15
portray the interactions between the atmosphere and the urban areas, SURFEX surface model is applied. Main objective of using SURFEX is downscaling the regional
projections for the future over Hungarian cities, and this paper is focusing on the
first step of this, i.e. the validation of the surface model.
1.3.3.2 Methodology
The SURFEX (SURFace EXternalisée; Le Moigne 2009) surface model consists of
four schemes for urban surface, sea, inland water and nature. Amongst these
schemes the Town Energy Balance (TEB) model (Masson 2000) describes interactions between urban surface and atmosphere by simulating turbulent fluxes. It follows local canyon approach, where canyon represents the road with buildings on the
sides. TEB considers three surfaces (roof, wall, road) with different energy budgets.
It takes several processes into account which are important in urbanized areas, e.g.,
it treats water and snow interception by roofs and roads, fog, runoff, radiative trapping, momentum and heat fluxes. The anthropogenic heat and moisture fluxes
derived from traffic, industry and domestic heating are also considered.
As input, SURFEX needs information about the atmospheric conditions, i.e. the
atmospheric forcing, which can be supplied either by measurements or an atmospheric model. The atmospheric model may be coupled with SURFEX and thus it
can get feedback from the surface scheme, but SURFEX running in offline mode
(i.e., without feedback) is feasible as well. It is noted that advection is not taken into
account in SURFEX, thus there is no interaction between grid points in offline
mode, which is only possible through the atmospheric model.
At HMS, the SURFEX studies started in 2010 (Vértesi 2011) for modelling
urban heat island (UHI) effect in Budapest. Some 10-year long experiments were
achieved over Budapest and Szeged. The atmospheric forcing was obtained from
ERA-40 re-analysis (Uppala et al. 2005) produced by ECMWF (European Centre
for Medium-range Weather Forecast). Re-analyses are three-dimensional climate
databases, which are created with data assimilation technique using as many observations as possible plus short-range weather forecasts. ERA-40 is a global dataset at
ca. 125-km horizontal resolution, which was downscaled by ALADIN-Climate
regional climate model (Csima and Horányi 2008) to a 10-km resolution domain
covering the Carpathian Basin for 1961–2000. These results were interpolated by a
special configuration of the model to two smaller areas around Budapest and Szeged
at 1 km resolution (Fig. 1.4) for the investigated periods. These served as inputs for
SURFEX, which was run in offline mode at also 1 km resolution. The information
for the fine surface coverage and physiography was derived from the ECOCLIMAP
database (Masson et al. 2003).
The first experiment was conducted over Budapest for 1961–1970. ECOCLIMAP
was created in 2006, thus it might not describe the surface characteristics of the
given period realistically, as several houses have been built since the 60s, especially
in the outskirt. Therefore, the experiment was repeated for 1991–2000 to see
1 Forecasting Models for Urban Warming in Climate Change
portray the interactions between the atmosphere and the urban areas, SURFEX surface model is applied. Main objective of using SURFEX is downscaling the regional
projections for the future over Hungarian cities, and this paper is focusing on the
first step of this, i.e. the validation of the surface model.
1.3.3.2 Methodology
The SURFEX (SURFace EXternalisée; Le Moigne 2009) surface model consists of
four schemes for urban surface, sea, inland water and nature. Amongst these
schemes the Town Energy Balance (TEB) model (Masson 2000) describes interactions between urban surface and atmosphere by simulating turbulent fluxes. It follows local canyon approach, where canyon represents the road with buildings on the
sides. TEB considers three surfaces (roof, wall, road) with different energy budgets.
It takes several processes into account which are important in urbanized areas, e.g.,
it treats water and snow interception by roofs and roads, fog, runoff, radiative trapping, momentum and heat fluxes. The anthropogenic heat and moisture fluxes
derived from traffic, industry and domestic heating are also considered.
As input, SURFEX needs information about the atmospheric conditions, i.e. the
atmospheric forcing, which can be supplied either by measurements or an atmospheric model. The atmospheric model may be coupled with SURFEX and thus it
can get feedback from the surface scheme, but SURFEX running in offline mode
(i.e., without feedback) is feasible as well. It is noted that advection is not taken into
account in SURFEX, thus there is no interaction between grid points in offline
mode, which is only possible through the atmospheric model.
At HMS, the SURFEX studies started in 2010 (Vértesi 2011) for modelling
urban heat island (UHI) effect in Budapest. Some 10-year long experiments were
achieved over Budapest and Szeged. The atmospheric forcing was obtained from
ERA-40 re-analysis (Uppala et al. 2005) produced by ECMWF (European Centre
for Medium-range Weather Forecast). Re-analyses are three-dimensional climate
databases, which are created with data assimilation technique using as many observations as possible plus short-range weather forecasts. ERA-40 is a global dataset at
ca. 125-km horizontal resolution, which was downscaled by ALADIN-Climate
regional climate model (Csima and Horányi 2008) to a 10-km resolution domain
covering the Carpathian Basin for 1961–2000. These results were interpolated by a
special configuration of the model to two smaller areas around Budapest and Szeged
at 1 km resolution (Fig. 1.4) for the investigated periods. These served as inputs for
SURFEX, which was run in offline mode at also 1 km resolution. The information
for the fine surface coverage and physiography was derived from the ECOCLIMAP
database (Masson et al. 2003).
The first experiment was conducted over Budapest for 1961–1970. ECOCLIMAP
was created in 2006, thus it might not describe the surface characteristics of the
given period realistically, as several houses have been built since the 60s, especially
in the outskirt. Therefore, the experiment was repeated for 1991–2000 to see
1 Forecasting Models for Urban Warming in Climate Change