Sulfide control optimization using sewer process models

INTRODUCTION
The Albuquerque Bernalillo County Water Utility Authority (Water Authority) owns and maintains over 2,000 miles of pipes in the collection system to convey approximately 50 MGD wastewater from the service area to the Southside Water Reclamation Plant (SWRP). The Water Authority has implemented a range of methods and technologies for controlling odors and corrosion in the collection system, resulting in significant institutional knowledge and experience of this branch of engineering science. Nevertheless, the sewer network still faces corrosion challenges due to sulfide generation with numerous pipes that are at risk of collapse. Comprehensive analysis of a complex collection system to allow for the optimization of sulfide control relies on the understanding and quantification of the processes impacting sulfide production and release (Revilla et al, 2016). To facilitate this intricated task, properly implemented sewer process models become a valuable tool. This implementation could also support the evaluation of the impacts of the sulfide control strategy on the SWRP process performance. The objective of this project was to marshal and organize the resource expenditure on odor and corrosion control so that it is optimized, coordinated across the collection system, and harmonized with the pipe rehabilitation program and the identified operational constrains of the SWRP. A novel approach using sewer process model (the WATS model) was implemented to develop a comprehensive sulfide control strategy, and to empower the Water Utility with a tool to support the migration from a reactive to a proactive approach on asset management while being able to communicate to SWRP staff the potential impacts in the treatment process.

METHODOLOGY
The study area for the Water Authority collection system included approximately 250 miles of pipe with a minimum diameter of 15 inch in four main interceptors (plus tributaries). A sewer process model was used to apprehend the study area as a whole and analyze physical, biological and chemical processes bearing on odor and sulfide corrosion. Data Collection. The Water Authority collected data on several parameters as part of the regular monitoring program for the collection system in 17 locations and the influent to the WWTP. These included grabs samples for wastewater temperature, pH, total and dissolved sulfide, total iron and continuous monitoring of headspace hydrogen sulfide. Additional monitoring at each of the most downstream location in the four interceptors was performed. Sewer Process Modelling. The WATS model (Hvitved-Jacobsen et al, 2013) was selected as the most appropriate software tool for this purpose. WATS is currently the most rigorous and complete sewer process model in the world and is well suited to assessing the Water Authority sewerage. The model was set up using the existing hydraulic model of the system and considering the operation of the collection system at the time of the selected calibration period in terms of flow diversions and chemical addition. Considerations for the optimization strategy in the collection system were in place to prevent negative impacts in the SWRP: 1) a minimum pH of 7 in the plant influent needed to meet the permitted effluent pH; 2) a maximum magnesium concentration in the influent to prevent formation of nuisance struvite (impacts use of magnesium hydroxide); 3) a limit in the use of solids produced in the potable water treatment plant that have the potential to be used for sulfide control if discharged to the collection system.

RESULTS AND DISCUSSION
Model Calibration The WATS model for the Water Authority collection system was calibrated to dissolved sulfide and headspace hydrogen sulfide in the 17 monitoring locations throughout the four interceptors. In addition, diurnal pH data was available to refine model calibration. Data for winter and summer were available and seasonal models were developed to allow for a more detailed strategy for optimizing sulfide control. Stochastic simulations (n=100) were run and the results compared to the complete data set for the calibration period. Figure 2 shows measurement-model agreement for station 10. Alternatives analysis for sulfide control optimization The calibrated WATS model allowed for the screening of a wide range of alternatives, including chemical dosing at various dose rates using different chemicals at various locations, and assisted in the prioritization of pipe rehabilitation given the relative cost of on-going chemical treatment versus rehabilitation. The screening resulted in novel and the most cost-effective strategies with potential to be implemented in each of the four major interceptors. Some of this approaches include: Addition of iron-rich water treatment plant solids. The San Juan Chama Water Treatment Plant can discharge residual sludge containing high levels of iron for sulfide control in the collection system. Bench-scale testing provided the input to the WATS model in terms of sulfide removal efficiency. It was determined that the amount of solids the SJCWTP was limited to discharge to prevent negative impacts at the SWRP was enough to provide comprehensive treatment for the Edith Interceptor and partial treatment at the Valley Interceptor. Comprehensive treatment at the Valley Interceptor was achieved by dosing Iron for the upstream end and by applying PRI-SC to reach sulfide control at the downstream end of it, providing the most cost effective approach. Iron addition with pH adjustment and Peroxide Regenerated Iron for Sulfide Control (PRI-SC). An evaluation was performed using the WATS model combining chemical dosing for a synergistic and cost-effective interaction between Mg(OH)2, ferric and PRISC. As shown in Figure 2, the Westside Interceptor relies on iron addition in the upstream end of the system with pH adjustment using Mg(OH)2, recognizing that sulfide precipitation rate increases as pH approaches 8 units (Hvitved-Jacobsen et al, 2013). This treatment targeted Station 64 where the PRISC process allowed for a more cost-effective solution to achieve comprehensive treatment to the nearest pump station. In addition to considering the limitations set by the SWRP influent characteristics to protect plant operation, a qualitative assessment of the impacts of the optimized strategy on the performance of the SWRP resulted in potential benefits to sulfide control in the solids train; SWRP better prepared for potential stricter phosphorus limits; and) an increase in soluble substrate for enhanced BNR. The optimization routine using WATS resulted in chemical selection and dosing maximizing existing infrastructure while maintaining wastewater characteristics in the SWRP influent within the constrains identified by plant operations.

CONCLUSION
The careful implementation of the WATS process model to the Water Authority collection system allowed for the understanding of the processes bearing in sulfide production and release, and for the screening and identification of alternatives for controlling sulfide in the system as a whole and in an optimized manner. The results confirmed that the presented methodology allows for the incorporation of modelling tools in the decision making for management of collection systems.