Methodology - Response of Flood Events to Land Use and Climate Change - page 80

Table 3.5 Advantages and disadvantages of low and high resolution of DEM

DEM Advantages Disadvantages Sources

Low Takes a short time for

simulation; in some flat
areas or on a global scale, it
is adequate to capture

Missing a larger number of
feature surface information
such as dyke; delineation of
drainage networks in
lowland flat terrain

[ 47 , 75 , 76 , 140 , 141 ]

High Small error; benefits the local

scale, contains most of
man-made feature surface

Takes a long time for
simulation

[ 47 , 75 , 76 , 140 ], [ 73 ]

Table 3.5 represents some advantages and disadvantages of low & high resolution
of DEM according to the experience of some investigators.
According to Alkema [ 3 ], the computation time of simulation depends on the
number of cells in the grid and consequently, a large number of grid cells for
the large area make it impossible to feed it directly into the flood model, because
the calculation time would become enormous.
Vaze [ 1 ] mentioned that the vertical accuracy of grid cell elevation is a critical
and important factor, as a small error in grid cell elevation can result in totally
different and incorrect model predictions of the values of the spatial indices
derived from DEM. He also indicated that the landscape process scale (global,
regional, local) is the key driver in determining a useful and more applicable grid
cell resolution scale.
The landscape process scale and the resolution of DEM have been discussed in
many papers. For example, Gallant and Hutchinson [ 79 ] point out that the grid
resolution of DEMs can profoundly influence the spatial patterns of attributes and
also the related models built from these attributes. Schoorl et al. [ 80 ] also dis-
cussed the influence of implications of varying DEM resolutions on the numerical
values of attributes derived from them.
In another example, Warren et al. [ 81 ] compared slopes measured in the field
with different resolutions of DEMs, and found that higher DEMs with 1 m reso-
lution produced much better and more accurate results than lower DEMs with
12 m resolution. They commented that this variation can lead to widely varying
estimations in environmental parameters such as soil erosion. As a general mes-
sage, Wilson and Gallant [ 82 ] stated that ''additional work is required to identify
the important spatial and temporal scales and factors that influence or control the
process and patterns operating at particular scales''.
Therefore, it is essential to quantify the impact of different resolution DEMs
application on hydrological modeling as an important variable which affects the
accuracy and reality of the results as the simulation is moved from a high reso-
lution to a lower resolution in comparison with observed and measured data Vaze
[ 1 ]. In this study, in the case of the DEM scenario, the aim was to compare
different resolutions of DEMs (5 m against 10 m) to understand how the grid
resolution of a DEM affects the flow characteristic results. To compare the flow