Water quality assessment of middle route of South-North water diversion project based on modi ﬁ ed Nemerow index method

It is of great signi ﬁ cance to ﬁ nd a scienti ﬁ c way to assess the water quality of the middle route of the South-North Water Diversion Project. In this paper, the spatio-temporal changes of the 26 water quality indicators in eight key monitoring stations along the middle route since the project was put into use were analyzed, and the modi ﬁ ed Nemerow index method was employed to assess the water quality of the route in its early stage of operation. The results show that the water quality remained good in the period of study. Important water quality indicators were identi ﬁ ed, including temperature, permanganate index, dissolved oxygen, total phosphorus, total nitrogen, mercury content, and density of planktonic algae. It is advised that potential ecological risks related to algae and shell ﬁ sh reproduction in the open canal deserve special attention. ﬁ compared. of


INTRODUCTION
The South-North Water Diversion Project of China is a grand strategic infrastructure project designed to alleviate the water shortage in northern China, optimize allocation of water resources, guarantee sustainable socio-economic development, and build a well-off society in an all-round way (Wei et al. ; Tang et al. ). Since the beginning of the construction of the Middle Route of the South-North Water Diversion Project (MR-SNWDP), its water quality has been an issue of top concern around the nation, and the water quality standards are high and difficult to meet (Zhao et al. ). Meanwhile, as the society moves

Building the assessment indicator system
The water quality assessment indicator system of the main canal of MR-SNWDP was built based on data of normal operation of the long-distance water diversion project and statistics from the MR Administration Bureau. The characteristics and risks of the MR-SNWDP were comprehensively considered, and new indicators including bio-toxicity and ecological indicators were added to the system, as shown in Table 1.

Data processing
The indicators were not comparable due to the different dimensions and magnitudes of the original data, and normalization was thus required to exclude the influences of different dimensions and magnitudes (Long et al. ). In particular, the larger the index value of a positive indicator, the better; while the smaller the index value of a reverse indicator, the better. Here, the 'Range 0 ∼ 1' method was used to process the data (Yang et al. ): suppose the indicators of the system to be X ij ¼ (x ij ) (i ¼ 1, 2, …, n; j ¼ 1, 2, …, m). Normalization processing is performed on X ij , and the normalization formula is as follows: where x min and x max are the miminum and the maximum values of the indicator x i .

RESULTS AND DISSCUSSION
Water quality characteristic analysis

General indicators
The indicators of the assessment system were analyzed first, and then a further characteristic analysis was performed on indicators with obvious changes, including the pH value, water temperature (T), permanganate index (PI), five-day biochemical oxygen demand (BOD 5 ) and dissolved oxygen (DO). Among nutritional salts, total phosphorus (TP), total nitrogen (TN) and ammonia nitrogen (NH 3 -N) were selected as the key indicators for in-depth analysis.
The annual average pH scale was between 8.1 and 8.4, indicating the water quality in the main canal of the MR-SNWDP remained alkaline throughout the monitoring and assessment period (Figure 2(a)). Water temperature (T ) showed a trend of overall decline from south to north, in conformity with the temperature change pattern of China, the smaller x i is, the better ( as shown in Figure 2(b). The maximum and minimum values were recorded at 32 C and 0.5 C, respectively, and the average temperature was 16-18 C, without significant spatial differences. The PI value displayed a gradual increase from south to north along the canal, as shown in Figure 2(c).
According to the limit of 2.0 mg/L for national standard   as shown in Figure 4(a); the values at several time points were lower than the detection limit of 0.0001 mg/L, but it requires long-term attention due to its high toxicity. The concentration of arsenic was all lower than the standard limit of Level I water, as shown in Figure 4

Ecological factor indicators
The ecological factors for assessment include comprehensive biotoxicity, fecal Escherichia coli and density of planktonic algae. The percentage concentration of luminescent bacteria was used to monitor the comprehensive biotoxicity, and there was no case where the concentration of luminescent bacteria was higher than 30%, but the comprehensive biotoxicity remains a very important indicator.
The overall average concentration of fecal E. coli in all monitoring stations was lower than 200 ind./L, meeting the requirements for Level I in GB, as shown in Figure 4(c).
The concentration tended to increase gradually from south to north along the canal, and this increasing trend is particularly prominent in the northern part of the waterreceiving area. Algae is another important ecological indicator. The annual average data in Figure 4(d) revealed that the density of planktonic algae (DPa) gradually increased from south to north, but not an increase in the strict sense.
This trend may be related to the impact of management measures such as manual salvage of algae.
Comprehensive water quality assessment based on the modified Nemerow index method With the modified Nemerow index method, comprehensive water quality assessment was carried out based on the standard limit of Level II water quality in the Environmental These values of water quality standard limits (for rivers) were then used to calculate the modified Nemerow index corresponding to various water quality levels, as shown in Table 2. When the calculation result of the modified Nemerow index is less than 1, the water quality is considered as A1; that is, the best water quality; when the calculation result of the modified Nemerow index is between 1 and 2, the water quality is categorized into A2, which means the water is 'safe and sensitive', and so on.

Impact of weight on the assessment results
In routine operation, the MR Administration Bureau used the single-factor assessment method based on the national standards for surface water assessment to select the worst water quality level as the basis for assessment (Figure 6(a)), but this method might fail to reach an accurate assessment of the overall water quality. The assessment result obtained by this method showed that the water quality of the monitoring stations was at a poor level, meeting the GB standards for Class IV or Class V water, as shown in Figure 6(a). (1) During the study period, the water quality of the MR-SNWDP has stayed at 'Level II' or 'Level I' water quality level; that is, a 'good' or 'excellent' level, of the GB Water Quality Standards (with the impact of TN concentration excluded); physicochemical and heavy metal indicators all met the GB standards of Level II water quality; water safety was assured during the monitoring and assessment period.
(2) Compared with the single-factor assessment method, the modified Nemerow index method delivers a more objective assessment result and better reflects the changes of water quality in spring and autumn; it also demonstrates the contribution of indicators like the heavy metal content and nutrient salt content to the overall water  quality, thereby providing a more comprehensive view of the impacts of different factors on the water quality than the single-factor method. Despite the different results these two methods reached, both showed that the water quality of the Middle Route remained at a good or excellent level since it has been put into service.
(3) Major water quality indicators that deserve special attention were defined, including temperature, permanganate index, total phosphorus, total nitrogen, mercury, arsenic and density of planktonic algae. Moreover, it is advisable that administrations pay more attention to environmental indicators like the temperature and TN/TP that might cause abnormal reproduction of algae and shellfish.