Sewer emissions are a notorious problem that water utilities have to deal with. The production and emission of hydrogen sulfide is a well-known problem for decades, which is the primary cause of sewer odors and corrosion. However, hydrogen sulfide is not the only harmful emission from sewer networks. Methane can also be generated in sewers through methanogenesis. Methane is a highly potent greenhouse gas, which is significantly contributing to climate change. Over a 100-year horizon, 1 ton of CH4 will induce a warming effect equivalent to 34 tons of CO2. It is also explosive with a lower explosive limit (LEL) of approximately 5% by volume, and thus poses a serious safety issue. Currently, very little attention has been paid to methane formation in sewer networks. Therefore, the overall aim of this thesis is to measure and understand methane emission from sewers.
The limited studies conducted so far on methane measurement in both gas and liquid phases in sewers have relied on manual sampling followed by off-line laboratory-based chromatography analysis. These methods are labor-intensive when measuring methane emissions from a large number of sewers, and do not capture the dynamic variations in methane production.
Therefore, the suitability of infrared (IR) spectroscopy-based online methane gas sensors for measuring methane in humid sewer air was investigated in both laboratory and field conditions. Under certain circumstances, humidity could become an issue, however, this can be solved by removing the humidity on the sensor probe surface. Also, IR sensors exhibit excellent linearity and can be applied with factory calibration. Furthermore, the detection limit of sensors is suitable for measuring methane gas in sewers. Field application of the sensors revealed that methane concentrations in sewer air are 3 – 4 orders of magnitude higher than in the atmosphere, confirming that sewers are a source of methane. The continuous measurement also revealed that methane concentrations in sewer are highly dynamic.
Complementary to the gas phase sensors, a new dissolved methane sensor was developed in this thesis. This device uses an online IR gas-phase methane sensor to measure methane under equilibrium conditions in a stripping chamber. The measured gaseous methane levels were then converted to liquid-phase methane concentrations according to Henry's Law. The detection limit and range was noted to be suitable for sewer applications. Good linearity was also obtained during field application. The newly developed online dissolved methane sensor was demonstrated through monitoring dissolved methane concentrations at the end of a rising main sewer network. This was done over two periods of three weeks each, in summer and early winter, respectively. Wide variations in dissolved methane concentrations were measured at 5–15 mg/L in summer and 3.5–12 mg/L in winter. This corresponded to significant average daily summer and winter methane productions of 24.6 and 19.0 kg-CH4/d, respectively, from a network with a daily average sewage flow of 2,840 m3/day. The dissolved methane concentrations demonstrated a clear diurnal pattern coinciding with flow and sulfide fluctuations and implying a relationship with the wastewater hydraulic retention time (HRT).
Contributions of sediments in gravity sewers to overall sewer emissions are poorly understood at present. Sediments collected from a gravity sewer were cultivated in a laboratory reactor fed with real wastewater for more than one year to obtain intact sediments for the study. Batch test results clearly showed significant sulfide and methane production from sewer sediments. Microsensor and pore water measurements of sulfide, sulfate and methane in the sediments, microbial community profiling along the depth of the sediments and mathematical modelling revealed that sulfide production takes place near the sediment surface due to the limited penetration of sulfate. In comparison, methane production occurred in a much deeper zone below the surface likely due to the better penetration of soluble organic carbon. Modelling demonstrated the dependency of sulfide and methane production on the bulk sulfate and soluble organic carbon concentrations, and these relationships were described well by half-order kinetics.
In order to control sewer emissions from sediments, the effect of nitrate dosing on sulfide and methane production in sewer sediments was investigated through laboratory studies. It was found that nitrate addition does not suppress sulfide production in sewer sediment, but it reduces sulfide accumulation through anoxic sulfide oxidation in the sediment and hence, also reduces sulfide accumulation in the bulk water. Microsensor measurement of sediment sulfide revealed the presence of sulfide oxidation and sulfide production zones with the interface dynamically regulated by the depth of nitrate penetration. In contrast, with nitrate dosing the methane production activity of sewer sediment was substantially reduced. This was likely due to the long-term inhibitory effects of nitrate on methanogens. Pore water measurements showed that methane production activity in the sediment zone was completely suppressed with frequent nitrate exposure, and consequently, the methane production zone re-established deeper in the sediment where nitrate penetration was infrequent.