Spatiotemporal water quality variations in the urbanizing Chongqing reach of Jialing River, China Open Access

Rivers are the key linkages between terrestrial ecosystems and reservoirs (lakes) and/or marine ecosystems, and historically provide the most valuable sources of water for human development (Ripl 2003; Grill et al. 2019). Rivers provide the amount of available freshwater to support the human activities, such as domestic life, industrial and agricultural production (Safari et al. 2021). However, the urbanization driving the rapid development of industry and agriculture has subjected rivers to an increased stress, and therefore unbalancing the river ecosystem and accelerating the degradation of water quality in rivers (van Vliet et al. 2017). Accordingly, identifying the water quality conditions in rivers during urbanization is conducive to making the protection strategies taken by the local government departments.

The pollutants degrading river water quality are generally stemmed from point and non-point sources, which are both strongly driven by human activities, e.g., urbanization (Walsh et al. 2005; Brooks et al. 2006; Schliemann et al. 2021). The point input sources, such as industrial operations and domestic sewage, commonly continually discharge to the rivers over time by a pipe or at a specific site (Ramsey & Szynkiewicz 2019) and significantly alter the water quality, while they are relatively simple to be monitored and controlled by treatment at the head (Varol 2012). In contrast, the non-point input sources like runoff from upper fertilized cropland are relatively more difficult to be quantified as a result of the dispersed distribution in a wide area and significant shifts with time owing to the weather effects (Carpenter et al. 1998; Voutsa et al. 2001).

Water quality in rivers is usually characterized by significant spatiotemporal variations due to the urbanization process, which causes the changeful intensities in point and non-point source pollutions. The high concentrations of pollutants in river systems are at specific time periods making polluted hot moments (McClain et al. 2003; Dwivedi et al. 2018), and at specific locations creating polluted hotspots (Luthy et al. 2015). These hot moments and spots collectively reflect the sharp changes in water quality in rivers under different spatiotemporal scales (Bernhardt et al. 2017). Previous studies have focused on the spatiotemporal water quality variations around the world (Ahipathy & Puttaiah 2006; Xie et al. 2013, 2015). For instance, Fraga et al. (2021) analyzed the temporal and spatial trends of surface water quality in the Doce River basin, Minas Gerais, Brazil. Schliemann et al. (2021) reported the water quality in the South Platt River in the Denver metropolitan area, Colorado, USA, and identified spatiotemporal hotspots for the study river. However, the spatiotemporal water quality variations are strongly dependent on the intensities of human activities and changing climate, and some conflicting findings are usually obtained due to the specific objectives in different study sites (Iida et al. 2011; Stevenson et al. 2012). Therefore, the characteristics of spatiotemporal variations in water quality are thus still unclear (Dragun et al. 2011; Safari et al. 2021; Schliemann et al. 2021), and more data are urgently needed to interpret the dynamic changes in water quality in some specific sites, which is expected to contribute greatly to the optimization of strategy for the universal sustainable management of water quality worldwide.

The Jialing River is one of the main tributaries in the upper reach of the Yangtze River, and is also the largest tributary of the Three Gorges Reservoir Area (TGRA). Therefore, the water quality conditions of Jialing River, to a large extent, play the pivotal role in ensuring the water environmental security of Yangtze River and TGRA (Wu et al. 2012a). The downstream of Jialing River flows through several urbanizing administrative districts of Chongqing City and directly discharges to the Yangtze River, resulting in substantial input of point and non-point pollutants to the waterbody and significant water quality degradation (Long et al. 2011). Liangtan River, the main tributary of lower Jialing River, strongly affects the water quality conditions of the trunk Jialing River (Gong et al. 2013). In recent years, the rapid urbanization, such as the development of urban industry and agriculture, has brought enormous negative impacts on the water quality of the Chongqing reach of Jialing River and its tributary (Long et al. 2011; Gong et al. 2013), which in turn may limit the future sustainable development in society, economy and ecology of the adjacent urban regions. Although protection measures have been implemented to prevent the water quality from deteriorating and to further achieve the improvement of water quality by the local government, the spatiotemporal water quality variations resulting from the urbanization driven by human activities in the urbanizing Chongqing reach of Jialing River, to the best of our knowledge, are not well determined.

In this study, a 6-yr multisite dataset (January 2010–December 2015) collected from the lower trunk Jialing River and its tributary Liangtan River were interpreted in detail to determine the spatiotemporal water quality variations in the urbanizing Chongqing reach of Jialing River. Considering vast areas of urbanization and required sustainable water quality management, this study will contribute greatly to understanding the extent of spatiotemporal variations in water quality and for optimizing strategies for advancing water resources management. The specific objectives of this study are to: (1) determine the spatial and temporal variations in selected water quality variables, and (2) disclose the potential relationships among these variables.

The Jialing River (102°33′–109°00′E, 29°18′–34°30′N) originates from the Qinling Mountain and flows through Shanxi, Gansu, Sichuan Provinces and Chongqing Municipality, presenting a drainage area of approximately 159,800 km² and a length of 1120 km (Figure 1). The Jialing River basin has a slow tilt from the northeast and northwest edge with altitudes of 1500–3000 m to the downstream (Wu et al. 2012a). The regional climate in Jialing River basin is characterized by a subtropical monsoon climate with an average temperature of 15 °C. The Beibei hydrological station, located in the export of Jialing River, recorded the average annual rainfall of 1098 mm with over 80% occurring in the wet season from May to September, the average annual evaporation of 709.4 mm, and the average annual runoff of 65.9 billion m³ (Wu et al. 2012b). The overlying soils in the basin mostly consist of brown soil, yellow-brown soil and purple soil. The upstream of Jialing River basin is related to the provinces of Shanxi and Gansu characterized by agriculture such as food production and breeding of livestock and poultry (Wu et al. 2012a, 2012b). The midstream is related to Sichuan Province which is the important production base of livestock, grain and economic crops. The downstream is across several urbanizing districts of Chongqing Municipality including Hechuan, Beibei, Yubei, Shapingba, Yuzhong and Jiangbei and flows into the Yangtze River at the Chaotianmen, being also the important producer of grain and economic crops (Wu et al. 2012a, 2012b).

The water network of Jialing River appears visually branched, and many tributaries are evolved along the trunk of Jialing River (Figure 1). Therein, the Liangtan River is the main tributary of the lower Jialing River, and has a drainage area of 525 km², a length of 79 km and average annual discharge of 6.43 m³ s⁻¹. The Liangtan River is also located in the advanced economic circle of Chongqing Municipality and is subjected to severe water pollution due to intensive human activities e.g., urbanization over the past decades (Gong et al. 2013). Its water quality was categorized to the weakest grade during 2001–2005 according to the Environmental Quality Standards for Water Quality (2002) (Table S1). The degraded aquatic ecosystems in Jialing River and its tributary have become a growing concern because of the demands for sustainable development of local society, economy and ecology. Thus, characterization of the spatiotemporal water quality variations in the lower Jialing River is thus very helpful for the local water environmental agency to make rational decisions on the water resources allocation and water quality protection.

To investigate the spatiotemporal water quality in the urbanizing Chongqing reach of Jialing River, a total of six observation sites were selected in the lower of Jialing River and its tributary (Figure 1). These observation sites were: site 1 (Xixiqiao), site 2 (Longfenghekou), site 3 (Jinzituo), site 4 (Beiwenquan), site 5 (Liangtuo) and site 6 (Daxigou), respectively. Among them, sites 1 and 2 are set in the tributary Liangtan River, and sites 3, 4, 5 and 6 are distributed in the lower trunk Jialing River, covering the starting point of Chongqing reach of Jialing River to the exporting point into the Yangtze River. Additionally, the Beibei wastewater treatment plant is situated at the upstream approximately 150 m away from site 2 and thus the water quality of the treated water was also able to be assessed.

Sample collection in each site was conducted monthly from January 2010 to December 2015 (Figure 2). In situ, the immediate measurements of pH and dissolved oxygen (DO) were preferentially implemented using a portable multi-parameter meter (Minisonde5x, Hach, USA) at a depth of around 0.5–1 m from the water surface of the platform of the intake tower. Simultaneously, water samples for further measurements of other water quality variables were collected using a 2.5 L grab sampler and then were filled into duplicate polyethylene bottles (∼500 mL) which had been pre-washed with the river water corresponding to the sampling depth at each site. All obtained water samples were sealed using a plastic bag and kept in the dark at 4 °C, until being processed immediately on return to the laboratory within 24 hr.

In the laboratory, the duplicate bottles of water samples experienced distinct test procedures. One bottle of the water sample was filtered through a 0.45 μm glass microporous fiber membrane to measure the ammonia nitrogen (NH₃-N) concentration by using the spectrophotometric method with salicylic acid. The other bottle of water sample was unfiltered and directly used to measure the permanganate index (COD_Mn) and total phosphorus (TP) using potassium permanganate titration after reduction by sodium oxalate and molybdenum blue colorimetric method after addition of potassium peroxodisuphate, respectively. These analytical methods for NH₃-N, COD_Mn and TP were performed following the standard Chinese protocols (SEPA 2002). In this study, the concentrations of DO, COD_Mn, NH₃-N and TP were measured in mg L⁻¹.

One-way analysis of variance (ANOVA) followed by Tukey's (or Tamhne's when equal variances were not assumed) post hoc multiple comparison test was conducted to determine the differences in water quality variables among sampling sites, and to compare the monthly, seasonal and annual changes in these variables, respectively. Spearman's rank correlation analysis was used to unclose the correlations among these study variables. Simple linear regression analysis was used to examine the relationship between TP concentration and other water quality variables after all variables were log₁₀-transformed. A P < 0.05 was considered statistically significant. All statistical analyses abovementioned were implemented using SPSS 22.0 (IBM, Armonk, New York, USA).

The spatial variations of water quality in the trunk Jialing River and its tributary are shown in Figures 3 and S1. During 2010–2015, pH varied significantly from the highest value in site 4 (Beiwenquan, 8.02) to the lowest value in site 6 (Daxigou, 7.66) (P < 0.05), with an average value of 7.86 ± 0.27 (Figure 3(a)). However, average pH was 1.2% higher in the trunk river than in the tributary (P < 0.05). DO concentration was 39.8% higher in the trunk sites 3–6 than those in the tributary sites 1–2 (P < 0.05), ranging from 8.97 ± 1.26 mg L⁻¹ (site 6, Daxigou) to 6.21 ± 1.86 mg L⁻¹ (site 1, Xixiqiao) with an average of 7.89 ± 1.98 mg L⁻¹ (Figure 3(b)). In contrast, the concentrations of COD_Mn, NH₃-N and TP were 66.9, 94.7 and 85.2% lower in the trunk sites 3–6 than those in the tributary sites 1–2, respectively (Figure 3(c)–3(e)) (P < 0.05). Specifically, COD_Mn concentration ranged from 6.84 ± 2.02 (site 1, Xixiqiao) to 2.04 ± 0.69 mg L⁻¹ (site 4, Beiwenquan) with an average value of 3.81 ± 2.50 mg L⁻¹. NH₃-N concentration varied from 3.49 ± 3.48 (site 1, Xixiqiao) to 0.13 ± 0.04 mg L⁻¹ (site 3, Jinzituo) with an average value of 1.24 ± 2.42 mg L⁻¹. TP concentration ranged from 0.45 ± 0.40 (site 1, Xixiqiao) to 0.06 ± 0.03 mg L⁻¹ (site 4, Beiwenquan), with an average of 0.21 ± 0.29 mg L⁻¹.

Monthly, pH and DO concentration were both the highest in February (8.06 ± 0.23 and 9.72 ± 2.42 mg L⁻¹, respectively) and the lowest in June (7.73 ± 0.27 and 6.91 ± 1.65 mg L⁻¹, respectively) (Table 1). COD_Mn was the highest in February (4.18 ± 3.78 mg L⁻¹) and March (4.18 ± 3.37 mg L⁻¹), and the lowest in December (3.19 ± 1.90 mg L⁻¹) (Table 1). NH₃-N concentration showed the highest value in February (1.93 ± 3.63 mg L⁻¹) and the lowest value in August (0.51 ± 0.72 mg L⁻¹) (Table 1). TP concentration presented the highest value in April (0.32 ± 0.53 mg L⁻¹) and the lowest values in June (0.15 ± 0.11 mg L⁻¹) and October (0.15 ± 0.12 mg L⁻¹) (Table 1).

Seasonally, pH was 1.3% higher in the dry season (7.91 ± 0.27) than in the wet season (7.81 ± 0.26) (P < 0.05) (Figure 4(a)). DO concentration was 18.2% higher in dry season (8.43 ± 2.10 mg L⁻¹) than in the wet season (7.13 ± 1.50 mg L⁻¹) (P < 0.05) (Figure 4(b)). COD_Mn concentration was slightly different between the two seasons (P > 0.05) (Figure 4(c)). NH₃-N and TP concentrations were respectively 102.8 and 32.5% higher in the dry season (1.57 ± 2.95 mg L⁻¹ and 0.23 ± 0.37 mg L⁻¹, respectively) than in the wet season (0.78 ± 1.22 mg L⁻¹ and 0.17 ± 0.13 mg L⁻¹, respectively) (P < 0.05) (Figure 4(d) and 4(e)).

In addition, pH did not show a significant inter-annual difference (P > 0.05), while the DO concentration showed significant inter-annual variations (P < 0.05), ranging from 8.78 ± 1.71 mg L⁻¹ in 2012 to 7.44 ± 2.10 mg L⁻¹ in 2014. Concentrations of COD_Mn, NH₃-N and TP also showed considerable inter-annual differences (P < 0.05) (Figures 5 and S1). Therein, COD_Mn and NH₃-N concentrations ranged from 4.97 ± 3.62 and 3.09 ± 4.69 mg L⁻¹ in 2010 to 3.30 ± 1.71 and 0.65 ± 1.07 mg L⁻¹ in 2014, respectively. TP concentration varied between 0.35 ± 0.56 mg L⁻¹ in 2010 and 0.16 ± 0.15 mg L⁻¹ in 2013.

Spearman's rank correlation analysis showed the different relationships among water quality variables within each sampling site and each month, season and study year. In particular, TP concentration was always closely correlated with pH, concentrations of DO, COD_Mn and NH₃-N in site 4 (P < 0.05), but was not in other sites (P > 0.05) (Table S2). TP concentration was significantly correlated with pH in November and December (P < 0.05), and was always considerably correlated with DO, COD_Mn and NH₃-N concentrations in all months (P < 0.05) (Table S3). TP concentration showed close relationships with DO, COD_Mn and NH₃-N concentrations (P < 0.05), but not with pH (P > 0.05) in both dry and wet seasons (Table 2). Further, TP concentration was always markedly correlated with pH, concentrations of DO, COD_Mn and NH₃-N in 2011, 2014 and 2015 (P < 0.05) (Table S4).

Additionally, simple linear regression analysis showed that TP concentration had significant negative relationships with pH (n = 364, P < 0.01) and DO concentration (n = 364, P < 0.001), but significant positive relationships with COD_Mn (n = 364, P < 0.0001) and NH₃-N concentrations (n = 364, P < 0.0001) across the whole spatial and temporal scales (Figure 6).

The water quality in the urbanizing Chongqing reach of Jialing River displayed significant spatial variations, showing relatively lower pH and DO concentration but higher COD_Mn, NH₃-N and TP concentrations in the tributary Liangtan River than those in the trunk Jialing River. This result suggested that more severe water quality degradation occurred in the tributary Liangtan River relative to the trunk Jialing River, due largely to the intensive human activities in the tributary. Gong et al. (2013) assessed the water quality of the Liangtan River and concluded that the pollutions in the river mainly originated from the adjacent urban wastewaters, wastewaters from industries and the intensively cultivated lands along the river. Moreover, the Liangtan River flows through the city center of the Gaoxin, Shapingba and Beibei Districts in which are burgeoning urbanizing areas, the substantial pollution, such as wastewaters from treatment plants and industrial operations, domestic sewage and runoff from impervious surfaces which were primarily caused by the urbanization, collectively resulted in the diminished water quality (Walsh et al. 2005; Brooks et al. 2006; Xie et al. 2013). Although the trunk Jialing River is also confronted with similar environmental circumstances, the large amount of flow discharge from the upstream could markedly dilute the pollutants and thus was conducive to preventing the water quality from further degrading (Vega et al. 1998; Dragun et al. 2011; Moatar et al. 2017).

Our results found an average pH value of 7.86 across all sampling sites, suggesting that the study river water overall had alkaline properties. This was likely related to the combined effects of anthropogenic sources, flow discharge and carbonate rocks distributed around the study area (Safari et al. 2021). Generally, the pH value of natural river water is mostly regulated by the equilibrium of carbonate-bicarbonate-carbon dioxide system, while the input pollutants (e.g., organic matters) unbalance the nature of water, therefore resulting in acidification in river water (Spry & Wiener 1991; Ramsey & Szynkiewicz 2019). Slightly alkaline pH is preferable in river water since some pollutants such as heavy metals can be removed by carbonate and bicarbonate precipitates (Ahipathy & Puttaiah 2006; Clark et al. 2014). In this study, the alkaline river water was also recorded overall which implied the rare existence of organic and inorganic oxidation substances. This was further supported by the observed higher DO concentration (average 7.89 mg L⁻¹) and lower COD_Mn concentration (average 3.81 mg L⁻¹) relative to the polluted waters referring to the Environmental Quality Standards for Water Quality (2002). This result was also responsible for the significant relationships between pH and DO as well as COD_Mn concentrations in this study.

Additionally, the river water showed the average NH₃-N concentration of 1.24 mg L⁻¹ and TP concentration of 0.21 mg L⁻¹, and significant relationships between NH₃-N and TP concentrations were found in several sampling sites. This result suggested that the potentially consistent sources, such as excess fertilizers and soil erosion occurring in the upper agricultural lands, contributed to the diminished water quality as a result of the heightened human activities and changing climate (Buda et al. 2015; Tan et al. 2015; Zhang et al. 2020). Gong et al. (2013) reported that a large tract of flower plantations that had been applied by substantial fertilizers was the pollution source degrading the aquatic ecosystem in the tributary Liangtan River. Similar conditions were also found in the trunk Jialing River (Wu et al. 2012a, 2012b). Iida et al. (2011), suggesting that ammonium nitrogen and phosphorus could both attach to soil particles and be transported into the downstream river through runoff discharge accompanied by severe soil erosion. Thus, an important way to reduce water quality degradation in rivers is the rational application of fertilizers and controlling soil erosion in the surrounding agricultural lands.

The water quality of Jialing River also presented significant temporal variations. The monthly changes in pH and concentrations of DO, COD_Mn, NH₃-N and TP showed similar trends, presenting almost the highest values in February, March and April, and the lowest values in June and August. The seasonal changes also proved these findings, presenting the lower pH and concentrations of DO, NH₃-N and TP in the wet season (May–September) rather than in the dry season (October–April). These findings were further supported by the relationships among these variables in monthly and seasonal scales. The results were partially consistent with the previous findings given by Stevenson et al. (2012) who found that DO concentration was lower in summer due to the high algal biomass which could deplete oxygen concentrations by respiration. However, they also found that pH increased with algal biomass owing to the algal uptake of CO₂ and the change in the carbonate equilibrium, which was in contrast with our findings. Parinet et al. (2004) found the negative relationship between pH and algal biomass because of the production of acidic compounds in the course of the alga decomposition in a tropical lake system. A similar mechanism was presumably found in river water. The reason for the decline in both pH and DO concentration in the wet season was probably because the relatively high temperature accelerated the decomposition of organic matters by enhancing the microorganism activities (Voutsa et al. 2001). This process could not only decrease the pH value of river water through acidification, but also consume a considerable amount of oxygen as a consequence of microbial metabolism, leading to the decline in DO concentration (Gong et al. 2013).

The dilution effects resulting from the upstream flow discharge decreased the NH₃-N and TP concentrations in the wet season (Tabari et al. 2011; Schliemann et al. 2021). This was also in support of the previous findings of Xie et al. (2015) who suggested that pollutant concentrations were not necessarily linked to river discharge, and the negative relationship between them was commonly found, resulting in the dilution of pollutants in the wet season. This result indicated that the increase in flow discharge from the upper reach was possibly conducive to ameliorating water quality (Iida et al. 2011). However, the COD_Mn concentration did not significantly differ by season. This was likely due to the obstinate pollutants originating from the wastewater treatment plants, industrial operations and domestic sewage, regardless of seasonal variability (Li et al. 2021; Schliemann et al. 2021).

Our results showed that the pH had no inter-annual difference, and DO concentration presented significant inter-annual variability, and concentrations of COD_Mn, NH₃-N and TP also had decreased trends from 2010 to 2015. This suggested the potential improvement of water quality due to the launching of controlling measures for the point source pollution in recent years. Gong et al. (2013) provided several strategies to improve maximumly the water quality, including reducing rural and urban domestic wastewater loads and industrial pollution loads, and reusing of solid wastes (e.g., municipal solid waste and agricultural residues). These strategies were helpful in preventing point source pollutants from transporting into the rivers, contributing to the reduction in water quality degradation for both the trunk Jialing River and its tributaries. In recent years, the non-point source pollution from overland flow accompanied by soil erosion has been demonstrated to contribute substantial nutrients to the aquatic ecosystem, degrading the water quality worldwide (Fraga et al. 2021; Schliemann et al. 2021). However, the measures, such as vegetative filter strips (Blanco-Canqui et al. 2004), conservation tillage (Issaka et al. 2019) and rational fertilization (Du et al. 2021), for controlling soil erosion and non-point source pollution are still not adequately assessed. Therefore, more attention should be paid to develop effective strategies for controlling soil erosion and non-point source pollution with the aim of improving downstream water quality in the future.

The spatiotemporal water quality variations were interpreted in the urbanizing Chongqing reach of Jialing River from January 2010 to December 2015. Relatively lower pH and DO concentration but higher COD_Mn, NH₃-N and TP concentrations were found in the tributary river relative to the trunk Jialing River, suggesting the relatively greater dilution effects occurring in the trunk river. Lower pH and concentrations of DO, NH₃-N and TP were found in the wet season (May–September) rather than in the dry season (October-April), and no significant difference in COD_Mn concentration was found in the two seasons, implying obstinate pollutants resulting from nature and human activities. The pH presented no pronounced inter-annual variation while DO showed a significant inter-annual difference. Concentrations of COD_Mn, NH₃-N and TP showed decreased trends from 2010 to 2015, indicating the potential improvement of water quality attributed to the implementation of point source pollution control measures in the urbanizing Chongqing reach of Jialing River. However, soil erosion and non-point source pollution are the key factors degrading water quality, and effective strategies for minimizing these effects are urgently required to improve river water quality in the near future.

This study was funded by the Fundamental Research Funds for the Central Universities, China (XDJK2020C070), and the Science and Technology Research Program of Chongqing Municipal Education Commission (Grant No. KJZD-K202100201, KJQN202100212). We thank Bishun Zeng for his help on the revision. Special thanks are given to three anonymous reviewers for their constructive comments for greatly improving the manuscript.

The authors declare that they have no conflict of interest.

Data cannot be made publicly available; readers should contact the corresponding author for details.