Towards a fine-scale urban climate simula on of Hong Kong...
1
Towards a fine-scale urban climate simula �on of Hong Kong using MesoNH-TEB with a detailed urban land cover descrip�on Yu Ting KWOK 1 *, Cécile DE MUNCK 2 , Robert SCHOETTER 2 1 School of Architecture, The Chinese University of Hong Kong; 2 CNRM, Université de Toulouse, Météo-France, Toulouse 1. Introduc�on and Objec�ves 2. Model Configura �on and Domains 3. Urban Surfaces and Building Geometry References 4. Building Archetypes 5. Extending Data Coverage with LCZs 6. Conclusion and Further Work Bueno, B., Pigeon, G., Norford, L.K., et al. (2012). Development and evalua�on of a building energy model integrated in the TEB scheme. Geoscien�fic Model Development, 5, 433-448. Ching, J., Mills, G., Bechtel, B., et al. (2018). World Urban Database and Access Portal Tools (WUDAPT), an urban weather, climate and environmental modeling infrastructure for the Anthropocene. Bulle�n of the American Meteorological Society. Kwok, Y. T., Schoe�er, R., & Masson, V. (2018). Defining building archetypes for urban climate simula�ons of the complex high-density environment in Hong Kong. 34th Interna�onal Conference on Passive and Low Energy Architecture, Hong Kong. 3, 982-983. Lac, C., Chaboureau, J., Masson, V. (2018). Overview of the Meso-NH model version 5.4 and its applica�ons. Geoscien�fic Model Development Discussions. Masson, V. (2000). A physically-based scheme for the urban energy budget in atmospheric models. Boundary-Layer Meteorology, 94(3), 357-397. Masson, V., Le Moigne, P., Mar�n, E., et al. (2013). The SURFEXv7. 2 land and ocean surface pla�orm for coupled or offline simula�on of earth surface variables and fluxes. Geoscien�fic Model Development, 6, 929-960. Salamanca, F., Mar�lli, A., Tewari, M., et al. (2011). A study of the urban boundary layer using different urban parameteriza�ons and high-resolu�on urban canopy parameters with WRF. Journal of Applied Meteorology and Climatology, 10(7), 2801-2831. Stewart, I.D., & Oke, T.R. (2012). Local climate zones for urban temperature studies. Bulle�n of the American Meteorological Society, 93(12), 1879-1900. Wang, R., Ren, C., Xu, Y., et al. (2018). Mapping the local climate zones of urban areas by GIS-based and WUDAPT methods: A case study of Hong Kong. Urban Climate, 24, 567-576. Acknowledgements We also thank all students who have helped with the mapping of building archetypes. The work was supported by a grant from the PROCORE-France/Hong Kong Joint Research Scheme sponsored by the Research Grants Council and the Consulate General of France in Hong Kong (Reference No. F-CUHK403/18), as well as the Vice-Chancellor's Discre�onary Fund of the Chinese University of Hong Kong. * [email protected] Ci�es face a mul�tude of urban climate issues which pose threats to the health and well-being of urban dwellers. Numerical models can simulate the altered surface energy balance due to buildings, impervious surfaces, and human ac�vi�es. They help us be�er understand the physical processes of urban climate and evaluate the poten�al impacts of climate change on ci�es . Hong Kong is an ideal study loca�on because Figure 1: Modelling challenges presented by Hong Kong's natural and urban landscapes (photo source: curiouscatontherun.wordpress.com) Hilly terrain Non-street-canyon Heterogeneous urban morphology High-density, high-rise buildings Ver�cal and volumetric Complex Strong Ac�ve atmospheric coastline convec�on mari�me influence geometry developments ▶ a detailed urban database is available; ▶ the weather sta�on network is well-established; ▶ it is challenging and interes �ng to model (Figure 1). Here, we present the model set-up and input parameters for describing urban land cover and building characteris �cs in prepara�on for the fine-scale simula�on of the thermal microclimate condi�ons in downtown Hong Kong (D4) during a week- long hot spell in September 2009. The non-hydrosta�c mesoscale atmospheric model Meso-NH, coupled with the single-layer urban canopy parametrisa�on Town Energy Balance (TEB) to solve the urban surface energy balance, will be employed for the urban climate simula�on TEB allows one to provide a detailed descrip�on of the street canyon geometry, urban vegeta�on, building construc�on materials, and (Figure 2). human behaviour related to building energy consump�on. ECMWF analysis data will be downscaled via 4 nested models to a spa�al resolu�on of 250 m for downtown Hong Kong (Figure 3). WATER SURFEX TOWN NATURE SEA MESONH Lowest atmospheric level TEB Averaged flux over 4 �les Atmospheric forcing ROAD BLD wall roof HVEG BEM HEIGHT LVEG NVEG GARDEN BLD (Lac et al., 2018) (Masson et al., 2013) (Masson, 2000) (Bueno et al., 2012) road BLD_ WOH = total wall area plan area of TOWN Figure 2: Schema�c representa�on of the MesoNH-SURFEX-TEB coupling and TEB input parameters. D1 D2 D3 D4 Figure 3: The model domains and their spa�al resolu�ons. D1 – The Pearl River Delta (PRD) region (8 km) D3 – Hong Kong and southern Shenzhen (1 km) D4 – Kowloon Peninsula and northern Hong Kong Island D2 – Major metropolitan areas in PRD (2 km) i.e. downtown Hong Kong (250 m) (image source: Google Earth) Precise data on land cover and urban morphology can significantly improve the performance of high-resolu�on model simula�ons (Salamanca et al., 2011). TEB input parameters for D4 are derived at 100 m resolu�on by aggrega�ng administra�ve data on building footprint, road networks, and land u�lisa�on, complemented by Figure 4: Building surface frac�on in D4. Figure 5: Impervious surface frac�on in D4. Figure 6: Urban vegeta�on cover in D4. Figure 7: Average building height (m) in D4. Figure 8: Street canyon aspect ra�o in D4. a digital eleva�on model and vegeta�on cover retrieved from treated satellite images. 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 10 25 50 100 200 404 0 1 3 5 7 9 20 0 Building characteris �cs (e.g. roof albedo, wall thermal conduc�vity, window-to-wall ra�o) and occupant behaviours (e.g. occupancy, cooling schedule and setpoint temperature) influence the urban radia�ve transfer, storage and anthropogenic heat fluxes, and are therefore of high relevance Using an "architect's exper�se" approach, 18 building archetypes are defined for Hong Kong as combina�ons of diff erent building typology, building use, and building age (Kwok et al., 2018; Figure 9). Each building archetype is described in TEB with a set of parameters related to building architecture and human behaviour. Figure 9: Spa�al distribu�on of the dominant building archetype for each urban grid point in D4 and example photos of selected building archetypes. Tong Lau/ Shophouse Public housing Commercial skyscrapers Commercial skyscraper Luxury hotel Shopping mall Old commercial building Industrial building Public housing Tong Lau/ Shophouse Private housing Newer private housing Modern private housing Mid-rise/ Village house School University/ Ins�tu�on Hospital Other public/ communal facility Transport facility Historical building Informal/ Rural infrastructure to city-atmosphere interac�ons. The WUDAPT ini�a�ve aims to provide a worldwide coherent descrip�on of the form and func�on of urban areas (Ching et al., 2018). Urban morphology is therein described through the local climate zone (LCZ) scheme (Stewart and Oke, 2012). The LCZ map of Hong Kong derived by Wang et al. (2018) is used to extrapolate the urban data coverage for D3 by means of a geospa�al analysis of the TEB input parameters in D4. Compared to the 'standard' ranges of surface cover proper�es for each LCZ, the urban areas in Hong Kong generally have higher impervious surface frac�ons and lower pervious surface frac�ons , par�cularly for open se�ngs (Figure 10). Figure 10: Sta�s �cal distribu�ons of building, impervious, and pervious surface frac�ons for the three most common LCZs in downtown Hong Kong compared to the 'standard' ranges given by Stewart and Oke (2012). Building Impervious Pervious 'standard' ranges LCZ scheme median mean 1 st quar�le 3 rd quar�le minimum within 1.5 IQR maximum within 1.5 IQR outliers (> 1.5 IQR) extreme outliers (> 3 IQR) LCZ 1 Compact high-rise LCZ 4 Open high-rise LCZ 5 Open mid-rise Surface frac�on 1.0 0.8 0.6 0.4 0.2 0.0 urban surface cover and morphology, as well as building architecture and human behaviour, employed for the ini�alisa�on of MesoNH-TEB. The comparison of detailed input parameters to 'standard' ranges for each LCZ demonstrates the importance of locally adap�ng relevant parameters for model applica�on. In the next steps, the model configura�on will be op�mised and model results will be evaluated against sta�on observa�ons. Ul�mately, the study aims to aid urban planning by simula�ng the spa�al and temporal varia�ons of the city's urban climate under diff erent development and mi�ga�on scenarios. The novelty of this study lies in the remarkably detailed descrip�on of the The urban data reflect the high building and road density , heterogeneous building height, and complex mix of building archetypes in downtown Hong Kong.
Transcript of Towards a fine-scale urban climate simula on of Hong Kong...
Towards a fine-scale urban climate simula�on of Hong Kong usingMesoNH-TEB with a detailed urban land cover descrip�on
Yu Ting KWOK1*, Cécile DE MUNCK2, Robert SCHOETTER2 1School of Architecture, The Chinese University of Hong Kong; 2CNRM, Université de Toulouse, Météo-France, Toulouse
1. Introduc�on and Objec�ves
2. Model Configura�on and Domains
3. Urban Surfaces and Building Geometry
Refe
renc
es
4. Building Archetypes 5. Extending Data Coverage with LCZs
6. Conclusion and Further Work
Bueno, B., Pigeon, G., Norford, L.K., et al. (2012). Development and evalua�on of a building energy model integrated in the TEB scheme. Geoscien�fic Model Development, 5, 433-448. Ching, J., Mills, G., Bechtel, B., et al. (2018). World Urban Database and Access Portal Tools (WUDAPT), an urban weather, climate and environmental modeling infrastructure for the Anthropocene. Bulle�n of the American Meteorological Society.Kwok, Y. T., Schoe�er, R., & Masson, V. (2018). Defining building archetypes for urban climate simula�ons of the complex high-density environment in Hong Kong. 34th Interna�onal Conference on Passive and Low Energy Architecture, Hong Kong. 3, 982-983.Lac, C., Chaboureau, J., Masson, V. (2018). Overview of the Meso-NH model version 5.4 and its applica�ons. Geoscien�fic Model Development Discussions. Masson, V. (2000). A physically-based scheme for the urban energy budget in atmospheric models. Boundary-Layer Meteorology, 94(3), 357-397.Masson, V., Le Moigne, P., Mar�n, E., et al. (2013). The SURFEXv7. 2 land and ocean surface pla�orm for coupled or offline simula�on of earth surface variables and fluxes. Geoscien�fic Model Development, 6, 929-960.Salamanca, F., Mar�lli, A., Tewari, M., et al. (2011). A study of the urban boundary layer using different urban parameteriza�ons and high-resolu�on urban canopy parameters with WRF. Journal of Applied Meteorology and Climatology, 10(7), 2801-2831. Stewart, I.D., & Oke, T.R. (2012). Local climate zones for urban temperature studies. Bulle�n of the American Meteorological Society, 93(12), 1879-1900.Wang, R., Ren, C., Xu, Y., et al. (2018). Mapping the local climate zones of urban areas by GIS-based and WUDAPT methods: A case study of Hong Kong. Urban Climate, 24, 567-576.
Acknowledgements We also thank all students who have helped with the mapping of building archetypes.The work was supported by a grant from the PROCORE-France/Hong Kong Joint Research Scheme sponsored by the Research Grants Council and the Consulate General of France in Hong Kong (Reference No. F-CUHK403/18), as well as the Vice-Chancellor's Discre�onary Fund of the Chinese University of Hong Kong.
Ci�es face a mul�tude of urban climate issues which pose threats to the health and well-being of urban dwellers.
Numerical models can simulate the altered surface energy balance due to buildings, impervious surfaces, andhuman ac�vi�es. They help us be�er understand the physical processes of urban climate and evaluate thepoten�al impacts of climate change on ci�es.
Hong Kong is an ideal study loca�on because
Figure 1: Modelling challenges presented by Hong Kong's natural and urban landscapes(photo source: curiouscatontherun.wordpress.com)
Hilly terrain
Non-street-canyon
Heterogeneous urban morphologyHigh-density, high-rise buildings
Ver�cal and volumetric
ComplexStrong
Ac�ve atmospheric
coastline
convec�on
mari�meinfluence
geometry
developments▶ a detailed urban database is available;▶ the weather sta�on network is well-established;▶ it is challenging and interes�ng to model (Figure 1).
Figure 1: Modelling challenges presented by Hong Kong's
Non-street-canyon
Here, we present the model set-up andinput parameters for describing urban land
cover and building characteris�cs inprepara�on for the fine-scale simula�on of
the thermal microclimate condi�ons indowntown Hong Kong (D4) during a week-
long hot spell in September 2009.
The non-hydrosta�c mesoscale atmospheric model Meso-NH,coupled with the single-layer urban canopy parametrisa�onTown Energy Balance (TEB) to solve the urban surface energybalance, will be employed for the urban climate simula�on
TEB allows one to provide a detailed descrip�on of the street canyongeometry, urban vegeta�on, building construc�on materials, and
(Figure 2).
human behaviour related to building energy consump�on.
ECMWF analysis data will be downscaled via 4 nested modelsto a spa�al resolu�on of 250 m for downtown Hong Kong(Figure 3).
WATER
SURFEXTOWN
NATURESEA
MESONH Lowest atmospheric level
TEB
Averaged flux over 4 �les Atmospheric forcing
ROADBLDwall
roof HVEG
BEMHEIGHTLVEG NVEG
GARDEN
BLD
(Lac et al., 2018)
(Masson et al., 2013)
(Masson, 2000)
(Bueno et al., 2012)
road
BLD_
WOH = total wall areaplan area of TOWN
Figure 2: Schema�c representa�on of the MesoNH-SURFEX-TEBcoupling and TEB input parameters.
D1
D2
D3
D4
Figure 3: The model domains and their spa�al resolu�ons.D1 – The Pearl River Delta (PRD) region (8 km)
D3 – Hong Kong and southern Shenzhen (1 km)D4 – Kowloon Peninsula and northern Hong Kong Island
D2 – Major metropolitan areas in PRD (2 km)
i.e. downtown Hong Kong (250 m)(image source: Google Earth)
Precise data on land cover and urban morphology can significantly improve the performance of high-resolu�on model simula�ons (Salamanca et al., 2011). TEB input parameters for D4 are derived at 100 m resolu�on by aggrega�ng administra�ve data on building footprint, road networks, and land u�lisa�on, complemented by
Figure 4: Building surface frac�on in D4. Figure 5: Impervious surface frac�on in D4. Figure 6: Urban vegeta�on cover in D4. Figure 7: Average building height (m) in D4. Figure 8: Street canyon aspect ra�o in D4.
a digital eleva�on model and vegeta�on cover retrieved from treated satellite images.
00.20.40.60.81
00.20.40.60.81
00.20.40.60.81
102550100200404
01357920
0
Building characteris�cs (e.g. roof albedo, wall thermal conduc�vity,window-to-wall ra�o) and occupant behaviours (e.g. occupancy, coolingschedule and setpoint temperature) influence the urban radia�ve transfer, storage and anthropogenic heat fluxes, and are therefore of high relevance
Using an "architect's exper�se" approach, 18 building archetypes aredefined for Hong Kong as combina�ons of different building typology,building use, and building age (Kwok et al., 2018; Figure 9).
Each building archetype is described in TEB with a set of parametersrelated to building architecture and human behaviour.
Figure 9: Spa�al distribu�on of the dominant building archetype for each urban grid point in D4and example photos of selected building archetypes.
Tong Lau/ Shophouse
Public housing
Commercial skyscrapersCommercial skyscraperLuxury hotelShopping mallOld commercial buildingIndustrial buildingPublic housingTong Lau/ ShophousePrivate housingNewer private housing
Modern private housingMid-rise/ Village houseSchoolUniversity/ Ins�tu�onHospitalOther public/ communal facilityTransport facilityHistorical buildingInformal/ Rural infrastructure
to city-atmosphere interac�ons.
The WUDAPT ini�a�ve aims to provide a worldwide coherent descrip�on of the form andfunc�on of urban areas (Ching et al., 2018). Urban morphology is therein described throughthe local climate zone (LCZ) scheme (Stewart and Oke, 2012).
The LCZ map of Hong Kong derived by Wang et al. (2018) is used to extrapolate the urbandata coverage for D3 by means of a geospa�al analysis of the TEB input parameters in D4.
Compared to the 'standard' ranges of surface cover proper�es for each LCZ, theurban areas in Hong Kong generally have higher impervious surface frac�onsand lower pervious surface frac�ons, par�cularly for open se�ngs (Figure 10).
Figure 10: Sta�s�cal distribu�ons of building, impervious, and pervious surface frac�ons for the three most common LCZsin downtown Hong Kong compared to the 'standard' ranges given by Stewart and Oke (2012).
Building
Impervious
Pervious
'standard' rangesLCZ scheme
medianmean
1st quar�le
3rd quar�le
minimum within 1.5 IQR
maximum within 1.5 IQRoutliers (> 1.5 IQR)extreme outliers (> 3 IQR)
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Impe
rvio
us s
urfa
ce fr
actio
n
1.00
.80
.60
.40
.20
.00 LCZ 1 Compact high-rise LCZ 4 Open high-rise LCZ 5 Open mid-rise
Surfa
ce fr
ac�
on
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Bui
ldin
g su
rfac
e fr
actio
n
1.00
.80
.60
.40
.20
.00
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Bui
ldin
g su
rfac
e fr
actio
n
1.00
.80
.60
.40
.20
.00
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Bui
ldin
g su
rfac
e fr
actio
n
1.00
.80
.60
.40
.20
.00
1.0
0.8
0.6
0.4
0.2
0.0
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Perv
ious
sur
face
frac
tion
1.00
.80
.60
.40
.20
.00
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Perv
ious
sur
face
frac
tion
1.00
.80
.60
.40
.20
.00
Local climate zone
5 Open mid-rise4 Open high-rise1 Compact high-rise
Perv
ious
sur
face
frac
tion
1.00
.80
.60
.40
.20
.00
urban surface cover and morphology, as well as building architecture andhuman behaviour, employed for the ini�alisa�on of MesoNH-TEB.
The comparison of detailed input parameters to 'standard' ranges for each LCZ demonstrates the importance of locally adap�ng relevant parameters for model applica�on. In the next steps, the model configura�on will be op�mised and model results will beevaluated against sta�on observa�ons. Ul�mately, the study aims to aid urban planning bysimula�ng the spa�al and temporal varia�ons of the city's urban climate under differentdevelopment and mi�ga�on scenarios.
The novelty of this study lies in the remarkably detailed descrip�on of the
The urban data reflect the high building and road density, heterogeneous buildingheight, and complex mix of building archetypes in downtown Hong Kong.
