2127
Numerical modelling of groundwater flow around contiguous pile retaining walls
Modélisation numérique des écoulements des eaux souterraines autour d’écrans de soutènement
de pieux contigusë
Wiggan C. A., Richards D.J., Powrie W.
University of Southampton, Southampton SO17 1BJ, United Kingdom
ABSTRACT: Pore water pressure constitutes a significant proportion of the lateral load acting on a retaining wall. Consequently,
guidelines often mandate that the worst case hydraulic conditions are applied in the design of retaining walls. This invariably dictates
that retaining walls are treated as impermeable unless special consideration is given to the maintenance of drainage systems or to the
prevention of infiltration. Contiguous pile retaining walls are, however, by their nature permeable unless considerable effort is
expended to prevent seepage through gaps. If allowed, this seepage can result in reduced active side pore water pressures. Numerical
simulations were conducted to determine the impact on pore water pressures of varying the pile gap (x) to diameter (d) ratio, x/d, in a
contiguous pile retaining wall. A relationship between x/d and the effective bulk wall permeability, k
p
was derived, and applied to
two-dimensional simulations representing a contiguous pile wall. The results show that pore pressures behind the retaining wall
reduced significantly with increased x/d.
RÉSUMÉ: La pression de l'eau interstitielle constitue une part importante des charges latérales agissant sur les parois d'un mur de
soutènement. Par conséquent, les règles de l’art imposent que les pires conditions hydrauliques soient considérées dans la conception
d’un mur de soutènement. Cela impose invariablement que les murs de soutènement soient considérés comme imperméables à moins
que des considérations particulières soient données à l'entretien des systèmes de drainage ou à la prévention des infiltrations. Les murs
de soutènement constitués de pieuxcontigus, sont cependant perméables (de par leur structure), à moins que des efforts considérables
soient déployés pour empêcher les infiltrations à travers les intervalles. Ces infiltrations peuvent entrainer une réduction des pressions
interstitielles effectives. Des simulations numériques ont été réalisées afin de déterminer l'impact des variations de l’espace entre
palplanches (x), de diamètre (d), de ratio, x /d, sur les pressions interstitielles d'un mur de soutènement constitué de pieux contigus.
Une relation entre le ratio x/d et la réelle perméabilité du mur, k
p
a été déduite et appliquée à un modèle à deux dimensions
représentant un mur en pieuxcontigus. Les résultats montrent que les pressions interstitielles derrière le mur de soutènement
diminuent significativement lorsque le ratio x/d augmente.
KEYWORDS: Pore water pressure, numerical modelling, retaining wall, seepage forces, surface settlement
1 INTRODUCTION
Guidelines generally require that the most onerous tenable pore
water pressure distribution is adopted for the design of
subsurface retaining structures. For example, Eurocode 7
recommends that, unless reliable drainage can be provided or
infiltration prevented, retaining walls should be designed with
the water table at the ground surface (BSI, 2004). This can
however cause over-conservative and unnecessarily expensive
engineering solutions which go against the ethos of sustainable
development. It would be advantageous if, based on the bulk
permeability of the structure, the hydraulic loads on retaining
walls could be treated as reduced.
There is however limited research into how the geometry
influences hydraulic loads on retaining walls although,
according to CIRIA 580, ‘economic advantages’ might be
derived if through-wall seepage is allowed (Gaba et al. 2003).
This is due mainly to the reduction in pore pressures because of
through-wall seepage. Research into ways of facilitating
through-wall seepage and quantifying its effect is necessary.
1.1
Research in hydraulic loads around retaining structures
Previous research has not generally sought to distinguish
between the long-term pore water pressure distributions around
different types of retaining walls. For example, Potts and
Burland (1983) and Hubbard et al. (1984) showed that the long-
term pore pressures behind a secant pile retaining wall
recovered to near their pre-construction values as might be
expected of an impermeable wall in fine soils. Powrie et al.
(1999) and Carder et al. (1999) observed a reduction in pore
pressures following construction of a contiguous pile retaining
wall at Woodford. The pore pressures at Woodford however,
did not return to their pre-construction values in the long-term.
This reduction was attributed at the time to under-drainage to
the more permeable chalk layer and therefore no consideration
was given to the possible contribution of through-wall seepage.
However Clark (2006) and Richards et al. (2007) have shown
that there was a reduction in long-term pore pressures measured
at a contiguous pile retaining wall in over-consolidated clay at
Ashford. The pore pressures have not returned to their pre-
construction values. Although there is underdrainage to the
more permeable Weald Clay at Ashford, there is evidence that
the long-term reduction in pore pressure can be attributed to
through-wall seepage.
In contrast to retaining walls, there has been significant
research into methods of reducing pore water pressures acting
on shallow tunnels and on tunnels acting as drains. Despite an
earlier assumption by Atkinson and Mair (1983) that
groundwater loadings do not change significantly in the
presence of varying hydraulic conditions, it has been shown by
numerical analyses that segmented tunnel linings do in fact
allow seepage of groundwater which contributes to reduced
pore water pressures around tunnels in fine grained soils. Pore
pressures at segmented lined tunnels approach atmospheric
values and increase with distance away from the tunnel (Shin et
