Wessex Water Leakage Datasets Scenario 2

Our distribution network comprises around 12,000 km of mains, supplying around 600,000 household properties, and around 45,000 commercial customers. Pipelines range from large trunk mains to smaller distribution mains, to service pipes. The company owned service pipe typically connects to the customer owned supply pipe at the juncture of the controlling tap or meter. This can vary by property type, such as for blocks of flats, or older terrace houses. The failure of pipes or fittings to point of leaking can be influenced by many factors including the pipe material and age, network operation, external environmental factors (such as hot summers and cold winters), and interaction between assets and the soil around them. We estimate leakage within zones of our network, but it is difficult to understand which pipes within each zone should be prioritised for replacement. Often, our work is focussed on responding to active repairs, whereas proactively identifying and replacing pipes will reduce the number of leakage incidents. Here you will find a number of datasets from one area of our network, known as a District Metered Area (DMA). The coordinates of all spatial files are modified to prevent the location of our infrastructure being identified. Each dataset is described below. This is a one-off data publication. Supply network and asset data: Description and collection: The Asset_Links and Asset_Nodes shapefiles provide a spatial representation of our network of pipes and assets with a DMA. Attached to each line object in the Asset_Links shapefile is an understanding of the Pipe Size, Pipe Size Units, Age of Pipe (Pipe Year), Pipe Material, Soil Type and Length (m). The Asset_Nodes shapefile is a point layer showing the location of assets within the network and includes the asset type and the identifying codes of any flow or pressure devices that can be found in corresponding datasets. The symbol angle field is the orientation of the symbols as shown in spatial software. For supply data, asset types such as non-return valves (NRVs) and Pressure Reducing Valves (PRVs) would be angled to show the direction of flow and how (for NRVs) flow cannot go backwards at this point. Data accuracy: The network data is an extract from our corporate GIS database. This is constructed from data captured from various sources since the 1980s. This means some of the lines, representing water mains, may not have been inspected since the day of capture, and therefore the accuracy cannot be confirmed. The original maps had limitations such as the thickness of the ink line versus the scale of drawing (1:2500) which would have been the same width as a road. Full network connectivity was only realised in the early 2000s, following the piecing together of multiple databases and tiles worth of data. This could have further implications of the accuracy of the pipe routes. A project of Positional Accuracy Improvement, also took place in 2009, which introduced other challenges to line up and update other data sources, typically at the edge of tiles. Since then, (and since paper records have begun), the network has had continuous addition and amendments to it, because of new development projects, operational incident observations, modelling queries and network analysis quality assurance. The accuracy of the network is reliant on any updates and changes being sent through from on-site teams. Dataset update: This data is a one-off extraction. Modelled network data: modelled pressure, model nodes and model links Description and collection: Modelled pressure (Model Pressure Data 24hr) has been extracted from the company hydraulic models for each sample District Metered Area (DMA). The data is presented in an Excel spreadsheet containing a set of 96 x 15-minute pressure values measured in metres (m), creating a time varying pressure profile for every nodal point in the DMA. The pressure profiles represent those predicted by the models to occur during an average demand scenario - pressure that results from customer demand set at typical levels. The field header for each pressure profile is the model node reference (format: P6677001 or H5689002) which can also be found populated in the node shape file (Model_Nodes). This enables each profile to be located spatially and an awareness of where it exists within the network (Model_links). Data accuracy: These data values are predicted by the hydraulic models and are not exact values that occurred on any particular day. They are representative of levels of service and considered to be a good match for the system behaviour under an average day scenario. The model network may be structured slightly differently to our supply network to accommodate node points which are necessary for the model to run. Small sections of missing lines do appear in the model network which coincide with assets and reflect edits to ensure these assets are represented in the model. Dataset update: The data is a one-off extraction, but it is a representation of pressure under a typical or average demand scenario. This representative data would only change if the physical network is subjected to change, like a system rezone that introduces a different pressure regime into the DMA, new assets or abandoned assets that affect system hydraulics within the DMA, or control changes like adjustments to pump or pressure control valve settings that introduce a different pressure regime into the DMA. Measured flow and pressure data Description: Available data from flow and/or pressure sensors has been extracted from our database within the scenario DMA. These data are available at a 15-minute timestep from 01/04/2025 00:00 and alongside the measured values the spreadsheet shows the type of logger, where it’s located in reference to the DMA and the units of measurement. This dataset will show how flow and pressure fluctuate at different locations across the network which may indicate changes in demand and/or leakage. IN – these are inflow loggers to the DMA. OUT – these are outflow loggers from the DMA. CUS – these are logged customers with high or known night time usage to ensure we understand what usage is genuine versus leakage so we can make accurate calculations. Data accuracy: The values presented may be raw verified data or they may be corrected due faults during data collection. Dataset update: This data is a one-off extraction. Planned and emergency job record Description: Where the network has undergone maintenance or emergency repairs we keep a record. This includes the Job Type, the date the job was resolves, and the size and material of the pipe. Jobs selected here have occurred since 01/04/2025 and are located within 5 metres of a pipe asset.