Case study:River Irwell Restoration Project

Jump to: navigation, search
(2 votes)

To discuss or comment on this case study, please use the discussion page.

Location: 53° 29' 22", -2° 11' 31"
Edit location
Loading map...
Left click to look around in the map, and use the wheel of your mouse to zoom in and out.

Project overview

Edit project overview
Status In progress
Project web site
Themes Flood risk management, Habitat and biodiversity, Hydromorphology, Social benefits, Urban
Country England
Main contact forename Oliver
Main contact surname Southgate
Main contact user ID
Contact organisation Environment Agency
Contact organisation web site
Partner organisations Groundwork, Natural England
This is a parent project
encompassing the following
Castle Irwell Urban Wetland, Clayton Vale, Goshen Weir removal project, River Roch, Bury, Philips Park
The Red River

Project summary

Edit project overview to modify the project summary.

The River Irwell Restoration Project plans to restore urban watercourses in an effort to achieve good ecological status for the watercourse. Such plans must be viewed against the considerable economic and physical constraints imposed on such rivers due to their setting, in particular the need to maintain or even enhance flood protection levels and to ensure infrastructure remains uncompromised. A restoration plan was developed for the heavily modified River Medlock at Clayton Vale and Philips Park, Manchester. The principal aims of the study were driven by opportunities to improve the hydromorphological and ecological status of the watercourse through naturalisation, working to develop a watercourse where the reinstated channel units function to temporarily store coarse sediment creating dynamic habitat within a restricted urban environment.

Known as the ‘Red River’ due to the brick lining along the study reach constructed in 1912, the present U-shaped 'flume' has created conditions with limited in-channel morphology (occasional berm top fine sediments) and flow diversity (a monotonous run due to the almost uniform width and depth along the reach). No significant sediment storage occurs, despite a strong coarse sediment supply, as a consequence of the unchanging steep gradient and immovable planform (the brick-lined banks have restricted any lateral movement of the watercourse locally) creating a uniform, high energy, transporting reach and preventing any sediment deposition on the channel bed. Under present flow conditions, the high velocities are considered to be a barrier to fish passage, due to velocities of >2m/s under low flow conditions and 3-4m/s during higher flow events within the channel.

The River Medlock at Clayton Vale and Philips Park is designated as a Heavily Modified Waterbody. At present the Water Framework Directive (WFD) defines the overall river status as Poor Ecological Potential, but with a target of reaching Good Ecological Potential by 2027. Restoration needed to be mindful of impacts locally, and upstream and downstream of Clayton Vale and Philips Park, including impacts on flood risk given the highly urbanised nature of the catchment. Removing the brick-lining and concrete base layer, without managing the steep gradient and high energy levels of the River Medlock could create uncontrolled destabilisation. Therefore, removal had be considered alongside morphological restoration and naturalisation through Clayton Vale and Philips Park to ensure a 'dynamically stable' restoration was implemented and that historic features lining the watercourse were not compromised.

Consultation throughout the restoration plan development highlighted the ‘stand-off’ between the objective of river naturalisation and the ‘need’ for stability. The River Medlock at Clayton Vale and Philips Park would be a dynamic, active single thread river if not constrained by the brick – lining flume. Upstream analogue features (including, rapids, riffles and pools) and hydraulic modelling were used to carefully design and size functional, dynamically ‘stable’ morphological features to manage the high energy system following removal of the concrete and brick lining. Engineering concerns remained over the potential for ongoing lateral erosion that could threaten local historic walls and public footpaths. Compromise was therefore necessary to satisfy the project board before works could be undertaken, however, the majority of the restoration objectives for naturalisation were approved and the results of the first stage of the project and initial river response is reported here.

Monitoring surveys and results

Edit project overview to modify the Monitoring survey and results.

The project will be monitored.

Lessons learnt

Edit project overview to modify the lessons learnt.

Utilise the material within the channel where possible to reduce the costs and impacts of importing foreign material.

Image gallery

Goshan after weir removal
Prestolee weir removal

Catchment and subcatchment


River basin district North West
River basin Irwell


River name River Medlock (Lumb Brook to Irwell)
Area category 10 - 100 km²
Area (km2)
Maximum altitude category 100 - 200 m
Maximum altitude (m) 121
121 m
0.121 km
12,100 cm
Dominant geology Calcareous
Ecoregion Great Britain
Dominant land cover Urban
Waterbody ID GB112069061152

Other case studies in this subcatchment: Clayton Vale, Philips Park


Name River Irwell
WFD water body codes GB112069061152
WFD (national) typology
WFD water body name River Medlock (Lumb Brook to Irwell)
Pre-project morphology
Reference morphology
Desired post project morphology
Heavily modified water body Yes
National/international site designation
Local/regional site designations
Protected species present No
Invasive species present No
Species of interest
Dominant hydrology
Dominant substrate Artificial
River corridor land use Urban
Average bankfull channel width category
Average bankfull channel width (m)
Average bankfull channel depth category
Average bankfull channel depth (m)
Mean discharge category
Mean annual discharge (m3/s)
Average channel gradient category
Average channel gradient
Average unit stream power (W/m2)

Project background

Reach length directly affected (m)
Project started
Works started
Works completed
Project completed
Total cost category
Total cost (k€)
Benefit to cost ratio
Funding sources

Cost for project phases

Phase cost category cost exact (k€) Lead organisation Contact forename Contact surname
Investigation and design
Stakeholder engagement and communication
Works and works supervision
Post-project management and maintenance

Reasons for river restoration

Mitigation of a pressure Barriers to fish migration, Flood protection maintenance works, Invasive species, Impoundments (not hydropower), Urbanisation
Hydromorphology Quantity & dynamics of flow, Width & depth variation
Biology Fish
Other reasons for the project


Structural measures
Bank/bed modifications Weir removal/modification for easement of fish passage, Bank reprofiling
Floodplain / River corridor
Planform / Channel pattern
Non-structural measures
Management interventions
Social measures (incl. engagement)


Hydromorphological quality elements

Element When monitored Type of monitoring Control site used Result
Before measures After measures Qualitative Quantitative

Biological quality elements

Element When monitored Type of monitoring Control site used Result
Before measures After measures Qualitative Quantitative

Physico-chemical quality elements

Element When monitored Type of monitoring Control site used Result
Before measures After measures Qualitative Quantitative

Any other monitoring, e.g. social, economic

Element When monitored Type of monitoring Control site used Result
Before measures After measures Qualitative Quantitative

Monitoring documents

Additional documents and videos

Additional links and references

Link Description -78w9AWtAQ Restoration Revolution - The River Irwell on You Tube BBC news clip maybe time limited The little Lever Weir Removal in the Irwell Pilot Catchment.pdf&ei=L85fUpfzHrOO7Aat3IGwAw&usg=AFQjCNFBkYd10J31gWM8uH-BSmL-P9dQuA&sig2=Nf2w8E xckngDf9nF5gIg Irwell is a Pilot Catchment RiverRestoration July2013%2FNash Monitoring River Irwell Restoration.pdf&ei=EOZfUqX2DpKI7AbRoYCACg&usg=AFQjCNE dQszysF5cJqnyok-x-l51lSvdA&sig2=lcqYz49Gxio2vXFuRpa0aQ&bvm=bv.54176721,d.ZGU monitoring plans for the River Irwell

Supplementary Information

Edit Supplementary Information

The River Irwell rises at Deerplay Moor, Cliviger (Lancashire) and runs south through Bacup, Rawtenstall, Ramsbottom, Bury and Kearsley before joining the Manchester Ship Canal in Salford, south of Irlam. Major tributaries include the Roch, the Croal, the Irk and the Medlock. This phase of the RRC‟s work (Phase 1) concentrated on the Irwell between Rawtenstall and Salford, and the Kirklees Brook between Hawkshaw and its confluence with the Irwell. The Croal and the Roch will be the subject of a similar study in the future (Phases 2 and 3).

The upper region is largely rural and fairly steep whereas the lower region is mostly urban and less steep. More than 30% of the catchment is urban including Manchester, Salford and Oldham in the south and Bury, Bolton and Rochdale in the centre (Environment Agency, 2008). The river has a long history of modification dating back to the industrial revolution, which includes walling, deepening, re-alignment, culverting, widening, dredging, straightening, and the construction of weirs. Many of these structures are now in poor condition and are crumbling or collapsing into the river channel, contributing to the supply of coarse, but unnatural, bedload material. Artificial structures such as weirs and culverts can also obstruct flow and result in accumulations of sediment within the channel, which in the past has been removed by channel dredging at some locations. Many of these structures were constructed as part of historical industrialisation of the area, and where they have collapsed there is evidence that the river has begun to naturally adjust, often with riffle features occurring downstream. In most cases there was little or no evidence (from the walkover survey) of any negative impacts in terms of additional fine sediment accumulation downstream.