Environmental Engineering Reference
In-Depth Information
to the urban environment. Various approaches
have been suggested, such as the use of a porosity
term (e.g. Hervouet 2007), the application of a
spatially varying roughness coefficient (N ´ elz and
Pender 2007), use of a coverage ratio and convey-
ance reduction factor (Chen et al. 2008), extraction
of subgrid-scale connectivity (Yu and Lane 2006)
and the above mentioned cut-cell approach
(Morris et al. 2006). Future developments in this
area will be expected to appropriately represent
building blockage effects (Lane and Yu 2008; San-
ders et al. 2008), but also overcome any issue
of over-parameterization.
This is facilitated by modern topographic data;
however, the decision on the choice of value to
apply to each category remains somewhat subjec-
tive. For high depth to grid size ratios, the lack of
appropriate theory to assist in setting appropriate
eddy viscosity values may become a significant
issue (Calenda et al. 2003; Liang et al. 2006).
Finally, localized topographic features of small
dimensions such as walls and fences have a sig-
nificant impact on local flow patterns, and data at
the appropriate resolution are seldom available.
Themodelling of buildings in high resolution is
a challenge per se, primarily because real buildings
affect flood flows in different ways, which are
often unpredictable. Some buildings with wide
openings or with openings forced by the strength
of the flood may become flooded with little or no
time delay, while others may remain dry at least
initially. Hydraulic storage space within a build-
ing cannot be inferred frommaps or aerial images,
as some buildings may be elevated in relation to
the surrounding ground, while others may have
basements. Several approaches are available to
include buildings in high-resolution models: the
use of porosity (Hervouet 2007), high roughness
(N ´elz and Pender 2008), or methods where one
of the building walls is artificially opened
(Syme 2008; Schubert et al. 2008).
Practical application of high-resolution regular
square grids faces the major difficulty of compu-
tational inefficiency. A 5-m model of an area of
several square kilometres can typically take hours
to run on computer hardware typical of that avail-
able to engineers at the end of the 2000 decade,
using any of the most widely used software pack-
age solving the full shallowwater equations. With
computational times inversely proportional to the
cube of the grid size (as a change in grid resolution
usually also implies a change in allowable time
step), the only practical approach often consists of
using a coarse grid, particularly in large-scale stud-
ies. The use of unstructured gridmodels can some-
what alleviate the issue, but time steps are still to
some extent governed by the size of the smallest
elements (Calenda et al. 2003). There is therefore
continued pressure on engineers and researchers
to design subgrid-scale methodologies applicable
Conclusions
A vast array of modellingmethodologies nowexist
topredictpathwayperformancewithintheflooding
system. 1D techniques remain appropriate for si-
tuations where a clearly defined one-dimensional
pathway exists, such as rivers and pipe systems.
In situations where no such clearly defined single
pathwayexists, recoursemustbemade to2Dmeth-
ods. Here a choice is required between those tech-
niques employing the full shallowwater equations
and those based on simplified equations, such as
diffusive or kinematic wave equations. The con-
sequences of this choice, when models are applied
to floodplains with complex geometries, are pres-
entlyunclear; however, ongoing research (Environ-
ment Agency, 2009) will soon clarify such choices.
Research is continuing to deliver improve-
ments in 2D modelling methods. In particular in
the areas of trans-critical flow simulation, hybrid
methods (linking 1D and 2D models) and discre-
tization techniques. Additionally, the systems
approach required by flood risk management and
the uncertainties associated with predicting pres-
sures and flood sources dictate a need for multiple
simulations to facilitate a probabilistic approach
to uncertainty analysis. There is therefore a press-
ing need for faster model predictions through
either the development of accurate model emula-
tion techniques or the use of parallel processing to
speed up 2D modelling methods. Research is on-
going in each of these areas.
