Geography Reference
In-Depth Information
F
2
Salix, Alnus,
Populus
5
Juglans, Acer
Type 5
Type 4 : less than 20 years
Type 3 : 20 to 40 years
Type 2 : 43 to 68 years
Type 1 : more than 68 years
F
1
−
1
0
5
−
1
Figure 10.7
Factorial map that shows stand composition shift through time in the Arve River floodplain forests.
for hydrodynamic roughness. Hydrodynamic roughness,
or flow resistance, determines the kinetic drag of the
water flow. The higher the roughness, the slower the
water will flow and, hence, the higher the water levels
will reach, increasing the flood hazard. The roughness
of the vegetation growing on the floodplains depends
on vegetation structure and has been described by many
different models that simplify different aspects of the
interaction between water flow and vegetation. Kouwen
(2000) lists the factors that are assumed to determine
vegetation roughness, including vegetation height and
density, rigidity of the stems, pattern of spacing of
the stems, deformation of a plant under flow, density
and orientation of foliage and the drag coefficient. In
addition to vegetation roughness, the ground surface will
also induce a friction force on the water. The problem
is that no accurate, spatially distributed and quantita-
tive method exists to parametrise the hydrodynamic
roughness of the floodplains as the input for models.
In practice, often a single roughness coefficient is used
for the whole floodplain section, which is subsequently
used as a calibration parameter to fit modelled water
levels to measurements. This may seriously misrepresent
local flow conditions relevant to sediment deposition
and scour. More detailed methods consist of land
cover classification, combined with a lookup table for a
roughness coefficient. Still, most of the variation in the
3D plant structure is not taken into consideration.
Various remote sensing data may provide information
on vegetation type and structure including their dynam-
ics. The key issue to overcome is the translation of remote
sensing information, i.e. the translation of the intensity
and pattern of reflected electromagnetic radiation into the
relevant parameters in order to be able to register patterns
of hydrodynamic roughness. Many studies have reported
successful and accurate mapping of natural vegetation
using multispectral or hyperspectral remote sensing data
(e.g. Mertes, 2002). Recently, spectral information has
been combined with height information in vegetation
classification schemes (e.g. Dowling and Accad, 2003).
Even though the spatial resolution and the level of detail
of the classification varies with the type of remote sens-
ing data, a lookup table is always required in order to
convert the vegetation classes to vegetation structure val-
ues, which leads to the undesirable loss of within-class
variation. In contrast, LiDAR enables the extraction of
the structural characteristics of vegetation such as height,
biomass, basal area, and leaf area index (see Straatsma and
Middelkoop, 2007). However, LiDAR data shows noise
with a standard deviation ranging from 1.5 to 4 cm mak-
ing the application in meadows and pioneer vegetation
unsuccessful.
Search WWH ::
Custom Search