Abstract:

Nitrogen (N) and phosphorus (P) are important contaminants that cause eutrophication and the deterioration of water quality. The transportation and distribution of N and P in the water column depends directly on suspended sediment (SS), which is the key medium for N and P cycling in natural rivers. Dam interception can significantly reduce the concentration and size of SS in reservoir discharge and regulate its spatiotemporal distribution within the drainage basin, resulting in considerable changes in the transport of these two pollutants in the lower reaches of the dam. Wudongde (WDD), Baihetan (BHT), Xiluodu (XLD) and Xijiaba (XJB), the four super cascade reservoirs located in the lower Jinsha River, have been impounded successively since 2012. Sichuan section mainstem (SSM) in the upper Yangtze River (UYR), located between the downstream of XJB Reservoir (the last one of the four cascade reservoirs) and the upper reach of the Three Gorges Reservoir (TGR), plays a crucial role in protecting the aquatic environment of the UYR and ensuring a safe water supply to the Three Gorges Reservoir. Since the impoundment of the Xiangjiaba Reservoir in 2012, the sediment regime in the SSM has changed completely, causing intense influences on the spatiotemporal distributions of N and P in the SSM. However, both these influences and the potential risks to the water quality safety of SSM and its downstream reaches are not well understood. Moreover, owing to the sample pretreatment method according to the Environmental Quality Standards for Surface Water (GB 3838—2002) in China, the total P (TP) concentration in most previous reports did not reflect the P content of the initial sample; instead, it measured the P content in the sample after settling for 30 min, namely, the TSSP, resulting in an underestimation of the true TP level. Based on the monitoring data of the TSSP and SS in the SSM and its six tributaries (including the Min River, Tuo River, and four small branches: Nanguang, Changning, Yongning, and Chishui Rivers) from 2010 to 2020, we rectified the TSSP for the settled sample into the TP for the initial sample. Moreover, the spatiotemporal distributions of ammonia nitrogen (NH₃-N) and TP in the SSM were quantified, the contribution rates of N and P inputs from the six tributaries to the SSM were evaluated, and the correlations of these variations with the concentration and grain size of SS were investigated. Significant changes in the spatiotemporal distributions of N and P in the SSM after the impoundment of XJB Reservoir in 2012 were identified as follows: 1) Due to a large amount of sediment being intercepted in the upstream reservoirs, in the inflow of Jinsha River into the SSM, the load (Q_SS), concentration (C_SS), and particle median size (D₅₀) of SS were found to have decreased by 98.2%, 98.7%, and 44.2%, respectively. Owing to the inflow of the six tributaries, the decreases in the Q_SS and C_SS at the outlet of the SSM were not as significant as those at the inlet of the SSM. However, they still reached 63.8% and 66.4%, respectively, indicating that the sediment regime along the SSM was largely regulated, and exhibited trends of decreased sediment concentration and particle refinement. 2) Along with significant changes in the sediment conditions in the upstream and downstream reaches of the SSM, the TP fluxes also underwent substantial declines of 87.1% and 62.9%, respectively, and those of NH₃-N were 44.5% and 53.7%, respectively. The decreases in TP fluxes were 2 and 1.2 times greater than those of NH₃-N, respectively. It is evident that both the SS and N and P adsorbed on the SS were significantly intercepted in the upstream reservoirs after impoundment, and more TP than NH₃-N was trapped. Seasonal variations in N and P fluxes in the SSM were also evidently affected after impoundment; N and P fluxes showed obvious declines during the flood, normal, and dry seasons, and the largest decreases occurred during the wet season (69.8% and 90.3%, respectively). 3) The Jinsha River provided the highest NH₃-N and TP to the SSM before impoundment, after which the Min River replaced the Jinsha River as the dominant supplier of NH₃-N and TP with contribution rates of 73.5% and 53.5%, respectively. In addition, the contribution rates of the Tuo River increased after 2012, ranking second and third in terms of the TP and NH₃-N inputs, respectively. The contributions of these four branches were relatively small. Therefore, the levels of N and P in SSM were determined by inputs from the Min River after 2012, and strengthening the prevention and control of pollution in the Min River Basin is crucial for the water quality safety of SSM and its downstream reach. 4) TP flux was found to have a strong, positive, and linear correlation with C_SS (r=0.9956) and a moderate correlation with particle size (r=0.5367). NH₃-N flux was moderately and positively correlated with C_SS (r=0.5209) and was independent of particle size. Therefore, compared with particle size, C_SS had a greater effect on the migration of N and P, and the change in C_SS had a stronger impact on the distribution of TP than on that of NH₃-N. 5) The proportion of TSSP in the TP decreased with increasing C_SS, and the two exhibited a strong negative correlation (r= −0.9939). Furthermore, TSSP represented only 48.9% of the TP at C_SS=540 mg·L⁻¹, indicating that it could not accurately reflect the actual P level, particularly in sediment-laden water. Therefore, it is necessary to revise the TSSP for the TP. 6) The endogenous release of N and P from sediments on the riverbed is a nutrient source for overlying water in the SSM. Although the changes in the hydraulic and suspended sediment conditions in the SSM alleviated the release after 2012, the TP flux resulting from internal release remained relatively high, and its contribution cannot be ignored. These observations provide a more detailed understanding of the effects of the impoundment of the XJB Reservoir on the transport and distribution of TP and NH₃-N in SSM. These findings can help lay a foundation for future work on the prevention and control of N and P pollution in SSM and UYR.

Key words: the upper Yangtze River, Sichuan section mainstem, impoundment, sediment, NH₃-N, TP, transfer

摘要：

长江上游干流四川段是长江上游重要的水源涵养地和三峡库区的直接水源补给区。2012年向家坝水库蓄水后，该段的水沙条件显著改变，进而对水中氮磷的迁移转化产生重大影响。系统分析了2010－2020年间四川段水沙输移特性的改变及其对氮磷通量时空分布的影响，结果表明，2012年后，四川段的含沙量大幅削减，上下游分别降低了98.7%和66.4%，同时，上下游的氮磷通量也显著降低，总磷分别减少87.1%和62.9%，氨氮分别减少44.5%和53.7%。2012年后，岷江取代金沙江成为四川段最大的氮磷贡献者，其输入的氨氮和总磷占比分别达到73.5%和53.5%，四川段的氮磷水平主要取决于岷江的输入。总磷和氨氮的通量与含沙量之间均呈现线性正相关，前者的相关性极强（r=0.9956），而后者的相关性相对较低（r=0.5209）；总磷与粒径呈现中等正相关，而氨氮与粒径基本不相关；相比粒径，含沙量的变化对氮磷的迁移影响更大，而含沙量的变化对总磷的影响比对氨氮的强。底泥的氮磷释放是四川段氮磷的一个重要来源，2012年后，受水力条件和水沙条件改变的影响，内源释放的氮磷通量减少，但总磷的内源释放通量仍然达到0.52×10⁴ t·a⁻¹（贡献率11.9%）。研究结果加深了对向家坝水库蓄水影响四川段氮磷迁移的认识，可为促进四川段流域的氮磷污染防控和保障长江上游干流水质安全提供决策参考。

关键词: 长江上游干流, 四川段, 水库蓄水, 泥沙, 氨氮, 总磷, 迁移