Renewable and Sustainable Energy Reviews 113 (2019) 109255 Contents lists available at ScienceDirect Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser A digital workflow to quantify regenerative urban design in the context of a changing climate T Emanuele Nabonia,∗∗, Jonathan Natanianb,∗, Giambattista Brizzia, Pietro Floriod, Ata Chokhachianb, Theodoros Galanose, Parag Rastogif a Institute of Architecture and Technology, The Royal Danish Academy of Fine Arts, Schools of Architecture, Design and Conservation, 1425, Copenhagen, Denmark Chair of Building Technology and Climate Responsive Design, Dept. of Architecture, Technical University of Munich, Arcisstraße 21 80333, Munich, Germany d Solar Energy and Building Physics Laboratory (LESO-PB), École Polytechnique Fédérale de Lausanne (EPFL), LE 0 02 (Bâtiment LE) Station 18, CH-1015, Lausanne, VD, Switzerland e Austrian Institute of Technology, Australia f Arbnco Ltd., Glasgow, UK b A R TICL E INFO A BSTR A CT Keywords: Regenerative design Climate change Outdoor microclimate Energy use Renewable energy Daylight Biophilia Parametric design The regenerative approach to design goes beyond limiting the environmental impact of the built environment and towards the enrichment of the ecosystem, adaptation to climate change, and the improvement of human health. This concept is being applied to buildings through new standards such as the Living Building Challenge, yet examples of implementation of regenerative design at the urban scale are rare. While this is a promising direction for sustainable design, in theory new metrics, design tools and workflows need to be developed to translate regenerative design concepts into practice effectively. Among other factors, barriers to implementation remain rooted in the shortcomings of existing urban simulation tools to evaluate a wide range of performance metrics simultaneously. This paper thus proposes a prototype workflow to evaluate regenerative performance using existing evaluation tools in a single digital workflow. A series of existing and customised plugins, most of which are already in use and open source, were integrated into a multi-parametric workflow based on the Grasshopper visual programming tool. The workflow was tested on Malaga as a case study and incorporated key performance indicators related to outdoor human thermal comfort, biophilia, daylight performance, and energy use and production, based on data exchange and synergies across the different tools. These indicators were evaluated for present and future climate scenarios obtained from a weather generator. This paper demonstrates the potential of this workflow to receive visual feedback on various aspects of regenerative urban design, thus enabling designers to more effectively pursue an evidence-based urban design process. 1. Introduction Cities are critical to modern life: approximately 55% of the world's population currently lives in urban areas and on average cities contribute more than 70% of most countries' Gross Domestic Products (GDP) [1]. The number of urban dwellers is projected to grow by 68% by 2050 [2]. Consequently, the built environment will continue to be a significant contributor to global greenhouse gas emissions, accounting for 50–60% of total emissions [3], 71–76% of emissions from final energy use [4] and consuming about 75% of global primary energy. At the moment, most urban dwellers (about 2 billion people) live in cities that have a high risk of mortality and economic losses associated with natural disasters [5], and extreme events disproportionately affect the ∗ approximately 1 billion “disadvantaged and vulnerable populations” [4]. Thus, while cities continue to grow in importance and size, improving living standards and prosperity [1], they remain inequitable, vulnerable, and polluted. This combination of population growth, increasing energy consumption and pollution, is expected to result in higher environmental stress on urban infrastructure [6,7]. These phenomena include the effect of both systematic changes in climate, such as warmer summers on average, as well as increased probability and severity of extreme events such as heat waves. For example, warmer days and nights along with higher air pollution levels can contribute to general discomfort, respiratory difficulties and illness, heat exhaustion, and heat-related mortality due to lack of night ventilation and trapped heat in the street canyons. Therefore, given the high Corresponding author. Corresponding author. E-mail address: jonathan.natanian@tum.de (J. Natanian). ∗∗ https://doi.org/10.1016/j.rser.2019.109255 Received 10 February 2019; Received in revised form 28 June 2019; Accepted 28 June 2019 1364-0321/ © 2019 Elsevier Ltd. All rights reserved. Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. proportion of total energy consumed, high global urbanisation rates, and the vulnerability of cities to climate change, the urban built environment is coming to be seen as both part of the problem as well as the solution. This indicates that the responsive design of cities needs to go beyond climate change mitigation, i.e., reducing greenhouse gas emissions by reducing energy intensity, to adaptation, by which cities will be transformed into catalysts of both health and wellbeing. This regenerative approach will enhance the creation of a thriving ecosystem system by focusing on the improvement of the interdependent relationships between humans and the built and natural environments. In terms of the regenerative built environment, this approach will require a shift away from the narrow focus on building energy performance, mitigation strategies, and minimisation of environmental impacts to a broader framework that enriches places and their inhabitants, ecology, and culture, and makes cities resilient to climate change and changing human needs. Although many professionals declare sustainability as one of their primary drivers, finding examples, where regenerative design has been implemented at the urban level, is difficult. Emerging codes and standards that incorporate a regenerative approach to building design include the Living Building Challenge, WELL building standard and One Planet Living. However, similar metrics do not exist for urban design, and rules defined for buildings cannot always be successfully scaled up to the urban scale, leading to a lack of Key Performance Indicators (KPIs) for urban regenerative design. This article thus responds to these concerns in a workflow incorporating KPIs based on the three pillars of regenerative design in practice proposed in Ref. [8]: Climate and Energy, Ecology and Carbon, and Human Wellbeing and Health. This paper begins by describing the overall concept of urban regenerative design. This is followed by a description of the areas of regenerative design and definition of KPIs that help address the challenges associated with its implementation. These metrics are computed for both historical and predicted future typical years. As the tools currently available for these evaluations do not cover all the KPIs defined here, a new and multi-layered parametric workflow is offered based on data exchange and synergies across the tools. The paper thus tests the potential of combining multi-domain objectives in design, focusing on the climatic and urban context of Malaga, Spain. The final sections discuss the capabilities, application potentials, as well as the limitations of the proposed workflow. and building codes, including energy codes. Whereas zoning and energy codes address several concerns related to health and sustainability, they have been used to separate the design of cities into two disconnected scales: a single building and its urban surroundings at different scales. At the city scale, insufficient attention has been devoted to the creation of pleasant microclimates and of the integration of vegetation and nature. At the building scale, the influence of urban microclimate on the indoor conditions has been ignored. Furthermore, a focus on energy optimisation has diverted attention away from the quality of the indoor environment, usually addressed with simplistic rules of thumb. This section presents a review of existing knowledge about the extents to which these approaches have long-lasting consequences for human health, sustainability, and climatic resilience [9]. 1.2.1. Climate change and urban heat islands The relationship between humans, buildings, urban microclimates and the global climate is complex. Urbanisation, particularly the densification of built spaces and populations, has given rise to the phenomenon of Urban Heat Islands (UHI). Furthermore, changes in landuse due to urbanisation are one of the two primary sources of anthropogenic effects on climate, the other being greenhouse gas emissions. For example, a study of surface temperatures over the continental United States found that half of the observed decrease in diurnal temperature range is due to urban and other land-use changes”, and that 0.27 °C of “mean surface warming per century [was] due to land-use changes [9]. Further, warmer summer temperatures due to climate change may exacerbate the effects of UHI. State of the art in building energy performance is to use a typical sample of weather based on historical records from the nearest weather station, which is often peri-urban or rural. Several recent studies have demonstrated the unsuitability of using only historical records for evaluating long-term design choices [10]. UHI may worsen indoor conditions or cause a further increase in energy demand, particularly for cooling and ventilation [12]. For example, Mauree et al. [10] estimated that a university campus on the shores of Lake Geneva could see an increase in cooling demand of 30% and 52% in 2069 and 2099, respectively. Oke identified three main causes of microclimatic change in cities [11]: (1) interception of short- and long-wave radiation by buildings, (2) reduced long-wave heat emissions due to reduced sky visibility, and (3) increased storage of sensible heat in buildings. A dense building stock also lowers mean wind speed, resulting in even higher mean air temperatures. Several strategies can be adopted to mitigate and adapt to UHI in a changing climate, promoting outdoor social life and potentially reducing building energy demand. For this, accounting for both climatic changes and local thermal characteristics is crucial in policies that address urban morphology, choice of materials and human activity. 1.1. Objectives This study aims to provide a strategy and tools for implementation of urban regenerative design in practice. This is broken down into five objectives: The first objective is to elaborate an urban regenerative design framework that addresses the priority areas of climate and energy, ecology and carbon, and human wellbeing identified above; the second is to establish, within these pillars, specific, quantifiable performative criteria for designers. These criteria are: building energy efficiency and solar electricity generation potential from Building Integrated Photovoltaics (BIPV) (energy and renewables), indoor access to daylight (daylighting), nature-connectedness (biophilia), and the impact of urban microclimatic conditions on outdoor comfort (microclimate and outdoor comfort). The third objective is to establish a set of KPIs that are understood by practitioners and reflect the shift from conventional metrics which are unfit to quantify regenerative design; the fourth objective is to create a digital workflow to calculate these KPIs, using scripts and tools that are increasingly used in practice, and may be customised to interlink various domains. The final objective is to apply the workflow to a real case study in Malaga to gain insights on limitations and opportunities for further developments. 1.2.2. Energy demand and renewable energy While certifications and standards for buildings, e.g., the European Union's Nearly Zero Energy Buildings (nZEBs) standards [12], promote renewables, the optimisation of energy demand is usually conducted in isolation without consideration of the urban environment or interactions with the surroundings. Yet, BIPV could potentially cover between 15% and 60% of the total electricity demand [13], and more than half of the global PV capacity needed to reach the 2050 goals may be installed on buildings [14]. These trends are encouraged by a continuous reduction in the price of solar technologies [15], increased efficiencies of new technologies [15], and the many environmental and economic benefits of distributed energy generation networks in comparison to centralised networks [16]. However, the pursuit of maximum solar power generation may conflict with other design intentions such as density or compactness due to shading or lower amount of exposed façade surfaces. Façademounted photovoltaics may also result in smaller window areas, which in turn may affect daylight levels and visual comfort. Thus, the trade- 1.2. Background The practice of urban and building design is shaped by city zoning 2 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. overall positive attitude towards the environment [27,28]. Along with advantages such as flood protection, noise reduction, visual impact attenuation and biodiversity [29], trees improve air quality [30,31] and outdoor hygro-thermal comfort [32]. Urban green spaces may also positively influence psychological wellbeing through proximity, encouraging exercise, and direct visual contact [33]. People with access to green spaces feel more relaxed and positive [34,35], and have a lower risk of cardiovascular disease [36]. The density and height of the urban fabric may block access to the sky vault, another connection with nature that is essential for human chronobiologic balance [37]. Space openness, in combination with natural elements, may reduce anxiety and impulsivity [38], while ensuring mitigation of the UHI effect [39,40]. Subjective assessment techniques exist to infer these feelings from surveys [41,42]. However, it is more useful for the urban design process to be able to estimate nature-connectedness from information on the distribution and attributes of vegetation, which stands for the main natural feature available in cities. Indicators have been proposed to correlate human health and wellbeing with urban greenery. For example, the Normalized Difference Vegetation Index (NDVI), the amount of near-infrared radiation reflected from green areas into airborne detectors, is correlated with enhanced wellbeing at the broad territorial scale [43]. Other useful indicators adapted to the regional scale are proximity to vegetation [44,45] and visual contact [46], but these are seldom considered in combination with wellbeing. Finally, the sky view factor, which refers to the sky seen by surfaces, is relevant for the assessment of UHI, whereas the sky exposure factor, refers to the sky seen by a point in space, is more significant in terms of psychological impact on personal wellbeing [47]. offs between energy demand and supply should be another critical reference in an environmental urban design workflow. This trade-off is usually expressed as the fraction of energy demand covered by renewables, coupled with the system payback period or the Levelized Cost of Energy (LCOE) throughout the building lifecycle. 1.2.3. Outdoor comfort Any approach to urban design must address the reality of an indoordominant modern lifestyle and proactively work to make outdoor public spaces as comfortable and accessible as possible. To improve the outdoor urban thermal conditions, pedestrian thermal comfort must be calculated, precisely and methodically. However, outdoor comfort is often neglected in the conventional practice of urban design [17] since, compared to indoor spaces, outdoor comfort is quite difficult to estimate. This is primarily due to the higher spatial and temporal variability of important microclimatic variables such as short- and long-wave thermal radiation, wind speed and direction, air temperature and humidity [18], and the complexity of the physical, physiological, or psychological processes triggered to adapt to these outdoor conditions [19,20]. To account for outdoor comfort, the most complete used metric is the Universal Thermal Climate Index (UTCI) [21], which provides a bio-meteorological assessment of the outdoor thermal environment based on the dynamic physiological response predicted by a model of human thermoregulation, coupled with an adaptive clothing model. The UTCI “equivalent temperature for a given combination of wind speed, radiation, humidity and air temperature is … the air temperature of the reference environment, which … produces the same response index value”, i.e., “… an equivalent dynamic physiological response.”. A UTCI assessment scale relates the UTCI range to different stress categories. In this paper, an application of UTCI in the regenerative urban design workflow is shown. 1.2.6. A framework for digital integration Urban modelling has often concentrated on isolated sustainable performance metrics [48]. Renewed interest in the impact assessment of urban form, building types, and overall design on energy demand and supply has generated a variety of methods supported by different tools in a range of resolution levels and capabilities [49,50]. This shift to the district and city scales is also reflected in the evolution of energy modelling tools – from Building Energy Modelling (BEM) to Urban Building Energy Modelling (UBEM) such as CitySim, UMI, CEA, and Ladybug Tools [51]. CitySim [52] can quantify the energy demand at the urban scale as well as the outdoor radiative environment, with high spatial resolution. The Urban Modelling Interface (UMI) [48], a Rhinobased urban modelling design tool, calculates operational energy use, embodied energy, daylighting and walkability at the urban scale. The City Energy Analyst (CEA) [53] was developed to evaluate urban energy use and energy system rationalisation. It was created to analyse different urban scenarios using energy, carbon emissions, and financial criteria, for which a Grasshopper interface is currently under development. The coupling of simulation engines for detailed and multi variable urban environmental performance analysis has been explored by a few studies; Mauree et al. [10] coupled a meteorological model (the Canopy Interface Model) with CitySim, and some studies have linked the Urban Weather Generator (UWG) [54] with energy simulation tools to account for the impact of urban microclimates on energy cooling demand and energy usage intensity [55,56]. Other examples of integration at the urban scale are the works of Natanian and Auer [57], who coupled Radiance and EnergyPlus in a Grasshopper parametric workflow to evaluate daylight potential, energy demand, and solar energy potential in numerous urban block scenarios. Nault et al. [58] coupled EnergyPlus and Radiance/Daysim to develop a predictive model for urban solar potential. Finally, Perini et al. [59] studied outdoor thermal comfort and energy by coupling Envi_Met with TRNSYS to account for radiant temperature in an urban canyon outdoor environment. Ladybug tools in grasshopper potentially offer added benefits to this coupling, which are yet to be fully explored. Mackey et al. [60,61] demonstrated 1.2.4. Daylight in urban residential buildings Implementing design and evaluation tools in urban design is important since decisions made at this scale, such as massing proportions and window to wall ratios impact the solar and daylight potential of the individual buildings [22]. Designing to maximise daylight area and access to natural light is one of the main ways of increasing human health and wellbeing in buildings [23]. Daylight availability has served as the main representative metric to evaluate daylight performance through computational modelling. In the case of residential buildings, one of the most commonly used metrics is the daylight factor, although it has recently been superseded by dynamic indices to represent annual conditions. Several dynamic performance metrics for daylight exist to evaluate the design of nonresidential buildings, e.g. Continuous Daylight Autonomy (CDA), Useful Daylighting Illuminance (UDI) or Spatial Daylight Autonomy (sDA). While these metrics have been widely used, they have several shortcomings: they do not distinguish between high and low illuminance ranges; they do not help evaluate direct sunlight except for detecting the potential for glare; and do not integrate the effects of seasonal daylight availability. In addition, multifunctional spaces in residential buildings with overlapping and diverse activities and social and cultural factors make it difficult to generate accurate schedules of occupancy for daylight analysis [24]. Despite these limitations, this work does not propose a new metric and uses the sDA metric, widely used in previous studies, because it can be used to evaluate daylight performance using one indicative number. 1.2.5. Wellbeing and contact with nature Improving urban inhabitants’ access to, and connection with nature is important for the “urban agenda” of this century [25]. The essential elements of direct biophilic experience are light, air, water, plants, animals, weather, natural landscapes, and fire [26]. An increased connection with nature may improve concentration and create an 3 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. a Ladybug-based workflow in which outdoor comfort was evaluated using microclimatic and energy modelling with OpenFOAM and EnergyPlus, respectively. In this paper, this workflow has been extended with several other algorithms that support the development of regenerative features. The next section introduces the regenerative design workflow, describes its components and exemplifies its application on an urban block in Malaga, Spain. modelling UHI, based on neighbourhood-scale energy balances. Processing the ‘future’ weather time series created by the synthetic weather generator through UWG is necessary to account for urban effects since the synthetic generator used to develop future typical files, Indra [62–64], works with weather data available from a weather station only and does not account for spatial effects. 2.1.3. Outdoor microclimate A suitable comfort model may be selected by addressing the following three aspects: (1) climatic conditions, (2) urban environment, and (3) physical characteristics of the pedestrians. The UTCI metric was selected here due to its capability to account for these three aspects, as well as due to its integration in the Ladybug toolkit [65,66]. The calculation of UTCI relies strongly on the accuracy of the outdoor Mean Radiant Temperature (MRT) calculation [60] since outdoor thermal comfort is linked to the indoor thermal behaviour of buildings (Fig. 4). The calculation first estimates longwave radiation based on surface temperatures using the EnergyPlus simulation engine. During this step, the hourly thermal demand of the buildings overlooking the given street canyon is also estimated. View factors for every surface are calculated using the ray-tracing capabilities of Radiance [67].The temperature of each building surface viewed from the face of a target point in the outdoor space is calculated as a weighted average temperature, where the weight is defined by the area fraction of surrounding surfaces, visible from that given point. Calculations also consider the sky temperature and the consequent radiative loss to the sky. The calculated longwave contribution is then augmented with the shortwave solar radiation that falls on people using the SolarCal model, which is a part of ASHRAE-55 thermal comfort standard [68]. In this respect, the model provides both the hourly dynamic thermal load of the buildings overlooking the canyon and the outdoor thermal performance measured with the Universal Temperature Climate Index (UTCI). In comparison to other existing approaches, the workflow proposed in this paper offers advantages by allowing for the complete modelling of both indoor and outdoor thermal fields in a continuous and contiguous manner, and in Grasshopper, a robust parametric design optimisation environment during early stages of urban design. 2. Methodology Naboni and Lavinga [8] have discussed a series of methods and tools to quantify regenerative design (Fig. 1). The idea is that regenerative design needs to be quantifiable using clear metrics to measure the impact of strategies and solutions. This suggests the creation of reasonable future climate estimates and KPIs for Urban Heat Island, outdoor comfort, energy, daylighting and biophilia. In this paper, these KPIs are translated and integrated into a digital workflow that responds to the needs of researchers and practitioners that are accustomed to digital processes supported by algorithmic and parametric design. 2.1. The urban regenerative design digital workflow The proposed conceptual workflow brings together many disciplines and spatial and temporal scales in a Grasshopper-based workflow. A toolchain is developed by assembling separate scripts handling: (1) the generation of future and urban weather files, (2) the calculation of UTCI, (3) the calculation of building energy demand and supply, (4) the simulation of indoor daylight, (5) the prediction of nature and sky view for pedestrians. The digital workflow proposed in this paper is summarised in Fig. 2. It is designed to leverage the synergies and interdependencies between the different aspects of urban and building performance evaluations. Fig. 3 names the specific plugins and explains the set of algorithms that are used by the tools. For example, Honeybee's dynamic thermal calculation is used for the determination of building energy demand as well as for simulating building surface temperatures that affect the shortwave heat exchanges and, consequently, the UTCI. The following sections describe the workflow components in detail. 2.1.4. Building performance and renewables The energy performance of the buildings was calculated using EnergyPlus [69] via the Grasshopper plugin Honeybee. Following the thermal zoning method, which was used in other district-scale energy assessment workflows and guidelines [48,57], this study relied on an automated division of each building to thermal zones. The first step is to divide each building mass into floors according to predefined floor height, and then to isolate each level into core and perimeter zones. This division enables the consideration of the different impact of solar gains on energy demand in different orientations. At this scale, it is acceptable to assign, to both core and perimeter zones, the same input parameters including occupancy patterns. 2.1.1. Future weather files The future climate files used here are created using the synthetic weather generation algorithm described in Refs. [62–64] and subsequently named Indra.1 The method creates samples of probable future weather using a combination of historical data and climate change model outputs (Global and Regional Climate Models). It maintains psychrometric consistency and correlation between the significant variables that affect urban and building comfort and energy use. In the original publications, the authors suggest the use of multiple plausible samples of future weather to obtain estimates of distributions of simulation outputs. However, due to computational constraints, this workflow limits itself to representations of ‘typical’ future weather – an issue to be rectified in future work. 2.1.5. Daylighting The daylight modelling was carried out using Honeybee parametric interface, based on Radiance and Daysim engines for raytracing [67]. The workflow quantifies the overall indoor daylight availability, accounting for urban effects while assuming uniform window-to-wall ratios and using the same or similar divisions into core and perimeter zones as in the energy modelling step. 2.1.2. Urban weather files To account for the urban microclimate, the Dragonfly plugin2 for the Urban Weather Generator (UWG) [54] in Rhinoceros 3D was used. UWG approximates the changes in conditions inside urban canyons compared to measurements at a weather station in an open area outside a city. The tool can be used alone (as a data pre-processing step) or coupled with existing programs (the co-simulation approach). UWG calculates the hourly values of urban air temperature and humidity by 2.1.6. Biophilia, sky and nature view factor The computation of the sky view factor is straightforward and computationally fast with the view analysis component featured by Ladybug. It is sufficient to provide the analysis surface (representing the surface where pedestrians may be positioned) and the overall context geometry, including building masses as input, to return the view or exposure factor on a raster grid sampled on the analysis surface at a 1 The latest iteration of the source code is available at https://github.com/ paragrastogi/SyntheticWeather. 2 https://github.com/chriswmackey/Dragonfly. 4 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Fig. 1. The regenerative design workflow proposed in this paper. Fig. 2. A schematic of the workflow and Key Performance Indicators (KPI). do with its spotlight. The script proposed here performs an isovist analysis through the dedicated Grasshopper native component: the overall fraction of Lines Of Sight (LOS) intercepted by the footprints of tree canopies throughout a complete round turn would give the planar tree view factor. This fraction of tree-intercepted LOS can be computed by filtering the indices of obstacles struck, as returned by the Grasshopper isovist component that matches the tree footprint polygons. custom resolution. To account for the nature view factor, the method accounts for the fraction of the visual field occupied by vegetation for several pedestrian positions on the public space. This fraction is calculated from the projection of the tree canopies on a spherical visual field in a 3D space [70], but quick and immediate feedback is given by a planar section of the pedestrian visual field at eye height. An advantage of this procedure is the absence of complex 3D manipulations of point clouds representing trees. A bi-dimensional isovist is the polygonal locus of all visible, unobstructed points from the generating viewpoint, namely the vantage point, in a plan view [71]. Lines of sight are projected by pedestrian observers in all directions on a horizontal plane as a lighthouse would 2.2. Simulation inputs and settings The workflow was tested using the Barrio of La Luz (Fig. 5) in Malaga, built around 1960. The regenerative performance of the site is 5 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Fig. 3. A schematic of the workflow used plugins and definition of inputs and outputs. Fig. 4. The UTCI calculation workflow here diagrammed into more details. The following descriptions focus on the inputs of the simulation settings in each one of the evaluation modules. The computational demands of both thermal and wind comfort calculations limited both the number of simulated wind directions and the temporal coverage of the final comfort assessment and climate change. The albedo values of materials used in these are given in Appendix C. Owing to computational constraints, the samples from the Indra future climate generator were post-processed to create hypothetical Typical Meteorological Year [59] files for the 2050s and 2080s. This is an incomplete assessment from the perspective of quantifying uncertainty due to future weather, and future work will improve the integration of sampling to obtain distributions. The historical TMY is not a matter of discussion in the paper, as the objective is to verify the workflow's ability to represent regenerative performance and the adaptability for practice. A brief description of the site can be found in Appendix A. The geometric model that served as a base for the full set of algorithms was created in Rhinoceros 6, a commercial NURBS 3D software. The five blocks within the area of interest were modelled explicitly, with automated Grasshopper parametric subdivision of floors and window-to-wall ratio. The urban landscape was shaped to include buildings up to 250 m away from the area of interest for Computational Fluid Dynamics (CFD) and UHI modelling, though buildings outside the area of interest were approximated as simple footprint extrusions (Fig. 6). 6 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. directions is generally low (< 1 °C), compared to using all available wind information [60]. The surrounding urban context was explicitly modelled only for a radius of 250 m around the area of interest, with the impact of the urban masses outside that radius being approximated using an appropriate roughness factor typical for urban areas, according to the Updated Wieringa Roughness Classification [72]. The CFD study presented here uses simplified urban models which consider massing and not smaller architectural details which have minimal impact on wind flow. Elements such as awnings and facade details, were not considered. Additionally, the effect of trees on wind movement was not included in the simulations both because of time constraints and the insufficient data available on the accurate location and 3D shape of trees. Finally, the impact of air temperature and buoyancy on the wind flow was not considered, and an isothermal environment was assumed for the whole area. Typically, the effects of temperature differences on the wind flow is minimal, especially when there are no large heat sources in the urban environment. As mentioned earlier, quantifying thermal comfort requires the calculation of a variety of parameters at different steps. First, the geometrical model of the area is developed. Then, the CFD cases are generated and run. Wind factor values at each point of the mesh are then calculated and extracted during run-time. Wind factors are calculated by dividing wind velocity values with the reference wind velocity at 10 m. The selection of height is not random but follows the requirements of the UTCI polynomial expression requiring input velocities at 10 m. A custom Grasshopper component, using meteorological data, translates the wind factors for each point to yearly wind velocity schedules, 8760 wind velocity values for each point (Fig. 7). The velocities which correspond to each analysis period are then used, along with MRT values for all surfaces, to quantify UTCI values at each point assessed. One advantage of this approach is that the output is a schedule of values, for all hours within the analysis period determined, allowing for various visualisations in the form of thermal comfort maps to be produced. Additionally, allowing for temperatures to be included would then mean that multiple, hourly wind simulations would be required to assess wind flow as temperatures vary greatly both due to climatic conditions and urban context. More CFD settings can be found in Appendix D. For the Building Energy Performance Simulation (Energy Use Intensity and Energy Generation), the depth of the perimeter zone was set to 6 m, corresponding to twice the height of the zone [48]. The geometrical model considers a window to wall ratio of 40%. Inputs parameters for the energy simulation were defined based on the Spanish energy code [73] for climatic zone 3A (Appendix E). The radiation analysis drove the selection of the envelope surfaces for the BIPV energy yield calculations. The following viability thresholds were assumed for annual irradiance rates: 800 kWh/m2 for facades and 1000 kWh/m2 for rooftops. Floors subdivided the facade surfaces, and these thresholds were tested against the average radiation levels for each storey (Fig. 8). The selected surfaces were then used for the PV Fig. 5. Perspective view of the site, La Luz in Malaga (google coordinates: 36.696504, -4.455054). Fig. 6. The 3D image of the simplification process for performance evaluation. labelled with the current year (2018) to indicate that, if climate change is not considered in the design and simulation process, this is the best estimate of weather data currently available. Specific settings can be found in Appendix B. Since the Indra algorithm uses the IPCC climate change projections, an increase in average annual temperatures, along with increases in the severity of extreme summer days [7], is expected for most locations on earth. The magnitude of the difference varies with location and time into the future. As the original publications point out, the algorithm does not itself project data into the future. Rather, it scales the historical distribution of weather parameters using the IPCC model outputs and resamples them. Wind simulations were done for the four prevailing trends in Malaga, according to local meteorological data. Previous studies have shown that the error in UTCI calculation using the four prevailing wind Fig. 7. Wind map for the four main directions selected for the outdoor microclimate analysis. 7 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. obstructions are constituted by the buildings and the tree canopies. The fraction of lines of sight over 360° from a viewpoint intercepted by the footprints of tree canopies would give the planar tree view factor. The computationally intensive intersections of 360 LOS (1 per degree) can be run independently per each viewpoint in a batch cycle that has been implemented in the current workflow, through the animation of a Grasshopper integer slider. 3. Simulation outputs 3.1. Urban weather file To account for the impact of Urban Heat Island on the simulations, Urban Weather Generator (UWG) [54] was used to modify the historical climate data obtained from Malaga airport and the future time series generated by the synthetic weather generator Indra based on this airport data. UWG approximates thermal conditions within an urban area using several key geometric and material variables including building height, coverage, facade-to-site ratio, and road, roof, wall, and window constructions [60]. UWG uses a generic urban canyon model to run an energy balance calculation and produce the urban-adjusted weather time series (Fig. 10). Fig. 8. Annual radiation analysis for potential photovoltaic studies. The roofs have more potential than facades, as expected. 3.2. Outdoor thermal comfort Thermal comfort assessment, on the other hand, was limited to the extremely Hot, Extremely Cold, and Typical Spring weeks (Fig. 11), with the aim to capture periods with both average and extreme conditions. 3.3. Building energy performance (Energy Use Intensity and Energy Generation) Figs. 12–15 show different energy demand, and energy balance evaluations produced from this workflow for three different climatic scenarios in different time frames (hourly and monthly). This analytical variability can be used for different evaluation proposes and can indicate the impact of future climate on energy demand in different time frames, as well as the impact of urban microclimate on the energy balance. Furthermore, energy evaluation can be conducted in conjunction with other environmental metrics and indicate the trade-off between them in various morphological and usage scenarios. Fig. 9. The Energy and Daylighting model, including the surrounding buildings that cast shadows or reflect light. energy calculation using 17% efficiency and 70% area coverage for each of the surfaces above the radiation thresholds. A DC to AC deratement factor of 0.85 was used, and the effect of trees was neglected. The daylighting simulation was based on the same model used for the energy simulation, which is an acceptable approximation at this scale. Only the context that generates shadows or contributes to light reflection is modelled (Fig. 9). The simulation settings regarding Radiance parameters and material definitions are summarised in Appendix F. For the calculation of view factors, viewpoints are sampled on the boundaries of sidewalks with a spacing of 5 m; the view 3.4. Daylighting The sDA (300 Lux/50%) ranges vary from 40% to 80% on the outer shell of the district, while this range drops to between 13% and 50% in Fig. 10. Representation of the local yearly temperatures measured at Malaga Airport, a rural location [top left], and the modified urban weather time series, at the urban project site [lower left] obtained from Urban Weather Generator (UWG). The chart at right zooms into a 24-h cycle. 8 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Fig. 11. A schematic overview of the calculation of the Universal Thermal Climate Index (UTCI). be considered as a low value, mainly associated with the minimal distance to a visible tree, which is, on average, 28 m (Figs. 18–19). Moreover, compared to the WELL standard, the assessed area features only 5% of vegetation, which corresponds to ⅕th of the requirement. 4. Discussion 4.1. Capabilities of the workflow The workflow allows for the consideration of interconnected, multiscale factors in conjunction with multiple predictions of future weather conditions to improve the estimation of urban resilience. Simulating many weather conditions should not increase confidence in characterising the probability of different weather conditions. Instead, the procedure helps distinguish the response of the urban environment to differing external climatic stressors, which in turn will help to build in resilience into the design. Furthermore, the ability to effectively and automatically generate thermal zones out of the district's geometry through Grasshopper can make outdoor comfort, energy demand and daylighting evaluation far more integrated into the design process. Adding energy production considerations can be achieved just as seamlessly, and lead to the critical energy balance evaluation between supply and demand. For biophilia, an easy-to-calculate metric quantifying the visual prominence of vegetation in outdoor urban areas offers an objective and numerical scale to determine whether the presence of trees and green areas is Fig. 12. A visualisation of the Cooling energy demand of each building. the inner core of the neighbourhood (Fig. 16). The simulations could be carried out for different climate periods. 3.5. Sky and nature view factor Fig. 17 shows that the smallest road sections in the study area feature a low sky-view factor of less than 30% while the central square has a factor higher than 80%: as a reference, Times Square in New York features a sky-view factor equal to 37%. Trees are currently not considered in the analysis of sky view. In the proposed case-study, on average 25% of the circular visual field is occupied by trees. This may Fig. 13. The monthly breakdown of energy demand across all buildings by end-user type. 9 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Fig. 14. Hourly energy demand curves for a cold (left) and hot (right) days in the historical, 2050, and 2080 typical files. The energy requirements for the hot day are larger for future files than for historical ones, while the energy requirement for a cold day decreases into the future. sufficient in a pre-design phase. This indicator can be effectively used to justify compliance with emerging building standards such as WELL. Moreover, it may add a psychological wellbeing metric to the current outdoor comfort indices. 4.2. Limitations The incorporation of uncertainty into our model is rudimentary for now because climate change integration is limited to the inclusion of ‘typical’ (median) future files. This has the potential to miss many features of concern, such as heat waves and extreme urban heat stress. While the synthetic weather generator can generate any number of plausible future year samples, including years with extreme events, the computational cost of urban calculations makes the use of many files prohibitively expensive. The same limitation applies to the characterisation of UHI, where computational resources and time represent a limit. This impacts the number of wind directions and the resolution Fig. 15. Monthly energy demand and supply for the three different climatic periods. Substantial energy use intensity level was recorded in summer, compared to negligible changes in PV production. Fig. 16. Continuous daylight autonomy and spatial daylight autonomy representations. Fig. 17. Sky View Factor. Sky view factor on a 1 m raster grid, computed in the plan (left) and axonometric (right) view. 10 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. performance during the design process. However, the generation of many design options that are informed by the corresponding performance measures creates a need for intelligent spatial design exploration to sort, quantify and select the best scenarios. 4.4. Applicability in practice Different tools in this prototype workflow are integrated by means of custom scripts (programs) written in Grasshopper. The advantage designers working with a visual tool is the ease of incorporating visualisations for KPIs, primarily through the creation of operational or functional displays. This workflow highlights the spatial and temporal aspects of the different KPIs. The prototype presented here shows that the visualisations are informative and intuitive since they relate a KPI directly back to geometry (Fig. 20). This could aid in the evaluation of trade-offs between energy demand and solar production hour-by-hour and seasonally, or between increasing vegetation and view of the sky or indoor daylighting. The potential to create a sustainable environment by integrative parametric urban planning is promising since this workflow allows the further incorporation of additional analyses and simulation. Although the framework does not yield results with the click of a button, the correlation or the simultaneous visualisation of multiple analyses can substantially help in performance-driven decision making. The integrative urban planning workflow still strongly relies on expertise in Grasshopper and environmental analysis, and reducing this reliance on the user's expertise is an area of future work. Fig. 18. Nature View Factor, construction. Lines of sight intercepted by the tree canopies footprints (blue) vs those caught by buildings (red). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) and extent of the domain for CFD simulations. Furthermore, simplifications in the UHI assessment, CFD simulations and energy simulations may affect the confidence in the outdoor comfort calculation. Finally, the calculation of visual biophilia and daylight indicators is limited by the computational intensity of raytracing of light paths and lines of sight. Including the geometry of trees would improve these calculations as well. Computational limitations may potentially be addressed by improving parallelisation and offloading the computation from a user's computer to cloud-based simulation services. In addition, some of the simulation and analytical processes of this methodology were carried out separately, where a process is started manually after the preceding one has finished. This, along with data management challenges associated with coupling these tools, needs to be addressed in future developments of this methodology. 5. Conclusion To ameliorate the thermal quality of the urban environment, cities should offer their residents healthy outdoor spaces, favourable microclimates, and buildings that operate with moderate energy costs and nearly-zero carbon emissions. Urban spaces should enhance environmental conditions such as contact with nature, view of the sky, interior daylighting and more. These goals should be achieved holistically and inter-connectedly; pursuing regenerative design at the urban scale cannot be fulfilled with a checklist of sustainable design rating systems, which may lead to a box-ticking approach without embracing the whole. This paper defined Key Performance Indicators for different aspects of human health, wellbeing and energy in the urban context and evaluated them in the context of a changing climate. This contribution stresses the importance of connecting several critical performance indicators to adapt cities to climate change while enhancing human health and connection to nature. The need for target-oriented, regenerative and efficient urban planning and design in cities is urgent due to relentless urbanisation. While regenerative approaches are attracting increasing interest among 4.3. Integration of the workflow in digital design processes The advantage of using parametric modelling means that it is possible to create design variations or climatic variations based on rulebased generation. The main benefit of this approach is the enormous time savings compared to modelling in a non-parametric way and thus the possibility to explore more design variants, which may lead to a reduction of design costs and an increase in the spatial design qualities. The performance-based computational design supports regenerative analysis, synthesis and evaluation that mainly focuses on the Fig. 19. Nature View Factor. The angular ratio between the number of lines of sight intercepted by tree canopies footprints and the total number of lines of sight at each viewpoint (left). Minimal distance from the closest visible tree to each viewpoint (right). 11 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Fig. 20. A schematic view of the presentation of results overlaid on the geometrical models. sustainability design practitioners, transitioning from energy-oriented green building practices to a regenerative urban and building design practice entails the challenges of incorporating new strategies, new performance indicators, new methods and integrated tools. Regenerative design requires blending several key performances into a holistic thinking framework to see interrelationships between them. Key to this is a digital environment, supporting informative decisions and facilitating the exchange of information between stakeholders. The application of this holistic approach is presented through a new and multi-layered parametric workflow, based on data exchange and synergies across different modelling tools. In addressing the issues raised by this integration through the selection of appropriate parametric simulation scripts, visualisation and interaction models, this paper introduces a prototype integrated parametric workflow. The various components provide quantitative feedback to the designer during decision-making. As the current software landscape is fragmented into single-function tools, this prototype workflow is necessary. The workflow includes analyses and simulations in Rhinoceros 3D using its parametric plug-in Grasshopper and the environmental plugins included in the Ladybug Tools suite. This workflow is applied to a specific case study in Malaga, Spain, to explore its ability to quantify various regenerative performance indices and is available in a unified tool ecosystem to promote ease of usage. The findings from this paper should improve the adoption of evidence-based urban design by helping designers make informed decisions. The value of this workflow is in making available, for the first time, a diverse and extensive set of analysis tools for regenerative and climate-appropriate design. The designers receive a spatial overview of the urban configuration, as well as on performance in different future scenarios and time frames. This can aid in the alignment of composition, layout and aesthetics to global climatic and ecosystem goals as well as help ensure human wellbeing, in a holistic design framework. Ultimately, the integration and interoperability demonstrated in this paper will help increase the development and application of parametric design studies and evidence-based optimisation of design through easy-to-use quantitative tools. This article is a first step towards providing the contemporary urban planning practice with clear goals and interactive digital workflows supporting informed decisions and facilitating the exchange of information between stakeholders. The results of this study are guidelines for urban designers to define the regenerative characteristics of cities and a workflow that supports key choices in the design process. In future work, this would be extended to a thorough treatment of the computational and conceptual impacts of uncertainty resulting from the evolution of the climate, microclimatic conditions, and urban development. Furthermore, the parametric workflow can be automated with improved interconnectedness and decision-making. Acknowledgements The article was developed with the support of the COST Action CA16114 ‘RESTORE: Rethinking Sustainability towards a Regenerative 12 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Economy”. The authors would like to acknowledge the help of Manuel de-Borja-Torrejón in gathering the background data. The second author gratefully acknowledges the financial support of the German Academic Exchange Service (DAAD) for his PhD research grant. The fourth author would like to acknowledge the SCCER – FEEB&D for providing research funding and Prof. Jean-Louis Scartezzini for the scientific support. Appendix A. La Luz, Malaga The site is a polluted heat-island disconnected from sea breezes, with extensive hardscape (built-up area) and few natural elements. The wellbeing of the residents is further compromised due to modifications by tenants that obstruct natural ventilation and light. Climate change may exacerbate these baleful outdoor and indoor conditions, stressing the overall need for scalable interventions. La Luz features 222 dwellings/ha and a population density of 33.100 inhabitants/km,2 one of the most densely populated neighbourhoods in Malaga. The percentage of the unbuilt area is 62%, and the availability of green areas is scarce despite the presence of two gardens. The result is a monotonous and dull landscape, spread over 8620 m2 of gross floor area and containing 1190 apartments. The height of the buildings and the width of the road creates an oppressive space and shaded areas with insufficient direct solar radiation for social interaction. The blocks should, by law, be spaced apart 1.5–2.5 times their height, but are not. Appendix B. Weather file generation input values Table B1 Input Parameter Selected Value Representative Concentration Pathway (RCP)3 Random Seed Sample years (used to create decadal TMY) Number of samples (used to create decadal TMY) 8.5 42 2051–2060, 2071–2080 100 r decade Appendix C. Albedo values Table C1 Surface Albedo Corrugated roof Colored paint Trees Asphalt Concrete Grass 0.15 0.35 0.18 0.2 0.4 0.25 Appendix D. CFD settings The setup of the CFD simulations followed recognised best practices for outdoor wind simulations for all relevant simulation parameters [74]. The wind tunnel generated was 2.5 km long, 1 km wide, and 200 m tall. The model used was the realisable KE, a RANS variant that has shown outstanding performance and convergence for outdoor studies. Simulations were run until the convergence criteria for all relevant parameters were satisfied. A summary of all relevant input parameters can be found in Appendix F. All four simulations were created with Butterfly, which allows for a rapid and accurate CFD case generation and runs with OpenFOAM engine. Table D1 Input Parameter Selected Value Climatic Parameters Wind Directions (degrees from North) South East (135) West (270) West North West (292) North West (315) 2.69 4.09 4.02 3.74 68.3 Reference Wind Velocity (m/s) Wind Coverage (% of hours) Simulation Parameters Reference Height (m) Ground roughness value Boundary Conditions Top and Sides Outlet Surface Inlet Surface 10 1.0 - Closed Slip (zero gradients) Zero Pressure Atmospheric Boundary Layer Inlet Velocity (continued on next page) 3 Representative Concentration Pathways have replaced the emissions ‘scenarios’ used in previous IPCC reports from Assessment Report 5 onwards. 13 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. Table D1 (continued) Input Parameter Selected Value Turbulence Model Discretisation Schemes Convergence Criteria Mesh Cell Size – Area of Interest (m) Windward Extension (m) Leeward Extension (m) Sides Extension (m) Height (m) Cell to Cell Ratio Wake Offset (m) Height Offset (m) Realisable-KE Model 2nd Order 10e-4 for all variables 4 400 1900 400 200 1.15 50 10 Appendix E. Energy Simulation Parameters Table E1 Parameter Value Zone loads Lighting Schedule Occupancy Schedule HVAC: Heating/cooling setpoints Schedule Coefficient of performance (COP) Ventilation Schedule Material properties: Exterior Walls Roofs Exterior Floors Windows Infiltration Floor Height 4.4 W/m2 Mon.-Sun. 10% 1–7, 30% 8–19, 100% 20–23, 50% 19,24 0.06 People/m2 Mon.- Fri (Weekends 100%) 100% 1–7, 25% 8–15, 50% 16–23. 100% 24 20°/25° Heating: Jan-May, Oct-Dec Cooling: Jun-Sep 2 (cooling), 0.98 (heating) 4 Air Changes per Hour (ACH) Jun-Sep; 1-8 U = 0.94 W/m2K U = 0.5 W/m2K U = 0.53 W/m2K U = 5.7 W/m2K, SHGC = 0.6 1.25 ACH 3m Appendix F. Daylighting radiance settings Table F1 Parameters Description Value -ab -aa -ar -ad -as context Interior walls Interior ceiling Interior floor Ambient bounces Ambient accuracy Ambient resolution Ambient division Ambient sample reflectance Reflectance Reflectance Reflectance 2 0.15 128 512 256 0.35 0.5 0.8 0.2 [9] Kalnay E, Cai M. Impact of urbanization and land-use change on climate. Nature 2003;423:528–31. https://doi.org/10.1038/nature01675. [10] Mauree D, Coccolo S, Perera A, Nik V, Scartezzini J-L, Naboni E, et al. A new framework to evaluate urban design using urban microclimatic modeling in future climatic conditions. Sustainability 2018;10:1134. https://doi.org/10.3390/ su10041134. [11] Oke TR. The energetic basis of the urban heat island. Q J R Meteorol Soc 1982;108:1–24. https://doi.org/10.1002/qj.49710845502. [12] European Parliament. Directive 2010/31/EU of the European Parliament and of the Council on the energy performance of buildings (recast). European Parliament; 2010. [13] International Energy Agency IEA. Potential for building integrated photovoltaics. 2002. [14] International Energy Agency IEA. Technology roadmap: solar photovoltaic energy. 2014. [15] Zhang HL, Van Gerven T, Baeyens J, Degrève J, Ve J. Photovoltaics: reviewing the References [1] [2] [3] [4] [5] UN-Habitat. Economy - UN-habitat. 2018. United Nations. World urbanization prospects: the 2018 revision. 2018. UN-Habitat. Energy - UN-habitat. 2018. UN-Habitat. Climate change - UN-habitat. 2018. United Nations. The world's cities in 2016 - data booklet. 2016. https://doi.org/10. 18356/8519891f-en. [6] Perrone D, Hornberger GM. Water, food, and energy security: scrambling for resources or solutions? Wiley Interdiscip Rev Water 2014;1:49–68. https://doi.org/ 10.1002/wat2.1004. [7] Dean AF, Diss TC, Wotherspoon AC, Cox T, Nevard C. Global warming of 1.5. IPCC; 2018. https://doi.org/10.1203/00006450-199705000-00009. [8] Naboni E, Havinga LC, editors. Regenerative design in digital practice: a handbook for the built environment. Bolzano: EURAC Research; 2019. 14 Renewable and Sustainable Energy Reviews 113 (2019) 109255 E. Naboni, et al. [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] European Feed-in-Tariffs and changing PV efficiencies and costs. ScientificWorldJournal 2014;2014:404913. https://doi.org/10.1155/2014/ 404913. Pepermans G, Driesen J, Haeseldonckx D, Belmans R, D’haeseleer W. Distributed generation: definition, benefits and issues. Energy Policy 2005;33:787–98. https:// doi.org/10.1016/J.ENPOL.2003.10.004. Naboni E, Meloni M, Coccolo S, Cucchi F, Macrelli G, Kämpf JH, et al. The integration of outdoor thermal simulation tools in architectural design. Des TO THRIVE Proc PLEA 2017 2017. Naboni E, Meloni M, Coccolo S, Kaempf J. An overview of simulation tools for predicting the mean radiant temperature in an outdoor space. Energy Procedia 2017;122:1111–6. https://doi.org/10.1016/J.EGYPRO.2017.07.471. Nikolopoulou M, Baker N, Steemers K. Thermal comfort in outdoor urban spaces: understanding the human parameter. Sol Energy 2001;70:227–35. https://doi.org/ 10.1016/S0038-092X(00)00093-1. Nikolopoulou M, Steemers K. Thermal comfort and psychological adaptation as a guide for designing urban spaces. Energy Build 2003;35:95–101. https://doi.org/ 10.1016/S0378-7788(02)00084-1. Bröde P, Fiala D, Błażejczyk K, Holmér I, Jendritzky G, Kampmann B, et al. Deriving the operational procedure for the universal thermal climate index (UTCI). Int J Biometeorol 2012;56:481–94. https://doi.org/10.1007/s00484-011-0454-1. Dogan T, Reinhart CF, Michalatos P. Urban daylight simulation calculating the daylit area of urban designs. Fifth Natl Conf IBPSA-USA; 2012. Naboni E, Danzo E, Ofria L. A parametric workflow to concieve facades as indoor and outdoor climate. Atlanta: SimAUD; 2019. Dogan T, Park Y. A critical review of daylighting metrics for residential architecture and a new metric for cold and temperate climates. Light Res Technol 2019;51:206–30. https://doi.org/10.1177/1477153518755561. World Health Organisation. Health as the pulse of the new urban agenda. Quito; 2016. Kellert SR. The biophilia hypothesis. Washington, DC: Island Press [etc.]; 1993. Kaplan R, Kaplan S. The experience of nature : a psychological perspective. Cambridge University Press; 1989. Nisbet EK, Zelenski JM, Murphy SA. Happiness is in our nature: exploring nature relatedness as a contributor to subjective well-being. J Happiness Stud 2011;12:303–22. https://doi.org/10.1007/s10902-010-9197-7. Tyrväinen L, Pauleit S, Seeland K, de Vries S. Benefits and uses of urban forests and trees. Urban for. Trees. Berlin, Heidelberg: Springer Berlin Heidelberg; 2005. p. 81–114. https://doi.org/10.1007/3-540-27684-X_5. Ni Y, Eskeland GS, Giske J, Hansen J-P. The global potential for carbon capture and storage from forestry. Carbon Balance Manag 2016;11:3. https://doi.org/10.1186/ s13021-016-0044-y. Toochi EC. Carbon sequestration: how much can forestry sequester CO2? For Res Eng Int J 2018;2:148–50. https://doi.org/10.15406/freij.2018.02.00040. Coccolo S, Pearlmutter D, Kaempf J, Scartezzini J-L. Thermal Comfort Maps to estimate the impact of urban greening on the outdoor human comfort. Urban For Urban Green 2018;35:91–105. https://doi.org/10.1016/J.UFUG.2018.08.007. Wolch JR, Byrne J, Newell JP. Urban green space, public health, and environmental justice: the challenge of making cities ‘just green enough. Landsc Urban Plan 2014;125:234–44. https://doi.org/10.1016/J.LANDURBPLAN.2014.01.017. Bratman GN, Daily GC, Levy BJ, Gross JJ. The benefits of nature experience: improved affect and cognition. Landsc Urban Plan 2015;138:41–50. https://doi.org/ 10.1016/J.LANDURBPLAN.2015.02.005. Irvine K, Warber S, Devine-Wright P, Gaston K. Understanding urban green space as a health resource: a qualitative comparison of visit motivation and derived effects among park users in sheffield, UK. Int J Environ Res Public Health 2013;10:417–42. https://doi.org/10.3390/ijerph10010417. Donovan GH, Michael YL, Gatziolis D, Prestemon JP, Whitsel EA. Is tree loss associated with cardiovascular-disease risk in the Women's Health Initiative? A natural experiment. Health Place 2015;36:1–7. https://doi.org/10.1016/J. HEALTHPLACE.2015.08.007. Münch M, Bromundt V. Light and chronobiology: implications for health and disease. Dialogues Clin Neurosci 2012;14:448–53. Bakolis I, Hammoud R, Smythe M, Gibbons J, Davidson N, Tognin S, et al. Urban mind: using smartphone technologies to investigate the impact of nature on mental well-being in real time. Bioscience 2018;68:134–45. https://doi.org/10.1093/ biosci/bix149. Unger J. Connection between urban heat island and sky view factor approximated by a software tool on a 3D urban database. Int J Environ Pollut 2009;36:59. https:// doi.org/10.1504/IJEP.2009.021817. Wang Y, Akbari H. Effect of sky view factor on outdoor temperature and comfort in montreal. Environ Eng Sci 2014;31:272–87. https://doi.org/10.1089/ees.2013. 0430. Nisbet EK, Zelenski JM, Murphy SA. The nature relatedness scale. Environ Behav 2009;41:715–40. https://doi.org/10.1177/0013916508318748. Mayer FS, Frantz CM. The connectedness to nature scale: a measure of individuals' feeling in community with nature. J Environ Psychol 2004;24:503–15. https://doi. org/10.1016/J.JENVP.2004.10.001. Taylor L, Hahs AK, Hochuli DF. Wellbeing and urban living: nurtured by nature. Urban Ecosyst 2018;21:197–208. https://doi.org/10.1007/s11252-017-0702-1. Thi Loi D, Anh Tuan P, Gupta K. Development of an index for assessment of urban green spacesat city level. Int J Remote Sens Appl 2015;5:78. https://doi.org/10. 14355/ijrsa.2015.05.009. [45] Van de Voorde T. Spatially explicit urban green indicators for characterizing vegetation cover and public green space proximity: a case study on Brussels, Belgium. Int J Digit Earth 2017;10:798–813. https://doi.org/10.1080/17538947.2016. 1252434. [46] Grinde B, Biophilia Patil GG. Does visual contact with nature impact on health and well-being? Int J Environ Res Public Health 2009;6:2332–43. https://doi.org/10. 3390/ijerph6092332. [47] Teller J. A spherical metric for the field-oriented analysis of complex urban open spaces. Environ Plan Plan Des 2003;30:339–56. https://doi.org/10.1068/b12930. [48] Reinhart CF, Dogan T, Jakubiec JA, Rakha T, Sang A. UMI – an urban simulation environment for building energy use, daylighting and walkability. Proc. Build. Simul. 201313th Conf. Int. Build. Perform. Simul. Assoc. Chambéry, FRA 2013. August 26-28. [49] Naboni E, Coccolo S, Harmanescu M, Hansen P. Rethinking sustainability towards a regenerative economy (RESTORE) within an adaptive neighbourhood design. 34th int. Conf. Passiv. Low energy archit., Hong Kong. 2018. p. 591–7. [50] Givoni B. Urban design in different climates. WMO; 1989. [51] Reinhart CF, Cerezo Davila C. Urban building energy modeling – a review of a nascent field. Build Environ 2016;97:196–202. https://doi.org/10.1016/J. BUILDENV.2015.12.001. [52] Robinson D, Haldi F, Kämpf JH, Perez D. Computer modelling for sustainable urban design : physical principles, methods and applications. Earthscan; 2011. [53] Fonseca JA, Nguyen T-A, Schlueter A, Marechal F. City Energy Analyst (CEA): integrated framework for analysis and optimization of building energy systems in neighborhoods and city districts. Energy Build 2016;113:202–26. https://doi.org/ 10.1016/J.ENBUILD.2015.11.055. [54] Bueno B, Norford L, Hidalgo J, Pigeon G. The urban weather generator. J Build Perform Simul 2013;6:269–81. https://doi.org/10.1080/19401493.2012.718797. [55] Palme M, Inostroza L, Villacreses G, Lobato-Cordero A, Carrasco C. From urban climate to energy consumption. Enhancing building performance simulation by including the urban heat island effect. Energy Build 2017;145:107–20. https://doi. org/10.1016/J.ENBUILD.2017.03.069. [56] Salvati A, Coch H, Morganti M. Effects of urban compactness on the building energy performance in Mediterranean climate. Energy Procedia 2017;122:499–504. https://doi.org/10.1016/J.EGYPRO.2017.07.303. [57] Natanian J, Auer T. Balancing urban density, energy performance and environmental quality in the Mediterranean: a typological evaluation based on photovoltaic potential. Energy Procedia 2018;152:1103–8. https://doi.org/10.1016/J.EGYPRO. 2018.09.133. [58] Nault E, Moonen P, Rey E, Andersen M. Predictive models for assessing the passive solar and daylight potential of neighborhood designs: a comparative proof-of-concept study. Build Environ 2017;116:1–16. https://doi.org/10.1016/J.BUILDENV. 2017.01.018. [59] Perini K, Chokhachian A, Dong S, Auer T. Modeling and simulating urban outdoor comfort: coupling ENVI-Met and TRNSYS by grasshopper. Energy Build 2017;152:373–84. https://doi.org/10.1016/J.ENBUILD.2017.07.061. [60] Mackey C, Galanos T, Norford L, Roudsari M. Wind, sun, surface temperature, and heat island: the critical variables for high‐resolution outdoor thermal comfort. San Francisco: BS 2017; 2017. [61] Evola G, Magrì C, Morgani G, Naboni E. Modelling outdoor thermal comfort and energy demand in urban canyons: validation of a novel comprehensive parametric workflow. 16° int. Build. Perform. Simul. Assoc., Rome. 2019. [62] Rastogi P. On the sensitivity of buildings to climate the interaction of weather and building envelopes in determining future building energy consumption. EPFL; 2016. https://doi.org/10.5075/EPFL-THESIS-6881. [63] Rastogi P, Andersen M. Incorporating climate change predictions in the analysis of weather-based uncertainty. ASHRAE IBPSA-USA SimBuild 2016 – Build Perform Model Conf 2016. [64] Rastogi P, Andersen M. Embedding stochasticity in building simulation through synthetic weather files. Proc BS 2015 2015. [65] Mostapha Sadeghipour Roudsari MP, Adrian Smith + Gordon Gill Architecture. Chicago USA. Ladybug: a parametric environmental plugin for grasshopper to help designers create an environmentally-conscious design. 13th conf int build perform simul assoc. 2013. p. 3129–35. [66] Ladybug Tools ¦ Homepage n.d. https://www.ladybug.tools/(accessed June 5, 2019). [67] Ward GJ. The RADIANCE lighting simulation and rendering system. Computer graphics. SIGGRAPH Conf.; 1994. [68] ANSI/ASHRAE. ANSI/ASHRAE standard 55-2013: thermal environmental conditions for human occupancy. 2013. https://doi.org/10.1007/s11926-011-0203-9. [69] LBNL. References for EnergyPlus v8.8. 2017. Berkeley. [70] Florio P, Munari Probst MC, Schüler A, Roecker C, Scartezzini J-L. Assessing visibility in multi-scale urban planning: a contribution to a method enhancing social acceptability of solar energy in cities. Sol Energy 2018. https://doi.org/10.1016/j. solener.2018.07.059. [71] Benedikt ML. To take hold of space: isovists and isovist fields. Environ Plan Plan Des 1979;6:47–65. https://doi.org/10.1068/b060047. [72] Wieringa J. Updating the Davenport roughness classification. J Wind Eng Ind Aerodyn 1992;41:357–68. https://doi.org/10.1016/0167-6105(92)90434-C. [73] de España Gobierno. Código Tecnico de la Edificación - documento básico - ahorro de energía. 2013. [74] Franke JJ, Hellsten A, Schlünzen H, Carissimo B. Best practice guideline for the CFD simulation of flows in the urban environment. 2007. 15