Assessment of Cau River water quality assessment using a combination of water quality and pollution indices

Water is a vital commodity, both to sustain life and for the global economy. However, the quality of global water has rapidly declined for decades due to the impact of both natural and anthropogenic factors (Vadde et al. 2018). Assessing water quality for different water use purposes, such as domestic use, irrigation, conservation and industrial usage, are an important strategy for food safety and human health. Water quality evaluation aims to identify the sources of water pollution and develop a strategy for sustainable water source management, maintaining and promoting human health and other social and economic growth (Carroll et al. 2006). Surface water quality indexes have been developed and introduced worldwide by researchers with various applications of the Nation Sanitation Foundation Water Quality Index (NSFWQI) (Carroll et al. 2006), the Water Quality Index (WQI) (Gupta et al. 2003; Chaturvedi & Bassin 2009; Rocha et al. 2015; Sener et al. 2017), the Comprehensive Pollution Index (CPI) (Matta et al. 2017), the Organic Pollution Index (OPI) (Mezbour et al. 2018), the Trace Metal Pollution Index (TPI) (Reza & Singh 2010), the Eutrophication Index (EI) (Liu et al. 2011) based on the the database of water monitoring parameters. In Vietnam, research on water quality assessment mostly focuses on comparing the concentration of pollutant to the national surface water quality standard (MONRE 2015). The WQI has been used for 10 years (VEA 2011), however, a combination of WQI with other water pollution indices was not applied widely for water quality assessment research.

Vietnam has abundant water resources with up to 2,360 rivers being over 10 km in length. There are 16 river basins in the whole country with an area larger than 2,500 km². The average annual surface flow of Vietnam's river basin ranges from 830 to 840 cubic meters per year and the annual rainfall is around 1,940 mm. Despite the abandance of natural water sources, Vietnam is still considered as a water shortage country due to three main causes. About 72% of the river basins are distributed outside the country, which accounts for 63% of the total national water source (equal to 520–525 billion cubic meter) and the volume of water coming from inside the country area only is 37%, around 310–315 billion cubic meters (MONRE 2006). Vietnam water resources are distributed unequally throughout geographical location and seasons. The Cuu Long River delta basin has the largest water proportion at 60%, the Hong and Thai Binh rivers have nearly a fifth of the national water volume at 16%, and less than a quarter of the total water volume belongs to other rivers. Vietnam also has seasonal precipitation in which the rainfall is concentrated mostly in the summer season and the other seasons have less than a third of the total rainfall. Therefore, many areas suffer drought for several months, especially in the winter (MONRE 2012). The other reason is water pollution. The water quality in Vietnam recently has been declining significantly under the pressure of population and economic growth.

Currently, the average water consumption is 9,560 cubic meter per capita annually. It is slightly lower than the middle water consumption country according to the average level of the International Water Resource Agency (IWRA). However, Vietnam accounts for a half (ca. 4,000 cubic meters) while other sources come from neighboring countries (MONRE 2012). The Vietnamese government has implemented various national water monitoring programs to protect the water environment. From 2008 to 2018, the general water monitoring programs of ten main river basins were established and implemented. In this study, we use a monitoring database in Cau River basin (from 2008 to 2018) to calculate the quality index and pollution index. The results of this study provide clear scientific evidence to regulate the situation of water quality of this river. The case of Cau River water evaluation could propose a solution for the Vietnamese government in applying water quality indices to managing and monitoring water sources in the future.

Cau River is the longest river branch of the Thai Binh River system at 288 km and its basin area covers 6,030 km² (Figure 1). The Cau River flows through four provinces including Bac Kan, Thai Nguyen, Bac Ninh and Bac Giang before being discharged into the Thai Binh River to the sea. The river's width changes following the flood season and drought season from 100 to 50 m, respectively. The river's bed is larger and the slope reduces to 0.05%. The downstream river is located in two provinces, Bac Giang and Bac Ninh, with an average elevation of 10–20 m (MONRE 2006).

According to the Center for Environmental Monitoring of Vietnam (CEM) water monitoring report (2018), there are 63 pollution sources discharging into Cau River which are scattered along four provinces: Bac Kan, Thai Nguyen, Bac Giang and Bac Ninh. The upper part is located in Bac Kan Province, and the population density is low at 65.65 persons per square kilometer. This area has ten pollution sources mostly related to agriculture production and residential activities. The middle part of the river flows through Thai Nguyen province, and it has a high population density as well as heavy industrial activities. CEM's water monitoring program at Cau River Basin has identified 16 sources of pollution in this area. The downstream of the river crosses two provinces, Bac Giang and Bac Ninh. These are populated places with diverse economies: agriculture, industrial and craft villages. The total waste water discharge into the river is estimated at 138.192 cubic meters per year (Table 1). The location of these pollution sources is the foundation for the establishment of water monitoring points.

A total of 22 monitoring points were selected for sampling of which six points are located in Bac Kan province, seven points in Thai Nguyen Province and the remainder belong to Bac Giang and Bac Ninh Provinces. A description of water sampling location is summarized in Table 2.

Water monitoring statistics of 22 points in the period of 2008–2018 from CEM was collected. Each water sample analysis contains 16 parameters which are described in Table 3. The pH, WT, turbidity and dissolved oxygen (DO) were analyzed by a portable pH meter, WT meter, turbidity meter and DO meter (D–50 Series, Horiba, Co. Ltd), respectively. Ammonium, nitrate, nitrite and phosphates ion concentrations were determined using a spectrophotometer (UV/VIS–EVOLUTION, Model EV0300PC). Heavy metal concentrations were determined using atomic absorption spectrophotometer (AAS) (EPA method). Coliform was analyzed by counting methods (ISO 10304-1:2007).

Based on the calculated score of WQI, water quality is classified into five categories:

1. Level 1: WQI score obtained from 0 to 25: Water is extremely polluted, emergency treatment is required before reuse.

2. Level 2: WQI score obtained from 26 to 50: water quality is suitable for transportation and equivalent purposes.

3. Level 3: WQI score obtained from 51 to 75: water quality is suitable for irrigation and equivalent purposes.

4. Level 4: WQI score obtained from 76 to 90: water quality is suitable for domestic usage.

5. Level 5: WQI score obtained from 91 to 100: water quality is suitable for domestic water supply.

CPI is classified into five categories:

1. Category 1: CPI from 0 to 0.20 (clean);

2. Category 2: CPI from 0.21 to 0.40 (sub clean);

3. Category 3: CPI from 0.41 to 1.00 (slightly polluted);

4. Category 4: CPI from 1.01–2.00 (medium polluted);

5. Category 5: CPI2.01 (heavily polluted).

In this study, we calculate CPI by using 12 water parameters: COD, BOD, TSS, amonium, phosphates, nitrate, nitrite, coliform, Fe, Cu, Zn and Cd. These parameters were analyzed in the Cau River water monitoring program.

DIN concentration is calculated by the total concentration volume of nitrate, nitrite, and ammonium, whereas the DIP is calculated by the concentration of phosphate in water. The EI is classified into two categories:

• 1.

  EI < 0: Zero eutrophication

• 2.

  EI > 0: Eutrophication

The TPI value is categorized into two groups:

• 1.

  Group 1: 0 < TPI 1 – none pollution

• 2.

  Group 2: TPI > 1 – pollution

In this study, TPI is calculated based on the average concentration of four trace heavy metals: Fe, Cu, Zn and Cd. These are four monitoring indicators in the general water monitoring program of Vietnam.

T-test analysis was used to evaluate the significant difference of water indices between the rainy season and the dry season in Cau River. ANOVA was used to identify the significant difference between upstream, midstream and downstream. Cluster analysis was applied to evaluate the water quality and pollution level between monitoring points. The points which have similar polluted levels were gathered into one group. The results of cluster analysis could assist managers to conduct better water monitoring and management plans.

The average values of ten consecutive years of water quality show that TSS and COD exceed the regulated level indicated in Vietnam National Standards QCVN08:2015/BTMT, column A1 – standard for domestic water usage (Table 4) (MONRE 2015). The TSS concentration was double the permission level, ranging from 49.46 to 54.52 mg/L compared to 20 mg/L of regulated value. The COD concentration was also higher ranging from 11.63 to 13.62 mg/L. Other parameters met the regulated levels. The table also shows the changes of water quality in the dry season and the rainy season. In comparison, parameters of TSS, DO and NH₄⁺ in the rainy season were higher than in the dry season, contrary to other parameters. This result was a consequence of the pollutants diluting due to the increase of river water volume in the monsoon season (MONRE 2006).

Table 5 presents the correlated relationship between monitoring parameters. The analysis shows the strong relations between physical and chemical parameters, especially the chemical parameters such as BOD, COD, NH₄⁺, NO₂^– and NO₃^–. The correlations were more significant in the dry season compared to the rainy reason. In a similar study, Jahin et al. (2020) employed multivariate analysis to develop an irrigation water quality index for suface water in Kafr El-Sheikh Governorate and found that the elements in waster have similar dynamics.

For overall quality assessment, data from 22 sampling points was calculated (Table 6). The average WQI scores were 67.52 and 69.67 in the dry and rainy season, respectively, indicating sufficient quality for irrigation supply. The WQI decreased from upstream to downstream indicating an accumulation of pollutants from discharge sources. Figure 2 shows that WQIs of Cau River were fairly stable in the rainy season and generally satisfied the quality for irrigation and similar purposes according to WQI's classification. However, the water quality in the dry season highly fluctuated due to geographical locations and the impacts of pollution sources. In the upper and lower part of the river, the WQI scores were mostly higher than the rainy season. Monitoring point number 5 nearly achieved the quality for domestic usage. This was a consequence of the steep river bed and the increase of turbidity in the monsoon period. Monitoring point number 10 was considered as extremely polluted. Water quality at other points in the river were categorized as suitable for transportation. The results show that the water quality index could be used effectively for water supply purposes. Similar research has been comprehensively conducted in many countries such as Egypt (Jahin et al. 2020), Sarayduzu Dam Lake, Turkey (Kükrer & Mutlu 2019), Amazonia Rivers, Brazil (Medeiros et al. 2017), and the Ganga River, India (Tripsthi & Singal 2019).

The results of water pollution indices calculation in both the dry and rainy season are summarized in Table 7.

The CPI data show the value of the entire river with no significant difference between dry and rainy seasons. In the dry season, the CPI of Cau River ranged from 0.50 to 1.57 with an average value of 1.08. According to the CPI's classification, this river was slightly and medium polluted. In the rainy season, the CPI of Cau River reached 0.66–1.37 with an average score at 0.96 and its quality was classified as the same level as the dry season.

Although the water quality of each water monitoring point tended to be better during the monsoon period, this difference was not statistically significant. However, the CPIs were different among the monitoring points upstream, midstream and downstream. The midstream had lower CPIs during the monsoon period in comparison to the dry season. We used the ANOVA test to analyze the significant difference levels of CPI in three parts of the river and the t-test to analyze the significant difference between the two seasons of the year. The results showed that the difference of CPI in three parts of the river was not statistically significant.

Similar to the CPI, the OPI in the dry season was higher than the rainy season (1.55 compared to 1.18). This result was significant with α = 0.1 (P= 0.0538). The OPIs of Cau River could be classified into two groups: Good (0 < OPI < 1) and Polluted (1 < OPI < 4). Regarding spatial location, the OPI of the upstream was 0.92 and 0.63 in the dry and rainy season respectively. These scores were in the good quality category (0 < OPI < 1). However, this difference was not statically significant. In the midstream of the river, the OPI was substantially higher in the rainy season (1.35 compared to 0.79 in the dry season) and this difference was statistically significant with α = 0.01 (P= 0.007559). The average OPI of the downstream in the dry season was 2.12, which was equal to the extremely polluted level, much higher than the OPI in the rainy season (1.83). However, the difference of OPI between the two seasons was not significant. Similar to the results of the CIP analysis, the OPI of the upstream had the highest quality and the downstream had the worst quality (Figure 3). However, the difference of OPI among three geographical areas was not significant.

The EI of the entire river was obtained in the range of about 100–400 for both dry and rainy seasons. This indicates that the river was at high risk of eutrophication. However, the EI in the dry season was much higher than the rainy season and this difference was statistically significant with α = 0.1 (P= 0.07977). According to geographical location, the EI of all monitoring points was higher than zero and increased from upstream to downstream of the river. There were significantly different average values of EI in both the dry season and the rainy season with α = 0.01 (P= 0.001376) and α = 0.001 (P= 0.000101) respectively. Specifically, the EI of the upstream was 128.57 and 85.54; the midstream was 224.49 and 101.64; downstream was 394.79 and 322.84 in the dry season and rainy season respectively. The EI in the dry season was always higher compared to the rainy season. However, this difference was only significant at midstream with α = 0.05 (P = 0.019287).

In contrast to other pollution indices, the TPIs of Cau River were low which clarified that this river was not polluted by trace heavy metals according to the TPI's classification. The rainy season also had higher TPI compared to the dry season, although this difference was not significant.

The TPI also slightly differed according to geographical locations. It was highest at the downstream and reduced in the upper part of the river. The difference of TPI in the upstream and downstream between the dry and rainy season was significant at α = 0.01 (P = 0.006916) and α = 0.05 (P = 0.026554). However, the average values of TPI at upstream, midstream and downstream monitoring points were not significantly different. The results of water pollution indices calculation determined that the level of pollution increased from the upper part to lower part of the river, specifically the CPI, OPI and EI, but not the TPI. The pollution level also tended to be higher in the dry season compared to the rainy season. This environmental status could be explained by the low density of pollution sources in the upstream. The midstream and downstream received more pressure from agricultural and industrial activities and populous areas. In addition, the accumulation of pollutants due to water flow in the downstream also contributed to higher concentrations of pollutants in this part of the river. The better water quality in the rainy season was a consequence of the increase of water volume which led to a higher dilution capacity. All trends of water pollution indices were reflected and are similar to the results of WQI calculations.

Cluster analysis was used in order to classify the monitoring points which have similar water quality characteristics. The computation categorized 22 points into five groups as summarized in Table 8.

According to Table 8, cluster analysis shows the order of water quality of which group 1 represents highest quality and group 5 the lowest. Representing the entire river, the polluted points (belonging to groups 4 and 5) accounted for around 13% (3 of 21 points in total). Nevertheless, in the dry season, the number of polluted points in groups 4 and 5 was double and the number in group 1 significantly reduced to six monitoring points. The results of cluster analysis stated that the water quality in monsoon time of Cau River was better compared to the dry season.

Grouping water monitoring points with similar characteristics could assist managers in making decisions related to water use planning for different purposes, such as domestic water supply, aquaculture cultivation, irrigation, or other purposes. Furthermore, the results of water quality clustering also support managers in designing water quality monitoring systems, especially addressing high attention to the serious pollution points. These benefits provide meaningful foundations for sustainable water source management.

This study has suggested a possible combination of quality and pollution indices based on monitoring environmental parameters for river quality assessment. At the current state, the water quality meets the requirements of Vietnam National Environmental Standard QCVN08-(A1 category), the standard for domestic water supply. TSS and COD concentrations are higher than the regulation for domestic water supply. The average concentration of pollutants was lower in the dry season, excluding TSS, DO and NH₄⁺. The results of WQI analysis indicate that the water of Cau River achieved the standard for irrigation purposes in both the dry season and rainy season. However, further study on bearing capacity of the river will be needed for water supply purposes.

The water quality indices varied depending on location and monitoring time. In the dry season, the water quality of the upstream was better than other parts of the river, followed by the midstream and downstream respectively.

Pollution Indices calculation indicates that the Cau River is polluted in different geographical conditions. Many locations of the river are contaminated by organic pollution with OIP > 1. The river water was at high risk of eutrophication as EI was above zero. The TPI of Cau River is in the safety level. All pollution indices in the Cau River tend to increase from upstream to midstream, then downstream. Cluster analysis grouped the water monitoring points into five groups with the quality reducing gradually from the first group to the fifth group. The classification of clustering analysis provided meaningful support for water pollution monitoring and appropriate solutions for the treatment of water for water supply.