The influence of large scale inhomogeneities on a construction dewatering system in chalk Dr Toby Roberts CTRL Thames Tunnel
CTRL 320 Thames Tunnel ( 133M) Twin 2.5 km 7.15 m diameter bored tunnel beneath River Thames. Dewatering required for approach structures, 400 m by 28 m in plan by up to 18 m below groundwater level. Project Management: RLE (Arup Bechtel Halcrow SYSTRA) Main Contractor: HOCHTIEF MURPHY JV Dewatering: WJ GROUNDWATER LTD
CTRL Thames Tunnel dewatering for approach structures Tank farm Northern approach Tunnels Southern approach
Southern approach Longitudinal Tunnel Section
CTRL Thames Tunnel Cross-section of approach Roof slab Discharge main Standing W.L Diaphragm wall Internal wall Target drawdown Wells
mod S1 S2 S3 S4 75m 135m 115m 120m m OD 0 Alluvium 0-10 Terrace Gravels -10-20 -30 Upper Chalk -20-30 Tunnel portal D-wall Toe Formation Cross cut-off Dewatering of a 445 m x 28 m structure with side support by diaphragm walls, divided into 4 sections (S1 to S4) Formation level from 18 to 6 mbgl Southern Approach Geological Section
Thames Thames N The extent of the Chalk outcrop Tunnel portal CONFINED AQUIFER Approach structure UNCONFINED AQUIFER Pump test location 0 500 m
Depth (mod) 0 0.01 0.1 1 10 100 m/d -5-10 -15-20 -25 South side pump test North side data Test in piezometers Packer tests -30-35 1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 Permeability (m/s) Chalk permeability depth profile
From CIRIA C515 Chalk Gravel Range of application of dewatering techniques
Design Information: Pumping Test: Terrace Gravels k = 3 10-3 m/s 260 m/d Pumping Test: Chalk: k h = 2 10-4 m/s 17 m/d Packer Tests (North Side) Chalk k decreases with depth Conventional wisdom suggests former spring line and elevated k zone along line of Chalk outcrop (cut-off?) Design Basis: Terrace Gravels k h = k v = 3 10-3 m/s 260 m/d Chalk k h = k v = 2 10-4 m/s 17 m/d
River River 3D Numerical model: Used as design basis Used to consider interaction between sections Used for sensitivity analysis Used to look at hydrostatic load on D- wall MODFLOW finite difference grid
Dewatering Scheme Design Section S1 S2 S3 S4 Length (m) 75 135 115 120 Dig depth (mod) -17-14 -11-7 Design flow (l/s) 192 168 72/120 72/120 No. of wells (12 to 20 l/s) 16 14 6 6 Total design flow: 600 l/s Total No. of wells: 42 No.
DESIGN FLOW S1 192 l/s S2 168 l/s S3 72 to 120 l/s S4 72 to 120 l/s N LEGEND: Designed dewatering well Piezometer 0 50m
Level (m OD) CL +2 0 MG AL -10 TG -20 CHK -30 Internal deepwell dewatering system
Aerial view of southern approach structure (Excavation in S1 and S2 underway)
Dewatering Wells
Water outfalls into the Thames via discharge main
Discharge flow (l/s) S1 S1, 2 S1, 2, 3 S1, 2, 3, 4 600 400 200 0 Nov 01 Feb 02 May 02 Aug 02 Nov 02 Feb 03 May 03 Aug 03 Flow Record
Comparison between design capacity and actual flows Section S1 S2 S3 S4 Length (m) 75 135 115 120 Depth (mod) -17-14 -11-7 Design flow (l/s) 192 168 72/120 72/120 Actual Flow (l/s) 12 206 131 243
ACTUAL FLOW S1 12 l/s S2 206 l/s S3 131 l/s S4 290 l/s N LEGEND: Designed dewatering well Piezometer Additional dewatering well 0 50m
S1 S2 S3 S4 SPECIFIC CAPACITY (m 2 /day) <25 200 to 400 Zone of highly 0 50m 25 to 50 400 to 800 permeable chalk? 50 to 100 > 800 100 to 200 NOTE: Specific capacity = flow/drawdown Variation in well performance along approach structure
HPZ
Thames Thames N Array of remote standpipe piezometers - Tidal fluctuations established prior to start of pumping by datalogging over 24 hr period. - Manual dipping at weekly to monthly intervals for duration of dewatering Chalk piezometer Gravel piezometer Chalk and Gravel piezometers 0 500 m
Depth (m OD) Depth (m OD) Outcrop Chalk High k Chalk 0 SCR TCR 0 SCR TCR -5-5 -10-15 -20-25 -30-10 -15-20 -25-30 -35 0 20 40 60 80 100 Percentage Variation in the quality of cores from the Chalk -35 0 20 40 60 80 100 Percentage
Thames 2500 m A Thames Confined Chalk Transition Zone HPZ Chalk Outcrop HPZ A 2500 m Extrapolated size and location of highly permeable zone of Chalk
Depth (m OD) Excavation (445 m) +3-9 -12-17 -23-26 Alluvium Terrace Gravels Weathered Chalk Transition Zone HPZ Chalk Outcrop Base Chalk -65 A 2500 m Model section along structure A
Design Basis: Terrace Gravels k h = k v = 3 10-3 m/s 260 m/d Chalk k h = k v = 2 10-4 m/s 17 m/d Best Fit Model: Terrace Gravels: Chalk Outcrop: k h = k v = 2 10-3 m/s k h = 6 10-4 m/s, k v = 6 10-5 m/s High k Chalk: k h = k v = 6 10-2 m/s 5,200 m/d Transition Zone: k h = k v = 5 10-4 m/s 40 m/d Weathered Chalk: k h = 4 10-4 m/s, k v = 1 10-6 m/s 0.09 m/d Base Chalk: k h = 2 10-5 m/s, k v = 2 10-7 m/s
mod 0-10 S1 S2 S3 S4 75m 135m 115m 120m m OD 0-10 -20 Low k v Transition HPZ -20-30 -30 Southern Approach Geological Section
Observations Site investigation did not identify high k zone High k zone expected but scale uncertain cut-off? Anisotropic conditions hard to identify in SI Coped with 2 orders of magnitude change in k Cost/programme impacts modest in this case Flow and drawdown data important in early identification of issues and solution Relatively simple model but no unique solution (piezo residuals +/-0.4 m average, max 0.9 m)