Integrated modelling and management of manganese for a conventional potable water treatment plant

Highlights

• Integrated model to simulate manganese transport from a monomictic reservoir to a drinking water treatment plant.

• Seasonal variation of manganese are mainly influenced by wind direction and speed.

• Elevated manganese levels were typically associated with the solids handling process in the treatment plant.

• A decision support system was developed to improve manganese management.

Abstract

A reliable simulation of the manganese cycle in water supply system is essential for timely and effective manganese treatment strategies. A data-driven model estimating manganese levels in the treated water was developed through a threshold-based approach using raw water and supernatant return manganese levels. This data-driven model and a three-dimensional manganese cycle model for the source reservoir were integrated to simulate the overall manganese variations from the lake to the treated water. A decision support system based on the developed data-driven model was established to allow operators to react to situations that could result in elevated levels of manganese in the treated water. Scenario analyses were completed to examine the manganese variations in the reservoir, raw water and treated water due to different precipitations and wind conditions and to allow the identification of optimal strategies for manganese management from the reservoir to the water supply system.

Introduction

Manganese (Mn) issues in water supply systems have been of concern to water authorities for more than half a century. Primarily, Mn concerns are aesthetic in nature, but health concerns have also been raised [[1], [2], [3], [4], [5], [6]]. The World Health Organization guidance level for Mn in drinking water is 0.5 mg/L, which is far above the level causing aesthetic issues [7]. The colour and taste of tap water can change considerably when the Mn levels are greater than 0.05 mg/L, due to the build-up of oxidised Mn species onto the pipe walls which can subsequently become dislodged with changes in water flow, even at this more widely adopted aesthetic guideline [8]. In distribution systems sensitive to elevated Mn levels, a treated water level of less than 0.02 mg/L Mn is typically adopted [3,9].

Four oxidation states of Mn can be found in either natural water systems or in water treatment processes: Mn(II), Mn(III), Mn(IV) and Mn(VII). Mn(II) is considered to be the common form of soluble Mn in natural water and the two oxyhydroxides of Mn (MnOOH and MnO₂) are found in the Mn cycle in aquatic systems [[10], [11], [12], [13]]. Soluble Mn(II) can be derived from reducing conditions developing in soils and bedrock such as occurs in the bottom sediments of a lake’s hypominetic zone. Having migrated into natural surface waters oxidation of soluble Mn back to the insoluble and readily filterable Mn(IV) dioxide (MnO₂) may take considerable time. If Mn(II) cannot be oxidised and removed in the water treatment process, it may slowly oxidised to insoluble Mn(IV) in the distribution system and attach to pipe walls as a brown/black sticky film. If dislodged, under changed hydraulic conditions, the sticky film causes black-brown coloured water particularly when mixed with insoluble iron oxides [14]. Inadequate Mn(II) removal is largely attributed to an elevation in the Mn level of the water source creating partial oxidation states that may not be treatable using conventional means. The understanding of the Mn cycle in a water reservoir is the foundation for effective Mn treatment [15]. In a water treatment plant (WTP), the soluble Mn can be oxidised through pH manipulation and chemical oxidation by chlorine, chlorine dioxide, ozone and potassium permanganate, and subsequently removed by clarification and filtration.

Routine water treatment processes, such as coagulation, flocculation and filtration are usually sufficient to manage Mn concentrations to required levels as the Mn source in raw water during much of the year is very low, frequently below the detection levels of the commonly used atomic absorption spectroscopic method. However, in certain monomictic reservoirs, Mn levels rise during natural destratification in winter, when the Mn-rich hypolimnetic water blends with the Mn deficient epilimnetic water [15]. As raw water is usually sourced from the epilimnetic layers of the reservoir, additional Mn removal processes within the WTP will need to be initiated. The removal of Mn typically involves the oxidation of soluble Mn[II] to insoluble Mn[IV] and the subsequent removal of the solid oxide form via filtration. The oxidation of soluble Mn in a drinking water treatment plant is practical only through the addition of an oxidising agent and the elevation of pH [16]. The oxidation of Mn to the insoluble oxide is reversible, subject to redox state, and therefore it is imperative that a very positive ORP environment is maintained throughout the treatment plant until the solid oxidised form of Mn is completely removed from the treatment stream. Special care is required if solids handling process streams are recycled to the WTP inlet as these can be a significant source of soluble Mn if not managed correctly.

Traditionally, Mn levels in water reservoirs are monitored by routine sampling and analysis programs, which are expensive and time-consuming. However, more recently, researchers have developed hydrodynamic or data-driven models to predict the variation of Mn in lakes or reservoirs [[17], [18], [19], [20], [21]]. The management of Mn in WTPs varies between different authorities; therefore, the ability to create a site-specific model that integrates a hydrodynamic, predictive model for Mn in the water reservoir as well as for the processes within the WTP into a Decision Support System (DSS) would be of considerable value to water treatment plant operators. The application of most environmental DSSs is to assist decision-makers in analysing different potential options based on available information and process capability [22,23].

The Tarago Water Treatment Plant (TWTP) in Victoria, Australia, has experienced increased soluble Mn concentrations in raw water due to the rising levels of Mn in the water sourced from the Tarago Reservoir, consequently it elevates Mn concentrations in the treated water. A three-dimensional (3D) hydrodynamic model and Mn cycle model of the Tarago Reservoir were developed and validated [24]. The purpose of this research work was to develop a DSS for Mn treatment in the TWTP and integrate it with the Mn cycle model, to better inform operations and hence better manage the Mn risk in the Tarago water supply system.