Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Eastern CFRAM

Study

HA09 Hydraulics Report Lower Liffey Model DOCUMENT CONTROL SHEET

Client OPW

Project Title Eastern CFRAM Study

Document Title IBE0600Rp0027_HA09 Hydraulics Report

Model Name Lower Liffey

Rev. Status Author Checked By Approved By Office of Origin Issue Date

D01 Draft M.Houston A. Sloan I. Bentley Belfast 12/02/2014

D02 Draft M.Houston A. Sloan S. Patterson Belfast 11/07/2014

D03 Draft A. Sloan S. Patterson G. Glasgow Belfast 13/01/2015

F01 Draft Final A. Sloan S. Patterson G. Glasgow Belfast 13/03/2015

F02 Draft Final A. Sloan S. Patterson G. Glasgow Belfast 13/08/2015

IBE0600Rp00027 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Table of Reference Reports

Report Issue Date Report Reference Relevant Section Eastern CFRAM Study Flood Risk December IBE0600Rp0001_Flood Risk 3.2.3 Review 2011 Review_F02 Eastern CFRAM Study Inception August 2012 IBE0600Rp0008_HA09 Inception Various Report UoM09 Report_F02 Eastern CFRAM Study Hydrology September IBE0600Rp0016_HA09_Hydrology 4.4 Report UoM09 2013 Report_F01 Eastern CFRAM Study HA09 Liffey November 2001s4884- SC2 Survey Report v1 Various Survey Contract Report 2012

IBE0600Rp00027 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

4 HYDRAULIC MODEL DETAILS

4.7 LOWER LIFFEY MODEL

4.7.1 General Hydraulic Model Information

(1) Introduction:

On the basis of a review of historic flooding and the extent of flood risk determined during the PFRA, the Eastern CFRAM Flood Risk Review (IBE0600Rp0027_Flood Risk Review) highlighted the River Liffey in Dublin City AFA as a HPW, subject to both fluvial and coastal flood risk.

The Lower Liffey model stretches from Islandbridge to Dublin Bay and represents the portion of the Lower Liffey in Dublin City which is significantly affected by tidal influence. The total contributing catchment of the Lower Liffey model from Pollaphuca dam to Dublin Bay is 1,020 km².

The model and AFA immediately upstream of the Lower Liffey HPW is referred to as the Lucan to Chapelizod AFA. The downstream boundary of the Lucan to Chapelizod model was extracted from the Lower Liffey.

A number of tributary catchments enter the Liffey within the model extents including the Camac, Poddle and Dodder Rivers. The Lower Liffey model receives its main freshwater inputs from the upstream Lucan model and these three tributaries. These tributaries have been studied as discrete HPWs under the Eastern CFRAM (in the case of the Camac and Poddle watercourses) and under the Dodder CFRAM Study.

The Dodder is the largest of these rivers and represents a contributing catchment of 113 km², flowing into the Liffey at Ringsend. The Dodder CFRAM predated the Eastern CFRAM, therefore an appropriate downstream control for the Dodder hydraulic model was established based on the ICPSS study which has more recently been used to establish the downstream conditions for the Lower Liffey model.

The Camac is the second largest tributary, representing a contributing catchment of 58 km² and flows into the Liffey via a large culvert at Heuston Station. The Poddle is the smallest of the three, representing a catchment of 12 km² and flows into the Liffey via a culvert at Wellington Quay. The Lower Liffey model was used to provide downstream controls for these two tributaries.

In addition to the main river inputs, run-off from Dublin City through the storm drainage network, combined storm overflows, other minor watercourses and direct from surface run-off, is estimated to account for up to 21 km² of urban catchment in total. The flow from this additional area is added to the model as a lateral inflow along the length of the Liffey main channel.

There are two water level gauging stations within this modelled reach of the Liffey, however both are tidally influenced and are not suitable for flow measurements. There is one river gauging station (09022)

IBE0600Rp00027 4.7.1 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

upstream of the model extents at the Leixlip hydro-electric power generation station, owned and operated by ESB. This station was not given a classification under FSU and is not a standard water level recorder station as operated by the OPW and the EPA. The flow data received have been derived from continuous recordings of flows and water levels through the various structures of the plant and converted into combined flow rates on request to ESB. The continuous flow records can be considered to have a high degree of confidence since they are derived from a combination of measured flows and fully defined water level - flow relationships (through the turbines, sluices and spillways) rather than based on an extrapolated relationship between water level and flow as is common at conventional gauging stations. However continuous recordings are only available for seven years from 2005 to 2011 inclusive and as such the

Qmed derived from the AMAX series of 61.64 m³/s can be considered to have a low degree of statistical confidence.

To ensure consistency with modelling throughout the greater Dublin area, this area has been modelled as 1D-2D using the Infoworks ICM flexible mesh software. A number of catchments in the Greater Dublin area have legacy drainage network models constructed in InfoWorks CS. Survey data gathered as part of the CFRAM study has been augmented with culvert and manhole information from these GDSDS models to allow a more accurate line and gradient of pipe networks to be represented in the model. Comprehensive data collection of existing sewer network records and survey of culverted reaches was undertaken for the GDSDS in order to capture detail in complex drainage networks i.e. changes to internal diameter and gradient. Therefore OPW and RPS selected ICM use to allow better representation of culverted river networks and enable better utilisation of, and future integration with, the existing sewer network models. ICM also provides a very stable 2D modelling regime for coastal inundation modelling, therefore ICM models (driven by a MIKE21 coastal model) were used for Dublin Bay coastal AFAs, to provide a consistent approach throughout the Greater Dublin area, facilitating integration with existing models.

In addition to the watercourses, the model also incorporates the entire Dublin Bay coastal frontage from Strand Road in Sutton to Merrion Strand at the southern end of Sandymount. However the focus of this report section is the River Liffey and Dublin Port area at the mouth of the River Liffey. The River Liffey has been modelled using a 1D-2D approach extending downstream to the entrance of Alexandra Basin in Dublin Harbour. From this location a 2D coastal inundation approach has been taken to model the flood risk to Dublin Harbour.

(2) Model Reference: HA09_LIFF2C

(3) AFAs included in the model: Dublin City

(4) Primary Watercourses / Water Bodies (including local names):

ID NAME

09LIFF LIFFEY

09BELL BELLEVUE MILL RACE

IBE0600Rp00027 4.7.2 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

09MILA MIL A

(5) Software Type (and version):

(a) 1D Domain: (b) 2D Domain: (c) Other model elements:

Infoworks ICM v5 Infoworks ICM Flexible Mesh N/A

4.7.2 Hydraulic Model Schematisation

(1) Map of Model Extents:

MILA Bellevue Mill Race

Figure 4.7.1: Map of model extents within Dublin City AFA

Figure 4.7.1 illustrates the extent of the modelled catchment, river centre line, HEP locations and AFA extents as applicable. The lower Liffey model contains 1 Upstream Limit HEP, 1 Downstream Limit HEP and 1 Intermediate HEP and 3 Trib HEPs.

(2) x-y Coordinates of River (Upstream extent):

River Name x y LIFFEY 311995 234161 BELLEVUE MILL RACE 312390 234210 MILA 312515 234227

(3) Total Modelled Watercourse Length: 7.3 km

IBE0600Rp00027 4.7.3 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

(4) 1D Domain only Watercourse Length: NA (5) 1D-2D Domain 7.3 km Watercourse Length:

(6) 2D Domain Mesh Type / Resolution / Area: Flexible / 1m2 - 25m2 / 17km²

(7) 2D Domain Model Extent:

Figure 4.7.2: 2D Domain Model Extent

Figure 4.7.2 represents the modelled extents and the general topography of the catchment within the 2D model domain. The ground elevation (based on LiDAR data used to generate a 2D flexible mesh) is shown to provide an overview of the modelled area topography. There was no further post processing of the data contained within the mesh required. Changes in the vertical scale of this map are outlined by the index; all levels have been set to OD Malin.

Figure 4.7.3 provides an overview of the model schematisation. Figure 4.7.4 and Figure 4.7.5 show detailed views. The overview diagram covers the model extents, showing the surveyed cross-section locations, AFA boundary and river centre line. It also shows the area covered by the 2D model domain. These diagrams include the surveyed cross-section locations, AFA boundary and river centre. They also show the location of the critical structures if applicable, along with the location and extent of the links between the 1D and 2D models.

IBE0600Rp00027 4.7.4 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

The upstream extent of the model of the Lower Liffey is Islandbridge Weir and extends to the entrance of Alexandra Basin at Dublin Port.

The 2D extent of the Lower Liffey model covers the majority of the coastline of Dublin Bay, however the focus of this section of the hydraulics report is the Lower Liffey and Dublin Port.

Alexandra Basin Gauging Station

O'More Bridge Gauging Station

Figure 4.7.3: Overview of Model Schematisation

IBE0600Rp00027 4.7.5 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Alexandra Basin Gauging Station

Figure 4.7.4: : Model Schematisation of Dublin Port

Figure 4.7.5: Model Schematisation from Merchants Quay to Alexandra Basin

IBE0600Rp00027 4.7.6 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

O'More Bridge Gauging Station

Figure 4.7.6: Model Schematisation from Heuston Station to Halfpenney Bridge

Islandbridge Weir

Figure 4.7.7: Model Schematisation from Islandbridge to Heuston Station

IBE0600Rp00027 4.7.7 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

(8) Survey Information

(a) Survey Folder Structure:

First Level Folder Second Level Folder Third Level Folder

Murphy_E09_M02C_WP6_120803_09LIFF_A GIS and Floodplain Structure Register Where: Photos

Murphy – Surveyor Name Surveyed Cross Section Lines E09 – Eastern CFRAM Study Area, Ascii Hydrometric Area 09

M02C – Model Number 2C

WP6 – Work Package 6

th Photos 120803 – Date Issued (12 Aug 2012)

09LIFF_A – River Reference Drawings and PDFs

(b) Survey Folder References: LIFFEY Murphy_E09_M02C_WP6_120803_09LIFF_A

BELLEVUE MILL RACE Murphy_E09_M02C_WP6_120803_09BELL

MILA Murphy_E09_M02C_WP6_120803_09MILA

2m resolution LiDAR of the entire modelled area was used to generate the 2D domain computational flexible mesh. The vertical accuracy of the LiDAR is quoted as 0.2m RMSE which was considered satisfactory for model application. A comparison of topographical survey levels along the quays and the LiDAR data for the same area was undertaken. The comparison indicated a good agreement between the datasets with level differences of between 0mm and 130mm found at the locations investigated.

The LiDAR data was augmented during the meshing process with the OSI building layer dataset to integrate the building footprints into the model. The buildings are represented in the computational mesh as porous polygons with a porosity value set to zero, allowing no flow to pass through them. No other amendments were made to the computational mesh.

IBE0600Rp00027 4.7.8 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

(9) Survey Issues: In June 2013 a survey query was raised in relation to incomplete survey information as follows. Not all parapets / defences are surveyed. Do the flood defence walls not continue west beyond 313861 234344. The survey was updated to provide the missing information and the data included in the model.

Figure 4.7.8 Liffey Survey Query

IBE0600Rp00027 4.7.9 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

In December 2014 a further survey was undertaken to provide data in the dock lands area where vegetation and storage had obscured LiDAR survey information.

Figure 4.7.9 Dublin Port Infill Survey

4.7.3 Hydraulic Model Construction

(1) 1D Structures in the 1D domain: See Appendix A.2

Number of Bridges and Culverts: 22

Number of Weirs: 3

The survey information recorded includes a photograph of each structure, which has been used to determine the Manning's n value. Further details are included in Chapter 3.5.1. A discussion on the way structures have been modelled is included in Chapter 3.3.4.

The 1D domain of the model was constructed from a combination of information gathered under the CFRAM Study Channel and Structure Survey and subsequent infill surveys. The topographical cross section information was used to define the river channel and associated structures. This information was also used to define the bank elevations. The level associated with the end point of each modelled cross section was taken as the bank level at that location with intermediate elevations being interpolated from one section to the next. The bank level information was augmented with LiDAR information if required.

IBE0600Rp00027 4.7.10 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Culvert inlet parameters were applied as per guidance contained in CIRIA report No. R168 "Culvert Design Manual" with the values for the closest approximation to the actual shape and form applied, based on survey geometry and photography.

All bridge units were constructed by inserting a discrete bridge opening unit for each actual bridge opening, no bridges have had piers defined.

Critical Structures:

Figure 4.7.10: 09LIFF0067: Islandbridge Weir is located at the upstream extent of the Lower Liffey model reach

IBE0600Rp00027 4.7.11 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.11: 09LIFF0067: Islandbridge Weir (upstream extent of the Lower Liffey model reach) southward view

The Islandbridge weir depicted in Figures 4.7.10 and 4.7.11 restricts the extent of tidal influence during higher probability coastal events reducing the potential flood risk due to extreme coastal water levels. However the influence of low probability tidal events can be seen to propagate upstream past the weir.

(2) 1D Structures in the 2D domain: None

(3) 2D Model Domain:

The following data sources were used to develop the 2D domain:

2m resolution LiDAR of the entire AFA was used to generate the computational flexible mesh. 2m LiDAR is considered sufficiently accurate for this type of modelling. This data was augmented during the meshing process with the OSI building layer dataset, to integrate the building footprints into the model. The buildings are represented in the computational mesh as porous polygons with a porosity of zero, allowing no flow to pass through them.

Additional survey data was requested for the Dublin Port area to supplement the available LiDAR data; Figure 4.7.12 shows the locations for the requested additional information. Due to access restrictions in the harbour area the additional data for the areas marked as requiring spot levels could not be acquired.

Upon receipt of the survey data and additional inspection via Google Earth some areas of raised ground were removed from the LiDAR data along the northern boundary of the harbour area between Location A and Location B on the following diagram. These areas consisted of raised flower beds and were deemed

IBE0600Rp00027 4.7.12 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

to be neither defences nor hydraulically significant structures.

A

B

Figure 4.7.12: Dublin Port Infill Survey

(4) Defences:

Type Watercourse Bank Model Start Chainage Model End (approx.) Chainage (approx.)

Wall, Formal Liffey Right 6210 5080

Wall, Formal Liffey Right 4520 2540

Wall, Formal Liffey Left 5020 4550

Wall, Formal Liffey Left 4550 2540

(5) Model Boundaries - Inflows:

Full details of the flow estimates are provided in the Hydrology Report (IBE0600Rp0016_HA09 Hydrology Report_F02 - Section 4.4 and Appendix D). The boundary conditions implemented in the model are shown in Table 4.7.1:

IBE0600Rp00027 4.7.13 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Table 4.7.1: Model Boundary Conditions

Node ID Boundary Description Branch Name HEP Type 09LIFF00695_BREAK Point Inflow US Boundary Lower Liffey 09_1870_14_RPS 09LIFF00508_BR_BREAK_US Point Inflow Tributary Camac 09_1872_9_RPS 09LIFF00326_BR_BREAK_US Point Inflow Tributary Poddle 09_1874_17_RPS 09LIFF00071_BR_BREAK_US Point Inflow Tributary Dodder 09_587_11

Figure 4.7.13: Inflow Hydrographs for River Liffey for the 1% AEP Event

Figure 4.7.13 illustrates the inflow hydrographs for the River Liffey 1% AEP event. Total lateral inflows have been calculated for river reaches between each HEP node, these flows were disaggregated and distributed pro-rata based on the model link length (river reach or conduit) along the watercourses.

(6) Model Boundaries – Coastal Water Level Boundary Downstream Conditions:

Outputs from the Irish Coastal Protection Strategy Study (ICPSS) have resulted in extreme tidal and storm surge water levels being made available around the Irish Coast for a range of Annual Exceedance Probabilities (AEPs). The locations of the ICPSS nodes for Dublin Bay along with the relevant AFA locations are shown in the figure below. The associated AEP water levels for each of the nodes is contained in Table 4.7.2.

IBE0600Rp00027 4.7.14 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

The values for Node 19 (see Figure 4.7.) were used to generate the downstream coastal boundary for both the River Liffey 1D river channel model and the 2D Dublin Harbour coastal inundation model. The values derived for Node 19 were selected to provide acceptably conservative coastal water levels throughout Dublin Bay. For additional detail on the determination of the water levels generated during the ICPSS project reference should be made to the ICPSS technical report (Irish Coastal Protection Strategy Study Phase 3 - North East Coastal, Work Packages 2, 3 & 4A - Technical Report).

Wave overtopping risk was assessed and not found to have significant influence and therefore was not included in this model.

Figure 4.7.14: ICPSS Node Locations (IBE0600Rp0016_HA09 Hydrology Report_F02)

IBE0600Rp00027 4.7.15 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Table 4.7.2: ICPSS Level in Close Proximity to HA09 AFAs / HPWs (IBE0600Rp0016_HA09 Hydrology Report_F02)

Elevation (m) to OD Malin for a Range of AEP AEP (%) NE_17 NE_19 NE_20 NE_21 NE_22 NE_23

50 2.52 2.44 2.42 2.46 2.46 2.43

20 2.65 2.57 2.54 2.58 2.58 2.55

10 2.75 2.67 2.63 2.67 2.67 2.64

5 2.85 2.77 2.73 2.76 2.76 2.74

2 2.98 2.90 2.85 2.88 2.88 2.86

1 3.08 3.00 2.94 2.97 2.97 2.95

0.5 3.18 3.11 3.04 3.06 3.07 3.04

0.1 3.41 3.34 3.26 3.27 3.28 3.25

Representative tidal profiles for Santry were extracted from a tidal model of the Irish Sea and Dublin Bay for a 70 hour period. The tidal model of Dublin Bay is part of a wider Irish Sea tidal model. As such the model is driven by global tidal predictions which are propagated into Dublin Bay; the model has been calibrated to all relevant tide gauge data, including the Dublin Port gauge at Alexandra Basin.

A normalised 48 hour surge profile was scaled based on the difference between the peak water level extracted from the tidal profile and the target extreme water level from the table above. The scaled surge profile was then appended to the tidal profile, with coincident peaks, to achieve a representative combined tidal and storm surge profile for the required AEP events, this calculation is detailed below.

ICPSS Extreme Water Level - Peak Astronomical Tide Level = Peak Surge Value

Peak Surge Value x Normalised 48Hr Representative Surge Profile = Scaled Surge Profile

Astronomical Tide Profile + Scaled Surge Profile = Extreme Coastal Water Level Profile

It should be noted that the peak water level values quoted in Table 4.10.2 included an allowance for the affect of seiching in Dublin Bay. Figure 4.10.15 illustrates the tidal profile, storm surge profile and resultant combined water level profile for the coastal boundary.

The appropriate water level profile was applied to the downstream node of the 1D Lower Liffey model (09LIFF00007J) as a level boundary and the 2D level boundary of the Dublin Port area. The application of the oscillating water level to these boundaries allows the model to simulate the ingress and egress of the coastal water levels.

The HA09 hydrology report concluded that correlation between total water levels and fluvial flood flow on

IBE0600Rp00027 4.7.16 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

the Liffey River can be considered to be negligible and it is proposed to follow a simplified conservative approach whereby the 50% AEP design event is maintained for one mechanism while the whole range of probabilities for the other mechanism are tested and vice versa, subject to sensitivity testing against average winter conditions to ensure the approach does not yield results which could lead to unrealistic flood extents or over design of measures.

Tide and Surge Profiles 4

3

2 Astronomical Tide (mAOD)

1 Normalised 48hr Representative Surge Profile 0.5 % Scaled Surge Elevation 0

0 20406080100

Water 0.5% AEP Water Level Profile ‐1

‐2

‐3 Time (hrs)

Figure 4.7.15: Extreme Water Level Profile for Lower Liffey Downstream Boundary

(7) Model Roughness:

(a) In-Bank (1D Domain) Minimum 'n' value: 0.040 Maximum 'n' value: 0.040

(b) MPW Out-of-Bank (1D) Minimum 'n' value: N/A Maximum 'n' value: N/A

(c) MPW/HPW Out-of-Bank Minimum 'n' value: 0.011 Maximum 'n' value: 0.045

(2D)

IBE0600Rp00027 4.7.17 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.16: Map of 2D Roughness (Manning's n)

Figure 4.7.16 illustrates the roughness values applied within the 2D domain of the model. Roughness in the 2D domain was applied based on land type areas defined in the CORINE Land Cover Map with representative roughness values associated with each of the land cover classes in the dataset.

IBE0600Rp00027 4.7.18 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

(d) Examples of In-Bank Roughness Coefficients

Figure 4.7.17: Photograph taken from Frank Sherwin Bridge looking towards Rory O'More Bridge

The Lower Liffey has a relatively straight clean channel; however it has a reasonably shallow gradient from Islandbridge Weir to Custom House Quay, at which point the bed begins to drop significantly towards the harbour area. A conservative Manning's n of 0.040 was initially applied to the entire reach of the Lower Liffey model in light of the shallow gradient (Figure 4.7.18) and low velocity flow. Verification runs based on a number of events as described in Section 4.7.5 (1) indicated that this Manning's n value was appropriate, with a good agreement between modelled and observed levels being achieved.

IBE0600Rp00027 4.7.19 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Islandbridge Weir Custom House Quay

Figure 4.7.18 - River Liffey Bed Profile

4.7.4 Sensitivity Analysis

To be completed.

4.7.5 Hydraulic Model Calibration and Verification

(1) Key Historical Floods (from IBE0600Rp0008_HA09 Inception Report_F02, unless otherwise specified):

(a) Nov 2009 Flooding occurred in parts of Kildare and Dublin in November 2009 following heavy rainfall. A press article following the event describes how the River Liffey burst its banks at several locations, including at Strawberry Beds. However, no additional information on flood flows, extents or levels were provided. Combined flow data from the Rye Water and Liffey at Leixlip was used to provide an estimation of the probability of the event at the confluence of the Rye Water and Liffey. The hydrograph was then scaled to achieve a peak flow at the upstream extent of the model for the same probability event. The November 2009 event was estimated to be a 7.5% AEP event. This flow data was used in conjunction with the corresponding tide gauge data at Dublin Port to simulate the event. Data from a Marine Institute level gauge located at O'More Bridge was used to assess the model performance.

IBE0600Rp00027 4.7.20 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

The peak recorded level at O'More Bridge during the November 2009 event was 1.87m OD, the peak modelled level was 1.8m OD. A comparison of the recorded and modelled water surface elevation profiles is shown in Figure 4.7.19.

This event was used to verify the model, the initial simulation achieved satisfactory agreement between the model output and recorded data and no adjustments were made to the model parameters.

2

1.5 AOD)

(m

1 Recorded Elevation

0.5 Modelled Surface

0 Water 0 500 1000 1500 2000 2500 3000

‐0.5 Time (minutes)

Figure 4.7.19: Comparison of the Recorded and Simulated Water Surface Elevation Profiles for the November 2009 event

(b) Feb 2002 The historical data indicated that flooding occurred in Dublin in February 2002 as a result of an exceptionally high tide. Around Dublin, the tidal flooding caused an increased water level in rivers and resulted in the Rivers Liffey, Dodder, Tolka and the Royal Canal all bursting their banks. Power cuts occurred due to flooding of ESB substations at Clontarf, Ringsend and East Wall. A large number of houses in the Ringsend and Irishtown areas were affected by the floods with three people hospitalised and over 100 residents evacuated from their homes. The Clontarf Road, Merrion Gates, Strand Road, North & South Quays and East Link Bridge were also flooded. Public transport networks were disrupted with the DART closed between Lansdowne Road and Dun Laoghaire. A report entitled “Dublin Coastal Flooding Protection Project (DCFPP) Final Report” by Royal Haskoning outlined peak flows of 10.5m³/s for the River Dodder at Waldron's Bridge Hydrometric Station and 8.26m³/s for the River Tolka at Botanic Gardens Hydrometric Station. A Dublin City Council Report entitled "Flood 2002, Interim Assessment Report" indicated that the tide level of 2.95mOD (Malin) was the highest reading on record and was higher than the flood of 1924, when the previous highest tide at Dublin Port was recorded.

IBE0600Rp00027 4.7.21 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

A model run was undertaken using a peak coastal water level of 2.95m at the downstream boundary of the River Liffey. The model predicted flooding at the East Link Bridge toll booths and along the South Quays. The model did not predict flooding along the North Quays, however the still water level predicted by the model is only 60mm lower than the quay wall in the locations recorded as flooding, Figure 4.7.20 presents a comparison of the observed and modelled flood extents. Flood mitigation measures have been put in place around the entrance to the Royal Canal to prevent the much of the flooding shown in Figure 4.7.20. This event was used to verify the model again with no further parameter adjustment considered necessary.

Entrance to the Royal Canal

Figure 4.7.20 Comparison of February 2002 obserbed and modelled flood extents

(c) Nov 2000 Extensive flooding occurred throughout large parts of Dublin and Kildare in November 2000 as a result of heavy rainfall, high tides and strong winds. In Dublin, 78.3mm of rainfall was recorded at Dublin Airport over a period of 40 hours beginning at 8.00am on 5th November, with 95.3mm of rainfall measured at Casement Aerodrome, Baldonnel for the same period. The Poddle, Dodder, Tolka and Liffey Rivers all overflowed their banks with coastal flooding also occurring at Clontarf and Dun Laoghaire. Insufficient data was available to use this event during verification.

(d) Feb 1994 The Dublin area experienced flooding in February 1994 as a result of heavy rainfall. The River Liffey overflowed, causing flooding in the Clondalkin area. Insufficient data was available to use this event during verification.

IBE0600Rp00027 4.7.22 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

(e) June 1993 Flooding was severe in the Dublin area following extremely heavy rainfall. The 12- hour rainfall had an AEP of 1% at Baldonnel Station while the 24 hour rainfall total had an AEP of 0.4% (further details on recorded rainfall amounts at various locations are available in an ESBI report – Reference 14). The total rainfall at some of the Dublin and Kildare rain stations were the highest recorded and even exceeded Hurricane Charlie in August 1986. Trains were prevented from leaving Heuston Station and a number of properties in the Dun Laoghaire Rathdown area were affected by flooding. Insufficient data was available to use this event during verification.

Additional events were chosen, for which data was available, to allow verification of the model. These events did not necessarily cause flooding from the River Liffey. Estimated flows for these events were generated as per the method described under event (a) above. Two water level gauges are present within the modelled reach at O'More Bridge adjacent to Victoria Quay and at the downstream extent of the model at Dublin Port. Data from these gauges relating to the events below and event (a) above was used during model verification.

(f) August 2008 The event in August 2008 had an estimated peak flow of 124m³/s at the upstream boundary of the model, this approximately equates to a 30% AEP fluvial event. The peak level recorded at O'More Bridge was 2.11m AOD, the modelled peak level was 2.16m OD, a comparison of the recorded and modelled water surface elevation profiles is shown in Figure 4.7.21.

2.5

2

1.5 AOD)

(m

1 Recorded Level Elevation

0.5 Modelled Surface

0

Water 0 1000 2000 3000 4000 5000 ‐0.5

‐1 Time (minutes)

Figure 4.7.21: Comparison of the Recorded and Simulated Water Surface Elevation Profiles for August 2008 event

IBE0600Rp00027 4.7.23 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

The model achieves satisfactory agreement with recorded data at peak tidal levels which is the focus of the verification process. There remains a degree of uncertainty regarding the river flows due to lack of flow gauging close to this location.

(g) August 2007 The event in August 2007 was not a significant fluvial event with an estimated flow of 84m³/s, a flow of this magnitude is likely to be exceeded once a year. The peak recorded level at O'More Bridge was 2.09m OD, the modelled peak level was 1.96m OD; a comparison of the recorded and modelled water surface elevation profiles is shown in Figure 4.7.22.

The close agreement supports the verification of the model without any requirement to adjust model parameters.

2.5

2 AOD)

1.5 (m

1 Recorded Elevation

0.5 Modelled Surface

0

Water 0 2000 4000 6000 8000 ‐0.5

‐1 Time (minutes)

Figure 4.7.22: Comparison of the Recorded and Modelled Water Surface Elevation Profiles for August 2007 event

(h) May 2007 The flow and tidal boundary record for 15 days during May 2007 was input to the model. This period provided a steady low flow throughout the month and a typical tidal profile enabling a low flow verification of the model. This verification run also enabled the response of the model to tidal variations to be made without a large fluvial influence. Analysis of the recorded and modelled profiles at O'More Bridge indicates that the model produces a realistic water surface profile with levels generally being within the 200mm calibration requirement with an average difference between the modelled and recorded levels of 0.079m; a comparison of the recorded and modelled water surface elevation profiles is shown in Figure 4.7.23.

IBE0600Rp00027 4.7.24 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

2

1.5 AOD)

1 (m

0.5 Recorded Elevation

0 Modelled 0 5000 10000 15000 20000 Surface

‐0.5 Water ‐1

‐1.5 Time (minutes)

Figure 4.7.23: Comparison of the Recorded and Modelled Water Surface Elevation Profiles for May 2007 event

Summary of Verification

Good correlation was achieved between the recorded levels and those generated by the hydraulic model for the events considered during the verification exercise. It should be noted that the model generates still water levels whilst the level gauge may experience a degree of wave action.

As a result of draft mapping review workshops, Local Authorities provided information on past flood events that contributed further to the model verification. The Local Authority review of the draft mapping indicated general agreement with the flood extent mapping. However, in light of flooding which occurred during the extreme coastal water level event in January 2014 a reach of flood defence along Victoria Quay was removed from the model as flood water bypassed the wall. Dublin City Council indicated that the wall in question may not be completely impermeable and as such should not be considered formal effective. Figure 4.7.24 shows the flooding which occurred in January 2014 along Victoria Quay.

IBE0600Rp00027 4.7.25 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.24: Photograph of flooding at Victoria Quay, January 2014

Figure 4.7.25 presents the predicted flooding along Victoria Quay during a 0.5% AEP coastal event when the wall which is shown in the image above is removed from the model.

Figure 4.7.25: Simulated flooding along Victoria Quay during a 0.5% AEP coastal event

IBE0600Rp00027 4.7.26 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Other information provided during the local authority workshop is summarised below:

 Flood defence works were undertaken at the mouth of the Royal Canal to prevent water levels in the canal rising during flood events on the River Liffey.

 Flood defences are to be installed along City Quay and Sir John Rogerson's Quay to a level of 4.0m AOD

 There are gaps in the quay walls at Halfpenny Bridge, this was checked to ensure the gaps were included in the model.

In addition, the ICPSS flood outlines were used to compare the maximum possible extents shown in Figure 4.7.26. Although comparison of the modelled extents with the ICPSS extents does not provide a reliable calibration method, it does provide supporting information on whether the model is over-predicting flood extents. As would be expected the different DTM datasets being used to generate the flood extents, resulted in small areas of additional flooding present in the CFRAMS modelled flood extents when compared to the ICPSS mapping. The ICPSS flood areas can therefore not be taken as an absolute maximum extent but rather an indicative guide and the overall comparison is in agreement supporting verification of the model. No adjustments were made to the model in light of this comparison.

Wolf Tone Quay

City Quay

IBE0600Rp00027 4.7.27 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.26: Comparison of Lower Liffey North Modelled 0.1% AEP Flood Extents and ICPSS 0.1% AEP Flood Extents

The mass balance assessment of the model is within acceptable bounds with a Mass Error Balance of 0.45% during the 1% AEP event.

(2) Public Consultation Comments:

Following informal public consultation in early 2015, it was noted that the South Docklands Campshire Works are currently ongoing, these are not included on flood hazard and risk mapping issued for the formal S.I. public consultation period. It was also noted that the Quay Walls, which are modelled as effective defences, may not afford the full level of protection indicated on the mapping due to seepage.

To be completed for final version of the report.

(3) Standard of Protection of Existing Formal Defences:

Halfpenney Bridge

Figure 4.7.27: Modelled flood defences and hydraulically significant structures from O'More Bridge to Talbot Memorial Bridge on the Lower Liffey

IBE0600Rp00027 4.7.28 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.28: Modelled flood defences and hydraulically significant structures from Islandbridge to O'More Bridge on the Lower Liffey

Figure 4.7. and 4.7.28 illustrate the locations and estimated SOPs of the flood defences included in the Lower Liffey model. A number of other structures were identified as having a potential impact on the flood extent and have been included in the model as hydraulically significant structures. The images below show examples of the flood defences and a hydraulically significant structure.

Figure 4.7. presents an example of defence walls with an estimated SOP of 0.1% AEP.

Figure 4.7.29: Flood Defence Wall with an estimated SOP of 0.1% AEP

IBE0600Rp00027 4.7.29 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.30: Flood Defence Wall with an estimated SOP of 0.5% AEP

Figure 4.7.30 presents an example of a defence wall with an estimated SOP of 0.5% AEP. The SOP of this section of wall is compromised by gaps in the wall left for pedestrian access to the cantilevered pedestrian boardwalks, lowering the SOP of the wall from 0.1% AEP to 0.5% AEP.

Figure 4.7.31: Flood Defence Wall with an estimated SOP of 1% AEP

Figure 4.7.31 shows the flood defence wall along Wolf Tone Quay which has an estimated SOP of 1% AEP. This low section of wall is predicted to be overtopped during a 0.5% AEP coastal event.

IBE0600Rp00027 4.7.30 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

A number of other structures were surveyed as part of the infill survey; however these were determined not to be flood defence structures. The wall shown in Figure 4.7.32 was included in the model as a hydraulically significant structure. The wall was deemed to be an ineffective flood defence structure as flood water is able to flow behind the wall via low points. It was retained in the model as a hydraulically significant structure as water may get trapped and pond behind it.

Figure 4.7.32: Hydraulically Significant Structure at Islandbridge

Another hydraulically significant structure has been included at Sarah Place, as water may get trapped behind these structures. Figure 4.7.33 shows the surveyed wall at Sarah Place.

IBE0600Rp00027 4.7.31 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.33: Hydraulically significant structure at Sarah Place

(4) Gauging Station:

There are two gauging stations located within the modelled reach. One is located at O'More Bridge and the other at Alexandra Basin. Both stations were used to provide validation data for the computational model, however neither are suitable for hydrological estimation as they are tidally influenced.

(5) Other Information: None

4.7.6 Hydraulic Model Assumptions, Limitations and Handover Notes

(1) Hydraulic Model Assumptions:

(a) Please refer to Section 3.4 for general assumptions using the Infoworks ICM modelling software

(b) In-channel roughness values have been selected based on normal bounds values which have been reviewed during the validation process.

(c) All culverts and bridges have been assumed to run clean during design events.

(d) Draft mapping assumed all formal and informal defences are effective.

(d) No specific afflux information is available for validation of headloss across bridges, as such all bridge coefficients have been considered at default values.

(e) The point of tidally dominant flooding was established based on a comparison of tidal flood levels and fluvial flood levels for the 0.1% AEP event. At the point indicated on the flood mapping levels upstream are greater during the 0.15 AEP fluvial event and greater downstream during the 0.1% AEP tidal event

(2) Hydraulic Model Limitations:

(a) Road and street networks have been defined by the inclusion of building polygons, but have not been specifically embedded in the 2D mesh.

(b) A mesh resolution of 1m2 to 25m2 has been applied.

(c) There are no major instabilities within the model.

(d) No flow gauging stations are located within the modelled reach, flows for validation purposes have therefore been estimated based on relevant information.

(e) Only flooding from the channels included in the model has been considered. Flooding from backing up of minor drainage systems has not been considered.

Hydraulic Model Parameters:

1D Domain

Timestep (seconds) 1

Min / Max Space Step 0.5m / 100m

Max Timestep Halvings 10

IBE0600Rp00027 4.7.32 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Max Iterations 10

2D Domain

Timestep (seconds) Dynamic

Timestep Stability Control 0.95

Maximum Velocity 10m/s

Theta 0.9

Inundation Mapping depth threshold 0.01m

(3) Design Event Runs & Hydraulic Model Handover Notes:

(a) A model timestep of 1 second has been applied to the model. It is possible to produce a stable run at a higher timestep, however this can result in a greater mass balance error.

(b) No drainage networks have been included in the model - as such, flows have been introduced directly to the 1D domain as point or lateral inflows as determined in the hydrological analysis.

(c) Two bridges are in close proximity at chainage 2560 (Butt Bridge and the Railway Bridge), the inclusion of both bridges within the model caused significant instabilities, causing model runs to fail, as there is only 8m between the downstream face of Butt Bridge and the upstream face of the railway bridge. This distance does not allow enough distance for the model to calculate the expansion and contraction losses of the bridges in question leading to failure of the model runs. Butt Bridge has a much more intrusive profile in comparison with the railway bridge (Butt Bridge has archsprung openings with soffit levels between 2m and 3.42m OD compared to the railway bridge square soffit openings at 8.4m OD) and was considered to be the controlling structure at this location; the railway bridge was therefore removed from the model to achieve model stability.

(d) No significant flooding is predicted during either the fluvial or coastal 10% AEP event.

(e) Flooding is predicted in the Bellevue area during the 1% AEP 0.1% fluvial events. Flood water from the Liffey enters the Bellevue apartments via a low point in the boundary wall. The 0.1% AEP fluvial event is also predicted to cause significant flooding to properties on the opposite bank along the Chapelizod Road.

(f) In addition to the flooding in Victoria Quay discussed in Section 4.7.5(1) and shown in Figure 4.7.24 and Figure 4.7.25, significant flooding is predicted in the vicinity of City Quay during the 0.5% AEP and 0.1% AEP coastal events. Figure 4.7. illustrates the modelled extent of the potential flooding during the 0.5% AEP event.

IBE0600Rp00027 4.7.33 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7.34: Modelled Extent of Potential Flooding at City Quay during a 0.5% AEP event

(g) Flooding is predicted in the northern portion of Dublin Port during the 0.1% AEP event and 0.5% AEP event as shown in Figure 4.7.. During the 0.5% AEP event flooding is predicted in the vicinity of the Irish Ferries Terminal with some minor flooding of the container storage area but no buildings are predicted to be at risk. More extensive flooding of the container storage area is predicted during the 0.1% AEP event, with the Irish Ferries Terminal building also predicted to be at risk. Flooding during the 0.1% AEP event is predicted at the eastern end of Alexandra Road affecting the dry docks and the Odlums factory. Some flooding is also predicted along the northern shore of this portion of Dublin Port, affecting industrial properties along Promenade Road and Tolka Quay Road.

IBE0600Rp00027 4.7.34 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Odlums Factory

Figure 4.7.35: Modelled Extent of Potential Flooding in Dublin Port during 0.5% & 0.1% AEP events

(h) No properties are predicted to be at risk during the 0.1% or 0.5% AEP event in the southern portion of Dublin Harbour, shown in Figure 4.7.. However a set of water treatment tanks are predicted to be inundated which may have associated environmental risk implications.

Figure 4.7.36: Modelled Extent of Potential Flooding Dublin Harbour South During 0.5% & 0.1% AEP events

IBE0600Rp00027 4.7.35 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

Figure 4.7. illustrates the comparison of the predicted defended and undefended flood extents. The model extents indicate that a number of the Quays in Dublin would experience flooding during the undefended 0.1% AEP event.

Figure 4.7.37: Predicted Defended and Undefended Flood Extents in 0.1% AEP Event

(4) Hydraulic Model Deliverables:

Model deliverables are supplied in an accompanying InfoWorks ICM transportable database containing all model files as required by the brief and the relevant network and event files.

Please see Appendix A.4 for a list of all GIS files provided with this report.

(5) Quality Assurance:

Model Constructed by: Andrew Sloan

Model Reviewed by: Andrew Jackson

Model Approved by: Grace Glasgow

IBE0600Rp00027 4.7.36 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

APPENDIX A.1 STRUCTURE DETAILS

Structure Details ‐ Bridges and Culverts: RIVER BRANCH CHAINAGE Shape WIDTH HEIGHT LENGTH MANNINGS N RECT 25.12 8 RECT 20.6 10.2 Lower Liffey 710 RECT 33.31 12.31 11.66 0.04 RECT 24.72 10.04 RECT 16.23 9.32 RECT 22.14 11.68 Lower Liffey 1600 27.43 0.04 RECT 87.18 11.03 RECT 26.94 9.6 Lower Liffey 2090 ARCH 39.79 10.31 4.45 0.04 RECT 22.06 8.77 RECT 18.19 6.06 Lower Liffey 2340 ARCH 32.64 7.71 22.8 0.04 RECT 21.45 5.86 ARCH 11.78 5.39 Lower Liffey 2540 ARCH 32.15 7.53 25.7 0.04 ARCH 10.57 5.71 ARCH 10.85 6.62 52.12 0.04 Lower Liffey 2860 ARCH 13.65 8.9 ARCH 11.09 6.8 Lower Liffey 3120 ARCH 42.67 9.19 3.66 0.04 Lower Liffey 3260 ARCH 46.48 6.58 4.5 0.04 ARCH 4.58 4.57 ARCH 12.37 6.39 Lower Liffey 3450 ARCH 13.74 6.53 20.08 0.04 ARCH 12.44 5.72 ARCH 4.58 4.57 ARCH 12.94 5.64 Lower Liffey 3740 ARCH 14.28 6.29 20.12 0.04 ARCH 12.84 6.32 ARCH 11.84 5.69 Lower Liffey 3980 ARCH 13.61 7.6 16.47 0.04 ARCH 12.32 5.81 ARCH 9.75 6.79 Lower Liffey 4310 ARCH 11.8 7.46 10.71 0.04 ARCH 9.5 6.84 Lower Liffey 4460 RECT 36.8 7.55 23.91 0.04 Lower Liffey 4550 ARCH 28.86 7.13 12.65 0.04 RECT 11.25 3.71 Lower Liffey 5020 ARCH 21.11 6.6 18.42 0.04 ARCH 11.2 6.35 Lower Liffey 5080 ARCH 31.13 6.73 7.7 0.04

IBE0600Rp00027 4.7.37 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

ARCH 4.72 4.31 ARCH 5.29 7.52 RECT 33.23 10.21 Lower Liffey 5910 9.89 0.04 ARCH 5.26 3.74 ARCH 5.36 3.34 ARCH 5.29 3.32 Lower Liffey 6210 ARCH 31.86 9.91 12.99 0.04 RECT 2.73 1.75 8 0.05 Belleview Millrace 220 RECT 2.73 1.75 Belleview Millrace 90 RECT 7.88 1 7.5 CW 6 Belleview Millrace 70 RECT 4.99 2.43 17 CW 1.5 RECT 1.69 1.67 Belleview Millrace 50 13 CW 1.5 RECT 1.37 1.37

Structure Details ‐ Weirs: Discharge RIVER BRANCH CHAINAGE ID Coefficient Type Liffey 6720 09LIFF00672_WEIR_BREAK_US.1 1.7 Irregular Weir Belleview Millrace 330 09LIFF00672_WEIR_BREAK_US.2 1.7 Irregular Weir Belleview Millrace 30 09BELL0003_BREAK.1 1.7 Irregular Weir

IBE0600Rp00027 4.7.38 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

APPENDIX A.2 FLOW COMPARISON

IBE0600 EAST CFRAM STUDY RPS PEAK WATER FLOWS

AFA Name Lower Liffey HPW Model Code HA09_Liff2C Status DRAFT Final Date extracted from model 25/02/2015

Peak Water Flows

Model Flow River Name & Chainage AEP Check Flow (m³/s) (m³/s) Diff (%) River Liffey 10% 183.13 276 47 09_631_D 1% 252.75 330 27 0.1% 338.01 435 24 1% no tide 252.75 248.78 1.5 River Liffey 10% 165 09_1874_17_RPS 1% 216 0.1% 282 1% no tide 208 River Liffey 10% 150 09_1872_9_RPS 1% 205 0.1% 274 1% no tide 205

The large tidal influence at the downstream boundary is reflected in the large differences seen in the peak estimated and modelled flows. An additional model run was carried out with a static downstream boundary level of 0m AOD. This run provided a much better comparison between estimated and modelled flows, which are included in the table above.

IBE0600Rp00027 4.7.39 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

APPENDIX A.3 LONG SECTION

Lower Liffey 1% AEP Long section

IBE0600Rp00027 4.7-40 Rev F02 Eastern CFRAM Study HA09 Hydraulics Report – DRAFT FINAL

APPENDIX A.4 GIS DELIVERABLES - HAZARD

Flood Extent Files (Shapefiles) Flood Depth Files (Raster) Water Level and Flows (Shapefiles) Fluvial Fluvial Fluvial E24EXFCD100C0 E24DPFCD100C0 E24NFCDC0 E24EXFCD010C0 E24DPFCD010C0 E24EXFCD001C0 E24DPFCD001C0 Coastal E24NCCDC0 Coastal Coastal E24EXCCD100C0 E24DPCCD100C0 E24EXCCD010C0 E24DPCCD010C0 E24EXCCD001C0 E24DPCCD001CO

Flood Zone Files (Shapefiles) Flood Velocity Files (Raster) Flood Defence Files (Shapefiles) E24ZNA_FCDC0 To be issued with Final version of this report N/A E24ZNB_FCDC0

GIS Deliverables - Risk

Specific Risk - Inhabitants (Raster) General Risk - Economic (Shapefiles) General Risk-Environmental (Shapefiles) Fluvial E24RIFCD100C0 E24RIFCD010C0 E24RIFCD001C0

Coastal E24RICCD100C0 E24RICCD010C0 E24RICCD001C0

IBE0600Rp00027 4.7-41 Rev F02