Narellan Creek Overland Flood Study (Vol 1) - Flipbook - Page 37
Narellan Creek Catchment
Flood Study
3.3.2
Grid Size and Topography
The TUFLOW software uses a grid to define the spatial variation in topography and hydraulic
properties (e.g., ground elevations and surface roughness) across the model area. As a
result, the choice of grid size can have a significant impact on the performance of the model.
In general, a smaller grid size will provide a more detailed and reliable representation of
flood behaviour relative to a larger grid size. However, a smaller grid size will take longer to
perform all the necessary hydraulic calculations. Therefore, it is typically necessary to select
a grid size that makes an appropriate compromise between the level of detail provided by
the model and the associated computational time required. A grid size of 2 metres was
ultimately adopted and was considered to provide a reasonable compromise between detail
and simulation time.
In addition, a TUFLOW feature called sub-grid sampling (SGS) was employed as part of the
model setup. When SGS is employed, TUFLOW will calculate water level versus storage
volume relationships based on a more detailed underlying terrain representation rather
than relying on a single elevation at the centre of the grid cell. Similarly, TUFLOW will
calculate water level versus discharge relationships across each cell side based on the more
detailed terrain rather than relying on the elevation at the midpoint of each cell to control
when water moves from one cell to the next. This feature allows storage and conveyance to
be represented in more detail than would have otherwise been allowed. The 1 metre DEM
derived from the LiDAR described in Section 2.4.1 was used for this purpose.
3.3.3
Roughness Coefficients
The TUFLOW software uses land use information to define the hydraulic (i.e., roughness)
properties for each grid cell in the model. The remote sensing information described in
Section 2.5.2 was used as the basis for defining the variation in land use across the TUFLOW
model (refer Figure 5). This land use information was combined with detailed vegetation
mapping around the Harrington Park Lakes area. This combined land use information was
used as the basis for assigning the variation in hydraulic roughness values across the model
area.
The roughness coefficient values listed in Table 7 were initially estimated based on values in
literature and were then refined as part of the model calibration process. Further details of
the TUFLOW model calibration are provided in Chapter 4. As shown in Table 7, some
materials/land uses were assigned depth varying roughness coefficients. For example, trees
were assigned a lower roughness coefficient when water depths were contained below the
tree canopy, transitioning to a higher roughness coefficient once water comes into contact
with the canopy to reflect the additional resistance to flow afforded by branches etc.
3.3.4
Building Representation
The Narellan Creek catchment incorporates significant urban areas. This urbanisation
creates many overland flow obstructions. One of the most significant impediments to
overland flow in an urban environment is buildings. Available research indicates that
buildings have a considerable influence on flow behaviour in an urban environment by
significantly deflecting flows irrespective of whether a building is flooded inside or remains
watertight (Smith et al, 2012). Accordingly, it was considered necessary to include a
representation of the obstruction afforded by buildings in the computer model.
23
