlated with Upots (r =0.825, p < 0.01), which evidences the importance of SS in the NH4+ uptake process. Other factors such as the concentrations of COD, TN and PN can also affect NH4+ uptake rates. This may be due to the significant positive correlations between COD, TN, PN and the REGs, which provides abundant substrates for the uptake of NH4+.

4.3 Higher planktonic NH4+ demand in the estuary than river section of Yangtze River

Many studies have reported that Yangtze River estuary is characterized as P limitation due to sufficient N inputs (Wong et al. , 1998; Liang and Xian, 2018). However, results in this study suggest that planktonic NH4+ limitation may also occur in the river-estuary continuum of Yangtze River. CBAD as the index of NH4+ limitation in water, addresses the vital question whether or not the NH4+ demand of plankton could be met by NH4+ recycling in the water column (Jiang et al. , 2019; Gardner et al. , 2017). Our results showed that CBAD in the estuary of Yangtze River were significantly higher than in the Anhui and Jiangsu sections (near zero) (Fig. 4) , where NH4+ regeneration and uptake were nearly balanced.
Significant higher CBAD in the estuary could be due to several possible reasons. First, high concentrations of SS can not only promote microbial nitrification, but also accelerate NH4+ regeneration (Xia et al. , 2013; Xue et al. , 2019). Our results further indicated that SS concentrations promoted Upots faster than REGs (Fig.6 ), resulting in higher CBAD in the estuary of Yangtze River. Second, higher water temperatures (averaged 29.2 ± 0.5℃) in the estuary than other sections (averaged 28.4 ± 0.7℃ in Jiangsu section, averaged 28.1 ± 0.8℃ in Anhui section) of Yangtze River promote the higher CBAD values and NH4+limitation. Culture incubations in Lake Taihu showed that Upots increased faster than REGs in response to increasing water temperature between 5.0 and 32.9℃ (Jiang et al. , 2019). Another possible explanation for the higher CBAD in estuary is the higher plankton biomass. Previous studies reported that approximately 57% of bacterial cells were retained on the 0.7 μm filters and confirmed the presence of small phytoplankton and archaea through DNA analysis (Sipler et al. , 2017; Connelly et al. , 2014). Our results showed that PN was significantly correlated with Chl-a (r =0.314, p < 0.05) (Fig. S3) . Thus, PN concentrations can approximately represent biomass N of plankton intercepted by filters (GF/F). Comparing with the river sections, the estuary showed higher PN concentrations and CBAD increased with increasing PN concentrations (r =0.44, p < 0.01), implying that organisms may be increasing NH4+-deprived as plankton biomass increases, which was similar to eutrophic lakes (Gardner et al. , 2017).

4.4 Implications for management

Our results suggest that SS is a key factor controlling NH4+ recycling rates. High concentrations of SS in the water column of Yangtze River can promote both NH4+ regeneration and nitrification rates, resulting in the persistence of algal growth and potential high levels of NO3concentrations. This is consistent with a previous study which showed that nitrification is one of the major sources of NO3 in the Yangtze River (Li et al. , 2010). Thus, effective measures to reduce SS concentrations in this turbid river may reduce the risk of N pollution. For example, restoration of shoreline vegetation, improvement of land use, and management of soil erosion are recommended to reduce SS concentrations in the lower reaches of Yangtze River.
Most previous studies in Yangtze River pay a special attention to NO3 exports into the East China Sea due to its high concentrations (Zhou et al. , 2008; Mulleret al. , 2008; Liang and Xian, 2018). As a more preferred N nutrition for most phytoplankton species, NH4+ were overlooked in river systems because of low ambient concentrations. Our study and more evidences have shown that NH4+ recycling rates can be high even with low ambient concentrations (Jiang et al. , 2019; Gardner et al. , 2017), which can still influence estuaries and its coastal areas profoundly. High NH4+ recycling rates and demands reflect rapid microbial metabolism including both algal uptake and bacterial activities, which closely related to the growth and persistence of microbes (Hampel et al. , 2019). Compared to eutrophic lakes, NH4+ recycling in the estuary of Yangtze River was influenced in a larger part by bacterial activities. As large amounts of SS and organic matters flow downside into the coastal areas, rapid internal NH4+ recycling may also support algal growth and contribute to bloom persistence. Thus, diverse countermeasures should be focused on reducing estuarine inputs of SS or labile organic matters that potentially support high NH4+ recycling. Future studies are expected to understand how different properties of SS and organic matter influence water column NH4+ recycling. Results of this study is not only beneficial to the management of this third longest and most economically valuable river in the world, but also valuable to other large rivers and river-estuary continuum systems.