1 In examining these issues, we drew on a variety of evidence sources.
2 We interviewed relevant individuals from organisations involved in the project, including: the Department, Ofwat, The Environment Agency, The European Commission, Thames Water, and Bazalgette Tunnel Limited.
3 We reviewed a range of documents to support our findings, including:
• legal advice received by the Department in relation to the Urban Waste Water Treatment Directive and some records of communications with the European Commission, to understand the UK's legal obligations;
• historical option appraisals and cost-benefit analyses relating to the Tunnel and alternative options, in order to understand why the Tunnel had been chosen as the preferred option;
• documentation on the Tideway modelling suite, including calibration and other quality assurance reports, and related emails, to understand the level of assurance which had been gained over the modelling; and
• project management reports detailing the state of progress and risk mitigation.
4 We performed data analysis on 2015 and 2016 sewage discharges data from five of the largest combined sewer overflows collected by Thames Water in order to estimate the number of spills which had lately occurred along the Thames Tideway from these overflows. Thames Water and the Environment Agency define a 'spill' differently:
• The definition used by Thames Water counts one spill for any 24-hour period in which there is some discharge. A new spill is counted if discharge occurs after a 24-hour period of no discharge.
• The definition used by the Environment Agency is that one or more discharge overflow events within a period of 12 hours or less will be considered to be one spill, one or more discharge overflow events extending over a period of greater than 12 hours up to 36 hours will be considered to be two spills. Each subsequent 24 hour duration counts as one additional spill and the whole of the 24 hour block is included.
For the purposes of estimating spills, we used the Thames Water methodology for consistency with the way estimated spills have been reported in Thames Water's updates along the project's lifetime. The choice in methodology has a material impact on the number of spills counted. For instance, the Tideway modelling forecasts that discharge counting at Hammersmith Pumping Station CSO with sewage treatment works improvements and the Lee Tunnel would result in 55 discharges with the Thames Water method and 93 discharges with the Environment Agency method.