July/August 2024 - 117

Forensic Study Observations
and Conclusions
The designer specified maximum 1 ft
(0.3 m) layer lift for compaction, but the
contractor actually used layers lift of 2
ft (0.6 m) (generally, the vertical space
between reinforcement layers). The
reason for the increase was to " ease
construction. "
Quality control in the field showed
that under the 10 T vibratory roller, Dr
was greater than 80% for the 2 ft (0.6 m)
layer lift. However, within the 3 ft (0.9
m) behind the wall, there was no
quality control at all. That is, there were
no field compaction tests done along
the seawall to ascertain that the
jumping jack can effectively compact 2
ft (0.6 m) of loose sand. A few such tests
were done at the beginning of the
construction of Millstream; however,
compaction there was much easier
than in the seawall where buttresses
and no-fine blocks restricted access. It
is a common practice to specify light
compaction equipment next to the
facing so as not to affect the wall
alignment; however, layer lift then is
limited to 6 in (152 mm). The general
impression from the construction notes
was that fill next to the facing had
secondary importance.
Most importantly, there were no
tests to ascertain that compaction next
to the geotextile will not drag the
geotextile down, possibly forming folds
and creases. It also was not part of the
quality control plan; no periodic
verification of proper installation of the
geotextile was conducted.
During construction, small sinkholes
( " ratholes " ) were observed next to
some joints. At that stage the seawall
was not submerged yet (it was " on the
dry " ). The contractor infilled these
holes with dry sand, ignoring the
possibility that the joints are not
properly sealed.
The geotextile apparent opening
size was sufficient to retain the Botany
Bay sand. Accelerated tests (conducted
after sinkholes formed and paid for by
the contractor) on large-scale model
walls generating simulated waves
demonstrated that in a period of 10
years, only a minute amount of sand
would escape through the joints,
certainly not sinkholes shortly after the
end of construction.
As a result, the forensic study drew
several conclusions.
First, it is unlikely to achieve
adequate consistent compaction using
lightweight equipment when having
layer lifts of 2 ft (0.6 m). This is even
more so when miles of such tedious
work need to be conducted under hot
weather at the site. Moreover, the space
within which such compaction was
needed was hardly accessible.
Quality assurance and quality control
to ensure that the geotextile was
affixed properly, regardless of the
thickness of a layer lift, is essential,
especially in a case of where bidirectional
water flow exists. Repeated tidal
water changes is a driving power that
moves sand.
It is apparent that the contractor
considered the 3 ft (0.9 m) behind the
panels as an unimportant portion of the
wall and therefore, spent very little
effort in that region.
Post failure excavations in 2002
at some locations where sinkholes
occurred clearly showed that the
geotextile had folds filled with coarse
sand and shell particles. These folds
extended between the retained sand
and the open joint. Aided by tidal
fluctuations and wave action, retained
sand could escape into the seaside of
the wall, resulting in sinkholes. Such
little to do with wave action or tidal
fluctuations. Its formation was likely
due to poor construction.
Folds were formed by downdrag
force occurring as the sand moves
down relative to the geotextile. If the
sand is loose, it will compress inservice
under tidal fluctuations. Alternatively,
folds could have formed
during inadequate compaction where
down-dragging occurs. In fact, sinkholes
occurred at only 30% of the joints.
Random quality construction usually
will lead to random performance.
Lessons Learned - Sydney
There are several takeaways from this
case study.
First, contractor and designer must
understand the function and importance
of all the elements comprising the wall.
And quality assurance and quality
control plans must ascertain that all
design specifications are attainable and,
indeed, achieved.
Problems discovered during construction
should be investigated by
experts. Infilling the " ratholes " discovered
during construction with sand
did not solve the problem; it simply
masked the problem.
Compaction is important everywhere
including close to the facing.
This is a fundamental geotechnical
requirement in construction.
And finally, redundancy in structures
is needed. In the Sydney wall, the
importance of the geotextile cannot be
underestimated. If it ceases to function
as a filter, the entire wall system would
fail. There was not a second front in
Aided by tidal fluctuations and wave action, retained
sand could escape into the seaside of the wall, resulting
in sinkholes.
folds in marine environment serve as a
" highway " for sand flow. It is noted that
folds filled with sand/shell were also
observed below the low tide zone thus
implying that the formation of folds had
case filtration by the geotextile becomes
ineffective. The viability of the
no-fine blocks is questionable; it certainly
was an obstacle in compaction
next to the facing.
DEEP FOUNDATIONS * JULY/AUG 2024 * 117
July/August 2024

Table of Contents for the Digital Edition of July/August 2024
