Water Journal : Water Journal December 2011
refereed paper water treatment water DECEMBER 2011 67 The most relevant parameter to monitor in terms of treatment efficiency is dissolved organic carbon (DOC), which is a quantitative measure that has direct connection to both coagulant and disinfectant demand. Source water DOC (Figure 3) showed rapid increases within short intervals in May and December 2010, consistent with the onset of the new water quality periods, but peaked between February and March 2011 when the highest levels of aquatic organic matter arrived with the flood waters, three months after the turbidity peak that described the beginning of the period. Actual DOC removal was less complete than indicated by colour reduction and clear differences in the performance of the different treatments were apparent (Figure 4). In terms of water quality parameters, the absorbance of light is one of the easiest to apply to monitoring practices, as it is non- destructive and requires no chemical processing, even offering the opportunity for real-time online application. While determination of colour is also dependent on the absorbance of light, intensity at 456nm in the visible light region is of relatively low energy compared to absorbance in the ultraviolet spectrum. UV absorbance at 254nm has long been applied to characterisation of organic matter (Krasner et al., 1996; Korshin et al., 1999) and as a surrogate for the character fractions of DOC that are involved in coagulation mechanisms and ion exchange, as well as the greatest reactivity with disinfectants (Chow et al., 2006, Korshin et al., 2009). Although UV absorbance at a single wavelength does not present a complete picture of the organic character (Del Vecchio and Blough, 2004), it is more sensitive for the monitoring of treatment effectiveness than more commonly applied water quality parameters like colour. The raw water UV254 data is shown in Figure 5, with reduction by the four treatment processes in Figure 6. As removal of UV-absorbing material is favoured by coagulation and ion-exchange mechanisms, UV254 reduction is generally higher than actual DOC removal but retains the ability to differentiate between the treatments. The broad span of reduction (min. 27% to max. 96%) through the conventional coagulation process is indicative of the high sensitivity of this parameter to treatment effectiveness, giving clear variations between sub-optimal and optimised conditions. Comparing colour and absorbance at 254nm as surrogates for DOC removal shows that while acceptable correlations may be obtained for both parameters when values are high, such as in the raw water, when increasing levels of treatment are applied, the relationship of colour to DOC diminishes and becomes unreliable (Table 1). This is especially apparent for MIEX/ Coagulation (Figure 7), MIEX/Coagulation/GAC and MF/NF, as evidenced by lower correlation coefficients (R2) and increasing p-values. The p-value represents the statistical probability that the relationship between the parameters does not exist and is expressed as a number between 0 (certain) and 1 (no relationship). Although the difference between a colour of 1 and 3HU is not able to be visually distinguished and may be dismissed as a simple Table 1. Comparison of colour and absorbance at 254nm (UV254) as surrogates for dissolved organic carbon (DOC) monitoring using data from July 2009 to June 2011. DOC vs UV254 DOC vs Colour n R2 p-value n R2 p-value Raw 69 0.68 <0.001 69 0.85 <0.001 Conventional 65 0.92 <0.001 65 0.65 <0.001 MIEX/Coag 68 0.93 <0.001 68 0.24 <0.001 MIEX/Coag/GAC 64 0.86 <0.001 64 0.09 0.016 MF/NF 68 0.01 0.410 68 0.00 0.574 0 5 10 15 20 25 1-Jul-09 1-Oct-09 1-Jan-10 3-Apr-10 4-Jul-10 4-Oct-10 4-Jan-11 6-Apr-11 7-Jul-11 DOC (mg/L C) Mt.Pleasant WTP Source Water DOC Figure 3. Raw water DOC (mg/L C) at Mt Pleasant WTP, July 2009 to June 2011. MAP = Mannum-Adelaide pipeline. 0% 25% 50% 75% 100% Conv MIEX/Coag MIEX/Coag/GAC MF/NF % removed DOC Removed - July 2009 to June 2011 Average Figure 4. Average, maximum and minimum DOC removed by each treatment technology. Conv = conventional coagulation; MIEX/ Coag = MIEX adsorption and coagulation; MIEX/Coag/GAC = MIEX, coagulation and GAC adsorption; MF/NF = microfiltration and nanofiltration. 0.00 0.25 0.50 0.75 1.00 1-Jul-09 1-Oct-09 1-Jan-10 3-Apr-10 4-Jul-10 4-Oct-10 4-Jan-11 6-Apr-11 7-Jul-11 Absorbance at 254nm (/cm) Raw Water UV254 Figure 5. Raw water UV absorbance (/cm at 254nm) at Mt Pleasant WTP, July 2009 to June 2011. MAP = Mannum-Adelaide pipeline. 0% 25% 50% 75% 100% Conv MIEX/Coag MIEX/Coag/GAC MF/NF % reduced UV254 reduction- July 2009 to June 2011 Average Figure 6. Average, maximum and minimum absorbance at 254nm (/cm) removed by each treatment technology. Conv = conventional coagulation; MIEX/Coag = MIEX adsorption and coagulation; MIEX/ Coag/GAC = MIEX, coagulation and GAC adsorption; MF/NF = microfiltration and nanofiltration.
Water Journal April 2012
Water Journal November 2011