NATIONAL CONSENT TRANSLATION PROJECT CHEMICAL OXYGEN DEMAND OF SEWAGE EFFLUENTS ('HARD COD') FINAL REPORT

Report No FR/CL0006

NATIONAL CONSENT TRANSLATION PROJECT CHEMICAL
OXYGEN DEMAND OF SEWAGE EFFLUENTS ('HARD COD') FINAL REPORT

FR/CL0006

Jul 1997

SUMMARY

EXECUTIVE SUMMARY

1.1 Background

The December 1994 report of the National Consent Translation Project (NCTP) (FWR 1994) included, an evaluation of the "percentage reduction" option for compliance assessment provided in the Urban Waste Water Treatment Directive (UWWTD). One of the conclusions from this part of the NCTP study was that:

'For COD, in contrast (to BOD), the 75% removal limit was a difficult target for the majority of works. Out of the 29 works with sufficient data for the assessment only nine complied. Against the effluent limit of 125 mg/l, however, 22 works complied. Six works failed both limits.'

This conclusion led FWR in consultation with its Research Advisory Group to formulate a brief for a follow-up study to investigate in more detail the extent and significance of the problem of COD removal. Invitations to tender for this research work on "Chemical Oxygen Demand of Sewage Effluents" were issued in October 1995 and the project formally started on 1 March 1996.

1.2 Objectives

The Objectives of the study as set out in the Terms of Reference are:

'To investigate the nature of COD in sewage effluents (containing minimal trade effluent) which appears not to be removed in conventional treatment processes: to consider the environmental implications of COD components in effluents discharged to rivers and potential impacts on water treatment plants and drinking water supplies.'

1.3 Organisational arrangements and responsibilities

The organisational arrangements and responsibilities can be summarised as follows:

Foundation for Water Research (FWR): The FWR was responsible for the management of the overall programme through the appointment of a programme manager whose role was to liaise with all the participants and undertake report production and dissemination.

Montgomery Watson Ltd: The review of published information, collection, collation and interpretation of data and preparation of the project report was undertaken by the research contractor Montgomerey Watson Ltd.

Regulators: The environmental regulators comprising the Environment Agency, and the Scottish Environmental Protection Agency and the Department of the Environment for Northern Ireland represented by SNIFFER (the Scotland and Northern Ireland Forum for Environmental Research) funded the programme management by FWR and the work of the specialist sub-contractor Montgomerey Watson Ltd.

Operators: The operators (comprising the Water Service Companies of England and Wales) together with the sewage treatment operators in Scotland and Northern Ireland. permitted the data collected for the initial NCTP project to be used for this study and provided additional data where these were available.

Steering Group: A Steering group representing all participants in the project was formed to oversee the study and approve outputs. A representative of OFWAT was also invited to join the Steering Group.

1.4 Findings

1.4.1 Influent and effluent quality

Typical levels of COD in sewage of mainly domestic origin are in the range 200-1000 mg/l. Average influent concentrations for UK STWs generally lie between 200 and 600 mg/l. Higher concentrations may occur in combined sewer systems after storm events particularly in the first flush of polluting matter.

Influent COD levels are well correlated with influent BOD and SS. Average COD:BOD ratios in influents generally lie in the range 2-4.5.

COD concentrations in sewage effluents vary widely but average values generally lie within the range 10-100 mg/l. The correlations between COD, BOD and SS are less strong in effluents than in influents but are still evident. Average COD:BOD ratios in effluents vary widely between works and may lie between about 5 and 20.

1.4.2 Compliance with UWWTD standards

The NCTP Phase l Study provided data to assess COD removal on 29 of the sample of 45 well-performing STWs making up the data set. Of the 29 works assessed, 7 failed the 125 mg/l limit for COD and 20 failed the standard for 75% removal.

When the limits are applied to each sample on an "either/or" basis as proposed in the draft DoE Guidance Note, the total number of failures reduces to 6 (STWs 13, 31, 64, 69, 70 and 78). Of the 29 works assessed for COD removal, 3 also fail one or other of the UWWTD standards for BOD but there are no failures of the BOD limits on an "either/or" basis. Two of the works that fail a BOD limit also fail both COD limits.

Analysis of the works failing either COD limit shows no clear pattern in terms of the works characteristics provided in the NCTP data (type of works, size of works, trade effluent %. BOD consent, SS consent, AmmN consent).

Analysis of the works failing both COD limits does show that none have consents for ammonia and that 3 out of 6 have relatively relaxed BOD:SS standards of 40:40 or more.

1.4.3 Treatabilitv

According to the literature, at least 50% of the influent COD is generally readily biodegradable. But there is a residual of 3-5% of the influent COD that is non-treatable and represents inert material.. The fraction between the readily biodegradable and the non-treatable is capable of treatment but may not be treated at a particular works.

There does not appear to be a rigorous definition of "hard COD". It is often used to describe the total residual of COD present in an effluent and as such would include both the inert non-treatable fraction and the fraction of more difficult to treat material not treated at the particular works. Using this definition, embracing both non-treated and non-treatable components of COD, the NCTP data indicate a typical level of hard COD for a well-performing STW of 40-70 mg/l. The absolute minimum value, representing the nontreatable fraction, appears to be about 10-20 mg/l.

The treatability of the COD in an influent depends on both the character of the influent and of the works treating it. No single model has been found that gives a good explanation of treatability. However, from the literature and the data analysis, the presence of one or more of the following factors increases the probability of poor treatability (% removal of COD):

a high COD:BOD ratio of more than about 2.5 (or conversely a low BOD:COD ratio of less than 0.4);
highly variable COD:BOD ratio due to effects of rainfall run-off or for other reasons;
works without nitrification;
works without denitrification;

These factors seem to be less common in large works than in small or medium sized works.

Characteristics of sewage treatment processes that are reported to favour COD removal are:

denitrification;
low sludge age;
low sludge loading rate;
elevated temperature.

A two stage process comprising a high rate process (high sludge loading rate) followed by extended aeration (low sludge loading rate, high sludge age) has been reported to be particularly effective at COD removal and should be capable of producing an ef fluent with mean COD concentration around 40 mg/l.

1.4.4 Nature and consequences of COD in effluents

COD is a broad measure of oxidizable material and may indicate the presence of a wide range of substances. There is little information on whether hard COD, either non-treatable (inert) material or non-treated material, is associated with particular substances. There are some pointers in the literature to inert COD being predominantly due to synthetic substances whilst "natural" sources of COD such as humic and fulvic materials are more readily biodegraded However, there is also evidence that inert COD can be created from the breakdown of more complex substances during treatment within STWs.

Without a better knowledge of the constituents of hard COD, the environmental significance is difficult to assess.

1.5 Recommendations

More detailed analysis is needed at individual STWs to understand the factors that affect COD removal and its variation. A simple model or indicator (such as COD:BOD ratio or denitrification) is needed to assess the number of works that may be at risk from COD failure on a national scale.

It is likely that the parameters and possibly the form of any predictive model (e.g. regression equation) will vary between different types of treatment plant. Daily composite samples are needed to allow the correlation of effluent quality with lagged influent quality to be investigated. Since the influent COD:BOD ratio is one of the suggested indicators of performance, it would be informative if the works selected for detailed analysis include variable influent COD:BOD ratios.

The six STWs that would fail the UWWTD standards should be studied to assess what scope there is for making process changes to ensure their compliance in terms of COD.

Laboratory scale or pilot plant work under controlled conditions would be of assistance in defining more closely how the characteristics of the most common processes (e.g. activated sludge, percolating filters) affect COD removal.

More detailed analysis of the constituents of COD in sewage effluents is needed to assess whether COD levels in effluents are of environmental significance.

Using the data derived in 2, and 3, in conjunction with an indicator or model of the type recommended in 1 and 2, an estimate could be made of the likely cost of complying with the COD provisions of the UWWTD.

Copies of the report are available from FWR, price £10.00, less 20% to FWR Members.