URBAN CLIMATOLOGY IX. Urban climate modelling Paper to read IntJ Biomeleorol (201 7) 01:527-539 DOI IU.lUU7,iOU4K4-01tl230-z ORIGINAL PAPER Enhancement of urban heat load through social inequalities on an example of a fictional city King's Landing Simulation of the urban climate of an imaginary city as an illustrative example to demonstrate that the residential areas with deprived socio-economic conditions can exhibit an enhanced heot lood at night, and thus more disadvantageous environmental conditions, compared with the areas of higher socioeconomic status. https://is.muni.cz/auth/el/sci/podzim2024/ZA311/um/67875456/09 modelling s00484-016-1230-z.pdf?lans=en 9.1 Modeling urban climate • Our knowledge on urban climates based on empirical studies is rather restricted (Urban) climatologists need some „laboratory" to experiment Models of various types and complexity may be such a lab : - O Radiations, building material, anthropogenic heal o model complexity Urban Climate model types Statistical models - simple to apply, restricted transferability, limited possibility of climate projections Physical models - requires careful design, provides experimental control, allows detailed study of selected features (wind tunnel), need of special facilities (expensive) Numerical models - account for different scales of urban climate, physically correct, however quite complex, provide controlled experiments and climate predictions j>* • -Tv-— ATU_R =f(population) Oke et ai, 2017, Urban Climates © Cambridge University Press 2017 The UrbClim mode! UroClfm output https://urban-climate.be/c/jrbclimDescription/ Hierarchy of numerical climate models Global circulation models (GCM) Earth System Models (ESM) (~100 km), GCM + IPCC scenarios • MUKLIMO_3 (DWD) • ENVI-Met (http://www.envi-met.com/' Urban climate numerical modeling Allows us to examine and test our understanding how cities affect climate and vice versa. Models of varying complexity: 1. Allows controlled experiments by isolating certain processes (effect of buildings on wind field modification, role of sky view factor on radiation budget etc. 2. Provides future climate projections (e.g. number and intensity of heat waves Mesoscale model domain Governing equations - core set of equations that ensures the conservation of momentum, mass and energy Small scale processes are solved through parametrization Boundary conditions defined by higher order models • Micro and local scale models - UCL, RSL • Mesoscale urban models - UBL and surface Oke etal., 2017, Urban Climates © Cambridge University Press 2017 20.11.2024 Model examples Rayman http://www.urbanclimate.net/ravman/ ENVI-Met http://www.envi-met.com/ The Urban Multi-scale Environmental Predictor (UMEP) http://www.urban-climate.net/umep/UMEP ALADIN-Climate SURFEX htt p://www. u m r-cn rm ,f r/su rfex/ Summer mean 2-m temperature CQ for 1991-2000 over Budapest according to the 10 km resolution ALADIN-Climate RCM and 1 km resolution SURFEX results MUKLIMO_3 (DWD) Comprehensive list of sources https://www.urbanclimate.net/E ltools.htm 9.2 Model MUKLIMO_3 • Mikroskaliges Urbanes Klima-AAodell, 3-dim • Developed in DWD (Deutscher Wetterdienst), intensively used in Austria weather service (ZAAAG) • Model simulates atmospheric flow fields in the presence of buildings (air temperature field, relative humidity and 3D field of wind speed an direction) • Horizontal resolution: 100 m, variable vertical resolution 10-100 m 4 Model MUKLIMO_3 - input parameters Model considers several parameters of buildings such as density of buildings, mean height of buildings, friction effects on building surfaces and turbulence generation etc. Precipitation, cloud processes , horizontal runoff and anthropogenic heat production are not considered The vegetation in the canopy model has three vertical layers: tree crown, tree trunk and low vegetation. Typical values of parameters for each LCZ are coded in Land Use Table Model MUKLIMO_3 • The ID model calculates the daily cycle of temperature, relative humidity and wind for the reference station located outside of the urban area 18.7.2015 REAL DATA 2 4 6 8 10 12 14 16 18 2D 22 24 Hour 28.8.2015 REAL DATA 2 4 6 B 10 12 14 16 18 20 22 24 18.7.2015 MODEL DATA 187.2015 DIFFERENCES 28.3.2015 MODEL DATA B 10 12 14 16 18 20 22 24 Hour 28.8.2015 DIFFERENCES Comparison of air temperature daily cycle from measured (left) and simulated (middle) data from five stations in Brno (for 18. 7. and 28. 8. 2015). Figures on the right side show differences between measured _and modelled air temperatures_ Model simulation of air temerature and wind filed, an ideal case, 16 h GMT Model MUKLIMO_3 • The ID simulation is run for 24 h after which the values for air temperature, relative humidity and wind are used to initialize the 3D model taking into account terrain height and soil type. • The meteorological fields given as the output of the 3D model are used for the analysis of the UHI effect and the calculation of climate indices. • Model is used to evaluate particularly the urban heat load in summer period. • For that purpose, the climate indices, such as mean annual number of summer days (Tmax > 25 °C), hot days (Tmax > 30 °C) and tropical nights (Tmin > 20 °C), are calculated. • The climate indices are calculated with the cuboid method. The method enables the calculation of heat load on a longer temporal scale by using a limited number of urban climate model simulations. CUBOID method • Model simulations in MUKLIMO are done only for eight corners of a cuboid. • These corners represent min and max values • Method uses 3D interpolation. (T<.mln.rhcm,n.vcmln) '-"oil/ / to, = 25°C) for three different decades of recent climate in Brno area Concentration - CO -rq. (incl. all forcing agenlsj 1250 9.3 Future climate • Modelling of future climate is based on the use of different scenarios that estimate future level of green house gass concentrations • Representative Concentration Pathways (RCPs) are four greenhouse gas concentration (not emissions) trajectories adopted by the IPCC for its fifth Assessment Report (AR5) in 2014 • They describe four possible climate futures, all of which are considered possible depending on how much greenhouse gases are emitted in the years to come • The four RCPs, RCP2.6, RCP4.5, RCP6, and RCP8.5, are named after a possible range of radiative forcing values in the year 2100 relative to pre-industrial values (+2.6, +4.5, +6.0, and +8.5 W/m2, respectively) Future climate In model AAUKLIAAO outputs from Regional Climate Models instead of contemporary real measurements may be used RCP4.5 a RCP8.5 resulting from the project EURO-CORDEX (Coordinated Downscaling Experiment) -European Domain were used to model the climate of Brno in 21st Century 1971-2000 2021-2050 2071-2100 Mean annual number of summer days (Tmax > 25°C) simulated for RCP8.5 scenario; awearage from an ensemble of eleven regional climate models Brno - no. of hot days in future climate 1981-2010 | 2021-2050 [RCP4.5] 2021-2050 [RCP8.5] 2071-2100 [RCP4.5] 2071-2100 [RCP8.5] Contribution of different LCZs to the increase of no. hot days Effect of trees on SVF and thermal comfort in Prague The PALM model - able to solve turbulence, energy processes and thermal conditions at the street level resolving individual buildings. • The model experiment was carried out by two simulations with: (1) the complete structure of the study domain; (2) the domain without individual shrubs and trees. • The differences in Universal thermal climate index (AUTCI) between these two simulations characterise the effect of trees on the thermal comfort of pedestrians. SIMULATIONS Jankü et ol. 2024 Effect of trees on SVF and thermal comfort in Prague a) >5\V f.-: m Calculated sky view factors (SVF*) for both simulations (a) without trees and (b) including trees, and (c) differences between both simulations (ASVF*), and (d) orthophoto of Prague-Dei vice district _Janku et al. 2024 Effect of trees on SVF and thermal comfort in Prague J Ifangs ÉIÉ1 SÄ- lllptlll Spatial distribution of the UTCI differences caused by tree SVF* reduction (AUTCI) within the study domain as hourly means at (a) 06-07 CEST, (b) 08-09 CEST, (c) 13-14 CEST, (d) 18-19 CEST, (e) 22-23 CEST and (f) dai ly mean. jan^ et a| 2024 Modelling changes in LU/LC distribution and their thermal characteristics response 14.26'E 1* 2 i* L 1*WE Average number of hot days per year in Klagenfurt (Austria) between 1981 and 2010 for simulations considering (a) the initial Urban Atlas map and (b) the proposed land use classification. Panel (c) provides the difference in the simulations in panels (a) and (b). Oswald et ol. (2020) 20.11.2024 Modelling changes in LU/LC distribution and their thermal characteristics response Daily average of Tn [°C] 10.D 12.5 15.0 L7.5 20.0 22.5 25.0 27.S JO.O Daily average of Tn [CTC] 10.0 12.5 15.0 1T.5 20.0 22.5 2S.D 27.5 Future climate projections shown as a probability density function (PDF) of the air temperature (To) taken from the bias-corrected EURO-CORDEX data set for (a) RCP4.5 and (b) RCP8.5 for the extended summer season (MJJAS) in Klagenfurt (Austria) Oswald et al. (2020) 9.4 Final remarks and questions • Models is able to simulate main features of spatial distribution of several climate indices which characterize potential heat load in cities • Parts of the city with the highest heat load correspond with the recent knowledge that is based on real measurements (model validation) • In controlled experiments models are able to quantify the role of selected urban features to heat load • Future climate simulations show significant increase of heat load especially for RCP8.5 at the end of the 21st century 12 20.11.2024 Final remarks and questions 1. What is the main purpose of urban climate models? 2. What aspects of urban climate would be useful to simulate? 3. Is there any other method how to do projections of future climate? 4. What is a difference between "projection" and "prediction"? 13