jalil javadi orte cheshme; mahmood kashefipoor
Abstract
Introduction: Nowadays, contamination of water is one of the problems that are more considered. Fecal Coliform (FC) is one of the most common indicator organisms for monitoring the quality of water. The problem that complicates the modeling of indicator organisms such as Fecal Coliform is determining ...
Read More
Introduction: Nowadays, contamination of water is one of the problems that are more considered. Fecal Coliform (FC) is one of the most common indicator organisms for monitoring the quality of water. The problem that complicates the modeling of indicator organisms such as Fecal Coliform is determining the appropriate amount and an optimum rate of decay. It has been reported by many scientists that the decay coefficient or mortality rate is significantly affected by environmental elements. In this study, the effect of environmental parameters such as temperature, turbidity, radiation and suspended sediment concentration on the coliform decay coefficient hasbeen verified to have a dynamic and variable decay coefficient for better and reliable estimations of fecal coliform concentartion values.
Materials and Methods: Karun River is the longest and largest river in Iran. In this study, due to the accumulation of pollutants from industrial and agricultural wastes near Ahvaz city and for existence of quality measurement stations along the river, the Mollasani station to Farsiat station was selected to simulate and evaluate the hydrodynamic and quality of the river. The FASTER model has been used for modeling of the flow, sediment and water pollution. In this study, the dynamic roughness Manning coefficient has been used for more accurate simulate the flow, that had been added to the model by Mohammadi and Kashefipour. In Coliform bacteria and sediment modeling, some other dynamic parameters such as longitudinal dispersion coefficient are important and increasing or decreasing of these parameters are very significant and the accuracy of the Advection-Dispersion Equation (ADE) depends on the choice of the theoretical and/or experimental relations of these parameters. It was previously found that the Fisher equation performs the best for Karun river in modeling coliform, and this equation was therefore used in this study to calculate the dispersion coefficient. In order to investigate the effect of suspended sediment concentration on coliform decay rates, first this parameter must be modeled. In this research, the von Rijn method was used for modeling the suspended sediment load. In order to modeling the caliform, all dates of measuring were firstly determined in Zargan station; for each date the model was run for several times. For each run the decay coefficient was selected accordingly, until the predicted concentration by the model has the least difference inthe corresponding measured values. After that, the measured amount of environmental parameters such as Temperature, TUrbidity, RAdiation and also, the modeled values of suspended Sediment concentration wasdetermined for the same dates. Then, using a statistical software a relationship was developed to describe the decay coefficient as follows:
(1)
Results and Discussion: Using a statistical software, an equationfor decay coefficient was derived as follow:
(2)
Where K is decay coefficient (hr-1), T temperature (°C), TU turbidity (NTU), RA radiation(mmH2o-Vaporizeable) and Se suspended sedimentconcentration (kg/m3). Equation (2) was then added to the FASTER model, so the model was able to calculate the decay coefficient using the calculated suspended sediment at any time of simulation and this equation (dynamic decay coefficient). To be able to compare the dynamic decay coefficient and constant decay coefficient, the model was performed repeatedly for the whole calibration period and each time one constant K was given to the model. The best constant decay coefficient for the period of calibration and validation patterns was obtained to be K= 0.05 hr-1.Tables (1) and (2) show the amount of accuracy in predicting the suspended sediment concentration and coliform in both calibration and verification patterns, respectively. Table (1) shows that the FASTER model was able to estimate the suspended sediment concentration relatively accurate. Table (2) compares the effect of a constant decay coefficient versus the dynamic decay coefficient inaccurate estimation of fecal coliform concentrations.
Table 1- Comparison of the estimated error and correlation of suspended sediment
Pattern R2 a %E RMSE
Calibration 0.85 0.95 29.81 0.039
Verification 0.87 1.3 30.52 0.059
Table 2- Statistical parameters for coliform concentrations predicted and measured
Perioud k R2 a %E RMSE
Calibration Relation (2) 0.97 1.2 19 1906
0.05 0.92 2 50 4341
Verification Relation (2) 0.94 1.4 20 3860
0.05 0.77 1.5 44 7384
Conclusions: Comparison of the predicted fecal coliform concentrations with the corresponding measured values in the calibration and verification periodsshowed that the error estimate improved respectively about 31% and 24% when the dynamic decay coefficient was used instead of a constant value (the best constant value was obtained 0.05hr-1). The concentration of coliform bacteria in Zargan station during the total time of studying is more than 1000 CFU/100ml. Due to coliform bacteria concentrations and compared them with the levels allowed by the Standards, Karun river water is not suitable for human's drinking, confined livestock drink, food industry, oyster farming, irrigation products that are consumed raw and recreational uses (contact with water) like swimming.
