Assessment of Some Water Quality Parameters in the Storm Sewer Network at Ramadi City, Iraq

The quality of storm water is being negatively impacted by urbanization as well as an increase in the frequency and severity of floods [1]. Storm water is being progressively contaminated by a wide range of biological and chemical contaminants that are the direct consequence of human activities in urban areas. Storm water runoff that has been polluted is one of the primary contributors to the transfer of contaminants into the water [2].

In urbanized settings, the storm water drainage system is critical. This system gathers rainfall and transports surface runoff over a long distance before discharging it to the water receiver [3]. A considerable quantity of pollutants settles in storm streams, forming sediments that induce the re-release of different sediment pollutants after being discharged into receiving waters, increasing the release of pollutants to receiving water bodies [4]. 

In order to maintain the system's right functioning and analyze pollution in stormwater sewer systems and whether it constitutes a risk to the environment or human health, the sediment must be collected on a regular basis and handled and correctly or disposed of the study [5]. The suspended particles present in the runoff are the most important in terms of quantitative and qualitative assessment of the sediment [6].

The long-term persistence of surface water runoff and unmanaged or improperly controlled discharges is one of the main causes of low-quality water. Drainage control has therefore become more crucial in recent years when it comes to sanitation issues. Additionally, big investments have been made to address these issues [7].

Because there are so many obstacles to overcome when trying to solve these problems, the creation and application of water quality models has become increasingly important. During the course of the last several years, a number of these models have been developed in order to simplify the management of water pollution control.

Powerful mathematical methods can be used to anticipate performance and improve operating control based on previous notes for certain important parameters. This can be done by looking at the relationship between those two factors. On the other hand, modeling water quality is challenging.

It is difficult to accurately characterize nonlinear behaviors using linear mathematical models [8] due to the complexity of the physical, chemical, and biological processes that take place in treated wastewater. The development of quality control programs for the quality of treated wastewater has the dual purpose of assisting planners and protecting freshwater resources already in existence. The quality of treated wastewater is the focus of the quality control programs. In most cases, determining the quality of water involves making a comparison of the current values to the standards that have been averaged over a certain amount of time. This allows for the extraction of important information about the quality of the water based on the spatial and temporal.

There have been a great number of investigations carried out all around the world to investigate the levels of pollution found in sewers water. The effectiveness of multivariate linear regression (MLR) and artificial neural network (ANN) models is examined in this study to forecast two crucial water quality metrics in a wastewater treatment facility. The indirect markers of organic content and the biochemical and chemical oxygen demands (BOD and COD) are indicative of the quality of wastewater [9].

Researcher's study intends to examine the possibility for using sewage water for irrigation and investigate the overall performance of Rustumiya Station in Baghdad, Iraq. Additionally, assess the environmental quality (EQI) of the wastewater at the water treatment facility to understand the impacts of wastewater at the Rustumiya STP on the Diyala River, and then create prediction models using the EQI ANN Model and Multiple Linear Regression (MLR) Model [10].

It aims to create a new water quality index (WQI) equation for Mirim Lagoon using multiple linear regression analysis models. It also assesses the suitability of these models for monitoring the water quality of this resource and whether the WQI values produced by the new equation can be used to accurately monitor the quality of the lagoon's waters [11]. The researcher's aims included establishing a connection between the origin of a pollutant and the nature of the pollutant, as well as determining how the distribution of heavy metals changed throughout transportation between dissolved and particulate forms of the pollutant [12]. The contamination of the Euphrates River was the primary focus of the researches conducted in Iraq. research studied temporal and geographical variations in Euphrates River Water Quality Index in Anbar Governorate [13]. A search for TDS concentrations during three years (2011, 2012, and 2013) and 3-year average values along the Euphrates River in Iraq [14]. Using satellite photos, the purpose of the research was to make a prediction about the regional distribution of water quality metrics in the Euphrates [15]. This research focused on the water quality standards of the Euphrates River in Iraq, namely in the towns of Hit and Ramadi, as well as the influence flow has on water quality [16].

The objective of this study is to assess the current status of the water quality of storm sewer system before it flows to the last estuary in the Euphrates River and Al-Warrar Canal in the Ramadi city. Water samples were taken from the rain collection holes (manholes), the main collection basin in the station, and the outlet hole before it flows into the river. This study was included, laboratory analyzed of, water quality parameters, and calibrate the WQI with a multiple linear regression model. This work intends to construct a novel WQI equation using multiple linear regression models and evaluate their feasibility for water quality monitoring of this water resource, as well as whether the resulting WQI results can be utilized accurately to monitor storm sewer quality.

The remainder of this article is structured as follows: Section 2 provides a description of the study area, and the third section describes the sampling sites, the method of collecting and analyzing samples, as well as describing the statistical analyzes and the method of knowing water quality index, and the fourth section describes the results of the analyzes and the most important elements causing water pollution the storm sewer , and describes the composition A new equation for the standardization of water quality, the fifth section describes the summary of the research and the most important recommendations reached by the researchers.

3.1 Sampling locations

Water samples were collected from three districts in downtown Ramadi city (Al-Moallemen District, Al-Malaab District, and Al-Shurta District). Samples were collected in each of the three sites: the first site is the rainwater collection hole (manholes), the second site is the station's main collection basin, and the third site is the storm sewer outlet opening before the water is discharged into the Euphrates River or the Warrar Canal. Figure 2 shows the sampling locations in each region. The samples were taken in two parts, the first time the samples were collected in the dry season before the rainy season, and the second time the samples were collected after the rainy season, but this year's rains were irregular and rare. During the past two years, Iraq has suffered from a lack of rain and its scarcity and climate change, in contrast to previous years, when drought increased and the rainfall rate decreased to less than 50% of its normal rates.

Table 1 shows the coordinate locations of the samples, where it denotes the sample taken from manholes (A), the sample taken from the main collection basin of stations (B), and the sample taken from an exit before it poured into the river (C).

2.png

Figure 2. Sampling sites in Ramadi city center

Table 1. The location coordinates of the regions from where the samples were obtained

3.2 Sampling collection

Water samples were collected in October and April. Samples were stored in sterile, hermetically closed glass plastic containers. The analyzes were carried out in the laboratories of the University of Anbar, the College of Science, and the laboratories of the Great Ramadi Water Project Department.

3.3 Analytical and statistical analysis

In this study, some physical and chemical properties of the water quality in storm drains were examined. They include (PH), electrical conductivity (E.C), alkalinity (Alk), total hardness (T.H), calcium (Ca), biological oxygen requirement (BOD₅), chemical oxygen requirement (COD), nitrate (NO₃), sulfate (SO₄), dissolved solids (TDS), suspended solids (TSS), and some heavy metals such as copper (Cu) and zinc (zn). Electrical conductivity and pH were measured immediately after sampling, total hardness (T.H), alkalinity (Alk) and chloride were measured by titration, Cu concentrations were determined by atomic absorption method (AAS). The Canadian CCME WQI model is characterized by its widespread and global use by researchers to assess water resources and determine the degree of their pollution [17]. This model does not care about the weight of characteristics. Parameters in which even one test deviation from the standard limits, but rather exceeds it to the weight of each measurement (test value) deviating from the standard limits, which gives high accuracy in evaluating the quality of the studied water [18].

The Canadian model (CCME WQI) was used in this study using the following equations and to find the values of the index by calculating three factors as follows:

(1) F1 (Scope) represents the percentage of variables that exceed the standard limits compared to the total number of variables (even once during the study period).

$\mathrm{F} 1=\frac{\text { number of failed variables }}{\text { total number of variables }} \times 100$             (1)

(2) F2 Frequency: the percentage of individual examinations exceeding the limits of the standard over the total number of examinations

$\mathrm{F} 2=\frac{\text { number of failed tests }}{\text { total number of tests }} \times 100$             (2)

(3) F3 (Amplitude) represents the amount of tests passed and is calculated in two stages.

The first stage: the number of times the individual concentrations exceed the standard limits, and it is called the (Excursion), and it is calculated as follows

Excursion $\mathrm{i}=\left(\frac{\text { failed test value } \mathrm{i}}{\text { objective } \mathrm{j}}\right)-1$             (3)

In the case that the exceeded test value is higher than the standard value, the calculation for it is done by inverting the ratio.

The second stage: the amount of the individual test set that is exceeded, and it is calculated by summing the individual deviations and dividing it by the total number of tests (overdone and non-override). This variable is called the (normalization of Excursion) and is symbolized by (nse).

$\mathrm{nse}=\frac{\sum_{i=1}^n \text { Excursioni }}{\text { numer of tests }}$             (4)

The Amplitude F3 is calculated from the following equation:

$\mathrm{F} 3=\frac{n s e}{0.01 n s e+0.01}$             (5)

After finding the three factors, the Canadian index is calculated from the following equation:

CCME WOI =$100-\left[\frac{\sqrt{F 1^2+F 2^2+F 3^2}}{1.732}\right]$             (6)

The constant 1.732 A, to adjust the result of the index value and make it confined between 100 - 0.0 [19].

Water quality is categorized into five categories as shown in the Table 2 [20].

Table 2. The Index for the quality of water Canadian council of ministers of the environment (CCME WQI)

3.3.1 Multiple linear regression (MLR)

The linear equation for multiple linear regression is [21].

$\mathrm{Y}=\alpha+\beta 1 \mathrm{X} 1+\beta 2 \mathrm{X} 2+\cdots+\mathrm{e}$            (7)

where, Y = dependent variable

α = Constant or Intercept

β1 = slope of the gradient y on the first independent variable

β2 = slope of the gradient y on the second independent variable

X1 = the first independent variable

X2 = the second independent variable

Once we have the outcomes of the multiple regression equation, the next step is to demonstrate that these coefficients are statistically acceptable, also known as statistically significant. It is important to note that the significance is determined on an individual basis for each parameter. The T-test and the probability level that corresponds to it are what we use to determine whether or not the regression coefficients have significant meaning. It should go without saying that the R and SPSS programs will automatically extract the T-test and the probability level that corresponds to it [9].

In addition to this, it will acquire statistics that are used for determining the overall relevance of the model, such as (R) (R2). The first R is known as the simple correlation coefficient, and it determines how strongly two or more variables are related to one another. The second R2 is known as the coefficient of determination, and it determines how well the estimated model explains the data (estimated equation) [11].