Session Overview
NOMTM1: Urban Canopy parameterizations I : Urban vegetation
Tuesday, 21/Jul/2015:
4:30pm - 6:15pm

Session Chair: Eric Scott Krayenhoff, University of British Columbia
Location: St-Exupéry Amphitheater


A multi-layer urban canopy model for neighbourhoods with trees

E. Scott Krayenhoff1, Andreas Christen1, Alberto Martilli2, Timothy R. Oke1

1University of British Columbia, Canada; 2CIEMAT, Spain

Addition of tall vegetation is a key design strategy for moderation of local urban climate, and many cities already boast extensive tree cover. Relative to shorter vegetation, urban trees have unique micrometeorological and climatic effects: they provide shade and shelter, interacting with buildings and streets to alter local climate and wind flow, affecting thermal comfort of residents, building energy demand, and pollutant concentrations in the canopy.

Urban canopy models (UCMs) predict neighbourhood-scale energy exchange and climate of the atmospheric layer between the buildings. Few UCMs represent the urban canopy with multiple layers, which permit more flexible and process-based representation of canopy physics. Most UCMs neglect vegetation, or incorporate its effects with a separate model, neglecting direct interaction between vegetation and built elements in the canopy.

We present BEP-Tree, the first multi-layer urban canopy model that explicitly includes trees and their interaction with buildings. It consists of an existing multi-layer UCM, a foliage energy balance model, and two major developments: firstly, a model that distributes solar and infrared radiation amongst tree foliage, road, roof, and wall elements at multiple heights, accounting for radiation ‘trapping’ and mutual shading; secondly, parameterization of building and tree foliage effects on flow, including generation and dissipation of turbulence, drag on the mean wind, and explicit consideration of sheltering. The combined model permits a wide range of building and tree configurations, and makes possible advanced assessment of impacts of trees on urban climate, air quality, human comfort and building energy loads.

BEP-Tree is compared with measurements from the Sunset neighbourhood in Vancouver, Canada. Urban trees principally channel sensible heat into latent heat (evaporation), shift surface-atmosphere energy exchange upwards, slow canopy wind, and dissipate turbulence more rapidly, especially if taller than nearby buildings. Effects of trees on neighbourhood-average canopy thermal climates are less clear; foliage clumping at the neighbourhood scale must be quantified with more fidelity, and we suggest this as an important future development for UCMs.



Estatio Gutierrez1, Jorge Gonzalez1, Alberto Martilli2, Mark Arend1, Robert Bornstein3

1City College of New York, United States of America; 2CIEMAT, Spain; 3San Jose State U., United States of America

There is a large need to better understand and quantify surface latent/sensible heat fluxes partitions from urban environments and their specific nature, whether natural or anthropogenic. In this research, a hydrology parameterization was implemented in the multilayer urban canopy model (BEP) of the Weather Research Forecasting (WRF) model to quantify latent heat fluxes contributions from impervious surface evaporation at ground and roof levels during and after precipitation events, and evapotranspiration from street vegetation in the surface energy balance. A cooling tower scheme was also incorporated in the building energy model (BEM) to represent the anthropogenic latent heat released by this type of technology for commercial buildings. The new modeling schemes for latent heat flux was tested in New York City (NYC) during summer seasons. City Planning data for buildings were assimilated at 250 meters to improve the representation of the city morphology.

An evaluation with heat flux measurements from a limited field campaign in nearby Baltimore, Maryland, indicates that the new formulation properly represents sensible and latent heat daily cycle particularly during rainy days. A comparison with surface weather station data in NYC for summer 2010 shows overall seasonal improvement in the foresting when the new surface heat fluxes parameterizations are introduced, with small biases to overestimate daily maximum temperatures and underestimate moisture content at nighttimes. The hydrology scheme introduces a slightly higher amount of sensible heat in the late afternoon and night in dry days. Latent heat produced by surface evaporation during rain events exceeds amount of heat flux produced by evapotranspiration. Evaporative cooling technology from buildings diminishes between 80 and 90% the amount of sensible heat which is transformed into latent heat in commercial areas. Streets constitute the main source of sensible and latent heat in residential areas. The increase in environmental moisture content from streets and roofs influences the lower boundary layer leading to modifications of the atmospheric stability in commercial areas. The new surface heat flux parameterization allows for the effective exploration of mitigating alternatives to environmental impacts of urban fluxes. Exploration of green and white roofs is addressed for the case of NYC.

The influence of tree crowns on urban thermal effective anisotropy

Daniel Dyce, James Voogt

Western University, Canada

The three-dimensional surface structure of urban areas imparts an anisotropic character to the upwelling longwave radiation. Trees contribute to the three-dimensionality of the urban surface and are therefore expected to influence urban thermal effective anisotropy. In this presentation, the SUM sensor view model (Soux et al. 2004) is modified to include trees in order to assess their contribution to the anisotropy. A gap probability approach to estimate foliage view factors is used so that tree canopies are represented as volumes of turbid media and a simple energy budget model for leaf surface temperatures is added. The vegetated model (SUMVEG) is tested against airborne thermal observations from a treed residential study area in Vancouver, BC. The inclusion of trees in the model results in an approximate doubling of the effective thermal anisotropy near midday with a smaller increase in the morning. The vegetated model shows a statistically significant improvement over the original model.

The model is then used to assess the impact of trees on the thermal anisotropy of a range of urban geometries. A simple array of cubic elements is used to represent buildings and identically sized tree crowns are located a fixed distance from the buildings. Temperatures for buildings are estimated from the TUF3d canyon model of Krayenhoff & Voogt (2007). The temperature for surface patches that are partially shaded by tree crowns are estimated by weighting between a fully sunlit and fully shaded TUF3d modeled temperature based on the probability of gap for direct and diffuse shortwave radiation through the shading tree crown. Tests are made for tree heights both less than and greater than that of the buildings.

Results show that trees both increase and decrease anisotropy as a function of tree crown plan area fraction and building plan area fraction. In open urban geometries, the shadows cast by trees act to increase the urban thermal anisotropy while in more compact geometries trees reduce anisotropy as the plan area of tree crowns increases. Many urban geometries have a critical value at which the impact of tree coverage on the effective anisotropy reverses. Increases of tree coverage where the trees are taller than the buildings act to increase anisotropy for all plan area geometries.


VTUF: An urban micro-climate model to assess temperature moderation from increased vegetation and water in urban canyons

Kerry Nice1,2, Nigel Tapper1,2, Jason Beringer1,3, Andew Coutts1,2, Scott Krayenhoff4

1Monash University, Australia; 2CRC for Water Sensitive Cities, Australia; 3University of Western Australia; 4University of British Columbia

With urban areas facing future longer duration heatwaves and temperature extremes, adaptation strategies are needed. Examining the role that increased tree cover and water availability can have on human thermal comfort (HTC) in urban areas as part of these strategies has been done using observations, but further work requires a modelling tool suited for this task. Sufficient model resolution is needed to resolve variables used to calculate HTC as well as the ability to model the physiological processes of vegetation and their interaction with water. The lack of such a tool has been identified as a research gap in the urban climate area and has impaired our ability to fully examine the use of vegetation and water for improved ​human thermal comfort.

A new model, VTUF​ (Vegetated Temperatures​ Of Urban Facets), addresses this gap by embedding ​the ​functionality of the MAESPA tree process model (Duursma & Medlyn 2012), ​that can model individual trees, vegetation, and soil components, within the TUF-3D (Krayenhoff & Voogt 2007) urban micro-climate model. ​An innovative tiling approach, allows the new model to account for ​important vegetative physiological processes and shading effects. It also resolves​ processes​ at sufficient​ly high​ resolution to calculate HTC ​and air and surface temperature, humidity, and wind speed across an urban canyon.

Model validations have shown performance improvements of the model and a suitability to use it to examine ​critical questions relating to the role of vegetation and water in the urban environment. Analysis using this model includes scenarios quantifying the impact each individual tree can have on temperatures in urban canyons as well the optimal arrangement and quantity of trees to maximize temperature moderation effects.

Street trees in the urban canopy model TEB

Emilie Redon1, Aude Lemonsu2, Marjorie Musy3, Valéry Masson2, Cécile De Munck4, Katia Chancibault5

1Universite de Toulouse, UPS, Meteo France, CNRM GAME, France; 2Meteo France / CNRS, CNRM GAME, France; 3Ecole Nationale Supérieure d’Architecture de Nantes, CERMA, France; 4INRA, Centre de Bordeaux, France; 5Institut Français des Sciences et Technologies des Transports, de l’Aménagement et des Réseaux, France

Urban vegetation can influence urban climate to the scale of a whole city by its radiative, aerodynamic, thermal and moisture properties. In recent years, the urban canopy model TEB1 (Masson 2000; Hamdi and Masson 2008) has been improved to explicitly represent vegetation in urban areas, especially small-scale interactions between mineral surfaces, vegetation and atmosphere. The TEB model, in the TEB-Veg version, now includes parameterizations dedicated to low vegetation in urban canyons (Lemonsu et al. 2012), vegetated roofs (de Munck et al. 2013) which were each evaluated from experimental data point, and a more realistic description of urban basement.

Urban trees in particular constitute an alternative technique in the mitigation of Urban Heat Island (UHI) and improvement of local thermal comfort thanks to shading and shelter effects. However, few urban canopy models are currently taken into account vegetation, even less tree layer (Lee and Park 2008; Lee 2011; Krayenhoff et al. 2013), which alter radiative and energy balances within the urban canyon by intercepting and absorbing incident radiation, shadowing and increasing relative humidity of the air by evapotranspiration. They also have an impact on the local dynamics by modifying the air flow in the street.

In this work, we attempt to set up in TEB-Veg model a parameterization including the following key processes : (1) shading of tree crowns on ground vegetation, walls or mineralized surfaces like roads and (2) summed evaporation of bare ground and transpiration of leaves (both lawns and trees). For this purpose, we employ the Soil Biosphere Atmosphere Interaction (ISBA) model to represent all natural covers (high and low vegetation strata).

Modelling the impact of green infrastructures on local microclimate within an idealized homogeneous urban canopy

Richard Tavares1,3, Isabelle Calmet1,2, Sylvain Dupont4,5

1LUNAM, CNRS, LHEEA UMR CNRS 6598, Nantes, France; 2LUNAM, Ecole Centrale de Nantes, LHEEA UMR CNRS 6598, Nantes, France; 3IRSTV, FR CNRS2488, Nantes, France; 4INRA, UMR 1391, ISPA, F-33140 Villenave d’Ornon, France; 5Bordeaux Sciences Agro, UMR 1391, ISPA, F-33170 Gradignan, France

Quantifying the impact of greening urban areas on urban meteorology, and therefore on human thermal comfort, is difficult due to the complex interaction between the energy fluxes coming from the various urban surfaces. Models give the opportunity to investigate this impact from local to regional scales. However, existing urban canopy models either represent the canopy through only one layer for city to regional scale applications or describe in detail buildings for only local scale applications. Single-layer canopy models may miss important mechanisms occurring within the canopy while detailed canopy models are too time consuming to be applied over several districts (neighbourhood scale). To overcome these limitations, we propose an intermediate approach with a multi-layer canopy model where meteorological fields are solved within and above the canopy and where the canopy roughness elements (building, vegetation) are represented through a drag approach. The model is further able to account for green infrastructures, such as, vegetation on building roofs and walls. The complete model, called ARPS-VUC, is thus applicable from neighbourhood to regional scale.

ARPS-VUC is used here to evaluate the impact of vegetation on the local microclimate of an idealized homogeneous urban canopy. To that purpose, we defined different canopy configurations with various degree of revegetation (small and high vegetation, green facades and roofs). Complete diurnal cycles were simulated for each configuration with meteorological conditions corresponding to a sunny summer day. Simulations show that the cooling effect of vegetation changes in amplitude and during the day following the type of vegetation.

Within the canopy, simulations reveal that trees induce a larger air cooling during daytime (-2 ºC) than small vegetation on ground (-1 ºC) and on building wall surfaces (-2 ºC). This is explained by the tree shadowing effect. This cooling is accentuated in presence of both trees and small vegetation on ground and building surfaces (-3 ºC). These first results are in agreement with recent studies (e.g., Leuzinger et al. 2010, Lindberg and Grimmond 2011, Hall et al. 2012).They further demonstrate the potentiality of our model to evaluate the impact of vegetation on local microclimate.

Hall J., J.F. Hanley and Ennos A. R. (2012) The potential of tree planting to climate-proof high density residential areas in Manchester, UK. Landscape and Urban Planning, 104, 410– 417.

Lindberg, F. and Grimmond, C.S.B. (2011) The influence of vegetation and building morphology on shadow patterns and mean radiant temperatures in urban areas: model development and evaluation. Theoretical and Applied Climatology, 105 (3-4). pp. 311-323. ISSN 1434-4483 doi: 10.1007/s00704-010-0382-8

Luzinger S., Vogt R. and Korner (2010) Tree surface temperature in an urban environment, Agricultural and Forest Meteorology 150, 56–62.


Modelling Radiative Exchange in a Vegetated Urban Street Canyon Model

Duick T. Young1, Peter A. Clark1, Maggie A. Hendry2, Janet F. Barlow1

1Department of Meteorology, University of Reading, United Kingdom; 2Met Office, Exeter, United Kingdom

One of the key findings of the recent international urban land surface model intercomparison (PILPS-urban) was that models do not capture the magnitude and temporal variability of the latent heat flux relative to observations. This is despite many of the schemes including a vegetation component, which is typically represented by separate vegetation tiles or, in a limited number of models, explicitly within the urban scheme. The inability to reproduce the latent heat flux suggests that many schemes do not accurately represent urban vegetation and do not account for the impact of urban surfaces on vegetation physiology. PILPS-urban did however suggest that there was an advantage in using an integrated vegetation scheme as the range in performances of models using the separate tile was larger. This raises the following question; can we improve model accuracy of urban moisture fluxes by including vegetation explicitly within an urban land surface scheme? To address this question an integrated vegetation scheme is being developed for the Met Office – Reading Urban Surface Scheme (MORUSES) to explicitly include vegetation in the form of urban trees and natural surfaces (e.g. grass). The new vegetation scheme is tested within a 2D infinitely long street canyon, with the aim of improving urban weather forecasts and to provide a tool to test the mitigation of extreme heat events through urban greening. This study presents the theory and initial results for the first aspect of the new scheme, radiative exchange within a vegetated urban street canyon. An analytical method was developed based on existing relations and applied to determine the viewfactors for calculation of the longwave radiation budget between the surfaces within a non-turbulent street canyon, with a range of aspect ratios, containing a representation of an urban street tree. Unlike previous methods for modelling radiative exchange, which often assume that the wall and road surface have the same equilibrium temperature, this work investigates the non-trivial radiative exchange problem of vegetated (tree and grass) and urban surfaces that are likely not to be in equilibrium due to the impact of vegetation physiology on canopy temperature.