Use of the Conduit Flow Process for the simulation of ...

Use of the Conduit Flow Process for the simulation of passive mitigation measures against the piezometric damming effect at the new underground High Speed railway station of Florence. Maria Filippini, Mario Martina, Stefano Menichetti, Luca Ranfagni, Francesco Palmiero, Filippo Alberto Rota, Alessandro Gargini

N°abstract 2106

Site location

N°abstract 2106

Station pit

Station design

The pit in Feb. 2016

N°abstract 2106

Hydrogeologic outlines Aquitard (k: 1x10-6 m/s 1x10-9 m/s)

Aquifer (k: 1x10-3 m/s 5x10-5 m/s)

Aquifer Top elevation:

LITHOSTRATIGRAPHIC HORIZONS IN THE FLORENCE SUBSOIL

[m a.s.l.]

46 m a.s.l. Recent alluvial deposits of the Arno River

35 m a.s.l.

Aquifer Bottom elevation: [m a.s.l.]

Backfill: sand with clay and pebbles.

Predominantly clay

Gravel with sandy matrix and sand lenses

Layered rocks: sandstone, limestone, marl, siltstone

Gravel with clayey matrix

Capecchi et alii, 1975

N°abstract 2106

Piezometric damming effect and active mitigation (temporary) Undisturbed water level

Damming effect

Wells Q

< Effective stress; Floods

Wells drying; Subsidence

0

N°abstract 2106

100

200 [m]

Piezometric damming effect and passive mitigation (permanent) Undisturbed water level

Damming effect

Drains

Q

Q

Objective: Hydraulic transparency

N°abstract 2106

Conduit Flow Process (MODFLOW) Shoemaker, W.B., Kuniansky, E.L., Birk, Steffen, Bauer, Sebastian, and Swain, E.D., 2008, Documentation of a Conduit Flow Process (CFP) for MODFLOW-2005: U.S. Geological Survey Techniques and Methods 6-A24, 50 p.

[m]

[m2/s] [m]

CFP:

[m]

Flow in the pipe (laminar or turbolent)

"0" = head not fixed

??? ( ̴ k of pipe wall) "1" = straight pipe Roughness: very low value assumed

from Moody diagram Flow from the pipe to the aquifer Flow from the aquifer to the pipe

N°abstract 2106

Numerical model – averaged head distribution

Constant head

Main groundwater Flow direction

1600 m 0 m

N

100

Pit and drains

300 m

k [m/s]

N

300 m

Area of pit influence on flow Hydrailc gradient=5.0E-3

Constant head

Steady- state flow N°abstract 2106

Numerical model – averaged head distribution - Results Hydraulic head VS conductance Drain 1: 2 m length 10 cm diameter

Drain 2: 18 m length 20 cm diameter

Flow rate VS conductance

Drain Conductance [m2/s] 1x10-2 1x10-3 1x10-4 1x10-5 1x10-6

N°abstract 2106

Seasonal variations – dry season Flow rate VS conductance

Hydraulic head VS conductance

N°abstract 2106

Seasonal variations – recharge season Flow rate VS conductance

Hydraulic head VS conductance

N°abstract 2106

Hydraulic gradient variation

N°abstract 2106

Hydraulic gradient variation – Results Hydraulic head VS hydraulic gradient

Drain 1: 2 m length 10 cm diameter

Drain 2: 18 m length 20 cm diameter

Flow rate VS hydraulic gradient Hydraulic transparency

Drain Conductance [m2/s] 1x10-3 1x10-4 1x10-5

Hydraulic gradient 1x10-3 5x10-3 1x10-2 1.5x10-2 2x10-2

N°abstract 2106

Conductance issue

- 1:1 physical model: Single-drain ( ̴ 15 m) Different drain wall porosities Different grain size distributions

-Field test on a section of the actual station pit

N°abstract 2106

Conclusions -Drains 18 m long, 20 cm diameter with conductance of 1e-3 m2/s would allow achieving the hydraulic transparency at the Belfiore underground station -Conductance is the most critical parameter for the numeric assessment of the drain system effectiveness -The Conduit Flow Process is an effective numeric tool for the simulation of man-made drains and pipes located below the water table

Open questions: How could we determine drain conductance? Is the achievement of hydraulic transparency essential at the Belfiore station?

Thanks! N°abstract 2106