Environmental Engineering Reference
In-Depth Information
Table 23.1
Flood Zones definitions
in unacceptably long run times. Consideration
was given to other ways of representing buildings
(such as using unfiltered LiDAR, imposing the
buildings back onto the filtered data through use
of MasterMap layers and using a 'porosity' func-
tion) but these were discounted as either imprac-
tical or unproved/inaccurate following initial
trials. Flooddefences had to be removed fromthese
data prior to the construction of themodel as Flood
Zones, by definition, show the undefended flood-
plain for various annual probabilities of flooding.
Flow routes were checked and schematized using
aerial photography.
The outputs from the 2D models were then
post-processed to remove small 'dry islands' (po-
tentially erroneous dry areas surrounded by flood-
ing, caused, e.g., by remnant man-made features
commonly present in LiDAR-derived DTMs).
Flood Zone
Annual probability of
ooding
1
'Low probability': less than 1 in 1000 (
0.1%)
<
2
'Medium probability: between 1 in 1000 (0.1%) and
1 in 100 (1%) for river
ooding, 1 in 200 (0.5%)
for
ooding from the sea
3a
'High probability': 1 in 100 or greater (
1%) for river
>
ooding and 1 in 200 (0.5%) or greater for
ooding from the sea
their annual exceedance probability of flooding
(Table 23.1) and depict the extents of flooding if
there were no flood defences. An associated map
is the 'area benefiting from defences' - a map
showing areas that would flood at the 1% annual
exceedance probability (0.5% for coastal flooding)
if formal defences were not in place/operational.
The Flood Zones and 'areas benefiting from
defences' maps for the Thames Estuary between
Teddington and Dartford were generated using a
combination of 1D and 2Dmodelling as described
below (Halcrow 2006).
Modelling methodology: areas benefiting
from defences
The Thames Estuary between Teddington and
Dartford benefits from defences that currently
provide protection against events with a greater
than 0.1% probability of occurrence. Such a high
standard of protection is probably an exception
nationally, and the Environment Agency has tak-
en the decision to show the area benefiting from
defences in London going out to the extreme flood
outline (i.e. the 1000-year event) rather than the
200-year event normally used for tidal flooding.
The same approach tomodellingwas used as for
the Flood Zones but with the flood defences added
to the model. The area benefiting from defences
was calculated as the difference between the with-
defences and no-defences flood extents.
Modelling methodology: flood zones
To generate the Flood Zones for the Thames Es-
tuary upriver of Dartford, hydraulically discrete
areas of tidal floodplain, known as embayments,
were modelled individually using a 2D floodplain-
only model (developed using the TUFLOW soft-
ware; www.tuflow.com).
Inflowboundaries were derived for each embay-
ment model using the 1D ISIS model of the
Thames Estuary, which used a tidal surge event
with peak water levels equivalent to the extreme
water levels calculated in the work described
above (see 'Extreme water levels'). These bound-
aries were then used to run the 2D 10-m grid
models. The numerical grids used in the 2D
modelling were constructed using a digital terrain
model (DTM) derived fromLiDAR (light detection
and ranging) data. Filtered LiDAR was used for
the floodplain modelling with a 10-m computa-
tional grid. Buildings in the floodplain were repre-
sented through the use of higher roughness values.
The use of a smaller grid size would have resulted
Appropriateness of modelling methods
For the flood mapping, non-linked 1D (channel)
and 2D (floodplain) models were considered most
appropriate, with each embayment modelled sep-
arately. This approach intentionally maximized
the simulated flood extent (there was no feedback
from the flow entering the floodplain to act to
reduce
inflow from the
river). Alternative
