Geoscience Reference
In-Depth Information
The parameters that make each of the tiles different are 4 :
A number of vegetation-related parameters: LAI, vegetation coverage, parameters related
to the canopy resistance (see Section 9.2.4 ), roughness length and the vertical root distri-
bution (needed to determine where the vegetation extracts the soil moisture)
The surface albedo which depends on snow cover
The skin layer conductivity (see
Section 2.3.6 on the soil heat lux for vegetated sur-
faces)
Those differences in parameters results in a situation in which each tile has its own
surface temperature. The vegetation-related parameters are selected for the dominant
high vegetation type and the dominant low vegetation type within the grid box. For
example: within a grid box both 'tall grass' and 'irrigated crops' may be present as
low vegetation types. But if the fraction of 'tall grass' is higher than that of 'irrigated
crops' the vegetation-related parameters for the low-vegetation tile will be assigned
as if all low vegetation in the grid box is 'tall grass'.
The water vapour transport, with the resistances involved, is depicted in Figure 9.17 .
The surface value of the speciic humidity equals the saturated value at the surface
temperature of the speciic tile ( q s, i = q sat ( T s, i ), where i signiies the index of the tile).
For all tiles the moisture transport encounters the aerodynamic resistance, whereas
some tiles have an additional resistance: canopy resistance, snow resistance and soil
resistance. The latter is modelled as:
()
r
=
r
f
θ
(9.33)
soil
soil,min
4
where r soil,min is the resistance when suficient water is available (taken equal to
50 s m -1 ), θ 1 is the water content of the upper soil layer and f 4 is identical to the
formulation in the canopy resistance (i.e., Eq. ( 9.28 )).
For heat transport the picture will be similar, except for the fact that in that case
only the aerodynamic resistance needs to be used, in combination with the surface
temperature of the tile under consideration.
Apart from the tiles discussed in the preceding text, some models also include an
urban tile, as urban surfaces have very distinct properties: pavement that prevents inil-
tration, large roughness, little or no vegetation, heat capacity of buildings, anthropo-
genic heat production (Arnield, 2003 ). Because the spatial extent of urban regions is
generally limited, the inclusion of an urban tile makes sense only for simulations at such
resolution that the urban areas make a signiicant contribution to the total surface. Mod-
els that make use of only one dominant land-use type per grid cell need an urban surface
parameterization only once the resolution becomes iner than the size of urban areas.
See Grimmond et al. ( 2011 ) for an overview of current urban land surface models.
4 For clarity we leave out one extra parameter, which is the fraction of shortwave radiation that directly reaches the
ground surface (between/under the vegetation).
Search WWH ::




Custom Search