The tool reads one weather file and turns its 8,760 hours into readings for design. Each entry below says what a
term means, why it matters to a designer, where it appears in the tool and what to keep in mind when reading it.
Entries link to each other; a term in a link has its own entry.
Numbers marked as the tool's own choice are conventions of this tool, not of a standard. The
README's "Assumptions to know" lists every one of them.
Weather data and climate classification
Every reading in the tool starts from one year of hourly weather. This group explains where that year comes from, how far it can be trusted, and the classifications and sums that condense it into a few figures a designer can compare.
EPW weather fileEnergyPlus Weather format
A plain text file of one year at one place: eight header lines (location, latitude, longitude, time zone, elevation, design conditions, ground temperatures, data periods) and then 8,760 rows, one per hour. Each row holds the dry bulb and dew point temperatures, relative humidity, pressure, the solar components, illuminances, long-wave sky radiation, wind direction and speed at 10 m, sky cover, visibility, rain and snow.
The values of an hour are those at the end of the hour for temperatures, and the sum or mean over the hour before for the sun.
For design. The file is the climate as far as the tool knows it. Check its source, its period and the station's distance and height from the site before drawing conclusions: an airport 20 km away and 300 m lower is another climate.
In the app. Chosen on the map (a station), opened with Open EPW… or dropped on the page. The header line under the title names the station and the source. Any variable, any hours shows any column of the file hour by hour.
Keep in mind. A file from PVGIS or a measured year has no rain, sky cover or illuminance; the cards that need them say so or estimate them.
A typical meteorological year is assembled from many measured years: for each calendar month, the one real month whose statistics (temperature, humidity, sun, wind) lie closest to the long-term distribution is chosen, and the twelve months are joined. The result looks like an ordinary year but none of its months is extreme.
IWEC: the International Weather for Energy Calculations files (ASHRAE), from about 1982 to 1999.
TMYx: typical years from recent records (for example 2009–2023) built by Climate.OneBuilding.Org for about 17,000 stations; the tool takes each station's newest.
National sets (CWEC in Canada, test reference years in Germany and others): the same idea by each country's method.
For design. A typical year is right for average energy use and for choosing strategies. It is wrong for peak loads, overheating risk in a hot summer and anything driven by extremes: use design conditions, a measured year or a morphed future year for those.
In the app. The map colours the stations by source (Colour by, Source of the file). Compare with another climate sets two files side by side.
Keep in mind. Files from different sources are made by different methods and periods: an older IWEC file of a city is usually cooler than its TMYx, and part of any difference between two files can be the method. TMYx files give 15 to 20 % less diffuse illuminance than IWEC files for the same cities, which changes the daylight targets.
A reanalysis runs a weather forecast model over the past and corrects it at every step with the observations then available (stations, balloons, satellites, ships). The result is a complete, gap-free record on a regular grid. ERA5, from the Copernicus Climate Change Service, covers 1940 to the present, hour by hour, in cells of about 30 km (9 km over land in ERA5-Land).
For design. Where there is no station, or where a long record is needed (trends, heatwaves, a particular hot summer), a reanalysis is the best data there is. It is smoother than a station: a city's heat island, a valley's cold pool and local winds fall inside one cell.
In the app. The Measured menu of the bar at the foot of the page opens any year from 1940 as a weather file; Heatwaves and cold spells in the measured record reads daily ERA5 since 1991. Both are fetched from Open-Meteo, which brings the cell to the point's own height.
Keep in mind. In the mountains a cell's mean height can be far from the site: for Santa Maria Val Müstair (1,463 m) the PVGIS year built on ERA5 has a mean of −2.6 °C, where the station's TMYx gives 4.0 °C.
Source. Hersbach, H. et al. (2020), The ERA5 global reanalysis, Quarterly Journal of the Royal Meteorological Society 146.
Typical year for a pointPVGIS TMY
The European Commission's PVGIS service builds a typical year for any point on land: the sun from satellite images (cut by the horizon of the terrain), air temperature, humidity and wind from ERA5, and months chosen from 2005 to 2023.
For design. Useful where the nearest station is far or in another setting (a coast against an inland site, a valley against a ridge). Compare it with the nearest station before trusting either.
In the app. A click on empty land on the map, or Exactly at … at the top of the search results.
Keep in mind. The file has no rain, sky cover or illuminance, so the moisture letter of Köppen, rain and driving rain are not read, and daylight targets are estimated. In the mountains the air is that of the ERA5 cell's mean height, often several kelvin too cold.
A year map draws all 8,760 hours as a coloured grid: the day of the year runs from left to right, the hour of the day from bottom to top (midnight at the foot, midnight again at the head). Each cell takes the colour of its value or class. The night is a band at the top and bottom, the day in the middle; seasons appear as changes from left to right.
For design. The map shows when, not only how much: whether heat comes in the afternoon or all day, whether cool nights follow hot days (so night ventilation can work), and how long each season lasts. A share of hours says how big a problem is; the map says when the building has to deal with it.
In the app. The temperature map beside Temperature by month, When each strategy is needed, Natural ventilation potential, hour by hour, and the year map of Any variable, any hours. The scales are the same for every climate, so two places can be compared by eye.
The difference between a day's highest and lowest air temperature, averaged over the days of a month. Dry and clear climates swing widely (15 to 20 K in deserts); humid, cloudy and coastal climates swing little (5 K or less).
For design. The range decides whether thermal mass pays: a heavy building averages out a large swing, so it stays near the daily mean; with a small swing there is little to average. As a rule of thumb, mass with night ventilation works well with a summer range of 10 K or more and poorly under about 8 K.
In the app. The bars of Temperature by month; the Thermal mass card; one of the "argues against" lines of This climate in short (summer daily swing under 8 K with over 3 % of hours at 27 °C).
The most used climate classification, from monthly mean temperature and precipitation. The first letter gives the main group: A tropical (every month at least 18 °C), B arid (rain short of a threshold set by temperature), C temperate, D continental, E polar. The second letter gives the season of rain (f no dry season, s dry summer, w dry winter; or W desert, S steppe for B) and the third the heat of the summer (a hot, b warm, c cool, d very cold winter; h hot, k cold for B). Milan is Cfa, London Cfb, Rome Csa.
For design. A quick label for comparing a site with places whose buildings you know. It says nothing about wind, sun or humidity directly, and two places of one class can need different strategies.
In the app. The map's Colour by menu (default), and This climate in short.
Keep in mind. The tool reads the file's own single year, and a typical year's rain is patchy: Copenhagen's TMYx comes out Dsb rather than Cfb. A file without rain gets its temperature letters only. The tool uses 0 °C between C and D, after Peel et al.
Source. Peel, M. C., Finlayson, B. L. and McMahon, T. A. (2007), Updated world map of the Köppen-Geiger climate classification, Hydrology and Earth System Sciences 11.
ASHRAE climate zoneASHRAE Standard 169
A classification made for building energy codes. The number (0 extremely hot to 8 subarctic) comes from cooling degree days above 10 °C and heating degree days below 18 °C; the letter says whether the place is humid (A), dry (B) or marine (C). Milan is 4A.
For design. ASHRAE 90.1 and the IECC set their prescriptive insulation and glazing values by this zone, so it tells you which column of the code applies.
In the app.This climate in short, the map's Colour by menu, and Prescriptive envelope for this climate zone. Where the file's header states a zone, it is shown beside the computed one.
Keep in mind. The dry test is the older form of the standard (yearly rain against mean temperature) and depends on the file's rain, as Köppen does.
A degree day sums how far and how long the outdoor temperature lies beyond a base temperature. For heating, each day adds the base less the day's mean temperature when the mean is below the base; for cooling, the mean less the base when it is above.
HDD = ∑ max(0, Tb − Td) CDD = ∑ max(0, Td − Tb)where Tb is the base temperature and Td the day's mean, summed over the days of the year
The base stands for the outdoor temperature at which the building needs neither heating nor cooling (the balance point). It is below the indoor set point because people, appliances and sun supply the rest.
For design. A building's heat loss through its envelope is roughly proportional to its heating degree days: a place with 3,000 K·d needs about twice the heating of one with 1,500 K·d for the same building. They set the code zone in many countries and the insulation a climate justifies.
In the app.Degree days (Temperature), with both bases editable (18 °C by default). Also used for the ASHRAE zone, the glazing target and the Insulation card.
Keep in mind. A well insulated building with high internal gains has a low balance point (12 to 15 °C), so degree days at 18 °C overstate its heating. Bases differ by country (15.5 °C in the UK, 20 °C for Italy's zones).
The temperatures and humidities used to size heating and cooling plant, given as annual percentiles of a long record. The 99.6 % heating condition is the temperature exceeded in all but 0.4 % of the year's hours (about 35 hours colder); the 0.4 % cooling condition is the dry bulb exceeded in 0.4 % of hours, with the mean coincident wet bulb. Values at 1 % and 2 % are less strict.
For design. They answer "how big must the plant be", where a typical year answers "how much energy over a year". They come from 20 or more years, so they include colder and hotter weather than the typical year shows.
In the app. Drawn as points on the psychrometric chart (Design days check box), from the EPW header or from a dropped .ddy file. Under a scenario they are morphed as an hour of the coldest or warmest month.
Source. ASHRAE Handbook of Fundamentals, chapter on climatic design information.
Ground temperature
Below the surface the yearly wave of air temperature is damped and delayed: at about 0.5 m the ground still follows the seasons with a lag of weeks; at 2 m the swing is roughly halved and late by one to two months; below about 4 to 6 m the ground stays near the annual mean air temperature. The EPW header usually gives monthly ground temperatures at three depths (often 0.5, 2 and 4 m).
For design. The ground is a heat source in winter and a sink in summer: earth-sheltered walls, earth tubes and ground-source heat pumps use it. The deeper and the larger the swing of the climate, the larger the gain.
In the app.Ground temperature (Temperature), from the file's header. Under a scenario they are shifted by the annual mean change.
Keep in mind. The header values are usually computed from the air temperature for a generic soil, not measured.
Psychrometrics describes moist air: how much water vapour it carries, how warm it is and how much energy it holds. Two values fix the state of the air at a given pressure; everything else follows. The psychrometric chart draws these relations, and the tool's main chart places every hour of the year on it.
Dry bulb temperatureTa, air temperature, °C
The temperature of the air as an ordinary thermometer reads it, shaded from the sun and from radiation. "Dry" distinguishes it from the wet bulb. Weather stations measure it at about 2 m above open ground.
For design. The first variable of every comfort and energy question, but never the only one: the same 28 °C is pleasant in dry, moving air and oppressive in humid, still air or in the sun.
In the app. The horizontal axis of the psychrometric chart; Temperature by month and its year map; any filter of Any variable, any hours.
The mass of water vapour carried per kilogram of dry air. It does not change when air is only heated or cooled; it changes only when water is added (evaporation, people, cooking) or taken out (condensation). Outdoor air ranges from under 1 g/kg in a cold winter to over 20 g/kg in humid tropics.
W = 0.622 pw / (p − pw)where pw is the vapour pressure and p the total atmospheric pressure; 0.622 is the ratio of the molar masses of water and dry air
For design. The quantity to watch for latent loads and dehumidification. Over about 12 g/kg, sweat evaporates poorly and comfort needs air movement or dehumidification; direct evaporative cooling loses most of its effect in such air.
In the app. The vertical axis (right side) of the psychrometric chart. The ASHRAE 55 comfort zone of the chart stops at 12 g/kg, and one "argues against" line of This climate in short counts hours above it.
The part of the air's total pressure due to its water vapour. At each temperature there is a largest possible value, the saturation vapour pressure, which rises steeply with temperature (about 611 Pa at 0 °C, 2,339 Pa at 20 °C, 4,246 Pa at 30 °C). Air holding that much is saturated.
For design. Vapour moves from high to low vapour pressure, so in winter it diffuses from a warm, humid interior out through a wall. Where it meets a layer colder than its dew point, it condenses. The moisture checks of walls work in vapour pressure.
In the app. A variable of Any variable, any hours; the basis of Moisture: condensation in the wall and mould at its surface.
Source. ASHRAE Handbook of Fundamentals, chapter 1, Psychrometrics (saturation pressure after Hyland and Wexler).
Relative humidityRH, φ, %
The vapour pressure of the air as a share of the saturation vapour pressure at its temperature. Because saturation rises with temperature, warming air lowers its relative humidity without changing its water content: winter air at 0 °C and 80 % heated to 20 °C falls to about 20 %.
RH = 100 × pw / pws(T)
For design. People feel relative humidity mostly at its ends: dry eyes and static below about 30 %, a sticky skin when it is high and warm. Materials respond to it directly: mould grows on surfaces kept above about 80 %, and salts and corrosion have their own thresholds.
In the app. The curved lines of the psychrometric chart; the humidity charts; the UTCI sensitivity maps.
Keep in mind. Comparing relative humidity between hours of different temperature says little; compare humidity ratio or dew point instead.
The temperature to which air must be cooled, at constant water content, to become saturated. Below it, water condenses on surfaces as dew, or in the air as fog. Since it depends only on the water content, it is a measure of absolute humidity in degrees.
For design. Any surface colder than the room's dew point will be wet: a cold window pane, a thermal bridge, a chilled ceiling, the inner face of a wall's outer layer. In summer, a dew point over about 18 °C feels muggy to most people.
In the app. On the chart, the dew point is read by moving horizontally to the saturation curve (see the figure under Psychrometric chart); Any variable, any hours lists it.
The temperature a wetted thermometer reaches in moving air: water evaporates from it and cools it until the heat it draws from the air balances the heat used for evaporation. The drier the air, the larger the difference from the dry bulb (the wet bulb depression). In saturated air the two are equal.
For design. The wet bulb is the lowest temperature that direct evaporative cooling can reach, so the depression measures its potential: a large depression in hot, dry air means strong cooling; in humid air there is none. Cooling towers and the heat stress of people working hard also depend on it.
In the app. The sloping straight lines of the psychrometric chart; the reading under the chart as the mouse moves; a variable of Any variable, any hours.
The heat content of moist air per kilogram of dry air, counted from 0 °C dry air: a sensible part that follows the temperature and a latent part held in the vapour (the heat that was needed to evaporate it).
h ≈ 1.006 T + W (2501 + 1.86 T)where T is the dry bulb in °C and W the humidity ratio in kg/kg
For design. The energy to cool and dry ventilation air is the mass flow times the drop in enthalpy. In humid climates the latent part is often larger than the sensible one, which is why heat recovery that also exchanges moisture (an enthalpy wheel) pays there. Lines of constant enthalpy lie almost on the wet bulb lines.
In the app. Each process drawn on the chart is named with its change in enthalpy, split into sensible and latent.
Air pressure falls with height, by about 11 % for every 1,000 m near sea level (101,325 Pa at sea level, about 89,900 Pa at 1,000 m). At lower pressure the same vapour pressure gives a higher humidity ratio, so the psychrometric chart changes with altitude.
For design. For sites up to a few hundred metres the difference is small. At 1,500 m or higher, a sea-level chart misplaces the hours, and air carries less heat per cubic metre, which matters for sizing ventilation.
In the app. The psychrometric chart is drawn at the site's standard pressure, from the station's elevation in the file.
A drawing of the states of moist air at one pressure. Dry bulb temperature runs along the horizontal axis and humidity ratio up the vertical one, so every point is one state of the air. Over that grid lie the other properties:
the saturation curve (100 % relative humidity) bounds the chart at the upper left; no air lies beyond it;
curves of lower relative humidity follow it below, at 90, 80, ... %;
sloping straight lines of constant wet bulb (and almost constant enthalpy) run down to the right;
the dew point of a point is read horizontally to the left on the saturation curve.
One state of the air, 30 °C and 40 %, read on the chart at sea level: its humidity ratio to the right, its dew point horizontally on the saturation curve, its wet bulb along the sloping line.
Building climatology adds zones to the chart: a comfort zone, and around it the states that a strategy (shading, mass, ventilation, evaporation, passive solar heat) can bring into comfort. Placing every hour of a year as a dot shows at once which strategies the climate asks for.
For design. The chart is the bridge between a climate and a design response. A cloud of hours far to the right and low (hot, dry) asks for mass and evaporation; to the right and high (hot, humid) for air movement and dehumidification; to the left for insulation and sun.
In the app. The large chart at the head of Comfort and strategies. Choose the overlay (Strategies, Comfort band, Predicted mean vote), the comfort model, the zones (Givoni or Szokolay), the months and hours; hours can be points or density. Moving the mouse reads the state of the air; the chart exports as SVG.
Source. ASHRAE Handbook of Fundamentals, chapter 1; Givoni, B. (1969), Man, Climate and Architecture. Further reading: 1.
Processes on the chartsensible and latent heat
A change of the air's state is a line on the chart. Heating or cooling without adding or removing water moves it horizontally (a sensible process). Adding or removing water at constant temperature moves it vertically (latent). Evaporative cooling follows a wet bulb line: the air cools as it takes up water, its enthalpy almost unchanged. Cooling below the dew point runs left to the saturation curve and then down along it, as water condenses on the coil.
Q = m Δhthe heat (kW) to take air from one state to another: the mass flow of dry air (kg/s) times the change in enthalpy (kJ/kg); split into a sensible part (temperature) and a latent part (water)
For design. Drawing the process shows what a system must do: whether cooling also has to dry the air, how far evaporation can cool, what heat recovery saves.
In the app. Click two states on the psychrometric chart: the line is drawn and listed under the chart with its change in temperature, humidity and enthalpy (sensible and latent). Clear processes removes them.
Every surface radiates heat as infrared, in proportion to the fourth power of its absolute temperature. The sky radiates back, but a clear, dry sky is much colder than the air (often 10 to 30 K colder), so a roof facing it loses heat at night. Clouds and humidity warm the sky and cut the loss. The EPW file holds this radiation on a horizontal surface for every hour.
For design. Night sky cooling (radiant roofs, cool roof ponds) works in clear, dry climates. The same loss chills roofs below the air temperature on clear nights, which brings frost and condensation on their underside.
In the app. Used for the dark and light roof temperatures of Weathering of materials. Under a scenario it follows the air temperature to the fourth power.
Keep in mind. In a measured year it is estimated from dew point and cloud cover, with coefficients the README says are written from memory.
Thermal comfort is "that condition of mind which expresses satisfaction with the thermal environment" (ASHRAE 55). It depends on six variables: four of the environment (air temperature, radiant temperature, air speed, humidity) and two of the person (activity, clothing). Two families of models judge it: heat-balance models for controlled buildings and adaptive models for buildings whose occupants open windows.
Heat balance of the body
The body produces heat by metabolism and must lose the same amount to keep its core near 37 °C. It loses heat by convection to the air, radiation to the surrounding surfaces, evaporation of sweat from the skin, and through breathing; what is not lost is stored and the body warms.
M − W = C + R + Esk + Cres + Eres + SM metabolic rate, W external work, C convection, R radiation, Esk evaporation from the skin, Cres and Eres heat lost by breathing, S heat stored; all in W/m² of body surface
For design. Each term points to a design lever: radiation to surface temperatures (insulation, glazing, shading, radiant systems), convection to air temperature and air speed, evaporation to humidity and air movement. Comfort is not only the air temperature.
In the app. The basis of PMV, SET, UTCI and PET, in Comfort of one person and the outdoor cards.
The heat the body produces per square metre of skin, in met. One met is a person seated at rest, about 105 W for an adult of 1.8 m² of body surface. Typical values: sleeping 0.7, seated reading 1.0, typing 1.1, standing relaxed 1.2, light work standing 1.6 to 2.0, walking at 0.9 m/s about 2.0, dancing 2.4 to 4.4.
For design. The use of a room sets its comfortable temperature: a gym or a workshop wants several kelvin cooler than an office or a living room. Values over 2 met take the heat-balance models outside their range of validity.
In the app. The met field of the main chart (with the ASHRAE 55 model) and the activity list of Comfort of one person, which can pass its activity, clothing and air speed to the main chart.
Source. ASHRAE 55-2023, Normative Appendix and its tables of typical activities; ISO 8996.
Clothing insulationclo; 1 clo = 0.155 m²K/W
The thermal resistance of clothing from skin to the outer surface of the clothes. Typical ensembles: shorts and a T-shirt 0.36 clo, trousers and a short-sleeved shirt 0.57, trousers and a long-sleeved shirt 0.61, a business suit about 1.0, winter indoor clothing 1.0. An office chair adds about 0.1 clo.
For design. Dress codes shift comfort: allowing summer clothing lowers cooling needs. Models assume 0.5 clo in summer and 1.0 in winter unless told otherwise.
In the app. The clo field of the main chart and the clothing list of Comfort of one person.
Keep in mind. When a person moves, the clothing's insulation falls (air is pumped through it). ASHRAE 55-2023 lowers it above 1.2 met, which the tool applies.
The uniform temperature of an imaginary black enclosure in which a person would exchange the same radiant heat as in the real room. It averages the temperatures of all surfaces the body sees, each weighted by how much of the body's view it takes (its view factor), and adds any sun falling on the body.
For design. Indoors the MRT is set by the envelope: a cold window, a poorly insulated wall or a sunlit floor moves it by several kelvin. Raising the inner surface temperatures through insulation lets the air be cooler for the same comfort. Outdoors, sun and hot paving can lift the MRT 20 to 30 K above the air.
In the app. An input of Comfort of one person, with SolarCal to add the sun. The main chart takes it equal to the air temperature. Outdoors see Mean radiant temperature outdoors.
A single temperature that combines air and mean radiant temperature as the body feels them together. In still air it is close to their mean; as air speed rises, the air counts more.
To = ATa + (1 − A) TrA is 0.5 below 0.2 m/s, 0.6 from 0.2 to 0.6 m/s and 0.7 from 0.6 to 1.0 m/s (ASHRAE 55)
For design. The temperature that comfort limits and overheating criteria are written in. A thermostat reads air temperature only, so a room with warm walls can be comfortable at a lower set point.
In the app. The horizontal axis of the chart in Comfort of one person; the adaptive models and A room through the year work in it.
Moving air carries heat from the skin by convection and speeds the evaporation of sweat. Indoors, still air is under 0.1 to 0.2 m/s; a ceiling fan gives 0.5 to 1.5 m/s. A moving person also creates air movement over the body: ASHRAE 55 adds 0.3 (met − 1) m/s for activities above 1 met, the relative air speed.
For design. Air movement is the cheapest cooling in warm and humid climates: ceiling fans and cross ventilation can extend the comfortable range by 2 to 3 K. In cold seasons the same movement is a draught.
In the app. The air m/s field of the main chart; the inputs of Comfort of one person, which also shows the air speed limits for occupants without control over it.
Fanger's model predicts the mean vote of a large group on the seven-point thermal sensation scale: −3 cold, −2 cool, −1 slightly cool, 0 neutral, +1 slightly warm, +2 warm, +3 hot. It solves the body's heat balance for the six variables and turns the imbalance (the load a person must make up) into a vote, using climate chamber studies of people in steady conditions.
For design. The model of air-conditioned and mechanically heated buildings with a steady environment. It shows how much clothing, activity, air speed and radiant temperature are worth in kelvin of air temperature.
In the app. The Predicted mean vote overlay and the ASHRAE 55 (PMV ±0.5) and EN 15251 comfort models of the main chart; Comfort of one person with ASHRAE 55-2023 or ISO 7730, where the zone is drawn on a chart of operative temperature and can be laid over the main chart.
Keep in mind. PMV was made for steady, uniform conditions and 0.8 to 4 met, 0 to 2 clo; in buildings where people control windows and clothing it tends to overpredict warmth (see adaptive comfort). On the main chart the hours are outdoor air with radiant temperature equal to air, and ASHRAE 55's zone is capped at 12 g/kg (the cap is this tool's, kept from earlier editions).
For design. A way to state a comfort target as an expected share of unhappy occupants, and to see why a target of zero complaints is unreachable with one temperature for all.
In the app.Comfort of one person gives PPD with PMV.
The temperature of a standard environment (relative humidity 50 %, still air, radiant temperature equal to air, standard clothing for the activity) in which a person would have the same skin temperature and skin wettedness, and so the same heat loss, as in the real one. It comes from Gagge's two-node model of the body (core and skin), which also accounts for sweating and shivering.
For design. SET puts very different environments on one scale: a warm room with a fan and a cooler one without can have the same SET. That is how ASHRAE 55 credits air movement.
In the app. Given in Comfort of one person; used for the cooling effect. Szokolay's comfort zone leans along SET lines.
Source. Gagge, A. P., Fobelets, A. P. and Berglund, L. G. (1986), A standard predictive index of human response to the thermal environment, ASHRAE Transactions 92. Further reading: 94.
Cooling effect of elevated air speedCE, K
Above 0.1 m/s ASHRAE 55 does not feed air speed into PMV directly. It finds the drop in air and radiant temperature that, in still air, gives the same SET as the moving air; that drop is the cooling effect. PMV is then computed at the lowered temperatures and still air.
The adaptive models add a fixed allowance to their upper limit at operative temperatures of 25 °C and above: 1.2 K at 0.6 m/s, 1.8 K at 0.9 m/s and 2.2 K at 1.2 m/s.
For design. The figure to quote when arguing for fans or cross ventilation in place of cooling: a ceiling fan at about 0.8 m/s is worth around 2 to 3 K.
In the app.Comfort of one person (ASHRAE 55-2023 and the adaptive models) and the ASHRAE 55 model of the main chart.
ASHRAE 55Thermal Environmental Conditions for Human Occupancy
The American standard for indoor thermal comfort, revised every few years (the tool follows the 2023 edition). It gives two methods: PMV for occupants with near-sedentary activity (1.0 to 2.0 met) in any space, with the cooling effect for elevated air speed and limits on local discomfort; and an adaptive method for naturally conditioned spaces whose occupants can open windows.
For design. The reference for comfort claims in rating systems and for sizing the comfort zone in early design.
In the app. The ASHRAE 55 (PMV ±0.5) and Adaptive (ASHRAE 55, 80 %) comfort models of the main chart; Comfort of one person.
The international standard for PMV, PPD and local discomfort. It grades the comfort a building aims at into three categories (named A, B and C in ISO 7730; I, II and III in EN 15251 and EN 16798-1):
Category
PMV
PPD
Typical use
A (I)
−0.2 to +0.2
under 6 %
high expectations, fragile occupants
B (II)
−0.5 to +0.5
under 10 %
new buildings, normal expectations
C (III)
−0.7 to +0.7
under 15 %
existing buildings, moderate expectations
For design. A stricter category narrows the band and raises plant size and energy use; category B (II) is the usual target.
In the app.Comfort of one person with ISO 7730; EN 15251 (category II) on the main chart.
Source. ISO 7730:2005, Ergonomics of the thermal environment.
Adaptive comfort
Field studies in real buildings found that people in naturally ventilated buildings accept a wider and warmer range in summer than PMV predicts. They adapt: they change clothing, open windows, use fans, and their expectations follow the recent weather. Adaptive models therefore tie the comfortable indoor temperature to the outdoor temperature of the past days.
Tc = 0.31 Tpma + 17.8 (ASHRAE 55) Tc = 0.33 Trm + 18.8 (EN 16798-1)Tc comfort operative temperature; Tpma, Trm the prevailing or running mean outdoor temperature. ASHRAE 55 accepts ±3.5 K for 80 % and ±2.5 K for 90 % acceptability, for prevailing means of 10 to 33.5 °C.
For design. The model for free-running buildings: it shows how much of the year a well designed naturally ventilated building can be comfortable without cooling, and it is the basis of most overheating criteria.
In the app.Adaptive (ASHRAE 55, 80 %) on the main chart; the adaptive choices of Comfort of one person, which take the outdoor mean of a day of the file; A room through the year; the dry bulb glyphs of Any variable, any hours; natural ventilation potential.
Keep in mind. Only for spaces where occupants can open windows and adjust clothing, with near-sedentary activity, and no mechanical cooling in operation.
Source. de Dear, R. and Brager, G. (1998), Developing an adaptive model of thermal comfort and preference, ASHRAE Transactions 104; Nicol, F. and Humphreys, M. (2010), Derivation of the adaptive equations for thermal comfort in free-running buildings in European standard EN15251, Building and Environment 45.
Running mean outdoor temperatureTrm, prevailing mean
A mean of the daily outdoor temperatures of the past days in which yesterday counts most and each earlier day less, standing for the weather people have grown used to. EN 16798-1 weights each day back by 0.8 times the day after it, over about a week:
Trm = (T−1 + 0.8 T−2 + 0.6 T−3 + 0.5 T−4 + 0.4 T−5 + 0.3 T−6 + 0.2 T−7) / 3.8the standard's rounded form; T−n the mean of the nth day before. The tool uses the exact weights, 0.8 to the power of the days back, over seven days.
For design. After a hot spell the limit has moved up, so a building that slowly warms with the weather is judged more kindly than one that heats up suddenly.
In the app. Behind the EN 16798-1 limits of Comfort of one person and A room through the year, from the file's daily means. For ASHRAE 55, whose prevailing mean may be a simple mean of 7 to 30 days, the tool takes the plain mean of the seven days before.
The European standard for the indoor environment (replacing EN 15251). Its adaptive method gives the comfort temperature 0.33 Trm + 18.8 °C, with limits by category:
Category
Upper limit
Lower limit
I
+2 K
−3 K
II
+3 K
−4 K
III
+4 K
−5 K
It applies for running means of 10 to 30 °C for the upper limit.
For design. The criterion European overheating assessments are made against; category II is the normal target for new buildings.
In the app.Comfort of one person (EN 16798-1, adaptive) and A room through the year, which counts occupied hours over each category.
Keep in mind. Outside 10 to 30 °C the room card holds the limits at their end values and counts those hours apart; whether the lower limits apply only from 15 °C, as in EN 15251, is not verified (README).
Local discomfortdraught, vertical difference, floor, radiant asymmetry
A person can be neutral overall and still uncomfortable in one part of the body. The standards limit four causes:
Draught: cool moving air on the neck and ankles. ISO 7730's draught rate gives the share annoyed from air temperature, mean speed and turbulence.
Vertical air temperature difference between head and ankles: ASHRAE 55 allows 3 K for a seated person.
Floor temperature: 19 to 29 °C for people in light shoes.
Radiant asymmetry: the difference in radiant temperature between two opposite sides. ASHRAE 55's limits show the order: a warm wall is tolerated up to 23 K, a cool ceiling 14 K, a cool wall 10 K, and a warm ceiling only 5 K.
For design. Why large single glazing causes complaints next to it even in a warm room, why radiant ceilings are kept moderate, and why a low-level supply of cool air needs care.
In the app. A fold of Comfort of one person: ankle draught and vertical difference after ASHRAE 55, draught rate and vertical difference after ISO 7730, floor temperature and the four radiant asymmetries.
Keep in mind. Beyond the standard's ranges the differences are held at the limit; ISO 7730's categories' allowed shares were not checked (README).
Sun through a window falls on part of the body and adds to its radiant gain. SolarCal works out the absorbed sun from the sun's altitude and direction relative to the person, the projected area of the body, the window's transmittance and the share of the sky the person sees, and turns it into an increase of mean radiant temperature.
For design. Direct sun on a seated person can add several kelvin, in strong sun over 10 K, of radiant temperature, enough to undo a well tuned air temperature. It argues for shading or for keeping desks out of the sun patch.
In the app. A fold of Comfort of one person, with the sun's position and direct normal irradiance taken from an hour of the file; its rise of MRT can be added to the inputs.
Keep in mind. The diffuse sky is taken as 0.2 of the direct normal irradiance, as the model does, whatever the file's diffuse.
Source. Arens, E. et al. (2015), Modeling the comfort effects of short-wave solar radiation indoors, Building and Environment 88.
Bioclimatic design strategies
How a climate's hours are sorted into those that are comfortable as they are, those a passive measure can make comfortable, and those that need a machine. These readings come before any building is drawn, and they set the brief for form, envelope and openings.
Bioclimatic design
Design that uses the local climate to keep people comfortable with as little bought energy as possible. The building's form, orientation, mass, openings and shading do the work first; heating and cooling machines cover only what is left. The approach goes back to Victor Olgyay's Design with Climate (1963) and Baruch Givoni's Man, Climate and Architecture (1969), which both read the climate on a chart of temperature and humidity.
The method is a sequence: describe the climate, find when people would be uncomfortable and why (too cold, too hot, too humid, too dry), then pick the strategies that move those hours into comfort.
For design. Read the strategy shares before sketching. A climate where natural ventilation carries most warm hours asks for shallow plans and cross openings; one where thermal mass and night cooling dominate asks for heavy, well shaded rooms that can be flushed at night.
In the app. The whole Comfort and strategies part: the psychrometric chart with its zones, When each strategy is needed, Strategies and Design guidelines. This climate in short summarises the result.
A psychrometric chart with zones drawn on it: a comfort zone in the middle, and around it the range of outdoor air that each passive strategy can still bring to comfort indoors. An hour of the weather file is a point on the chart; the zone it falls in names the strategy it needs. Givoni drew the zones from the indoor temperatures that buildings of each type reach for a given outdoor state.
The tool uses fourteen numbered zones: conventional heating (1), active solar heating (2), passive solar heating (3), internal gains heating (4), solar protection (5), humidification (6), comfort (7), evaporative cooling (8), natural ventilation cooling (9), high thermal mass cooling (10), high thermal mass with night cooling (11), air conditioning (12), air conditioning and dehumidification (13) and dehumidification (14).
For design. The shape of the cloud of hours matters more than any single number: a cloud lying low and to the right (hot, dry) favours mass and evaporation; a cloud lying high and to the right (hot, humid) leaves ventilation and dehumidification.
In the app. The main chart (menu Show: Strategies) with the menu Zones set to Givoni (fixed). The zone outlines follow a published drawing of Givoni's chart and are drawn at the site's pressure.
Keep in mind. The zones are fixed for every climate and are not those of other programs, whose limits differ, so the hour counts will not match them one for one. The zones describe what a well designed building can do; a poorly designed one does less.
Source. Givoni, B. (1969), Man, Climate and Architecture; Givoni, B. (1992), Comfort, climate analysis and building design guidelines, Energy and Buildings 18(1). Further reading: 1, 91.
Comfort zone
The range of air states, drawn on the psychrometric chart, in which most people in light clothing and at rest feel neither too warm nor too cold. In Givoni's chart it spans about 20 to 27 °C, between 20 % and 80 % relative humidity, narrowing at the warm end where high humidity makes heat harder to lose.
Other definitions give other zones: the PMV zone of ASHRAE 55 depends on clothing, activity and air speed; the adaptive zone moves with the outdoor temperature of recent days; Szokolay's zone is centred on each month's neutral temperature.
For design. The hours in the comfort zone need nothing from the building but to not spoil them: no overheating by sun, no draughts. A large share of comfortable hours is a reason to keep the building open to the outdoors.
In the app. The menu Comfort above the chart chooses the model: Givoni comfort zone, ASHRAE 55 (PMV ±0.5), EN 15251 (category II) or Adaptive (ASHRAE 55, 80 %). The tile of comfortable hours and every strategy count follow it.
Keep in mind. The chart places outdoor air. Calling an outdoor hour comfortable means that air, brought indoors unchanged, would be comfortable; it says nothing about sun or wind on a person outdoors (for that see UTCI).
Heating a building with the sun that enters through its windows and is stored in its floors and walls, with no pumps or fans. It works when the air is cool (in Givoni's chart from about 8 to 20 °C) and there is sun to collect. The three parts are collection (glazing facing the equator), storage (exposed mass in the sunlit rooms) and retention (insulation and airtightness, often with night insulation of the glass).
For design. Face most glazing within about 30° of the equator, keep it free of winter shade (neighbours, trees, overhangs too deep), and put dense floors or walls where the low sun falls. Too much glass without mass overheats on sunny winter days and loses heat at night.
In the app. Strategy 3 on the chart and in Strategies. With daily checks on, it counts only on days with at least 1 kWh/m² of sun on the horizontal. Orientation and form: sun on the facades and Solar energy by tilt and direction show how much winter sun each facade gets.
The heat given off by people, lighting and equipment is enough to keep a well insulated, airtight building comfortable when the outdoor air is only a little below comfort, about 12.5 to 20 °C in Givoni's chart. A seated person gives off roughly 100 W; offices with many people and computers can need cooling at outdoor temperatures well below 20 °C for this reason.
For design. Keep the heat in: a compact form, continuous insulation, few thermal bridges, an airtight envelope and heat recovery on the exhaust air. In dense, busy buildings the same gains turn into a cooling load, so plan for both.
In the app. Strategy 4 on the chart and in Strategies.
Keeping the sun off glazing, and off people, when the air is already warm. It does not make an hour comfortable by itself, so the tool counts it as a support: it is needed on top of whatever strategy carries the hour. The tool counts hours of at least 20 °C with at least 120 W/m² of sun on the horizontal (this tool's choice of thresholds).
For design. Shade outside the glass where possible: external devices stop most of the sun's heat before it enters, internal blinds much less. Overhangs suit the facade facing the equator; east and west need vertical fins, movable shading or fewer windows, since the sun there is low.
In the app. Strategy 5 and its tile; the Sun and sky part sizes the devices: Shading benefit by orientation, Optimal shading angles by orientation, Test a window and its shading.
Air moving across the body raises the rate at which it loses heat by convection and by evaporating sweat, so a warm room feels cooler. Givoni's zone extends comfort to about 32 °C at moderate humidity, provided the building gives an appreciable indoor air speed. It also flushes heat out of the building when the outdoor air is cooler than the indoor.
For design. Openings on opposite walls, inlets towards the prevailing wind of the warm months, shallow rooms (cross ventilation works to about five times the ceiling height), and high outlets for stack ventilation on still days. Ceiling fans give the same air speed with little energy.
In the app. Strategy 9; the Wind and air part shows whether the wind and the hours suit it: Wind by season and by time of day, Natural ventilation potential by month and Natural ventilation potential, hour by hour.
Keep in mind. The zone needs air movement that the building must provide; a still day with closed windows gives none. Noise, pollution, insects and security often close windows in practice.
Heavy walls, floors and roofs absorb heat during the day and release it later, so indoor temperatures swing less than outdoor ones and the indoor peak arrives hours after the outdoor peak. It works where days are hot but nights are cool, that is, where the daily range is large, typically in dry climates. Givoni's zone reaches about 37 °C at low humidity.
For design. The mass must be exposed to the room (no suspended ceilings or raised floors over it), shaded from the sun outside, and insulated on its outer face in climates with cold seasons. A daily range under about 8 K gives mass little to work with.
In the app. Strategy 10; the Thermal mass card in the Building envelope part reads the daily swing and the time lag of a chosen wall.
Thermal mass with night ventilationzone 11, night flushing
The building's mass is cooled by opening it to the cool night air, often with fans or stacks, and then closed during the day so the cooled mass absorbs the day's heat. It extends the mass zone to hotter days (to about 44 °C at low humidity in Givoni's chart), but only where nights are cool enough.
For design. Night openings that are secure and rain proof, a clear air path past the exposed mass, and a daytime regime of closed windows and shaded glass. In humid climates night air may be too moist to be welcome.
In the app. Strategy 11. With daily checks on, an hour counts only if its day's minimum is at most 20 °C (this tool's threshold). A room through the year shows what night ventilation does to the overheating hours of one room.
Evaporating water takes heat from the air: about 2.45 MJ for each kilogram of water. In direct evaporative cooling the air passes over water (wetted pads, fountains, pools) and becomes cooler and more humid, moving along a line of constant wet bulb temperature on the chart. In indirect cooling a heat exchanger cools the supply air without adding moisture to it.
It works only with dry air: the drier the air, the lower its wet bulb temperature and the more it can be cooled. Givoni's zone reaches about 41 °C at low humidity.
For design. Courtyards with water and planting, wind catchers with wetted surfaces, evaporative coolers. It needs a water supply and works poorly in humid air, where the wet bulb temperature is close to the dry bulb.
In the app. Strategy 8 on the chart; draw a process along a wet bulb line to see the temperature drop.
The hours that neither comfort nor a passive strategy covers need a machine. The tool names the measure from where the hour lies. Below 20 °C: conventional heating, or active solar heating (solar collectors and storage) for hours of 4 to 8 °C on sunny days. From 20 to 27 °C: dehumidification when the air is over 80 % relative humidity or above the comfort zone's humidity, humidification under 20 %, otherwise heating under 23.5 °C and air conditioning above. Over 27 °C: air conditioning, with dehumidification when the air is also humid.
For design. These hours size the plant and drive the energy bill. A design that adds passive hours shrinks them; the remaining share is the honest answer to "how much will the systems run".
In the app. The active rows of Strategies, coloured in When each strategy is needed. The Technologies part shows which efficient systems suit the climate.
Keep in mind. The thresholds for naming the active measure are this tool's rules; they sort the leftover hours by kind and are not a load calculation.
Strategies overlap: an hour may lie in the zones of both natural ventilation and thermal mass. The tool picks a set in order: first the passive strategy that makes most of the uncomfortable hours comfortable, then, among the hours still left, the one that adds most, and so on until the next would add under 0.5 % of the hours (this tool's threshold). What is left goes to an active measure.
For design. The first two or three strategies usually carry almost all the passive hours; they are the ones to design around. A strategy that adds little after the others is not worth compromising the plan for.
In the app.Strategies lists the hours each strategy adds in order. Tick strategies by hand to try another set; Back to the automatic selection restores it. When each strategy is needed shows the year, hour of day against day of year, in the colour of the strategy used.
A chart of single hours cannot see what a day is like. Two strategies depend on the day: passive solar heating needs a sunny day to collect from, and night cooling needs a cool night. With the checks on, the tool counts passive solar heating (and names active solar heating) only on days with at least 1 kWh/m² of sun on the horizontal, and mass with night cooling only on days whose minimum is at most 20 °C. Both thresholds are this tool's choice.
For design. Compare the strategy shares with the checks on and off: a large drop means the strategy depends on days the climate rarely gives (cloudy winters, warm nights).
In the app. The switch Daily checks above the chart. It does not apply with Szokolay's zones, whose zones are already sized from each month's sun and swing.
The temperature at which people of a place feel neither warm nor cool. Field studies show it follows the climate people are used to: it is higher in warm climates and seasons. Szokolay uses Auliciems' relation with the month's mean outdoor temperature:
Tn = 17.6 + 0.31 × To,av where To,av is the month's mean outdoor temperature (°C); comfort is taken as Tn ± 2.5 K.
For design. In a hot month the comfort target is several kelvin above the target of a cool month, which makes free-running buildings more feasible than a fixed 20 to 26 °C band suggests.
In the app. Behind Szokolay's zones and the shading criterion Neutrality (Szokolay). The adaptive models use a similar idea with other coefficients.
Source. Szokolay, S. V. (2004), Introduction to Architectural Science, Section 1.2.4 and method sheet M.1.7. Further reading: 3.
Szokolay's control potential zonesCPZ
A version of the bioclimatic chart in which the zones are drawn for each month of the climate rather than fixed. The comfort zone is centred on the month's neutral temperature (±2.5 K along lines of equal standard effective temperature, between 4 and 12 g/kg). Each control potential zone extends it by what a strategy can achieve that month:
passive solar heating: down by an amount proportional to the month's mean daily sun on a vertical wall facing the equator;
thermal mass: up by 0.3 of the month's mean daily swing, or 0.6 with night ventilation;
air movement: up by a function of the effective air speed (0.8 m/s by default);
evaporative cooling: along the wet bulb lines from the comfort zone's corners.
For design. The zones grow and shrink with the climate's own sun and swing, so a cloudy climate gets a small solar zone and a humid one a small mass zone. It is a sharper reading than the fixed chart for climates far from where Givoni's zones were drawn.
In the app. Menu Zones: Szokolay (per month). The chart shows the zones of the warmest and coldest month chosen; each hour is counted against its own month's zones. The Comfort menu is disabled while it is chosen.
Keep in mind. Some edges of the zones are this tool's reading of the method sheet, as the README lists. There is no internal gains zone. Checked against the book's worked figures for Brisbane, Canberra, Cloncurry and Mombasa.
Source. Szokolay, S. V. (2004), Introduction to Architectural Science, Section 1.5.1 and method sheet M.1.7. Further reading: 3.
Olgyay's bioclimatic chart and bioclimatic potential
Olgyay's chart (1963) has relative humidity on the horizontal axis and temperature on the vertical, with a comfort zone in the middle and lines showing how much sun, wind or moisture would restore comfort outside it. Košir and Pajek (2017) turned it into a monthly method: each month is drawn as a line from its mean daily minimum temperature (at the month's highest humidity) to its mean daily maximum (at its lowest humidity), and the share of the line in each zone is the share of the month needing that measure.
The zones: mechanical cooling, hot and arid measures, ventilation, thermal mass, comfort with shading, comfort with the sun, passive solar heating and conventional heating. A modified reading lets the sun on the body shift the comfort zone downwards in cool months.
For design. A quick monthly profile: which months a building can be open, which need sun let in and which need it kept out.
In the app.Bioclimatic potential, in the Comfort and strategies part; it follows the scenarios.
Keep in mind. Monthly means, not hours. The method applies to light activity, indoor clothing and sites below about 300 m. The zone outlines were digitised from the paper's figure (to about 0.3 K and 1 % RH).
Source. Olgyay, V. (1963), Design with Climate; Košir, M. and Pajek, L. (2017), BcChart v2.0: a tool for bioclimatic potential evaluation, ISES Solar World Congress, doi:10.18086/swc.2017.21.04. Further reading: 4.
Mahoney tables
A set of worksheets published for the United Nations (Koenigsberger, Mahoney and Evans, 1971) that turn twelve months of temperature, humidity, rain and wind into sketch design advice. Each month is judged day and night against comfort limits that depend on its humidity group; the results are counted as six indicators, each the number of months it holds:
H1 air movement essential, H2 air movement desirable, H3 rain protection needed;
The indicator totals then select recommendations on layout, spacing, air movement, openings, walls, roofs and outdoor spaces.
For design. A fast first answer to "long and thin or compact, heavy or light, big or small windows". Example: A1 in 3 or more months gives heavy external and internal walls, and in 6 or more heavy roofs with over 8 hours of time lag; H1 in 3 or more months gives single banked rooms with permanent provision for air movement.
In the app.Building form for this climate: the indicators, the recommendations, and drawings of building archetypes that fit them.
Keep in mind. Written for tropical housing. The tables' numbers were taken from an open implementation that cites the manual and were not compared with the book itself.
Source. Koenigsberger, O. H., Mahoney, C. and Evans, M. (1971), Climate and House Design, United Nations; Koenigsberger, O. H., Ingersoll, T. G., Mayhew, A. and Szokolay, S. V. (1974), Manual of Tropical Housing and Building. Further reading: 2.
Building form and archetypes
Climates have produced recognisable building types: the courtyard house of hot dry regions, the raised, open, light house of the humid tropics, the compact, heavy, small-windowed house of cold ones, the earth-sheltered house. An archetype is such a type reduced to its climatic principle: compactness, mass, openness, shading, contact with the ground.
For design. Compactness (little envelope for the floor area) saves heat in cold climates; openness and shallow plans let air through in humid ones; courtyards give shade and a cool reservoir of night air in dry ones.
In the app.Building form for this climate scores archetypes against the Mahoney indicators; Forms compared by simulation (a pilot) shows simulated results of house archetypes and prototype buildings for the studied climate nearest to the file.
Keep in mind. The archetype fit scores are this tool's own judgement; the simulation study is meant to replace them. The simulations use a code envelope only and the nearest studied climate, not the file's.
Short design measures at five scales (district, site, building, building parts, systems), each with a small drawing, ranked by how much the climate calls for them. The ranking reads the weather file: shares of cold, warm, humid and dry hours, winter sun, daily range, wind and the strategies' hours.
For design. A checklist to bring to the first sketch, ordered by the climate. Read the top of the list as the measures with the most leverage, and check each against the site.
In the app.Rules of thumb for this climate (119 measures; Show all lists every one) and Design guidelines, which gives the guidance for each selected strategy.
Keep in mind. The texts, drawings and scores are this tool's own: simple readings, not a simulation of a building.
Where the sun stands at each hour, how much energy it brings to a surface, and what the sky is like. These readings decide orientation, the size and kind of shading, and how much sun a facade or a solar panel can collect.
Solar time and the equation of time
Solar time is the time by the sun: at 12:00 solar time the sun crosses the meridian and stands highest. Clock time differs from it for two reasons. First, a time zone spans about 15° of longitude, so a site east or west of the zone's meridian sees noon earlier or later, by 4 minutes per degree. Second, the Earth's tilted axis and elliptical orbit make the sun run fast or slow through the year by up to about 16 minutes; this is the equation of time (EoT).
solar time = clock time + (4 (L − 15 Z) + EoT) / 60 where L is the longitude (degrees, east positive), Z the time zone in hours from UTC, EoT in minutes; a weather file keeps standard time all year, with no summer time. The tool takes EoT = 9.87 sin 2B − 7.53 cos B − 1.5 sin B, with B = 360 (n − 81) / 364 degrees and n the day of the year.
For design. Shading is designed in solar time: a device symmetric about solar noon is symmetric about the sun's path. In the far west of a wide time zone, solar noon can fall after 13:00 clock time, and with summer time later still.
In the app. Every sun position in the tool is worked out from the hour of the file in this way. Sun angles for this site works in solar time; the sun path on the map uses local mean solar time, so the equation of time appears as the figure-eight of the analemma.
The latitude at which the sun stands overhead at noon. Because the Earth's axis is tilted by about 23.45°, the declination swings from +23.45° at the June solstice (sun overhead on the Tropic of Cancer) through 0° at the equinoxes to −23.45° at the December solstice. The tool uses Cooper's approximation:
δ = 23.45° × sin(360° × (284 + n) / 365) where n is the day of the year (1 on 1 January).
For design. Declination and latitude together give the noon sun: its altitude is 90° − |latitude − δ|. At 45° N that is about 68° at midsummer, 45° at the equinoxes and 22° at midwinter: the three numbers that size an overhang.
In the app.Noon sun at the solstices and equinoxes: the overhang section and Sun angles for this site.
Source. Cooper, P. I. (1969), The absorption of radiation in solar stills, Solar Energy 12(3).
Solar altitude and azimuth
Two angles fix the sun in the sky. The altitudeα is its height above the horizon (0° at sunrise and sunset, 90° overhead); its complement is the zenith angle. The azimuth is its compass direction, measured in this tool from north, clockwise (90° east, 180° south, 270° west). Both follow from the latitude φ, the declinationδ and the hour angle H (15° per hour from solar noon):
sin α = sin φ sin δ + cos φ cos δ cos H
Left, the altitude in section; right, the azimuth in plan (here about 135°, south-east).
For design. Altitude decides how deep sun reaches into a room and how long shadows are (a shadow is 1 / tan α times the height of what casts it); azimuth decides which facades it strikes.
In the app. The Sun path card, Sun angles for this site, and the sun path on the map (header button Sun path on the map), whose key gives the sun's height, azimuth and shadow direction for the chosen moment.
Keep in mind. Azimuth conventions differ between sources: some measure from south, some anticlockwise. Check which before taking a number from another table.
Sun path diagramstereographic, orthographic, cylindrical
A drawing of the sun's track across the sky through the day, for chosen days of the year, with the hours marked. Two kinds of drawing are common. The cylindrical diagram plots altitude against azimuth like a panorama unrolled. The polar diagram shows the sky dome seen from above, north up and the zenith in the middle; in an orthographic projection the rings of equal altitude crowd towards the horizon, in a stereographic projection they are spaced more evenly, giving the low sun near sunrise and sunset more room.
For design. Lay a shading mask or the outline of neighbours over the diagram to read when a window is in sun. Colour the hours by temperature to see which sun is wanted and which is not.
In the app.Sun path: the flat view on the left, the round view on the right (menu: orthographic or stereographic), every hour of the file with the sun up as a dot. Colours: air temperature, radiation benefit, the shading period (Szokolay) or the cumulative sky. The header button Sun path on the map draws a stereographic diagram around any point of the map.
The figure traced by the sun when it is observed at the same clock time every day for a year. Its height range comes from the declination; its sideways spread comes from the equation of time, which makes the sun early or late. With clock time in local mean solar time the result is a figure-eight.
For design. The analemma of a working hour shows the full range of sun positions at that hour across the year: useful for a device that must shade one hour, say a window used at 15:00, through all seasons.
In the app. The sun path on the map draws the analemma of each whole hour, the chosen hour solid; the play button runs the sun along it through the year.
Irradiance is the power of sunlight falling on a surface, in W/m². A weather file gives three components each hour. Direct normal irradiance (DNI) is the beam from the sun's disc, measured on a surface facing the sun. Diffuse horizontal irradiance (DHI) is the light scattered by the sky and clouds, falling on a horizontal surface from the whole dome. Global horizontal irradiance (GHI) is all of it on the horizontal:
GHI = DHI + DNI × sin α where α is the solar altitude.
A clear noon in summer gives a GHI of roughly 800 to 1,000 W/m²; an overcast day can give under 100 W/m², almost all diffuse.
For design. Direct sun can be blocked by a shading device; diffuse light comes from the whole sky and mostly cannot. In cloudy climates shading saves less and daylight depends on the sky, not the sun.
In the app.Sun on a horizontal surface and the mean day of each month (diffuse and direct stacked to the global, the direct normal as a dashed line). Any variable, any hours shows each component hour by hour.
Irradiance summed over time: the energy a square metre receives in a day, a month or a year. A year's global horizontal irradiation is about 900 to 1,100 kWh/m² in northern Europe, around 1,300 to 1,900 in southern Europe and above 2,000 in deserts. One hour at 1,000 W/m² gives 1 kWh/m².
For design. The yearly sum on a surface sets what a solar panel or collector can produce; the monthly sums on facades set winter gains and summer loads. The daily sum tells whether a day was sunny enough for passive solar heating.
In the app. Hover a month in the mean day chart for its kWh/m²; Solar energy by tilt and direction gives yearly sums on tilted surfaces. The header button Solar on the map shows the yearly global horizontal irradiation over land from the Global Solar Atlas.
A surface that is not horizontal receives three parts: the beam, at the angle of incidence θ between the sun and the surface's normal; the share of the sky it sees; and light reflected from the ground. In the simplest model the sky is equally bright everywhere (isotropic):
It = DNI cos θ + DHI (1 + cos β) / 2 + ρ GHI (1 − cos β) / 2 where β is the tilt from the horizontal (90° for a wall) and ρ the ground's reflectance (albedo); a wall sees half the sky and half the ground.
For design. At mid latitudes a wall facing the equator receives more sun in winter than in summer, because the winter sun is low; east and west walls receive most in summer. This is why equator-facing glazing suits passive heating and east and west glazing overheats.
In the app.Orientation and form: sun on the facades, Shading benefit by orientation, Glazing by facade and Solar energy by tilt and direction all use this model with a ground albedo of 0.2.
Keep in mind. The real sky is brighter near the sun and the horizon, so the isotropic model slightly underestimates surfaces facing the sun. Yearly sums agree with a reference isotropic calculation within 0.5 % on six orientations.
How clear the sky is can be read from the sun data alone. The Perez clearness index compares the direct and diffuse light, corrected for the sun's height:
ε = ((DHI + DNI) / DHI + 1.041 Z3) / (1 + 1.041 Z3) where Z is the solar zenith angle in radians. ε is 1 under a fully overcast sky and rises above 6 under a very clear one.
Perez et al. divide ε into eight bins. The tool groups them in three: overcast under 1.23, intermediate, and clear from 4.5 (this tool's grouping), for each daylight hour with the sun over 5°.
For design. A climate of mostly overcast skies is designed for the overcast sky: tall windows, light from above, high reflectances. A clear sky climate is designed for the sun: control of glare and heat, light redirected deep into rooms.
In the app.Sky type (share of each type by month) and What the sky asks of daylighting, whose advice is this tool's (overcast from 50 % of the hours, clear from 40 %).
Source. Perez, R., Ineichen, P., Seals, R., Michalsky, J. and Stewart, R. (1990), Modeling daylight availability and irradiance components from direct and global irradiance, Solar Energy 44(5). Further reading: 26.
Sky covercloud cover, tenths
The share of the sky hidden by cloud, recorded in tenths (0 clear, 10 fully covered). Weather files hold two values: total sky cover, all cloud, and opaque sky cover, cloud thick enough to hide the sky behind it. Many files fill these fields from models or nearby stations, and some leave them empty.
For design. Cloud lowers the sun's heat by day and raises the sky's long-wave radiation at night, so cloudy nights are warmer and sky cooling works poorly under them.
In the app.Sky cover, in the Sun and sky part.
Keep in mind. A typical year for a point from PVGIS has no sky cover; a measured year takes total cloud from ERA5 as opaque cover too.
The sun of many hours summed onto patches of the sky dome, so that each patch shows how much energy came from that part of the sky over a period. A common division is Tregenza's 145 patches, in bands of 12° of altitude. The tool sums only the direct sun of the hours shown and colours each patch by welcome less unwelcome sun (the page's 20 and 27 °C bands).
For design. It shows at a glance which parts of the sky a shading device should block and which it should leave open: a device that hides the red patches and leaves the blue is doing its job.
In the app.Sun path, colour cumulative sky (145 patches).
Keep in mind. Diffuse light is left out, so it shows the sun worth having or keeping out, not the sky's whole radiance. The patches here start at north, which may differ from Tregenza's own offsets.
Source. Tregenza, P. R. (1987), Subdivision of the sky hemisphere for luminance measurements, Lighting Research & Technology 19(1).
Sunshine hours
The hours in which the sun is bright enough to cast a clear shadow. The World Meteorological Organization defines a sunshine hour as one with direct normal irradiance of at least 120 W/m². Sunshine hours are counted in a year, a month or a day, and often given as a share of the hours the sun could shine.
For design. Planning guidance on sun in gardens and on facades is set in sunshine hours, and they show how much a neighbour's building takes away. A climate with few sunshine hours in winter gives little to passive solar heating.
In the app.A building opposite (sun hours lost by month) and The site and its neighbours (sunshine hours of the open ground and of facades) count hours of DNI of at least 120 W/m² from the file.
The same sun is a gain on a cold day and a burden on a hot one. Radiation benefit colours each hour's direct sun by whether it is wanted: welcome when the air is under 20 °C, unwelcome over 27 °C and neutral between (the page's own temperature bands, not a building's balance temperature).
For design. On a sun path coloured this way, a shading device should cover the unwelcome hours and leave the welcome ones open. Where the two overlap in the sky (spring and autumn share the same sun positions), a fixed device must compromise; a movable one need not.
In the app.Sun path, colour radiation benefit; the scale ends at 800 W/m² for every climate.
Keep in mind. It is the sun on a surface facing the sun, not on a particular facade, and it judges by outdoor air only.
How much sun each facade direction receives over the year, and when. At mid latitudes the facade facing the equator gets much winter sun and, with the high summer sun, little in summer; east and west facades get strong, low sun on summer mornings and afternoons; the facade facing the pole gets almost only diffuse light.
For design. In heating climates lay the long axis east to west and put the main glazing towards the equator; in hot climates the same orientation is easiest to shade. Keep east and west glazing small or well shaded. A slab turned some 20° from the ideal loses little.
In the app.Orientation and form: sun on the facades (sixteen facade directions and the best way to turn a slab) and Glazing by facade.
Two angles describe the sun as seen from a window. The horizontal shadow angle (HSA) is the difference in plan between the sun's azimuth and the direction the facade faces; when it exceeds 90° the sun is behind the facade. The vertical shadow angle (VSA, also called the profile angle) is the sun's altitude projected onto a section through the window, perpendicular to the facade:
tan VSA = tan α / cos HSA where α is the solar altitude.
Section through a window and an overhang: the overhang shades the whole glass while the sun's profile angle is above the angle from the sill to the overhang's edge.
An overhang of depth d above glass of height h shades it fully while tan VSA ≥ h / d; so d = h / tan VSA for a chosen cut-off angle. Vertical fins work in plan with the HSA in the same way.
For design. Choose the cut-off VSA from the sun positions you want to block (summer, warm hours) and to admit (winter); then the depth follows. On a facade facing the equator the VSA at noon is high in summer, so a modest overhang works; on east and west facades the sun is low and the VSA small, so overhangs would need to be impractically deep.
In the app.Optimal shading angles by orientation gives the cut-off angles, Noon sun at the solstices and equinoxes: the overhang section draws the section, and Sun angles for this site answers the classic angle questions for the latitude.
Fixed external shading devices. An overhang (a horizontal projection over the window) cuts off high sun; vertical fins beside the window cut off sun from the side; louvres (many slim blades, horizontal or vertical) do the same job as a deep device with less depth, at some cost to view and daylight. An egg-crate combines both directions.
For design. Overhangs for the facade facing the equator; fins, or fins with an overhang, for facades turned well towards east or west; movable devices where the same sun position is wanted in spring and unwanted in autumn. Extend an overhang beyond the window's sides, or the sun slips round its ends in the morning and afternoon.
In the app.Test a window and its shading places one window with an overhang, fins and louvres sized by the user, draws its shadow at any day and hour, and tests it against every hour of the file: a month by hour map of the glass shaded, crossed with hot and cool hours, and the monthly solar gain with and without the device.
Keep in mind. The tester takes the device as not cutting diffuse light, so its saving is understated a little; frames, reveals and the change of g-value with angle are left out.
The part of the sky that a shading device hides from a point on the window, drawn on a sun path diagram. An overhang gives a segment-shaped mask bounded by its VSA; fins give a mask bounded by lines of constant HSA. Every sun position inside the mask is a time when the point is shaded.
For design. Draw the mask over the sun path coloured by temperature: a good device's mask covers the hot hours and leaves out the cold ones. Comparing masks is quicker than comparing devices.
In the app.Sun path, the mask menu (N to NW): the mask of the optimal practical overhang and fins for that facade, by the criterion chosen in Shading benefit by orientation.
To size fixed shading one must first say when shade is wanted. The tool offers five criteria, each read from the weather file:
outdoor temperature: shade over 27 °C, keep the sun under 20 °C;
degree days: shade through the cooling period, set evenly about midsummer;
sun angle: shade from the spring to the autumn equinox;
solar gain: shade when the facade receives over 250 W/m²;
neutrality (Szokolay): shade over the warmest month's neutral temperature, likely over the coldest month's, and keep the sun below the lowest lower comfort limit.
For design. The criterion changes the answer: the equinox rule ignores the climate, while the temperature rules follow it. Try two and see whether the recommended angle moves.
In the app. The menu of Shading benefit by orientation; the sun path colour shading period (Szokolay).
Keep in mind. The cut-off angles are those that shade most of the unwanted sun for least of the wanted sun, counting a device as shading wholly above its cut-off and not at all below: a rule of thumb of this tool. In the Szokolay criterion, an hour between the likely limit and the must counts as half a reason to shade (this tool's weighting).
Source. Szokolay, S. V. (2004), Introduction to Architectural Science, method sheet M.1.9. Further reading: 3.
Solar energy by tilt and direction
The yearly sun on a surface depends on its tilt and the direction it faces. For a fixed panel the most energy comes from facing the equator at a tilt a little less than the latitude, since the summer half of the year carries more sun; but the curve is flat near the optimum, and a panel a few tens of degrees off still collects most of it. Vertical facades collect much less, especially at low latitudes where the sun is high.
For design. Roofs that face east and west at a low tilt lose only a little against the optimum and allow more panels; facade panels pay off better at high latitudes and where winter yield matters.
In the app.Solar energy by tilt and direction (a table of yearly irradiation from the file). The header buttons PV on the map and PV tilt on the map show the yearly yield of a free-standing plant at the optimum tilt and that tilt, from the Global Solar Atlas.
Keep in mind. The map's PV figures are for a free-standing plant in open ground; facades and roofs with panels flat on them give less, as do panels shaded by neighbours.
How much daylight reaches a room, judged by the light of the sky rather than the sun, and how windows, room depth and the buildings opposite change it. The weather file sets the targets; the room's geometry decides whether they are met.
Illuminance and luminous efficacyE, lx; lm/W
Illuminance is the visible light falling on a surface, in lux (lumens per square metre). It weighs radiation by the sensitivity of the eye, so it is a different quantity from irradiance in W/m². An overcast sky gives about 5,000 to 20,000 lx on the open ground, direct summer sun around 100,000 lx; a desk needs some 300 to 500 lx.
Luminous efficacy converts one to the other: lumens per watt of solar radiation. Diffuse skylight has a higher efficacy than the direct beam, and both vary with cloud, water vapour and sun height.
For design. Daylight design works in illuminance: the ratio of indoor to outdoor illuminance (the daylight factor) only becomes a number of lux, and so a judgement of a room, once the outdoor illuminance of the climate is known.
In the app.Daylight outdoors (Sun and sky) gives, month by month, the share of working hours (8 to 18 h) with at least 10,000 and 20,000 lx of global illuminance outdoors. The EN 17037 targets of Daylight factor indoors start from the file's diffuse illuminance.
Keep in mind. Some files hold no illuminance (PVGIS and measured ERA5 years). The tool then estimates diffuse illuminance at 120 lm/W of diffuse irradiance: an assumption.
A model of the brightness of a fully overcast sky, adopted by the CIE: the sky is brightest at the zenith and three times as bright there as at the horizon, and its luminance does not depend on direction (azimuth). The sun plays no part.
L(θ) = Lz (1 + 2 sin θ) / 3where Lz is the zenith luminance and θ the altitude above the horizon.
For design. Because it has no sun and no direction, a daylight factor under this sky is the same for every orientation and every hour: a fixed property of the room. That is why daylight rules use it; the sun is judged separately, by sunshine hours or sun patches.
In the app. The daylight factor grid of Daylight factor indoors, the VSC of A building opposite and the facades of The site and its neighbours all integrate this sky. A building opposite also draws the sky from the window shaded by its weight.
The illuminance at a point inside a room as a percentage of the illuminance on an unobstructed horizontal plane outdoors at the same moment, under the CIE overcast sky. A daylight factor of 2 % means that when the open ground receives 15,000 lx, the point receives 300 lx.
It falls steeply with distance from the window: high near the glass, low at the back. It has three parts: the sky seen directly through the window, light reflected from outside surfaces, and light reflected inside the room (see the split flux method).
The three components of the daylight factor at a point on the work plane.
For design. Rough readings: below 2 % a room looks gloomy and needs electric light most of the day; 2 to 5 % is well daylit; above 5 % there is plenty of light but often glare and heat from large glazing. Room depth, window head height, glass transmittance and obstructions move it most.
In the app.Daylight factor indoors (Sun and sky) gives the targets for the climate and checks a room lit from one wall: a grid of daylight factors on the work plane, drawn in an isometric view of the room.
Keep in mind. The tool's grid is its own split-flux estimate, not a simulation: the window reveal is ignored, glass transmittance is taken at normal incidence for every angle, and the obstruction is a long parallel block given a fifth of the sky's luminance.
The diffuse horizontal illuminance outdoors that is exceeded for half of the daylight hours of the year. EN 17037 uses it to turn a target illuminance indoors into a target daylight factor for a given climate: D = Etarget / Ev,d,med.
Bright climates have a high median, so the same indoor target needs a lower daylight factor. Northern Europe's capitals lie around 13,000 to 15,000 lx; southern ones higher.
For design. A room that is well daylit in Athens may be under-lit with the same window in Helsinki. The median ties the window size to the climate.
In the app.Daylight factor indoors computes it from the file (the median over the 4,380 hours with the most diffuse light) or takes the value of EN 17037's Table A.3 for the capital of the file's country.
Keep in mind. On IWEC files the tool reproduces the standard's table (Helsinki 13,550 against 13,500 lx; Dublin equal). TMYx files give 15 to 20 % less for the same cities, so their targets come out higher.
Source. EN 17037:2018 Daylight in buildings, Annex A.
EN 17037Daylight in buildings
The European standard for daylight in buildings. For daylight provision it asks a target illuminance over a share of the reference plane (the work plane, 0.85 m above the floor, less a margin along the walls) for at least half of the daylight hours. Three levels are given:
Level
Over 50 % of the plane
Over 95 % of the plane
Minimum
300 lx
100 lx
Medium
500 lx
300 lx
High
750 lx
500 lx
Divided by the climate's Ev,d,med, each illuminance becomes a target daylight factor. The standard also covers view out, exposure to sunlight and glare, not judged here.
For design. The 50 % target is about the room's general brightness; the 95 % target catches the dark back corners. A deep room usually fails the second first.
In the app.Daylight factor indoors gives the target daylight factors for the climate and judges the room's grid on its median and its 5th percentile, as the standard asks.
Keep in mind. The levels are taken as secondary sources give them (checked 2026-10-29). The margin along the walls is 0.5 m here; UK practice reports use 0.3 m.
A hand method that splits the daylight factor at a point into three parts and adds them:
Sky component: light from the patch of sky seen through the window.
Externally reflected component: light from obstructions outside, such as a building opposite.
Internally reflected component: light that has bounced off the room's floor, walls and ceiling; spread fairly evenly over the room.
Near the window the sky component dominates; at the back of a deep room the internally reflected part may be most of what is left, which is why pale surfaces matter.
For design. Raising the window head adds sky component deep in the room; pale finishes raise the reflected component everywhere; a tall building opposite cuts the sky component and gives back only a fraction as reflected light.
In the app. The grid of Daylight factor indoors: the sky component integrated exactly in altitude and summed across the window, the obstruction at a fifth of the sky's luminance, and the BRE split-flux formula for the internal part with one mean reflectance.
Keep in mind. The internally reflected part is taken as even over the plane, and the coefficient by obstruction angle comes from the BRS table as Robbins and Hunter (1983) reproduce it.
The mean daylight factor over the room's surfaces, given by a short formula (Littlefair, BRE):
ADF = TAwθ / (A (1 − R²))where T is the glass transmittance, Aw the net glazed area, θ the angle of visible sky in degrees seen from the window's centre, A the area of all room surfaces and R their mean reflectance.
For design. A quick first sizing of a window: more glass, cleaner and clearer glass, more visible sky and paler surfaces all raise it. The UK long used about 2 % for kitchens, 1.5 % for living rooms and 1 % for bedrooms as targets.
In the app. Shown with the room check in Daylight factor indoors; the frame share (20 %) and glass transmittance (0.68) are defaults to change. A building opposite and The site and its neighbours can send their obstruction angle to it.
Keep in mind. An average is not the median that EN 17037 asks for: a room with bright windows and a dark back can pass on the average and fail on the median. BR 209's 2022 edition judges new rooms by EN 17037 instead.
The BRE rule for the deepest a room lit from one side can be before its back half looks gloomy, whatever the window size:
L/W + L/H ≤ 2 / (1 − Rb)where L is the room depth, W its width, H the window head height above the floor and Rb the mean reflectance of the back half of the room.
For design. If the rule fails, a larger window does not help: raise the window head, make the room shallower or wider, use paler surfaces, or add light from a second side.
In the app. The room check of Daylight factor indoors reports it with the room's dimensions.
A rule of thumb: daylight from a side window is useful up to a depth of a few times the height of the window head above the floor. Reinhart's simulation study (2005) found 1.5 to 2 head heights with venetian blinds and up to 2.5 head heights without a shading device.
For design. With a 2.7 m window head the daylit zone ends around 4 to 7 m from the facade: a plan deeper than about twice that from two facades will have a core that relies on electric light.
In the app. Drawn on the floor of the room in Daylight factor indoors, both on the daylight factor grid and on the sun patch view.
Source. Reinhart, C. F. (2005). A simulation-based review of the ubiquitous window-head-height to daylit zone depth rule of thumb. Building Simulation 2005.
Obstruction angle
The angle above the horizontal, measured at the window's centre (or the reference point), up to the top of an obstruction opposite. A block 10 m higher than the window and 20 m away stands at atan(10/20), about 27°. The visible sky angle θ of the average daylight factor is 90° minus the obstruction angle for a vertical window.
For design. Up to about 25° an obstruction leaves good daylight to a conventional window; beyond that, windows need to be larger or rooms shallower. Spacing and height of buildings along a street are set by this angle.
In the app.A building opposite and The site and its neighbours compute it; "Use in the daylight room" sends it to the room of Daylight factor indoors.
Keep in mind. The angle is taken in section, on the window's normal: a building to one side does not change it, though it lowers the VSC.
The illuminance on a vertical window from the visible part of a CIE overcast sky, as a percentage of the unobstructed horizontal illuminance. It ignores reflected light. A window with nothing in front of it reaches 39.6 %, the most a vertical surface can see; behind a long obstruction at 25° it falls to about 27 %.
For design. BRE's bands: with a VSC of 27 % or more a conventional window can give good daylight; between 15 and 27 % it needs to be larger; 5 to 15 % makes good daylight very hard; under 5 % it is often impossible. It is the usual test of a new building's effect on its neighbours' windows.
In the app.A building opposite gives the VSC of a window and up the whole facade (where 27 % is reached). The site and its neighbours gives it per facade panel of real buildings, and the site's climate view tests a proposal against BR 209.
Keep in mind. The building opposite is dark (reflects nothing) and the only one; for blocks of limited width the tool's VSC is checked only for consistency, not against a published figure.
Source. BRE (2022). Site layout planning for daylight and sunlight (BR 209), 3rd edition.
BRE site layout guideBR 209
The UK guide to daylight and sunlight in site layout, widely used in planning. Its main tests, as quoted in planning reports:
Daylight to an existing window is noticeably affected if its VSC falls below 27 % and below 0.8 times its former value.
Sunlight to windows within 90° of due south: at least 25 % of the annual probable sunlight hours, 5 % of them between 21 September and 21 March.
Gardens and open spaces: at least half the area with 2 hours of sun on 21 March.
For design. The 0.8 rule means a building may take some light from neighbours, but not a fifth or more of it where they are already near the limit. Massing that steps back or rises away from neighbours passes more easily.
In the app.The site and its neighbours and its site's climate views test a proposed building on the facades and ground around it.
Keep in mind. The thresholds are taken as planning reports quote them, not read in BR 209 itself. The file's sunshine hours stand in for BRE's probable sunlight hours, and winter is taken as October to March in whole months.
The area where direct sun falls through a window onto the floor and walls at a given moment, found by carrying the window's corners along the sun's rays until they meet a surface. The patch moves and changes shape through the day and the year.
For design. A sun patch on a desk or screen means glare; on a dark massive floor in winter it means stored heat. Low winter sun reaches far into a room facing the equator; the high summer sun stays near the window and is easily kept out by an overhang.
In the app. The room view of Daylight factor indoors switches between the daylight factor grid and the sun patch for a chosen window orientation, day and time.
Keep in mind. The window is a plain opening: no frame, reveal or obstruction outside.
Typical arrangements of rooms, windows and devices with known daylight behaviour: side lighting, clerestories, light shelves, top lighting, shading for clear skies. A pattern catalogue lets a designer start from a tested layout rather than a blank wall.
For design. The sky decides which patterns suit: overcast skies favour high windows and top light to reach deep; clear sunny skies favour shaded, diffusing openings that keep the direct sun off the task.
In the app.Daylight factor indoors links patterns of the Daylighting Pattern Guide (New Buildings Institute) when the room is too deep, below the minimum level or obstructed by 20° or more, and when the climate has much hot direct sun. What the sky asks of daylighting reads the sky types.
Keep in mind. The guide is only linked; its pages are all rights reserved and the one-line summaries are the tool's.
Where the wind comes from, how strong it is near the ground, and what that means for ventilating buildings, for people walking past them and for rain on the facades. The weather file measures wind at 10 m over open ground; everything else is carried from there.
Wind rose
A polar chart of how often the wind blows from each direction, split into speed classes. Each spoke points to the direction the wind comes from; its length is the share of hours. Calm hours are counted apart.
For design. Read two roses, not one: the summer or afternoon rose tells where to open for cooling breezes; the winter rose tells where to shelter entrances, terraces and the most exposed facades. A rose of the whole year can hide that the two come from opposite sides.
In the app. The year's rose opens the Wind and air part; Wind by season and by time of day gives four seasonal roses (named for the file's hemisphere) and three by time of day (morning, afternoon, night by the sun), in sixteen sectors on one fixed scale so that they compare.
Keep in mind. The rose is the airport's or station's wind at 10 m. Between buildings both speed and direction change; a rose does not show channelling or eddies.
Hours in which the wind is too weak to have a direction worth counting; in the tool, under 0.5 m/s at 10 m. Some stations record many calms, especially inland valleys and stations with stalling anemometers.
For design. A climate with many calm, warm hours cannot rely on cross ventilation: buoyancy (stack) ventilation, ceiling fans or high ceilings must carry the still hours.
In the app. Given with the roses; the pedestrian wind comfort under Wind comfort (Lawson) leaves calms out of the directional fits.
Wind profile: power law, logarithmic law, roughnessα, z0, d
Wind slows towards the ground because the ground drags on it. Rougher ground (a city) slows it more and over a thicker layer than smooth ground (open water). Two common descriptions:
v(z) = vref (z / zref)αthe power law: α grows with roughness, about 0.10 over water, 0.14 over open country, 0.22 over suburbs and woods and 0.33 in city centres (ASHRAE Fundamentals' terrain values).
v(z) = (u* / κ) ln((z − d) / z0)the logarithmic law: u* the friction velocity, κ ≈ 0.4, z0 the roughness length and d the displacement height (where the flow "sees" the ground over buildings).
For design. The 10 m wind of a weather file is for open ground. At the windows of a low house in a city the mean wind may be less than half of it; at the top of a tower it may be more. Ventilation openings and wind turbines should be sized with the wind at their own height and terrain.
In the app. The wind profile card in Wind and air draws the file's wind at heights over four terrains. Comfort in a street carries the wind up to a blending height and down the city's profile (displacement 0.7 H, roughness 0.1 H, rules of thumb after Grimmond and Oke 1999). Pedestrian comfort uses the power law with 0.14 to 1.5 or 1.75 m.
The usual statistical shape of wind speeds: many moderate hours, few strong ones. Two parameters describe it, the scale c (m/s, near the mean speed) and the shape k (around 2 for many sites; higher means steadier wind). The share of hours above a speed V is:
P(v > V) = exp(−(V / c)k)
For design. A fitted curve gives the chance of strong winds more smoothly than one year of hours can, which matters for rare thresholds such as the safety limits of wind comfort.
In the app. Under Wind comfort (Lawson): the file's wind in twelve directions of 30°, each fitted by maximum likelihood, with fitted and counted shares side by side.
Keep in mind. In a file with many calms the fit misses strong winds (Milan: 0.2 % fitted, 0.5 % counted above the limit), so every verdict takes the larger of the two.
Whether the wind at pedestrian height is acceptable for what people do there: sitting at a café needs calm air, walking past tolerates more. Criteria pair a threshold speed with the share of hours it may be exceeded, for each activity, plus a safety limit for strong gusts that can blow people over.
For design. Tall and slab-like buildings bring fast upper air down to the street, and gaps and corners accelerate it. Entrances, terraces and seating should sit where the criteria allow sitting or standing; podiums, canopies, setbacks and trees break the downdraught.
In the app.Wind comfort (Lawson) rates the open site by month; under it the open site is judged by Lawson, NEN 8100 and Davenport, with the speed-up a building could add before each class is lost.
Keep in mind. The tool reads the climate the site is offered, not the wind around a design: that needs a wind tunnel or a flow simulation. The criteria often disagree with each other (Janssen, Blocken and van Hooff, 2013).
The criteria most used in UK planning. Each activity has a mean wind speed at pedestrian height that may be exceeded in no more than 5 % of hours:
Activity
Limit, exceeded under 5 % of hours
Sitting
4 m/s
Standing
6 m/s
Strolling
8 m/s
Business walking
10 m/s
The safety limit is 15 m/s exceeded less often than about once a year (0.022 % of hours, the City of London's variant).
For design. A café terrace needs the sitting class in the months it is used; a station forecourt can live with strolling.
In the app.Wind comfort (Lawson), month by month for the open site: the speed exceeded in 5 % of each month's hours, at 1.5 m (the 10 m wind taken down with the power law, exponent 0.14).
Keep in mind. Limits as environmental statements citing Lawson (1990) give them (secondary). In a one-year file even one hour above 15 m/s suggests a likely safety failure.
The Dutch standard for wind comfort and danger. It counts the share of hours with wind above 5 m/s at 1.75 m and grades the place from A (under 2.5 %) through B (5 %), C (10 %) and D (20 %) to E (above). For each class, sitting, strolling and traversing are graded good, moderate or poor. Danger is judged at 15 m/s: below 0.05 % of hours no risk, below 0.3 % a limited risk.
For design. One threshold makes the grades easy to compare between places and schemes.
In the app. The table under Wind comfort (Lawson).
Keep in mind. Taken from secondary summaries, consistent with each other; not read in the standard.
Older North American criteria: an activity is comfortable if its speed is exceeded in at most 1.5 % of hours: sitting long 3.6 m/s, sitting short or standing 5.3 m/s, strolling 7.6 m/s, walking fast 9.8 m/s; danger at 15.1 m/s exceeded in 0.01 % of hours.
For design. Stricter in the share of hours allowed (1.5 % against Lawson's 5 %), so a site may pass one and fail the other.
In the app. The table under Wind comfort (Lawson), read on hourly means at 1.75 m.
Keep in mind. Summaries of Davenport's criteria disagree (one gives a weekly limit, one 5 %), and none says whether the speeds are means or gust equivalents. Not read in Davenport (1972).
The ratio of the wind at a point near a building to the wind at the same height over open ground. Corners of tall blocks, gaps between them and passages under them often reach 1.5 to 2; sheltered courtyards fall well below 1. Only a wind tunnel test or a flow simulation (CFD) gives it for a design.
For design. Knowing how much speed-up the climate can absorb tells whether a tall scheme needs wind studies early, and where.
In the app. Under Wind comfort (Lawson): for each class the largest speed-up that keeps it, and a box to try one.
The share of hours in which outdoor air can cool a building by being let in: warm enough not to chill the occupants, cool enough to stay within the adaptive comfort limit, and not too dry or too humid. After Causone (2016): from a lower limit (12 °C here; the paper's own is 10 °C) up to the day's adaptive limit (neutral temperature plus 3.5 K), with relative humidity within 30 to 70 %.
For design. A long season of potential favours operable windows, cross ventilation and a mixed-mode strategy; a short one points to a sealed, well-shaded building with heat recovery.
In the app.Natural ventilation potential by month and Natural ventilation potential, hour by hour (Wind and air); the lower limit and the humidity range are inputs, and humidity can be judged on the air once indoors or as it is outdoors.
Keep in mind. Only the paper's summaries were read; the humidity-ratio form is the tool's reading. The wind speed itself is not part of the test.
Two ways air moves through a building without fans. Cross ventilation is driven by wind pressure: air enters on the windward side and leaves on the leeward side. Stack ventilation is driven by buoyancy: warm indoor air rises and leaves through high openings, drawing cooler air in low down; it works in still air and grows with the height between openings and the temperature difference.
For design. Common rules of thumb: single-sided ventilation reaches about 2.5 times the ceiling height into the room, cross ventilation about 5 times. Stack ventilation needs a tall path (stairwell, atrium, chimney) and openings high and low. In warm, still climates the two are combined.
In the app.Rules of thumb for this climate scores cross ventilation, stack ventilation, high ceilings and openings high and low from the file.
Rain carried sideways by wind onto walls. The amount on a wall grows with the wind speed, the rain intensity and how directly the wind blows at the wall. ISO 15927-3 gives an index from hourly wind and rain:
wall rain ∝ v · r8/9 · cos(D − Θ)summed over hours with rain and the wind on the wall; v wind speed, r rain intensity, D wind direction, Θ the direction the wall faces.
For design. The wettest facade needs the most protective cladding, deeper eaves and sills, and more caution with internal insulation or masonry exposed to frost.
In the app.Driving rain on the facades (Wind and air), on eight facade directions.
Keep in mind. It is the airfield index of open ground; the terrain, the building's height and its shape (corners and top edges take most) change it. A typical year has patchy rain, and files from PVGIS have none.
A world map of the long-term mean wind speed at 10 m above ground, from reanalysis downscaled with a flow model over terrain (Global Wind Atlas 4, DTU and the World Bank).
For design. Shows where a site sits in the regional wind field: a ridge or coast versus a sheltered valley, before choosing a weather station.
In the app. The header button "Wind on the map", over land, in classes of 1 m/s.
Keep in mind. Cells are about 2.4 km at the equator, so narrow valleys and coasts are smoothed; it is a modelled mean for the atlas's own roughness, not a measurement and not the wind of the open file.
How warm or cold a person feels outdoors, where sun, wind and the surfaces around them count as much as the air. These entries lead from the indices to the street and the site, where the designer can change what people feel.
Universal Thermal Climate IndexUTCI, °C
An index of how the outdoor environment acts on the body, given as a temperature: the air temperature of a reference environment that would cause the same physiological strain as the actual one. The reference has radiant temperature equal to air, wind of 0.5 m/s at 10 m and a moderate humidity; the person walks at 4 km/h and their clothing adapts to the temperature.
It rests on a detailed model of human heat transfer and temperature regulation (Fiala et al. 2012); in practice it is computed from a polynomial fitted to that model (Bröde et al. 2012), from four inputs: air temperature, mean radiant temperature, wind speed at 10 m and humidity.
For design. UTCI lets a designer compare a sunny, sheltered square with a shaded, windy one on one scale. In hot climates shade usually changes it most (often by 10 K or more); in cold ones shelter from wind and access to sun.
In the app.Outdoor comfort: Universal Thermal Climate Index (Outdoor comfort): every hour of the file in four conditions, by season and hour, with what to provide outdoors. Also in Comfort in a street, the site's climate views and Any variable, any hours. UTCI on the map colours every station by the share of its year's hours with no thermal stress, in one of the four conditions.
Keep in mind. The polynomial is defined for air from −50 to 50 °C, radiant temperature from 30 K below to 70 K above air, and wind from 0.5 to 17 m/s; outside that the tool holds the wind at the limits and greys out the sensitivity maps.
Source. Jendritzky, de Dear and Havenith (2012); Bröde et al. (2012), International Journal of Biometeorology 56(3). Further reading: 13, 69.
UTCI stress categories
The index is read on a fixed scale of ten categories of thermal stress:
UTCI, °C
Stress category
above 46
Extreme heat stress
38 to 46
Very strong heat stress
32 to 38
Strong heat stress
26 to 32
Moderate heat stress
9 to 26
No thermal stress
0 to 9
Slight cold stress
−13 to 0
Moderate cold stress
−27 to −13
Strong cold stress
−40 to −27
Very strong cold stress
below −40
Extreme cold stress
For design. "No thermal stress" is broad; people sitting still feel the edges of it sooner than people walking. Count the hours of a period that fall in each band rather than looking at a mean.
In the app. The colours of every UTCI map. The activity limits (sitting 9 to 26 °C, walking 0 to 26 °C, play and sport −13 to 26 °C) are this tool's own reading of the categories, not a standard.
The uniform temperature of an imaginary enclosure that would exchange the same radiation with a person as the real surroundings do. Outdoors it adds the direct sun, the diffuse sky, sunlight reflected from the ground and walls, and the long-wave radiation of the sky, ground and walls. In full summer sun it can be 20 to 30 K above the air temperature; under a clear night sky it falls below it.
For design. MRT is the variable a designer controls most: shade from trees and canopies, cool or light paving, and walls that do not bake in the sun all lower it. Air temperature is hard to change outdoors; MRT is not.
In the app. The UTCI card uses the balance of VDI 3787-2 for a standing person (absorption 0.7, emission 0.97, ground albedo 0.2); under cover the surroundings are at air temperature. Comfort in a street weighs six directions after Thorsson et al. (2007) and Lindberg et al. (2008).
Keep in mind. In the UTCI card the ground is at air temperature, so open paving there reads cooler than in the street and site models, which heat it in the sun.
The same hour of weather feels very different depending on where a person stands. The tool computes the UTCI of every hour in four conditions: in the sun and in the wind; in the sun, sheltered; under cover, in the wind; under cover, sheltered. Sheltered means the lowest wind the index is defined for (0.5 m/s at 10 m).
For design. The condition with most comfortable hours in a season tells what an outdoor space needs then: sun traps and windbreaks in a cold spring, shade and breeze in a hot summer. A good space offers more than one condition so people can choose.
In the app. The four maps and the comparison of Outdoor comfort: Universal Thermal Climate Index, with the gain of sun over cover and of wind over shelter by season.
Another outdoor index: the air temperature of a typical indoor room (radiant equal to air, still air, moderate humidity) at which a reference person would have the same skin and core temperatures as outdoors. The person is that of VDI 3787-2: 35 years, male, 75 kg, 1.75 m, standing lightly active, 0.9 clo, clothing fixed.
For design. PET is widely used in European urban climate studies, so it helps when comparing with literature. Its scale differs from UTCI's; do not mix their numbers.
In the app.Comfort in a street can give PET instead of UTCI for the person in the section.
Keep in mind. The categories (limits at 4, 8, 13, 18, 23, 29, 35 and 41 °C) are those of Matzarakis and Mayer (1996) for Central Europe; other climates have their own scales. Because clothing is fixed, PET exaggerates cold stress in winter compared with UTCI.
Maps of UTCI over two variables at a time, air temperature against relative humidity, against mean radiant temperature and against wind, with the others held at a reference. They show where in the climate a change of one variable matters: humidity matters little in the cold and much in the heat; wind cools strongly when the air is below skin temperature and less, or even warms, when it is very hot.
For design. They answer which measure is worth most here: shade (radiant temperature), shelter or breeze (wind), or nothing that design can easily touch (humidity).
In the app.How each variable moves the index (Outdoor comfort), with references of 50 %, radiant equal to air and 3 m/s by default; the file's hours near the references are dotted over the maps.
The share of the sky hemisphere that a surface or point sees, from 0 (no sky) to 1 (open ground). A vertical wall facing an open field sees 0.5. In the middle of the floor of an endless street with height equal to width it is about 0.45.
SVF = cos β, β = atan(2H / W)for the middle of the floor of an endless street of height H and width W.
For design. A low sky view keeps long-wave heat in at night (part of the urban heat island) and shades by day; a high one lets surfaces cool at night but exposes them to the sun. Daylight on facades and ground also follows it.
In the app. The site's climate view "sky view factor" in The site and its neighbours, from each point's skyline; Comfort in a street uses the view factors of its section.
Keep in mind. For a standing person the tool splits the upper half of the view by the ground point's sky view factor, a simplification of a body's view.
A street between two rows of buildings, described by the ratio of building height H to street width W. The ratio sets how much sun reaches the street floor, how much sky it sees, and how wind above the roofs reaches into it.
A street section: the sky seen from the street's middle narrows as H/W grows; low sun lights only the upper facade.
For design. Shallow streets (H/W under about 0.5) are sunny and windy; deep ones (H/W of 2 or more) are shaded and still, cool by day in summer but slow to cool at night and dark in winter. In hot dry climates deep narrow streets have a long tradition; in cold climates wider streets let in winter sun. The street's orientation matters as much: an east–west street has one sunny and one shaded side.
In the app.Comfort in a street labels H/W and direction under the section; drag roofs and edges to change them and watch the UTCI profile across the street.
Keep in mind. In the street model only the wind along the street counts, reduced by 2/π · exp(−H/4W) (Masson 2000); a wind across a deep street leaves it near calm.
The tool's own model of comfort in a street, drawn as a cross-section. For every hour of the file it traces the sun and sky in the section plane (exact for a street with no end), heats each strip of ground and wall by its sun, sky and reflections, conducts heat into them, and then finds what a standing person feels at any point across the street.
The steps: surfaces in cells of about 0.5 m; reflections solved among groups of cells; conduction in layers, half an hour a step; a person's radiant temperature from six directions; UTCI or PET from that, the air temperature and humidity of the file, and the wind brought down into the street.
For design. It compares sections: a tree row against a canopy, grass against asphalt, a wider sidewalk on the shaded side, a canal. The difference between two designs is more reliable than either absolute figure.
In the app.Comfort in a street (Outdoor comfort): drag strips, trees and canopies into the section, turn the compass, move the person; presets include a 15 m street, an old town lane, a boulevard and a street with a tram and canal.
Keep in mind. Not compared with measurements. Trees, grass and water cool surfaces and people by radiation only; the air temperature stays the file's everywhere (a point model cannot change it). Some soil and evaporation values are from memory (README).
The share of incoming solar radiation a surface reflects, from 0 (absorbs all) to 1 (reflects all). What is not reflected is absorbed and heats the surface.
Surface
Albedo used in the tool
Asphalt
0.10 (fresh 0.05)
Water
0.06
Grass
0.20
Brick
0.30
Concrete
0.35
Cool paving
0.60
White render
0.65
For design. Light surfaces stay cooler in the sun, which lowers their long-wave heat on people and the heat stored for the night. But a bright pavement reflects sun onto people and facades, so on a sunny square it can raise the mean radiant temperature: shade first, then light surfaces.
In the app. The strips of Comfort in a street and the ground surfaces of the site's climate views.
Keep in mind. Asphalt and concrete after the LBNL Heat Island Group; light stone (0.45) has no source found.
How well a surface gives off (and absorbs) long-wave heat radiation compared with a perfect black body, from 0 to 1. Most building materials, paints of any colour, vegetation and water lie around 0.9 to 0.96; bare polished metals are low (often below 0.1), which is why low-e coatings and foil insulation work.
For design. A cool roof needs both a high albedo (to reflect sun) and a high emissivity (to shed heat). A shiny metal roof reflects sun well but holds its heat.
In the app. Surfaces of Comfort in a street (0.90 to 0.96, from memory); the person's emission of 0.97 in the UTCI card.
The temperature of a pavement, wall or roof, set by its energy balance: absorbed sun and long-wave radiation in; long-wave out, convection to the air, conduction into the material and evaporation out. In summer sun asphalt can reach 50 to 60 °C while shaded grass stays near the air temperature.
For design. Hot surfaces radiate onto people for hours, including after sunset. Shade, light colours, vegetation and water keep them down; heavy materials delay their heat into the evening.
In the app.Comfort in a street lists each strip's surface temperature under the section (Milan in July, 12 to 15 h, an east–west 15 m street: road lane 57 °C, grass 39 °C, water 21 °C, shaded sidewalk under a tree 25 °C).
A tree cools a person mainly by shade: its crown lets through only a small share of the sun when in leaf. Konarska et al. (2014) measured 1.3 to 5.3 % in leaf and 40 to 52 % bare in Gothenburg. A crown is also cooler than a roof or wall in the sun, so it radiates less heat.
For design. Deciduous trees give summer shade and let most winter sun through; evergreens shade all year. A crown that is wide and high enough to shade the sidewalk at the hottest hours does more than many small trees.
In the app.Comfort in a street (crowns at 5 % in leaf and 50 % bare, deciduous in leaf April to October); the site's climate views add trees by species (21 species from the GALK street tree list) and the mapped trees of OpenStreetMap.
Keep in mind. Trees here cool by shade and their cool crowns only; their evaporation does not change the air temperature. The list's permeability read as a transmittance is this tool's reading.
Water that evaporates takes heat with it (about 2.45 MJ per kg), cooling the surface it leaves. Grass and trees evaporate water through their leaves (transpiration); open water evaporates from its surface. The cooling is strongest in dry, sunny, windy weather.
For design. Water features and irrigated planting keep the surfaces around people cool; in humid climates the gain is smaller and the added moisture can make it feel muggier.
In the app.Comfort in a street (grass evaporating at 0.3 of a wet surface, water as a mixed 0.3 m layer, by the Lewis relation: values from memory) and the site's climate ground surfaces.
Keep in mind. The tool lets evaporation cool surfaces, not the air.
Cities are warmer than the countryside around them, mostly at night: by day surfaces store heat, deep streets see little sky to cool to at night, there is less evaporation, and traffic and buildings give off waste heat. The difference is often a few kelvin and can reach more on calm clear nights.
For design. Night ventilation and night cooling of mass work less well in a dense centre than the airport's weather suggests. Trees, water, light surfaces and open sky help.
In the app. Not modelled as air temperature. The sky view factor, surface temperatures and radiant temperatures of the street and site cards show its causes at the scale of a street.
Keep in mind. Most weather files come from airports outside cities, and reanalysis (ERA5) smooths a city into its cell, so both usually miss the heat island.
The site's climate (shadows, sun hours, proposals)
Readings of a real place: the buildings within a radius of an address (from OpenStreetMap), their shadows at a moment, hours of sun on a day, sunshine hours from the file, the sky view factor of the open ground, UTCI on it hour by hour, sun and photovoltaics on roofs and facades, and the effect of a proposed building on its neighbours.
For design. Tests a massing early: where the ground keeps sun in spring, which facades lose daylight, where trees or canopies would help most.
In the app.The site and its neighbours (Sun and sky) and its four further views; tools place proposed buildings, trees, canopies, pergolas and ground surfaces, kept in the browser per site.
Keep in mind. Building heights are tagged where known, else 3 m a storey or 9 m: assumptions. Flat roofs, level ground; the wind around buildings is not modelled.
The envelope sets how much heat, sun, light and air pass between the climate and the rooms. These entries cover the quantities used to size walls, roofs and windows from the climate alone, before any building is simulated.
Thermal conductivityλ, W/(m·K)
The heat flow through a metre thickness of a material for one kelvin of temperature difference across it, per square metre. It is a property of the material, not of a layer: still air trapped in small cells conducts very little, dense stone and metal a great deal.
Material
λ, W/(m·K)
Expanded polystyrene, mineral wool
about 0.035
Softwood
0.13
Brickwork
0.84
Dense cast concrete
1.4
Granite
2.3
For design. A low λ means less thickness for the same insulation. Moisture raises λ: a wet insulant or masonry loses part of its value, so keep insulation dry (see condensation).
In the app. The masonry values above are those of the Thermal mass card's wall materials and the moisture card's build-ups (Szokolay's data sheet D.1.1, declared values, not corrected for moisture). The Insulation card takes 0.035 W/(m·K).
Source. EN ISO 10456 (design values of materials); Szokolay, S. V. (2004), Introduction to Architectural Science, data sheet D.1.1. Further reading: 3.
Thermal resistance and surface resistancesR, Rsi, Rse, m²K/W
A layer of thickness d resists heat flow by R = d / λ. Resistances of layers in series add up. The thin films of air at each face add their own: the inner surface resistance Rsi and the outer Rse, which stand for convection and long-wave radiation at the surface.
RT = Rsi + Σ di / λi + Rsewhere RT is the total resistance from air to air. For a wall (horizontal heat flow) EN ISO 6946 gives Rsi = 0.13 and Rse = 0.04 m²K/W.
For design. The surface resistances matter for thin elements: a single pane of glass is almost all surface resistance. For a well insulated wall they are a small share.
In the app. The moisture card uses 0.13 and 0.04 (and 0.25 at the inner surface for the mould check, as EN ISO 13788 asks); the time lag of the Thermal mass card uses 0.123 and 0.055 m²K/W, the values behind Szokolay's tables.
Source. EN ISO 6946 (thermal resistance and transmittance of building components).
U-valuethermal transmittance, W/(m²K)
The steady heat flow through one square metre of an element for one kelvin between the air on both sides: the inverse of the total resistance, U = 1 / RT. The heat lost through an element is U × area × temperature difference; over a season it is U × area × degree days × 24 h.
Typical values today: walls and roofs of new buildings in Europe 0.15 to 0.35 W/(m²K), windows 0.7 to 1.4, an uninsulated solid brick wall about 2.
For design. The U-value is the number that codes ask for. Halving a U-value halves the loss through that element, but each further centimetre of insulation saves less than the one before, so the first centimetres pay most (see insulation thickness).
In the app.Prescriptive envelope for this climate zone gives code U-values; Insulation plots loss against thickness; Glazing specification by facade gives a window U target; the cost-optimal card tests whether a package meets the national code's U-values.
Keep in mind. A U-value is a steady-state figure. It says nothing of time lag or of the sun on the surface, and it leaves out thermal bridges unless they are added.
Adding insulation to a wall lowers its U-value along a curve: U = 1 / (R0 + d / λ). The loss saved by each added centimetre falls with the square of the total resistance, so the curve flattens. Past a certain thickness the extra material costs more (in money and embodied carbon) than the heat it saves.
For design. Size insulation to the climate's heating demand: a cold climate justifies 20 to 30 cm, a mild one far less. In hot climates, roof insulation matters more than wall insulation because the roof takes the strongest sun.
In the app.Insulation (Building envelope) draws a square metre of wall's yearly loss against thickness: insulation of 0.035 W/(m·K) on a wall of 0.3 m²K/W with its surface resistances, losing heat in every hour below 20 °C. It marks the "knee", the thickness beyond which another 5 cm saves under 2 kWh/m² a year, and a band of target U-values by heating degree days.
Keep in mind. The knee criterion and the target bands are this tool's choice, a first sizing; codes and cost-optimal studies set the real targets.
A place where heat bypasses the insulation: a balcony slab running through the wall, a window reveal, a corner, a steel fixing. A linear bridge is described by ψ, the extra loss per metre of its length; a point bridge by χ. Bridges add to the loss and, more seriously, cool the inner surface locally.
For design. Keep the insulation layer continuous: wrap the slab edge, set windows in the insulation plane, use thermal breaks for balconies. The cold inner surface of a bridge is where mould appears first.
In the app. The moisture card checks typical bridges by their temperature factor (insulated outer corner 0.85, uninsulated reveal 0.72, balcony slab 0.62, uninsulated solid corner 0.55); the cost-optimal card adds 0.05 W/(m²K) for bridges.
Keep in mind. The bridges' temperature factors are typical values from memory, not calculations; a real junction needs a two- or three-dimensional calculation (EN ISO 10211).
Solar factorg-value, total solar energy transmittance; SHGC in North America
The share of the solar energy falling on a glazing that ends up inside as heat: the part transmitted directly plus the part absorbed in the panes and given off inwards. Clear single glass has g about 0.85, a double low-e unit 0.5 to 0.65, solar control glass 0.25 to 0.4.
Qsol = g × Aglass × Iwhere I is the irradiance on the glass (W/m²) and A the glass area; frames, dirt and angle of incidence reduce it further.
For design. A high g is welcome where the winter sun heats; a low g where the sun comes mostly in warm hours. External shading cuts gains far more than a low g, and keeps the view and the winter sun.
In the app.Glazing by facade balances a square metre of glass's winter gain (g × 0.75 for frame, dirt and curtains) against its loss; Glazing specification by facade picks a solar factor class by the share of each facade's sun that comes in hours over 24 °C; the window tester and the room model use g at normal incidence.
Source. EN 410 (luminous and solar characteristics of glazing).
Light transmittanceτv, VLT, Tvis
The share of visible light, weighted by the eye's sensitivity, that a glazing lets through. A double low-e unit transmits about 0.7 to 0.8, a triple 0.65 to 0.75, strong solar control glass 0.5 to 0.6.
The ratio of light transmittance to solar factor (light to solar gain) shows how selective a coating is: it lets light in while keeping heat out. Above about 1.25 a glass is commonly called spectrally selective; the best coatings come near 2.
For design. Daylight depends on τv directly (it enters the average daylight factor), so a dark tinted glass that lowers heat also darkens the room. Prefer a selective coating to a tint.
In the app. The units of Glazing specification by facade list it; the daylight room takes 0.68 for the glass by default (a value to change).
An insulating glass unit is two or three panes sealed around cavities filled with argon or krypton. A low-emissivity (low-e) coating, a thin layer of silver, reflects long-wave heat and so cuts the radiation across the cavity; the U-value of a double unit falls from about 2.8 to 1.0 to 1.1 W/(m²K). Surfaces are numbered from outdoors: #1 outer face, #2 and #3 the faces of the first cavity.
Where the coating sits matters. On #3 (the inner pane) it keeps the absorbed solar heat inside: high solar gain. On #2 it rejects more of the sun: lower g. Double and triple silver coatings are selective solar control.
For design. Choose the unit per facade: high gain on sunny cold sides, solar control where sun arrives in warm hours, triple glazing where heating degree days are high.
In the app.Glazing specification by facade takes a U target from heating degree days (0.6 W/(m²K) above 3,500, 0.7 above 2,500, 1.1 above 1,500, 1.3 above 700, else 1.6), double or triple to meet it, and one of four solar factor classes per facade, then names a unit, its coating and the surface it sits on.
Keep in mind. The units' values are typical catalogue figures: four agree with manufacturers' data, the others are from memory. The U steps and class limits are this tool's.
The window-to-wall ratio is the share of a facade's area that is glazed, frame included. A window is at once a heat loss (its U-value is several times a wall's), a solar gain and the source of daylight and view, and the balance between them depends on the facade's orientation and the climate.
For design. On a facade where winter sun is strong, more glass can be a net gain; on a dull or north facade, size glazing for daylight only. In warm climates, more glass means more shading.
In the app.Glazing by facade counts, over the hours below 15 °C, what a square metre of double (U 1.1, g 0.62) or triple glass (U 0.6, g 0.5) gains from the sun and loses to the cold, per facade and for a rooflight, and the sun on it in hours above 24 °C. Its advice: a net gain with triple glazing over 40 kWh/m², glaze 40 to 60 % of the wall; near balance, 30 to 40 %; a loss, 20 to 30 % for daylight only.
Keep in mind. The thresholds and ratios are this tool's rules of thumb; the room model (5R1C) tests one window hour by hour.
Heat capacity and thermal massc, J/(kg·K); κ, kJ/(m²K)
Specific heat capacity c is the heat that warms a kilogram of material by one kelvin; multiplied by density ρ it gives the heat stored per cubic metre. Thermal mass in buildings means exposed heavy materials (concrete, masonry, stone, earth) that store heat by day and give it back later, which damps the indoor temperature swing.
Only the first 50 to 100 mm or so of a surface take part in the daily cycle, and only if they are exposed to the room: a suspended ceiling or a carpet cuts the effect.
For design. Mass helps where the days are hot and the nights cool (a large daily range), especially with night ventilation to empty it. Where nights stay warm, mass stores heat that cannot be shed.
In the app.Thermal mass classes every day: mild (no cooling asked), hot days (maximum over 27 °C) split into mass with night flushing (night minimum at most 20 °C), mass alone (a warm night, but a mean of at most 27 °C) and beyond mass (mean over 27 °C and minimum over 20 °C), and gives the indoor swing for light, medium and heavy construction as 0.75, 0.45 and 0.25 of the outdoor swing (assumed damping factors).
When the outdoor temperature swings through the day, a heavy wall passes the swing on late and small. The time lag is the delay between the outdoor peak and the inner surface's peak; the decrement factor is the ratio of the inner swing to the swing a weightless wall of the same U-value would pass.
The outdoor wave and the wave at the inner surface: smaller (decrement) and later (time lag). The bars at the right are the two swings.
Typical figures: 220 mm solid brick about 6 h; 300 mm of dense concrete about 8 h. Insulation on the outside lowers the decrement more than the same insulation inside: 220 mm of brick with 50 mm of insulation gives about 0.22 outside and 0.30 inside, against 0.54 bare (the card's method).
For design. In a hot dry climate, aim for a lag of 8 to 12 h on the sunny walls and roof, so the heat of the afternoon arrives indoors at night, when windows can be opened.
In the app.Thermal mass has a wall (material, thickness, 50 mm of insulation inside, outside or none): it gives the U-value, time lag and decrement, and the inner surface's swing and peak hour for the three warmest months of the file.
Keep in mind. The inner swing is the outdoor air's mean daily swing times the decrement; the sun on the wall (sol-air temperature) is left out, so sunny walls swing more.
A way to compute how a layered wall responds to a periodic temperature wave. Each layer is described by a 2 × 2 matrix of complex numbers that links temperature and heat flow on its two faces for a wave of one period (24 h); multiplying the matrices of the surface films and layers gives the whole wall. From the result come the periodic thermal transmittance, the time lag, the decrement factor and the areal heat capacity.
For design. It shows why the order of layers matters: the same materials give a different lag and decrement with the insulation outside or inside, which a U-value cannot tell.
In the app. The wall of the Thermal mass card is computed this way. It reproduces Szokolay's data sheet D.1.3 for solid brick (105, 220, 335 mm) and dense concrete (75, 150, 200 mm) within 0.1 h of time lag and 0.01 of decrement. Szokolay's simpler formula for one homogeneous layer leaves out the surface films and is not used.
Source. ISO 13786 (thermal performance of building components: dynamic thermal characteristics).
Sol-air temperature
The outdoor temperature that would give, without sun, the same heat flow into a surface as the real air temperature with the sun and the long-wave exchange with the sky. A dark roof in summer sun can reach a sol-air temperature 30 K or more above the air.
θsa = θe + (αG − ε ΔR) / hewhere θe is the air temperature, α the surface's solar absorptance, G the irradiance on it (W/m²), ε its emissivity, ΔR the net long-wave loss to the sky (W/m²) and he the outer surface coefficient (W/(m²K)).
For design. A light-coloured roof or wall (low α) lowers the sol-air temperature and the heat that reaches the room; this matters most for roofs in hot, sunny climates.
In the app.Weathering of materials computes the surface temperature of a dark and a light roof in this form (see roof surface temperature). The time lag of the Thermal mass card leaves the sun out.
Prescriptive envelopeASHRAE 90.1, IECC, national codes
Energy codes can be met in two ways: prescriptively, by meeting a table of maximum U-values (or minimum insulation R-values), maximum window shares and solar factors for each climate zone; or by performance, by showing with a calculation that the whole building uses no more than a reference. ASHRAE 90.1 (non-residential) and the International Energy Conservation Code (IECC) are prescriptive tables by ASHRAE climate zone; European national codes set U-values by their own zones, often by heating degree days.
For design. The prescriptive table gives a quick first target for each element. It is a minimum: a cost-optimal or zero-energy design often goes further.
In the app.Prescriptive envelope for this climate zone (Building envelope) shows the values for the file's zone from ASHRAE 90.1, the IECC or the national code, and Prescriptive assemblies the insulation by class of construction.
Keep in mind. The zone is computed from the file (Standard 169) and can differ from the zone a code assigns by county or municipality; check the code's own map for a real project.
National laws and city building regulations add rules that bear on climate design: minimum daylight and sunlight, window areas, overhang limits, solar access, insulation and solar control. They differ widely between countries and even between neighbouring cities.
For design. Read them before concept design: a minimum window share or a sunlight hour rule can fix the depth and spacing of buildings more firmly than any energy target.
In the app.Local codes and standards (Comfort and strategies) lists the rules that apply at the file's place: by country, by region, or by city (a city rule shows within its circle plus 30 km, labelled by the distance from the station). Each names its source and how far its wording was checked: read in the document, found by a search of the text, or from memory.
Keep in mind. Entries marked "from memory" must be confirmed in the document before use; codes change, and a weather station can be outside the municipality whose rules apply to the site.
Water in its three states is the main cause of damage to buildings. The first entries follow vapour from the room into the wall; the rest read from the weather file what rain, frost, salts and sun do to the outside of a building over the years.
Surface and interstitial condensation
Air holds water vapour up to a limit that rises steeply with temperature (the saturation vapour pressure). When warm indoor air meets a surface colder than its dew point, water condenses on it: surface condensation. Vapour also diffuses through a wall from the warm, humid side to the cold side; where it meets a layer cold enough, it condenses inside the construction: interstitial condensation.
Long before water runs down a surface, a relative humidity of about 80 % at the surface, held for days, lets mould grow.
For design. Keep inner surfaces warm (continuous insulation, few thermal bridges), keep indoor humidity down by ventilation, and arrange layers so that vapour meets less resistance as it moves outwards (tight inside, open outside).
In the app.Moisture: condensation in the wall and mould at its surface (Building envelope) runs both checks month by month from the file.
Source. EN ISO 13788 (hygrothermal performance of building components: internal surface temperature to avoid critical surface humidity and interstitial condensation).
Temperature factor and mouldfRsi
The temperature factor tells how warm an inner surface stays compared with the indoor and outdoor air. It is 1 for a surface at room temperature and 0 for one at outdoor temperature.
fRsi = (θsi − θe) / (θi − θe)where θsi is the inner surface temperature, θi the indoor and θe the outdoor air temperature. For a plain wall fRsi = 1 − U Rsi, with Rsi = 0.25 m²K/W for this check.
EN ISO 13788 finds, for each month, the lowest surface temperature at which the surface's relative humidity stays below 80 %, and from it the temperature factor that month asks. The month that asks most is the critical month; the wall and each junction must exceed its factor.
For design. A well insulated wall has fRsi above 0.9; the risk lies at junctions. DIN 4108-2 asks at least 0.70 of every thermal bridge in a dwelling.
In the app. The moisture card gives the factor each month needs, the critical month, the wall's own factor, and typical bridges (insulated outer corner 0.85, uninsulated reveal 0.72, balcony slab 0.62, uninsulated solid corner 0.55) against it.
Keep in mind. The bridges' factors are typical values from memory, not calculations; a real junction needs EN ISO 10211.
The moisture risk depends on how much vapour the occupants add. EN ISO 13788 describes this as an internal vapour pressure excess, the difference between the vapour pressure inside and outside, by humidity class: largest when it is cold outside (the windows are shut), falling to nothing at 20 °C outdoors.
Class
Use
Excess at 0 °C and below
1
Storage areas
270 Pa
2
Offices, shops
540 Pa
3
Dwellings with low occupancy
810 Pa
4
Dwellings with high occupancy, sports halls, kitchens
1,080 Pa
5
Laundries, breweries, swimming pools
1,350 Pa
For design. A swimming pool or a crowded flat needs a far tighter vapour control than an office; ventilation that removes vapour lowers the class in practice.
In the app. The moisture card's menu chooses the class (or a fixed 50 % indoors) and the heating set point; a safety factor of 1.10 is applied to the excess. Class 3 is the default and gives about 60 % indoors in winter, on the high side for an ordinary dwelling.
Keep in mind. The class values and the safety factor are from memory (the 2001 form of Annex A), not checked against the printed standard.
A steady-state check of interstitial condensation. Through the wall, two lines are drawn against the vapour resistance of the layers (sd): the saturation vapour pressure, which follows the temperature line through each layer, and the actual vapour pressure, a straight line from inside to outside. Where the actual line would cross above the saturation line, it is pulled down to touch it, and condensation forms at that interface.
Inside insulation without vapour control: the vapour line (dashed) is pulled down to touch the saturation line at the cold face of the insulation, where water collects. Layers are drawn by vapour resistance, not thickness.
Month by month the condensate collects in winter and dries in summer. The wall passes if all of it dries out within the year and the most that collects stays small: DIN 4108-3 takes 1,000 g/m² as a yardstick (500 g/m² at a face that does not absorb water).
For design. The method shows at once why inside insulation needs a vapour control layer, and why an outer layer should be more open to vapour than the inner ones.
In the app. The moisture card draws the temperature, saturation and vapour pressure lines through a preset or edited build-up (outside insulation, inside insulation with and without vapour control, timber frame with and without it, a filled cavity wall) for the month of most condensation or any month, across the layers' vapour resistance or their thickness, and adds up condensation and drying over the year.
Keep in mind. Glaser ignores liquid transport, moisture storage, rain, the sun on the wall and air leaking through it. It judges a build-up; it does not predict its moisture content. For inside insulation and timber, a hygrothermal simulation (EN 15026) is the next step. The splitting of layers and the 1,000 g/m² yardstick are from memory.
The vapour resistance factor μ says how many times a material resists vapour diffusion more than still air of the same thickness: about 1 for mineral wool, 5 to 10 for brick, about 100 for dense concrete, very large for metal and plastic sheets. The equivalent air layer thickness sd = μ d (in metres) gives the resistance of a layer of thickness d: the thickness of still air that would resist the same.
For design. A rule of thumb for a wall that dries outwards: the sd of the inner layers several times that of the outer ones.
In the app. Each layer of the moisture card's build-up has a μ (typical of EN ISO 10456's tables) or, for a membrane, an sd; the Glaser diagram can place the layers by sd or by thickness.
Keep in mind. The μ values are typical and from memory; real products vary, and many have different dry and wet values.
A vapour control layer (often called a vapour barrier) is a sheet on the warm side of the insulation that keeps indoor vapour out of the construction. A breather membrane on the cold side keeps wind and rain out while letting vapour escape. Their resistances differ by a factor of a thousand or more.
For design. Put the tight layer on the warm side and the open one on the cold side; seal its joints, since air leaking through a gap carries far more vapour than diffusion. In hot humid climates with air conditioning the warm side is outside, and the order turns round.
In the app. The moisture card offers a vapour control layer (sd 20 m), a polyethylene sheet (100 m) and a breather membrane (0.02 m); compare a preset with and without the layer.
Keep in mind. The three sd values are typical and from memory.
Water in the pores of brick, stone, mortar or concrete expands by about 9 % when it freezes. Each freeze and thaw of a wet material can crack and spall it; the damage depends on how wet the material is when it freezes and how many times this happens.
For design. In climates with many cycles, choose frost-resistant masonry and pavings, keep facades and copings shedding water (drips, overhangs, sloped sills), and be cautious with inside insulation of solid walls, which leaves the masonry colder and wetter.
In the app.Weathering of materials (Building envelope) counts cycles of the air from −1 °C down and back over +1 °C (so a temperature hovering around zero counts once), and the "wet" cycles with at least 1 mm of rain in the 72 hours before the freeze; today, in 2041–2060 and in 2081–2100 under the scenario on show.
Keep in mind. These are cycles of the air, not of the material: a sunlit wall thaws more often, a sheltered or insulated one less. The thresholds are this tool's choice.
Soluble salts in masonry (from the ground, the sea, de-icing or the materials themselves) dissolve when the air is humid and crystallise when it dries. Each crystallisation can break the pores near the surface, giving powdering and flaking. Sodium chloride takes up water from the air above 75.3 % relative humidity (its deliquescence point) and crystallises below it.
For design. Where cycles are many, prevent salts from entering (damp-proof courses, drainage, no de-icing salt splash) and use renders and mortars that let salts crystallise in a sacrificial layer.
In the app.Weathering of materials counts each rise of the air's relative humidity through 75.3 %.
Keep in mind. It reads the outdoor air, not the pores; other salts have other thresholds.
Source. Greenspan, L. (1977), Humidity fixed points of binary saturated aqueous solutions, Journal of Research of the National Bureau of Standards 81A(1).
Time of wetnessEN ISO 9223
The hours a year in which a metal surface is likely to carry a film of water, taken as the hours with relative humidity over 80 % at a temperature over 0 °C. With pollution (sulphur dioxide, chlorides from the sea) it sets the corrosion rate of steel, zinc, copper and aluminium.
Class
Hours a year
τ1
10 or fewer
τ2
10 to 250
τ3
250 to 2,500
τ4
2,500 to 5,500
τ5
over 5,500
For design. A high class, especially near the sea, asks for thicker galvanising, stainless grades or coatings, and details that let metal dry.
In the app.Weathering of materials gives the hours and the class.
Keep in mind. The full corrosivity category also needs the deposition of pollutants, which the weather file does not hold.
Source. EN ISO 9223 (2012), corrosion of metals and alloys: corrosivity of atmospheres.
Wetting and drying spells
Many kinds of decay follow the alternation of wet and dry rather than wetness alone: the swelling and shrinking of timber and clay, the leaching and deposit of salts, the cracking of coatings. A simple measure is the number of days of rain that follow a dry day.
For design. Protect timber and earth from frequent wetting (roof overhangs, plinths, ventilated claddings).
In the app.Weathering of materials counts the days with at least 1 mm of rain after a dry day.
Keep in mind. A typical year's rain is patchy, and the scenarios do not change rain.
Carbon dioxide from the air reacts with the calcium hydroxide in concrete and lowers its alkalinity. When the carbonated front reaches the reinforcement, steel is no longer protected and can corrode. The depth grows roughly with the square root of time, x = k √t. The rate is highest at moderate humidity: in dry concrete the reaction lacks water, in saturated concrete the gas cannot enter.
ke = ((1 − RH5) / (1 − 0.655))2.5the environmental factor of fib Bulletin 34, with RH the mean relative humidity as a fraction and 0.65 the reference; the rate scales with √ke.
For design. Enough cover and a dense concrete matter most in sheltered, moderately humid places; rain-washed surfaces stay wetter and carbonate more slowly.
In the app.Weathering of materials gives the mean humidity and the rate against the reference (65 %) for sheltered concrete, today and under the scenario.
Keep in mind. The constants are from memory; CO₂ concentration and the concrete's own properties, which also rise or vary, are not included.
Source. fib Bulletin 34 (2006), Model code for service life design.
Roof surface temperaturecool roofs
An exposed roof takes the full sun by day and loses heat to the sky at night. Its surface temperature follows from a balance of absorbed sun, long-wave exchange with the sky and convection to the air. A dark roof can stand about 30 K above the air at noon in summer, a light one only a little above it, and both fall below the air on clear nights.
Ts = Tair + (αG − ε (σTair4 − L↓)) / hewith α 0.9 (dark) or 0.3 (light), ε 0.9, he 25 W/(m²K), G the global horizontal irradiance and L↓ the sky's long-wave radiation from the file.
For design. A cool (light, high albedo) roof cuts heat gains below it and the ageing of membranes; a green roof does the same by shade and evaporation.
In the app.Weathering of materials gives the range of a dark roof's surface temperature, its mean daily swing and its hours at 60 °C or more, and the range of a light roof.
Keep in mind. The long-wave term is linearised at the air's temperature; the roof is taken as having no heat capacity.
Ultraviolet light breaks down the polymers of coatings, sealants, membranes and plastics; heat speeds the reactions. The yearly solar irradiation is a first measure of the dose a horizontal surface receives.
For design. In sunny climates choose UV-stable materials, protect sealants from direct sun and plan for shorter service lives of exposed coatings.
In the app.Weathering of materials gives the yearly global horizontal irradiation as the dose, today and under the scenario.
Keep in mind. It is total sun, not UV alone, and a facade receives less than the horizontal.
From the climate to energy: how a room's temperature follows the weather, what a building needs to stay comfortable, and which systems and investments meet that need at the lowest cost and carbon over its life.
Energy balance of a building
A building's air stays at a steady temperature when the heat coming in equals the heat going out. Heat leaves by transmission through the envelope and by ventilation and infiltration; it comes in from the sun through windows and on opaque surfaces, and from people, lights and equipment (internal gains). Heating or cooling supplies the difference.
QH = (Qtr + Qve) − η (Qint + Qsol)the heating need of a month in the monthly method of EN ISO 13790: transmission and ventilation losses less the gains, the gains reduced by a utilisation factor η (not all gains are useful: some come when the room is already warm). Qtr = Htr (θi − θe) t, with Htr = Σ U A + thermal bridges.
For design. The balance tells which lever matters: a compact, well insulated house is dominated by gains (watch overheating); a leaky, glazed one by losses.
In the app. The cost-optimal card computes this monthly balance for every package; A room through the year computes it hour by hour.
Source. EN ISO 13790:2008 (energy performance of buildings: calculation of energy use for space heating and cooling), now replaced by EN ISO 52016-1.
The 5R1C modelEN ISO 13790, simple hourly method
A room reduced to an electric circuit: three temperature nodes (the air, a combined surface node, and the mass), five heat transfer coefficients between them and the outdoors (ventilation, windows, air to surface, surface to mass, mass to outdoors), and one heat capacity, the thermal mass. Gains from the sun and from people are split among the nodes. The model steps through every hour of the year, so the mass's temperature carries heat from one day to the next.
The operative temperature, what an occupant feels, is taken as 0.3 of the air node plus 0.7 of the surface node.
For design. It is the simplest model that shows thermal mass, night ventilation and shading working hour by hour, at a cost of about 1 ms for a year: fast enough to try many variants.
In the app.A room through the year: overheating and what each measure does runs a shoebox room (by default a 4 × 6 m office, 2.7 m high, its facade south, 40 % glazed, medium mass) free-running or heated and cooled, and takes each passive measure it has away in turn (an overhang where one is set, thermal mass, night ventilation, windows opened by day, movable shading) to show what each does.
Keep in mind. The mass classes (Am 2.5 to 3.5 Af; Cm 80 to 370 kJ/(m²K)), his 3.45, hms 9.1 W/(m²K), At 4.5 Af and the sky loss are written from memory, not checked against the printed standard. Only one facade is exposed; the occupants' use of blinds and windows is assumed. Checked for energy balance and limits, not validated against BESTEST.
A building, or a period of the year, without heating or cooling: its indoor temperature floats with the weather, the sun, the gains and what the occupants do with windows and blinds. Comfort in a free-running building is judged by the adaptive models, since people in it adapt their clothing and expectations to the season.
For design. A good passive design keeps the free-running temperature within the adaptive limits for most of the year, so that mechanical systems are small or absent.
In the app.A room through the year runs free by default and counts the occupied hours above and below the EN 16798-1 categories; the "ideal" mode instead holds the air between set points and reports the heating and cooling energy.
A free-running room overheats when its operative temperature rises above what occupants accept for long or by much. CIBSE TM52 tests three criteria over May to September, occupied hours, against the upper limit of the adaptive comfort band (category II):
Hours of exceedance: no more than 3 % of the hours at 1 K or more above the limit.
Daily weighted exceedance: on no day the sum of hours times kelvins above the limit (counted from 1 K) over 6.
Upper limit: the temperature never more than 4 K above the limit.
A room failing two of the three overheats.
For design. Shading, exposed mass with night ventilation and openable windows are the measures that move all three; criterion 3 catches heatwave peaks that averages hide.
In the app.A room through the year gives the three criteria and the verdict, and the hours over each EN 16798-1 category, with each measure taken away.
Keep in mind. The criteria are written from memory, not from the guide at hand. A typical year has milder heatwaves than real ones: try a measured hot summer (measured years) or a scenario.
Source. CIBSE TM52 (2013), The limits of thermal comfort: avoiding overheating in European buildings.
Heating and cooling need
The heat that must be added or removed to hold a building at its set points, in kWh a year (or kWh/m² a year of floor area). It is a property of the building and climate. The energy delivered is the need divided by the system's efficiency (or COP), plus auxiliary energy for pumps and fans.
For orientation: a passive house is designed to a heating need of at most 15 kWh/m² a year; an unrenovated European house of the 1960s often needs 150 or more.
For design. Lower the need first (envelope, shading, compactness), then supply it efficiently, then cover it with renewables: in that order each step is smaller and cheaper.
In the app. The cost-optimal card gives heating and cooling needs for each package; A room through the year in its ideal mode gives the room's; the Degree days card gives a first, building-free measure.
Delivered energy is what crosses the building's boundary (the gas, the electricity from the grid). Primary energy counts what was taken from nature to deliver it: a kWh of grid electricity needs more than a kWh of fuel at the power plant. The primary energy factor converts one into the other. European codes usually rate buildings by non-renewable primary energy, in kWh/m² a year.
For design. The factors decide whether electricity or gas looks better on paper: a high electricity factor penalises heat pumps, a low one favours them. Photovoltaic electricity exported is often counted back at the same factor, which allows "net zero" primary energy.
In the app. The cost-optimal card uses 2.06 for electricity and 1 for gas (Pagliano's financial perspective), and 0.2 for pellets (Italy's non-renewable factor for solid biomass); its x axis is net primary energy, appliances included.
Keep in mind. Factors are set by each country and change as grids decarbonise; the card's are Italian and fixed over the years.
Source. EN ISO 52000-1; Leutgöb, K., Pagliano, L. and Zangheri, P. (2013), eceee Summer Study proceedings.
Heat pumpsCOP, SCOP, Carnot efficiency
A heat pump moves heat from a cold source (outdoor air, the ground, water) to a warmer sink (the heating water) using electricity. Its coefficient of performance (COP) is the heat delivered per unit of electricity, typically 3 to 4 over a season (SCOP). The ideal limit is the Carnot efficiency, which falls as the lift between source and sink grows:
COPCarnot = Tsink / (Tsink − Tsource)temperatures in kelvin. Real machines reach roughly 40 to 50 % of it.
For design. Low-temperature emitters (floor heating at 35 °C rather than radiators at 55 °C) and a warm source (the ground in winter) raise the COP. A heat pump in a mild climate works better than the same one in a cold climate.
In the app.Active systems and what this climate gives them estimates air- and ground-source heat pumps at 45 % of Carnot with water at 35 °C. The cost-optimal card takes the Danish Energy Agency's seasonal efficiencies (air 3.1, ground 3.35 with radiators) and scales them by the ratio of the Carnot efficiency here to Copenhagen's: Milan gives air 3.06 and ground 3.65.
Keep in mind. Both ways are this tool's scaling, a first estimate; defrosting, part load and the real emitter temperature change the result.
A photovoltaic module's rated power, in kW peak (kWp), is its output under standard test conditions: 1,000 W/m² of sun at 25 °C cell temperature. A module of 20 % efficiency gives 0.2 kWp per m². The specific yield is the electricity a year per kWp (kWh/kWp); the performance ratio (PR) is the yield divided by what the module would give at its rated efficiency from the sun it received, and covers losses from heat, dirt, wiring, the inverter and shade. A well-built system reaches about 0.8.
E = Pp × H × PRwhere E is the yearly electricity (kWh), Pp the peak power (kWp) and H the yearly irradiation on the module plane (kWh/m², numerically equal to hours at 1 kW/m²).
For design. Orientation and tilt set H; ventilation behind the module and avoiding shade set PR. In Central Europe a well-oriented roof gives about 1,000 kWh/kWp a year, in the south of Spain about 1,600 to 1,700.
In the app.Solar energy by tilt and direction gives the sun and the yield of 20 % modules at PR 0.8; the PV layers of the map give the Global Solar Atlas's yield; the site card's PV view and the cost-optimal card (30° south roof, PR 0.8) size systems.
A module turns a fifth of the sun into electricity and most of the rest into heat, so its cells run well above the air: 20 to 40 K in strong sun. Crystalline silicon loses about 0.3 to 0.4 % of its power per kelvin above 25 °C. The cell temperature depends on the sun, the air, the wind and how the module is mounted: a module on an open rack is cooled on both faces, one built into an insulated facade hardly at all.
Tc = Tair + αG (1 − η) / (U0 + U1v)a heat balance of Faiman's form, with the absorbed share α (1 − η) turned into heat: G the irradiance on the module, η its efficiency, v the wind speed, U0 and U1 heat loss coefficients that depend on the mounting.
For design. Leave a ventilated gap behind roof and facade modules where possible; in hot climates the temperature loss can be a tenth of the yield.
In the app. The site card's plain PV model uses −0.35 %/K and heat loss factors of 29 W/(m²K) on racks, 20 ventilated and 15 insulated; the Photovoltaics built in card computes each module's heat balance.
Keep in mind. The plain model's figures (21 % efficiency, −0.35 %/K, 14 % system losses) are from memory.
Source. Faiman, D. (2008), Assessing the outdoor operating temperature of photovoltaic modules, Progress in Photovoltaics 16(4).
Building-integrated photovoltaicsBIPV
Modules that are part of the building's skin: roof tiles, facade cladding, glazing with cells, shading devices and balustrades. They replace another material (and its cost and carbon) but usually receive less sun than an optimally tilted plant and run hotter, and in a city they are shaded and lit by neighbouring buildings. Which technology suits which place depends on the light it receives (direct or diffuse, its colour), its temperature and how it handles partial shade.
For design. Facades are worth covering where the sun is low (high latitudes, east and west facades for morning and evening); thin-film and some newer technologies cope better with diffuse light and heat; the carbon saved over the building's life, not the peak power, is the fair comparison.
In the app.Photovoltaics built in: which technology where ranks photovoltaic technologies for ways of building modules in, by urban setting and orientation, by energy per m², per kWp and carbon saved over the building's life, from the file's climate.
Keep in mind. The card shows only where its data are present; it is not on the published site. Many newer technologies' responses are generic, and the settings are idealised blocks.
Collectors that heat water or air directly: flat plates (a dark absorber under glass) and evacuated tubes, which lose less heat and keep working in cold weather. They convert a larger share of the sun than photovoltaics, often 40 to 60 % at the temperatures of domestic hot water, but heat is worth less than electricity and is hard to store across seasons.
For design. Size for summer hot water (oversizing gives surplus heat in summer); tilt steeper than for PV to favour winter.
In the app.Active systems and what this climate gives them gives a figure for solar thermal from the file; the cost-optimal card's solar water heater gives 35 % of the sun on a 45° collector, up to 60 % of the hot water need (an assumption).
Low-energy techniquesearth tubes, sky cooling, PCM and others
Techniques that use a free source or sink in the climate:
Earth tubes: supply air drawn through buried pipes, warmed in winter and cooled in summer by the ground.
Sky (radiative) cooling: surfaces that lose heat to a clear night sky, cooler than the air (see long-wave sky radiation).
Phase-change materials (PCM): layers that melt and freeze near room temperature, adding thermal mass to light buildings.
Heat recovery: ventilation that passes the warmth of exhaust air to the fresh air.
Night ventilation and evaporative cooling, daylight-linked lighting, small wind, rainwater use, cool and green roofs, dehumidification.
For design. Each works only in some climates: evaporative cooling needs dry air, sky cooling clear nights, earth tubes a ground cooler than the summer air, PCM a daily swing across its melting point.
In the app.Active systems and what this climate gives them (Technologies) gives each a figure read from the file and a fit for the climate.
Keep in mind. The figures are indications from climate alone: no loads, no site, no prices.
The EU's method for setting energy requirements: for a reference building, every combination of measures (insulation, windows, airtightness, ventilation, heat generator, renewables) is costed over its life, and each package is plotted by its global cost against its primary energy. The lower edge of the cloud of points is the cost curve; its lowest point is the cost-optimal level. National requirements should not be far less ambitious than it.
The curve is typically flat near its minimum: a wide range of energy levels costs about the same over the building's life, so going from the optimum towards zero energy often costs little.
For design. It shows which measures pay for themselves over the building's life and in what order, and what a zero-energy house costs over the cheapest.
In the app.Cost-optimal and zero energy: what a house costs over the years (Technologies) costs about 560,000 packages of the JRC's Milan house, with the energy worked out from the open file (so it follows the scenarios), and draws the curve at each year of its life, the optimum, the cheapest zero-energy package and the payback against the base house.
Keep in mind. The house and its costs are the JRC's Milan prototype everywhere; the country of the file changes prices, grid carbon and the code tested, not building costs. Cooling is likely high (no night ventilation). Several inputs are assumptions listed in the README.
Source. Directive 2010/31/EU; Delegated Regulation (EU) 244/2012; D'Agostino, D. and Parker, D. (2023), JRC106218; Leutgöb, Pagliano and Zangheri (2013), eceee.
Global cost and discount rate
The global cost is the investment at the start plus every later cost (energy, maintenance, replacements) brought back to its present value by discounting: a euro spent in i years is worth 1 / (1 + r)i today, with r the real discount rate.
CG(τ) = CI + Σj Σi=1..τCa,i(j) / (1 + r)i − VfCI the initial investment, Ca,i(j) the yearly cost of component j in year i, τ the period and Vf the residual value at its end (EN 15459).
For design. A high discount rate favours cheap buildings with high running costs; a low one favours investment in efficiency. Rising energy prices act the opposite way to discounting.
In the app. The cost-optimal card discounts at 4 % with prices rising 2.5 % a year (both real, Pagliano's financial perspective) and counts no residual value. Running cost is energy only by default (the lines meet near zero energy); the other choice adds maintenance and replacements as Regulation 244/2012 does.
Source. EN 15459 (economic evaluation procedure for energy systems in buildings).
Nearly zero and zero energy
The Energy Performance of Buildings Directive (2010/31/EU) asked that new buildings be nearly zero-energy buildings (nZEB) from 2021: a very high energy performance, with the small remaining need covered to a very significant extent by renewable energy on site or nearby. Each country set its own number. A net zero-energy building produces over the year as much energy as it uses (in primary or delivered terms). The 2024 recast of the directive moves new buildings to zero-emission buildings.
For design. Net zero over a year still draws from the grid in winter and exports in summer; the hours do not balance. Lowering the need first keeps the photovoltaic area needed within the roof.
In the app. The cost-optimal card marks the cheapest package at or below zero net primary energy, and draws energy classes under the axis (A1 to G, with plus classes for a house that gives energy back).
Keep in mind. Electricity is netted over the year, which flatters heat pumps with photovoltaics. The energy classes are those of the author's doctoral figure; their source scheme is not verified.
Source. Directive 2010/31/EU, Article 2 and Article 9.
Payback
The time an investment takes to repay itself through what it saves. The simple payback divides the extra investment by the yearly saving; the discounted payback counts discounting and price rises, and so is longer at a positive real discount rate.
tsimple = ΔCI / ΔCyear
For design. Payback is easy to explain but ignores everything after it: two measures with the same payback can differ greatly in life-cycle value. Compare by global cost where possible.
In the app. The cost-optimal card gives the year each package's cumulative cost falls below the base house's; the site card's PV view gives a simple payback (no discounting, degradation or replacement).
Keep in mind. The site card's PV costs (1,200 €/kWp on a roof, 400 €/m² on a facade) are round figures from memory.
The greenhouse gases emitted per kWh of electricity generated in a country, in grams of CO₂ equivalent. Life-cycle figures include building the plants and extracting the fuels. It ranges from a few tens of grams in grids of hydro, nuclear and wind to several hundred where coal and gas dominate.
For design. On a clean grid, an electric heat pump emits far less than a gas boiler; on a coal grid the gap narrows. Grids are getting cleaner, so a building lasting 60 years will mostly run on a future grid.
In the app. The file's country sets the grid factor (Ember's life-cycle carbon intensity via Our World in Data) in the cost-optimal and PV cards. Gas counts 202 g/kWh burnt (IPCC 2006), pellets 30 g/kWh (assumed, supply chain only); surplus electricity counts as grid carbon avoided.
Keep in mind. The grid is held as it is today over the years; gas's supply chain and methane leaks are not counted, which flatters gas against the grid's life-cycle factor.
A decision under uncertainty can be judged by its regret: in each future case, how much more it costs than the best choice for that case. The robust choice has the smallest worst regret over all cases considered (minimax regret). It need not be the cheapest in the expected case, but it is never far from the best.
For design. Energy prices and interest rates over 30 years are unknown; a robust package keeps its value whichever way they go.
In the app. The cost-optimal card's recommendations include the most robust package: the least worst regret over 27 cases (electricity, and gas and pellets, each 30 % lower, as is or higher; the discount rate 2 points lower, as is or higher), among the packages within 15 % of the cheapest. The others are the cheapest, the cheapest that meets the code, and the lowest carbon within a margin of the cheapest.
Keep in mind. The climate scenarios are not among the cases yet; the scenario on show is the climate of all of them.
A building designed today will stand in a warmer climate than the one its weather file records. These entries explain how the tool turns the file into future years under the IPCC scenarios, how it sets the file against measured past years, and how extremes are counted.
IPCC AR6Sixth Assessment Report, 2021–2023
The Intergovernmental Panel on Climate Change assesses the published science on climate change; it runs no models of its own. Its Sixth Assessment Report has three Working Group reports: the physical science (WG I, 2021), impacts, adaptation and vulnerability (WG II, 2022) and mitigation (WG III, 2022). Each opens with a Summary for Policymakers (SPM), the most cited and most carefully agreed text of the report.
The tool takes from WG I the best estimates of global warming by scenario and period (Table SPM.1, against 1850–1900) and the global mean sea level rise (Table 9.9). The framing of risk as hazard, exposure and vulnerability is WG II's.
For design. AR6 is the reference a client, a planning authority or a taxonomy assessment will accept for "the future climate". Quote its scenario and period, not "climate change" in general.
In the app. The scenario bar at the foot of the page; Hazards at the site (Hazards) for sea level and the risk framing.
Source. IPCC (2021), Climate Change 2021: The Physical Science Basis, Contribution of Working Group I to the Sixth Assessment Report. Further reading: 80.
A scenario of AR6 has two parts in its name. The first, SSP1 to SSP5, is a story of how society develops: population, economy, technology, cooperation between countries. The second is the radiative forcing reached by 2100 in W/m², the extra energy the greenhouse gases and aerosols hold at the top of the atmosphere. SSP2-4.5 is the "middle of the road" society reaching 4.5 W/m².
Scenario
In short
2021–2040
2041–2060
2081–2100
SSP1-2.6
strong mitigation
1.5 K
1.7 K
1.8 K
SSP2-4.5
middle of the road
1.5 K
2.0 K
2.7 K
SSP3-7.0
regional rivalry
1.5 K
2.1 K
3.6 K
SSP5-8.5
fossil-fuelled development
1.6 K
2.4 K
4.4 K
The figures are the best estimates of global surface warming against 1850–1900 (AR6 WG I, Table SPM.1). The near term hardly depends on the scenario; the choices made now show after mid-century.
For design. A building's life runs to the last column. SSP2-4.5 is a common central case; SSP5-8.5 is a stress test for overheating and cooling plant, not a forecast. Checking a design against two scenarios shows whether a decision is robust.
In the app. The scenario bar at the foot of the page switches every card to one of the four scenarios and three periods (keys [] and -=). This climate in short looks ahead to SSP2-4.5 in 2050.
Source. IPCC AR6 WG I (2021), Table SPM.1. Further reading: 80.
Global warming levelGWL, K above 1850–1900
The rise of the global mean surface temperature above the pre-industrial period 1850–1900, in kelvin. Many local changes (the warming of a city's summer, heavy rain) scale roughly with it, whichever scenario or year reaches it. So "a 2 K world" is a useful unit: the scenarios say when it may come.
A weather file already holds some warming. The tool takes the warming of the file's own years as about 1.09 K for 2011–2020, falling by 0.2 K a decade into the past, and adds only the difference to reach the scenario's level.
ΔTglobal = TSPM.1(scenario, period) − Tfile yearswhere the second term is the observed warming at the median year of the file.
For design. A file made of 2004–2018 months is already about 1 K warmer than pre-industrial; one made of 1980s months is cooler, so the same scenario shifts it further.
In the app. The note of the scenario bar gives the step applied to the open file.
Keep in mind. The file's years are placed by its median year only; a typical year is built of months from many years.
CMIP6 climate modelsCoupled Model Intercomparison Project, phase 6
Global climate models solve the physics of atmosphere, ocean, land and ice on a grid of cells some 100 to 250 km wide. CMIP6 is the coordinated set of their runs under the same scenarios, on which AR6 rests. Models differ in how much they warm and in where; their mean is more reliable than any one of them.
The tool uses one run of each of 14 model families (from the Pangeo copy of CMIP6), regridded to 2.5° and averaged without weights. From them it keeps, for every cell and month, how the local climate changes per kelvin of global warming: mean temperature, mean daily range, relative humidity, solar radiation, wind speed and cloud.
For design. The models give the pattern: a Mediterranean summer warms faster than the global mean, a coast slower than inland. That pattern, more than the global figure, sets cooling needs at a site.
In the app. Used by every card once a scenario is chosen; the page loads the 30° tile of its point once.
Keep in mind. A 2.5° cell (about 280 km) mixes land and sea, so a coastal site warms a little less than inland in the tool. Hot models only shape the pattern: the global level comes from AR6, which already constrains it. Two of the 14 models carry non-commercial licences (README).
A shortcut that treats each local change as proportional to global warming. For each model, scenario and cell, the change from 1995–2014 to three later periods is fitted by a line through the origin against that model's own global warming. The slope, say +1.6 K of July temperature per K of global warming, is the pattern; multiplied by the scenario's global step it gives the local change.
Δxlocal, month = pmonth × ΔTglobalwhere p is the pattern (per K), for temperature an addition, for humidity, sun, wind and cloud a relative change.
For design. It lets one tool serve any scenario and period from a few kilobytes per site, and it shows that the scenarios differ mainly in how far, not in which direction, a climate moves.
In the app. Behind the scenario bar. Without the patterns (no connection) the tool falls back to one uniform step times a land factor by latitude, from 1.1 in the tropics to 2.2 at 80°, an approximation of this tool; the bar says so.
Keep in mind. Linearity holds well for temperature and less well for humidity, sun, wind and cloud, and where feedbacks are not linear (sea ice, aerosols in SSP3-7.0). Rain is not changed at all.
Morphing turns a present weather file into a future one while keeping its sequence of weather: the same sunny days, the same windy nights, but warmer. After Belcher, Hacker and Powell (2005), each hour of a month is changed by that month's projected change:
Shift moves every hour by the change of the monthly mean.
Stretch scales a variable by a ratio. Temperature is stretched about the month's mean by the change of the daily range over the file's own mean daily range, so the days can swing more or less. The tool holds this stretch within ±0.5 and the other ratios at 0.5 or more.
T′ = T + ΔTm + αm (T − T̄m)where ΔTm is the month's mean change, T̄m the file's monthly mean and αm = Δ(daily range) / (file's mean daily range).
One day before (black) and after morphing (blue): the mean rises by the shift; the swing about the mean grows by the stretch.
Relative humidity, the three solar components and both illuminances, wind speed and cloud are stretched by their ratios; humidity ratio and dew point follow from the new temperature and relative humidity; sky long-wave radiation follows the air temperature to the fourth power.
For design. A morphed file can be run through any card, and through a simulation, like the original. It answers "how does this design fare in the 2050s?" with the site's own weather.
In the app. The scenario bar: every chart switches at once, since all twelve combinations of scenario and period are worked out ahead after a file opens.
Keep in mind. The weather sequence stays the file's: heatwaves become hotter but not longer or more frequent than the file's own. Morphing changes means and spreads, not the shape of extremes. Rain is not changed.
Source. Belcher, S. E., Hacker, J. N. and Powell, D. S. (2005), Constructing design weather data for future climates, Building Services Engineering Research and Technology 26(1). Further reading: 88.
Measured years
Instead of the typical year, a single past year as it was: the hourly ERA5 reanalysis at the file's latitude and longitude, from 1940 to the last whole year, fetched from Open-Meteo's archive and written as an EPW. A hot summer that happened, such as 2003 or 2022, can then be set beside a projected one.
For design. A measured extreme year is a test a design should pass; it is also concrete evidence for a client ("the summer of 2022"). Comparing a measured hot year with SSP2-4.5 in 2050 shows how ordinary such a summer may become.
In the app. The "Measured" menu of the scenario bar. Heatwaves and cold spells in the measured record (Temperature) can open the worst summer or coldest winter directly.
Keep in mind. ERA5 cells are about 9 to 30 km: a city's heat island and local wind are smoothed. Sky long-wave radiation is estimated from dew point and cloud; there is no illuminance. The first choice of a year waits for the download.
The pth percentile is the value below which p % of the values lie: the median is the 50th, the 90th is exceeded by one value in ten. A quantile is the same idea on a scale of 0 to 1. Between two sorted values the tool interpolates linearly (Hyndman and Fan's definition 7, the usual default).
For extremes the percentile is taken per calendar day: the 90th percentile of the daily maximum on 15 July is found from the days of a 15-day window around 15 July in every year of the base period 1991–2020. So "hot" means hot for that time of year at that place.
For design. Percentiles describe a spread without being pulled by one odd value, which is why design conditions are given as percentiles (the 0.4 % or 99.6 % hour) and why the tool's statistics show medians and the middle half.
In the app.Statistics by month (Any variable, any hours); the heatwave and cold spell definitions in Heatwaves and cold spells in the measured record.
Source. Hyndman, R. J. and Fan, Y. (1996), Sample quantiles in statistical packages, The American Statistician 50(4).
Heatwave
There is no single definition. The tool counts two:
Relative: at least 3 consecutive days with the daily maximum above that calendar day's 90th percentile (base 1991–2020), in the extended summer (May to September in the north, November to March in the south), after Perkins and Alexander (2013).
Fixed: at least 3 consecutive days with the maximum at 30 °C or more, at any time of year.
The relative one says what is unusual for the place, to which people and buildings are adapted; the fixed one says what is hot in absolute terms. Excess heat, the sum of the maxima above the 90th percentile over the heatwave days, measures intensity in degree days.
For design. Overheating should be checked for a hot spell, not a typical week: night cooling fails when nights stay warm for days. A cool refuge room and shading that works without the occupant matter most then.
In the app.Heatwaves and cold spells in the measured record (Temperature), on request; its counts feed the heat wave row of Hazards at the site.
Keep in mind. The months of the extended summer are from memory of the paper, not checked. ETCCDI's correction for years inside the base period is not done.
Source. Perkins, S. E. and Alexander, L. V. (2013), On the measurement of heat waves, Journal of Climate 26.
Cold spell
The mirror of a relative heatwave: at least 3 consecutive days with the daily minimum below that calendar day's 10th percentile, in the extended winter. This is the tool's own choice; the international index of cold spell duration (ETCCDI's CSDI) asks 6 days in any season.
For design. Cold spells size heating plant and test the freezing of pipes and water; in mild climates they expose houses built without heating or insulation.
In the app.Heatwaves and cold spells in the measured record: counts by year, trend, the coldest week, and the coldest winter opened as a measured year.
Two of the ETCCDI's fixed-threshold indices. A tropical night is a day whose minimum stays above 20 °C; a frost day one whose minimum falls below 0 °C. The tool also counts hot days, with the maximum at 30 °C or more.
For design. Tropical nights are the enemy of passive cooling: night ventilation and thermal mass need cool nights to discharge. A rising count of them is often the first sign that a building which works today will need cooling tomorrow. Frost days set the risk of freeze–thaw damage and of ice on paths.
In the app.Heatwaves and cold spells in the measured record; Hazards at the site counts frost and ice days in the file and its scenarios for its cold row.
A straight line fitted by least squares through a yearly series (the number of heatwave days, say) gives the change per decade. Its 95 % interval, from the standard error of the slope, says how sure that change is: an interval that includes zero means the record cannot tell a trend from chance.
For design. A clear trend in the measured record of the site is the most persuasive local evidence that the past is no longer a guide.
In the app.Heatwaves and cold spells in the measured record, for each index.
Keep in mind. The interval ignores serial correlation (one hot year making the next likelier), so it is somewhat too narrow. Some 35 years is short for rare events.
A return period of T years means a chance of 1/T in any one year: a 100-year flood has a 1 % chance each year, not one occurrence per century. The return level is the value reached at that period, such as the hottest day expected once in 20 years. Over a 50-year building life, the 100-year event has about a 40 % chance of happening at least once.
The tool fits a Gumbel distribution to the yearly maxima by L-moments (Hosking 1990):
x(T) = ξ − α ln(−ln(1 − 1/T))where α = l2 / ln 2 and ξ = l1 − 0.5772 α, from the first two L-moments of the yearly maxima.
For design. Codes size gutters, drains, flood levels and loads by return periods; a student should read "1 in 100" as a yearly probability.
In the app.Heatwaves and cold spells in the measured record; the wettest day once in 20 years in Hazards at the site.
Keep in mind. The fit is stationary: it assumes the climate does not change, so under warming the hot return levels come out low.
Source. Hosking, J. R. M. (1990), L-moments, Journal of the Royal Statistical Society B 52(1).
What a typical year leaves out
A typical year is stitched from the most ordinary months of a record. By design it leaves out the hottest summers and coldest winters, and often the heaviest rain. A heatwave in it is milder and rarer than the record holds; its rain is patchy, which can even change its Köppen class.
For design. Use the typical year for energy and for the character of a climate; use measured extreme years, or a scenario, for overheating, cooling plant and resilience. Ask "how does this fare in a hot year?", not only "in a normal one?".
In the app.Heatwaves and cold spells in the measured record ranks the typical year's hottest and coldest weeks among the measured years (Milan's IWEC year: 25th of 35 for its hottest week).
A site's climate is also what can go wrong there: floods, fires, storms, drought. These entries explain how the tool screens a site for climate-related hazards today and under a scenario, which maps and indices it reads, and what a screening can and cannot say.
Hazard, exposure, vulnerability, risk
In the framing of IPCC AR6 WG II and ISO 14091, risk arises where three things meet. The hazard is the physical event or trend (a 100-year flood, a heatwave). Exposure is what is in harm's way (the building, its users, its contents). Vulnerability is how badly they suffer when it happens (a basement flat floods, a raised floor does not). Remove any one and the risk falls.
Risk lies where hazard, exposure and vulnerability overlap. The tool rates the first circle only.
For design. Designers act mostly on exposure (where and how high to build) and vulnerability (materials, raised floors, passive survivability). A high hazard does not condemn a site; it sets the brief.
In the app.Hazards at the site (Hazards) rates hazards only, each with a one-line design response.
Keep in mind. Some sources the card reads (Aqueduct's drought and flood risk) already include exposure and vulnerability of people; they are marked as risk.
Source. IPCC AR6 WG II (2022); ISO 14091:2021, Adaptation to climate change: guidelines on vulnerability, impacts and risk assessment.
EU Taxonomy climate hazardsAppendix A, Regulation (EU) 2021/2139
The EU Taxonomy says when an investment counts as sustainable. To claim that a building "does no significant harm" to climate adaptation, its climate risks must be assessed against the list of Appendix A of the Climate Delegated Act. The list sorts hazards into four kinds (temperature, wind, water, solid mass) and two speeds: chronic (slow shifts such as rising temperature, sea level rise, water stress) and acute (events such as heatwaves, storms, floods, landslides).
For design. Investors and public clients increasingly ask for this assessment; a screening organised by Appendix A shows which hazards are material and need a closer study.
In the app.Hazards at the site lists its rows in the classes of Appendix A, with earthquake, volcano and tsunami listed beside them (not climate-related, so outside the appendix).
Source. Commission Delegated Regulation (EU) 2021/2139, Appendix A.
Hazard screeningClass 0
A screening is a first, broad look: which hazards could matter at all at this site, from data at hand, before any detailed study. Arup's universal taxonomy for natural hazard and climate risk (2024) calls this Class 0, a basic screening; later classes add site studies and models.
The tool rates 31 climate-related hazards low, medium or high in three columns: today, 2041–2060 and 2081–2100 under the scenario on the bar (SSP2-4.5 when none is chosen). The evidence comes from the weather file and its morphed years, from global hazard maps read at the site, and from IPCC AR6's sea level rise. Where GFDRR's ThinkHazard! covers the site's district, its level is shown as a second opinion.
For design. Use it to decide what to ask a specialist about, not as the answer: a "high" calls for the local flood map, the national annex of a Eurocode or a geotechnical survey.
In the app.Hazards at the site; its "Method and thresholds" fold lists every threshold. The same maps can be shown on the world map (header menu "Hazard on the map").
Keep in mind. The thresholds between low, medium and high are this tool's conventions, not a standard's. The maps are cells of 7 to 20 km, not the plot.
The temperature rows of the screening are read from the weather file and its scenarios:
Changing temperature: the projected rise of the mean (medium from 1 K, high from 2 K).
Heat stress: hours of strong heat stress, UTCI of 32 °C or more in the sun (medium from 50 h, high from 300 h), with the highest wet bulb temperature.
Heat wave: days in spells of 3 or more at 30 °C (medium from 1, high from 15); the measured record's median when it was fetched.
Cold wave and frost: frost days, ice days (maximum below 0 °C) and the lowest temperature (medium from 30 frost days or a lowest of −10 °C, high from 10 ice days or −20 °C).
For design. Heat is the hazard most certain to grow. Its design answers are shading, cross ventilation, night cooling, cool outdoor places and a cool refuge room.
In the app. The Temperature rows of Hazards at the site.
Keep in mind. A typical year understates heat waves; the levels are this tool's conventions.
A river flood is water leaving a river's channel after long or heavy rain or snowmelt upstream. Flood maps show the area under water in a flood of a given return period; the 100-year flood has a 1 % chance in any year. The tool reads the Copernicus GloFAS map (v2.1.2) as the share of each cell's area under water in a 100-year flood: medium from 0.5 %, high from 5 %. Aqueduct's river flood risk is shown beside it.
For design. Build out of the floodplain, or raise habitable floors and services above the flood level; make the ground floor resilient (materials that survive wetting, sockets high, no sleeping rooms below the line).
In the app.Hazards at the site, row River flood; "Hazard on the map", river flood.
Keep in mind. A cell's share does not locate the water: a site on a terrace in a flooded cell can be safe, and a site by a stream in a dry cell may not be. The national flood map is the reference.
Coastal floods come from storm surges and high tides on top of a rising mean sea. AR6 gives the global mean rise by scenario; the tool reads it at 2030, 2050 and 2090 for its three periods (Table 9.9). Whether a site is coastal comes from Aqueduct's coastal sub-basins; its height is the weather station's height in the file. A coastal site at 5 m or less is rated high for sea level rise, up to 10 m medium. Under SSP2-4.5 the median global rise against 1995–2014 is 0.20 m by 2050 and 0.48 m by 2090.
For design. Floor levels above the surge plus the future sea; no habitable basements on low coasts; salt-resistant renders and mortars where saline water reaches walls.
In the app.Hazards at the site, rows Sea level rise, Coastal flood, Saline intrusion and Coastal erosion (the last two a proxy from height and coast).
Keep in mind. The height is the station's, not the site's; the rise is the global mean, and local land movement can add or subtract much.
Source. IPCC AR6 WG I (2021), Table 9.9; Aqueduct 4.0 (World Resources Institute). Further reading: 80.
Water stress
Baseline water stress is the ratio of water withdrawn by people (homes, industry, farms) to the renewable water available in a river basin. Aqueduct 4.0 classes it from low (under 10 %) to extremely high (over 80 %); "arid, low water use" marks basins with little water and little use, which the tool rates medium. The tool rates 20–40 % medium and 40 % or more high.
For design. Rainwater harvesting, greywater reuse, low-flow fittings and drought-tolerant planting; in high stress, water is a design constraint like energy.
In the app.Hazards at the site, row Water stress; "Hazard on the map", water stress.
Keep in mind. Aqueduct's projections have no SSP2-4.5: the tool takes the worse of its optimistic (SSP1-2.6) and business-as-usual (SSP3-7.0) cases.
A drought is a long shortage of water against what is normal: of rain (meteorological), of soil moisture (agricultural) or of rivers and groundwater (hydrological). The tool reads Aqueduct 4.0's drought risk (0 to 1; medium from 0.4, high from 0.6) and gives the longest dry spell in the file.
For design. Water storage and reuse; planting that survives dry summers; on shrinkable clay soils, foundations deep enough to resist the ground's shrinking.
In the app.Hazards at the site, row Drought; "Hazard on the map", drought.
Keep in mind. Aqueduct's drought index is a risk (it weighs the people and activities exposed), not hazard alone.
The Canadian Fire Weather Index (Van Wagner 1987) rates how readily a fire would start and spread from the weather alone, fuel aside. It runs day by day on noon temperature, humidity, wind and the 24 h rain, through three moisture codes of the fuel (fine litter FFMC, loose organic layer DMC, deep organic layer DC), a spread index (ISI) and a fuel build-up index (BUI), to one number. The tool counts days at 21.3 or more (high) and 38 or more (very high).
For design. In fire weather: defensible space around the building, a non-combustible envelope and roof, ember-proof vents, gutters kept clear.
In the app.Hazards at the site, row Wildfire, from the file and its scenarios, with observed fires beside it.
Keep in mind. The equations are written from memory and checked against the system's standard first-day example; the day-length factors are from memory too; the year is run twice from standard start values instead of from snowmelt. A file without rain cannot give the index.
Source. Van Wagner, C. E. (1987), Development and structure of the Canadian Forest Fire Weather Index System, Canadian Forestry Service; Wang, Y., Anderson, K. R. and Suddaby, R. M. (2015), Updated source code for calculating fire danger indices in the Canadian Forest Fire Weather Index System, Natural Resources Canada.
Observed wildfire
Fires seen from space: the yearly chance that a 4 km cell around the site had a fire detected by the MODIS satellites in 2001–2020, from CLIMADA's data. Medium from 1 %, high from 10 %. It complements the Fire Weather Index: the index says the weather allows fire, the record says fires happen here.
For design. Near wildland, the same measures as for high fire weather, and access for fire engines.
In the app.Hazards at the site, row Wildfire; "Hazard on the map".
Keep in mind. The record includes fires set on purpose, such as burning of fields.
Storms that form over warm tropical seas, called hurricanes in the Atlantic and typhoons in the western Pacific, bringing extreme wind, rain and storm surge. The tool counts tracks of hurricane strength passing within 100 km per decade, 1980–2025, from the IBTrACS record: medium from 0.25 a decade, high from 1.
For design. Wind loads by the local code, impact-rated openings, roofs tied down to walls and walls to foundations, a safe room.
In the app.Hazards at the site, row Tropical cyclone; "Hazard on the map", tropical cyclones.
Keep in mind. IBTrACS winds are the US agencies' one-minute winds, or the WMO ten-minute wind divided by 0.88 where those are missing (a common conversion). The count is of the past; future cyclones are not projected here.
Mid-latitude storms of the cold season bring the strongest winds of Europe. The tool reads CLIMADA's ERA5 storm set: the gust reached once in 10 years (the third highest storm-day gust of 30 winters, 1980–2010, on a 0.5° grid), medium from 30 m/s, high from 35 m/s. For 2081–2100 it states the median change of CMIP6 models by 2070–2100 beside it. It also counts the hours with a mean wind of 15 m/s or more at 10 m in the file (medium from 5 h, high from 50 h) and takes the higher of the two levels; outside Europe that count is all there is.
For design. Design wind comes from the national annex to EN 1991-1-4; gusts are not in a weather file, which holds hourly means.
In the app.Hazards at the site, row Storm (straight-line wind); "Hazard on the map", winter storms.
Keep in mind. SSP1-2.6 is shown with SSP2-4.5's change and SSP3-7.0 with SSP5-8.5's; 2041–2060 has no projection.
Heavy rain is rated from the wettest day: once in 20 years in the measured record when it was fetched (medium from 60 mm, high from 120 mm), else the wettest day of the file (30 and 60 mm). Snow is the water that falls at 0.5 °C or below, and freezing rain the hours of precipitation between −3 and 0.5 °C: rough proxies from temperature.
For design. Gutters, drains and flat roofs sized for the storm, parapets and thresholds detailed against overflow; snow loads from the national annex to EN 1991-1-3.
In the app.Hazards at the site, rows Heavy rain, Snow, Ice storm.
Keep in mind. The scenarios do not change precipitation, so these rows have no future values. A typical year understates heavy rain.
How prone the ground is to slide, from slope, geology, faults, roads and forest loss; heavy rain is the usual trigger. The tool reads NASA's global landslide susceptibility map in five classes from very low to very high (medium for moderate, high from high).
For design. On susceptible ground: a geotechnical survey, drainage of slopes, retaining walls, and no building at the foot or crest of unstable slopes.
In the app.Hazards at the site, row Landslide; "Hazard on the map", landslides.
Keep in mind. Susceptibility is not likelihood: it says where slides could start, not how often.
Not a climate hazard, but listed for completeness. The tool counts shallow earthquakes of magnitude 5.5 or more within 100 km since 1973, from the USGS catalogue: medium from 0.2 a decade, high from 1.
For design. Seismic design follows EN 1998 (Eurocode 8) and its national annex, or the local code; it shapes structure, form and regularity from the first sketch.
In the app.Hazards at the site, row Earthquake; "Hazard on the map", earthquakes.
Keep in mind. A count of past quakes is not a seismic hazard model, which gives the ground acceleration with a probability; the open global models carry non-commercial terms and are not used.
A classification of Europe's regions at NUTS 3, the EU's statistical regions of roughly 150,000 to 800,000 people, by their climate hazards, exposure, sensitivity and adaptive capacity (Carter et al. 2018). Each region gets a typology class and a value for each indicator. The tool converts each hazard indicator into a z-score, its distance from the European mean in standard deviations, and calls a hazard group relevant from z = 0.5, this tool's choice.
For design. It places the site within its region's profile: which hazards stand out here against the rest of Europe.
In the app. Heads Hazards at the site for sites in Europe.
Keep in mind. A region is large; the typology describes it, not the plot.
Source. Carter, J. G., Hincks, S., Vlastaras, V., Connelly, A. and Handley, J. (2018), European Climate Risk Typology (RESIN project); method: Hincks, S., Carter, J. G. and Connelly, A. (2023), Global Environmental Change 83.
Copernicus climate indicators
Indicators of the Copernicus Climate Change Service for Europe's regions: hot days, tropical nights, heatwave days, days of high UTCI, frost days, high fire danger, extreme precipitation, dry spells, droughts and extreme wind. Present values come from ERA5 for 1991–2020 by NUTS 2 (NUTS 3 where offered); future changes are the median of the EURO-CORDEX regional model runs by NUTS 2, for 2041–2060 and 2081–2100.
For design. Regional models resolve mountains and coasts better than global ones, so these are a useful check on the tool's own figures from the weather file.
In the app. A folded table in Hazards at the site, for sites in Europe.
Keep in mind. The regional runs use the older RCP scenarios: RCP4.5 stands for SSP1-2.6 and SSP2-4.5, RCP8.5 for SSP3-7.0 and SSP5-8.5. A few regions do not match between NUTS editions and show nothing.
Books, papers and standards ranked by how often the reading lists of leading courses set them (MIT, UCL, TU Delft, Melbourne, SPA Vijayawada, IIT Tirupati, UFSC and others, read in October 2026), then by citations and by their place behind the methods of standards. The number is the rank in the whole list of 100. Most leading schools keep their lists behind a login, so the ranking is indicative; editions superseded since a list was written are named under Standards and their editions.
Climate and bioclimatic design17 titles
Givoni, B. (1969). Man, Climate and Architecture. Amsterdam: Elsevier; 2nd ed. 1976, London: Applied Science Publishers. Link Taught at: SPA Vijayawada 2025/26 (dated 1992 there), IIT Tirupati 2022 (dated 1998 there, probably conflated with Givoni 1998), Washington 2008, TKM 2022, Florida.
Koenigsberger, O. H., Ingersoll, T. G., Mayhew, A., & Szokolay, S. V. (1974). Manual of Tropical Housing and Building. Part One: Climatic Design. London: Longman (Orient Blackswan/Universities Press edition from 1975; Spanish: Viviendas y edificios en zonas cálidas y tropicales, Paraninfo, 1977). Co-authors (detail unconfirmed). Taught at: IIT Tirupati (required textbook), SPA Vijayawada, UFSC (Spanish ed.).
Szokolay, S. V. (2004). Introduction to Architectural Science: The Basis of Sustainable Design. Oxford: Architectural Press (detail unconfirmed); 2nd ed. 2008, DOI 10.4324/9780080878942; 3rd ed. 2014, Routledge, DOI 10.4324/9781315852409. Link Taught at: Melbourne ABPL90086 2026 (prescribed, 2008 ed.), IIT Tirupati (required, 2008 ed.), Florida; reported for UTS 11225 and UniSA.
Olgyay, V. (1963). Design with Climate: Bioclimatic Approach to Architectural Regionalism. Princeton, NJ: Princeton University Press; new and expanded ed. 2015, DOI 10.1515/9781400873685. Link Taught at: UFSC, Washington 2008, TKM 2022 (2015 ed.), Miami 2021; title of EPFL studio AR-201(bc); frames SPA Vijayawada's course.
Kwok, A. G., & Grondzik, W. T. The Green Studio Handbook: Environmental Strategies for Schematic Design. 2nd ed. 2011 (Architectural Press); 3rd ed. 2018 (Routledge). Editions and subtitle (detail unconfirmed). Link Taught at: Melbourne ABPL20036 2026 (recommended, 2nd ed.), NJIT 2025 (recommended, 3rd ed.).
Krishan, A., Baker, N., Yannas, S., & Szokolay, S. V. (Eds.) (1999). Climate Responsive Architecture: A Design Handbook for Energy Efficient Buildings. New Delhi: Tata McGraw-Hill (some lists give 2001). Link Taught at: IIT Tirupati (reference), TKM 2022.
Watson, D., & Labs, K. (1983). Climatic Design: Energy-Efficient Building Principles and Practices. New York: McGraw-Hill; reissued 1993 as Climatic Building Design (subtitle (detail unconfirmed)). Link Taught at: SPA Vijayawada 2025/26, Washington 2008.
Yanovshtchinsky, V., Huijbers, K., & van den Dobbelsteen, A. (2025). Architectuur als klimaatmachine (2nd ed.). ISBN 9789024468867. In Dutch. Link Taught at: TU Delft BKB1TE1 2025/26 (course book).
ASHRAE (2025). ASHRAE Handbook: Fundamentals, ch. 1 Psychrometrics and ch. 14 Climatic Design Information. Atlanta: ASHRAE. Link Taught at: Miami 2021 (2009 ed.). Free: through university Knovel subscriptions.
Evans, M. (1980). Housing, Climate and Comfort. London: Architectural Press. Link Taught at: SPA Vijayawada.
Lechner, N., & Andrasik, P. (2021). Heating, Cooling, Lighting: Sustainable Design Strategies Towards Net Zero Architecture (5th ed.). Hoboken, NJ: Wiley. ISBN 978-1-119-58574-9 (1st ed. 1991). Link Taught at: Miami 2021 (4th ed., 2015), USC 2019 and 2023 (recommended, snippets).
Pelsmakers, S. (2015). The Environmental Design Pocketbook (2nd ed.). London: RIBA Publishing. Link Taught at: UCL BENV0055.
Rivero, R. (1985). Acondicionamento Térmico Natural: Arquitetura e Clima. Porto Alegre: Editora da Universidade / Luzzatto. In Portuguese. Link Taught at: UFSC ECV5161.
Kottek, M., Grieser, J., Beck, C., Rudolf, B., & Rubel, F. (2006). World map of the Köppen-Geiger climate classification updated. Meteorologische Zeitschrift, 15(3), 259–263. DOI 10.1127/0941-2948/2006/0130. Link Free: open-access journal (not checked in the research).
Beck, H. E., Zimmermann, N. E., McVicar, T. R., Vergopolan, N., Berg, A., & Wood, E. F. (2018). Present and future Köppen-Geiger climate classification maps at 1-km resolution. Scientific Data, 5, 180214. DOI 10.1038/sdata.2018.214. Link Free: open-access journal (not checked in the research).
ANSI/ASHRAE Standard 169-2021, Climatic Data for Building Design Standards, with Addendum a (approved 28 November 2025), which fully revises the data. Atlanta: ASHRAE. Link
DeKay, M., & Brown, G. Z. (2014). Sun, Wind, and Light: Architectural Design Strategies (3rd ed.). Hoboken, NJ: Wiley. ISBN 978-0-470-94578-0 (1st ed. Brown, 1985). Link Taught at: ASU 2019 (secondary).
Nicol, J. F., & Humphreys, M. A. (2002). Adaptive thermal comfort and sustainable thermal standards for buildings. Energy and Buildings, 34(6), 563–572. DOI 10.1016/S0378-7788(02)00006-3. Taught at: none found.
de Dear, R. J., & Brager, G. S. (2002). Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55. Energy and Buildings, 34(6), 549–561. DOI 10.1016/S0378-7788(02)00005-1. Taught at: none found.
Nicol, F., Humphreys, M., & Roaf, S. (2012). Adaptive Thermal Comfort: Principles and Practice. London: Routledge. DOI 10.4324/9780203123010; companion: Humphreys, Nicol & Roaf (2015), Adaptive Thermal Comfort: Foundations and Analysis, DOI 10.4324/9781315765815. Link Taught at: TU Delft AR0137 2025/26 (chapters 1 to 5 and 8, one of two alternatives).
Bluyssen, P. M. (2009). The Indoor Environment Handbook: How to Make Buildings Healthy and Comfortable. London: Earthscan. Link Taught at: UCL BENV0055.
Frota, A. B., & Schiffer, S. R. (1988). Manual de Conforto Térmico. São Paulo: Nobel. In Portuguese. Link Taught at: UFSC ECV5161.
Kurvers, S., & Leyten, J. Indoor Climate and Adaptive Thermal Comfort (in press). Link Taught at: TU Delft AR0137 (alternative to item 23). Free: Dutch version free on Klimapedia.
Parsons, K. (2003). Human Thermal Environments (2nd ed.). London: Taylor & Francis (the syllabus gives Blackwell, Oxford). Link Taught at: TU Delft AR0137 (ch. 2).
ANSI/ASHRAE Standard 55-2023, Thermal Environmental Conditions for Human Occupancy, with Addendum a (approved 31 August 2026). Atlanta: ASHRAE. Link
EN 16798-1:2019, Energy performance of buildings. Ventilation for buildings. Part 1: Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics. Module M1-6. Brussels: CEN. Link
ISO 7730:2025, Ergonomics of the thermal environment. Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria (4th ed.). Geneva: ISO. Link
Givoni, B. (1992). Comfort, climate analysis and building design guidelines. Energy and Buildings, 18(1), 11–23. DOI 10.1016/0378-7788(92)90047-K.; the revised chart limits most software draws. (Crossref90047-K))
de Dear, R. J., & Brager, G. S. (1998). Developing an adaptive model of thermal comfort and preference. ASHRAE Transactions, 104(1), 145–167. ASHRAE RP-884. Link Free: eScholarship.
Fanger, P. O. (1970). Thermal Comfort: Analysis and Applications in Environmental Engineering. Copenhagen: Danish Technical Press; New York: McGraw-Hill, 1972 (subtitle (detail unconfirmed)). Link
Gagge, A. P., Fobelets, A. P., & Berglund, L. G. (1986). A standard predictive index of human response to the thermal environment. ASHRAE Transactions, 92(2B), 709–731. Link
Höppe, P. (1999). The physiological equivalent temperature: a universal index for the biometeorological assessment of the thermal environment. International Journal of Biometeorology, 43(2), 71–75. DOI 10.1007/s004840050118. Link Taught at: none found.
Bröde, P., Fiala, D., Błażejczyk, K., Holmér, I., Jendritzky, G., Kampmann, B., Tinz, B., & Havenith, G. (2012). Deriving the operational procedure for the Universal Thermal Climate Index (UTCI). International Journal of Biometeorology, 56(3), 481–494. DOI 10.1007/s00484-011-0454-1. Link Taught at: none found.
Jendritzky, G., de Dear, R., & Havenith, G. (2012). UTCI: why another thermal index? International Journal of Biometeorology, 56(3), 421–428. DOI 10.1007/s00484-011-0513-7. Link
Matzarakis, A., Rutz, F., & Mayer, H. (2010). Modelling radiation fluxes in simple and complex environments: basics of the RayMan model. International Journal of Biometeorology, 54(2), 131–139. DOI 10.1007/s00484-009-0261-0. Link
Błażejczyk, K., Epstein, Y., Jendritzky, G., Staiger, H., & Tinz, B. (2012). Comparison of UTCI to selected thermal indices. International Journal of Biometeorology, 56(3), 515–535. DOI 10.1007/s00484-011-0453-2 (online 2011). Link
Fiala, D., Havenith, G., Bröde, P., Kampmann, B., & Jendritzky, G. (2012). UTCI-Fiala multi-node model of human heat transfer and temperature regulation. International Journal of Biometeorology, 56(3), 429–441. DOI 10.1007/s00484-011-0424-7. Link
Duffie, J. A., & Beckman, W. A. (2013). Solar Engineering of Thermal Processes (4th ed.). Hoboken, NJ: Wiley. DOI 10.1002/9781118671603; 5th ed. with Blair (2020), Solar Engineering of Thermal Processes, Photovoltaics and Wind, DOI 10.1002/9781119540328. Link Taught at: MIT 4.430 2012 (3rd ed., 2006, for solar geometry).
Perez, R., Ineichen, P., Seals, R., Michalsky, J., & Stewart, R. (1990). Modeling daylight availability and irradiance components from direct and global irradiance. Solar Energy, 44(5), 271–289. DOI 10.1016/0038-092X(90)90055-H. Taught at: none found.
Reda, I., & Andreas, A. (2004). Solar position algorithm for solar radiation applications. Solar Energy, 76(5), 577–589. DOI 10.1016/j.solener.2003.12.003. Link
Schittich, C. (Ed.) (2003). In Detail: Solar Architecture. Basel: Birkhäuser. Link Taught at: Miami 2021 (secondary).
Reinhart, C. (2014). Daylighting Handbook I: Fundamentals, Designing with the Sun. Cambridge, MA: Building Technology Press. ISBN 9780692203637 (subtitle (detail unconfirmed)). Link Taught at: MIT 4.430 2012 (main text, chapters 1 to 8 mapped to lectures), MIT 4.401 2018.
Wienold, J., & Christoffersen, J. (2006). Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras. Energy and Buildings, 38(7), 743–757. DOI 10.1016/j.enbuild.2006.03.017. Link Taught at: MIT 4.430 2012.
Rea, M. S. (Ed.) (2000). The IESNA Lighting Handbook: Reference and Application (9th ed.). New York: Illuminating Engineering Society of North America (subtitle (detail unconfirmed)); also Harrold, R., & Mennie, D. (2003), IESNA Lighting Ready Reference. Link Taught at: MIT 4.430 2012 (recommended), Miami 2021 (Ready Reference).
Baker, N., & Steemers, K. (2002). Daylight Design of Buildings: A Handbook for Architects and Engineers. London: James & James / Taylor & Francis. Link Taught at: TU Delft AR0137 (ch. 10).
Fontoynont, M. (Ed.) (1999). Daylight Performance of Buildings. London: James & James [publisher and year U]. Link Taught at: MIT 4.430 2006 (recommended; snippet only).
Guzowski, M. (1999). Daylighting for Sustainable Design. New York: McGraw-Hill. Link Taught at: MIT 4.430 in 2006 (Andersen) and 2012 (Reinhart).
Inanici, M. (2006). Evaluation of high dynamic range photography as a luminance data acquisition system. Lighting Research and Technology, 38(2), 123–134. Link Taught at: MIT 4.430 2012.
Knoop, M., et al. (2019). Daylight: what makes the difference? Lighting Research & Technology. DOI 10.1177/1477153519869758. Link Taught at: UCL BENV0027.
Lam, W. M. C. (1986). Sunlighting as Formgiver for Architecture. New York: Van Nostrand Reinhold. Link Taught at: MIT 4.430 2012.
Larson, G. W., & Shakespeare, R. A. (2004). Rendering with Radiance: The Art and Science of Lighting Visualization. Booksurge (subtitle (detail unconfirmed)). Link Taught at: MIT 4.430 2012.
O'Connor, J., Lee, E., Rubinstein, F., & Selkowitz, S. (1997). Tips for Daylighting with Windows: The Integrated Approach. Berkeley, CA: Lawrence Berkeley National Laboratory (co-authors beyond the third and subtitle (detail unconfirmed)). Link Taught at: MIT 4.430 2012. Free: LBNL report (free availability not checked).
Reinhart, C. F. (2004). Lightswitch-2002: a model for manual and automated control of electric lighting and blinds. Solar Energy, 77(1), 15–28 (OCW credits Reinhart and Wienold; sole authorship (detail unconfirmed)). Link Taught at: MIT 4.430 2012.
Reinhart, C. (2018). Daylighting Handbook II: Daylight Simulations, Dynamic Facades. Cambridge, MA: Building Technology Press. ISBN 9780578407098 (subtitle (detail unconfirmed)). Link Taught at: MIT 4.401 2018.
Reinhart, C. F., & Weissman, D. A. (2012). The daylit area: correlating architectural student assessments with current and emerging daylight availability metrics. Building and Environment, 50, 155–162. Link Taught at: MIT 4.430 2012.
Reinhart, C. F., & Wienold, J. (2011). The daylighting dashboard: a simulation-based design analysis for daylit spaces. Building and Environment, 46(2), 386–396. Link Taught at: MIT 4.430 2012.
Littlefair, P., et al. (2022). Site Layout Planning for Daylight and Sunlight: A Guide to Good Practice (BR 209, 3rd ed.). Watford: BRE. Link Free: free PDF from BRE.
EN 17037:2018+A1:2021, Daylight in Buildings. Brussels: CEN. Link
Perez, R., Seals, R., & Michalsky, J. (1993). All-weather model for sky luminance distribution: preliminary configuration and validation. Solar Energy, 50(3), 235–245. DOI 10.1016/0038-092X(93)90017-I. (Crossref90017-I))
Reinhart, C. F., Mardaljevic, J., & Rogers, Z. (2006). Dynamic daylight performance metrics for sustainable building design. LEUKOS, 3(1), 7–31. DOI 10.1582/LEUKOS.2006.03.01.001. Link
Ward, G. J. (1994). The RADIANCE lighting simulation and rendering system. Proceedings of SIGGRAPH '94, 459–472. DOI 10.1145/192161.192286. Link
Tominaga, Y., Mochida, A., Yoshie, R., Kataoka, H., Nozu, T., Yoshikawa, M., & Shirasawa, T. (2008). AIJ guidelines for practical applications of CFD to pedestrian wind environment around buildings. Journal of Wind Engineering and Industrial Aerodynamics, 96(10–11), 1749–1761. DOI 10.1016/j.jweia.2008.02.058. Link
Blocken, B. (2015). Computational Fluid Dynamics for urban physics: importance, scales, possibilities, limitations and ten tips and tricks towards accurate and reliable simulations. Building and Environment, 91, 219–245. DOI 10.1016/j.buildenv.2015.02.015. Link
Krautheim, M., et al. (2014). City and Wind: Climate as an Architectural Instrument. Berlin: DOM Publishers. Link Taught at: Miami 2021 (secondary).
Givoni, B. (1998). Climate Considerations in Building and Urban Design. New York: Van Nostrand Reinhold. Link Taught at: Washington 2008.
Stull, R. B. (1988). An Introduction to Boundary Layer Meteorology. Dordrecht: Kluwer. DOI 10.1007/978-94-009-3027-8. Link
Stewart, I. D., & Oke, T. R. (2012). Local Climate Zones for urban temperature studies. Bulletin of the American Meteorological Society, 93(12), 1879–1900. DOI 10.1175/BAMS-D-11-00019.1. Link
Arnfield, A. J. (2003). Two decades of urban climate research: a review of turbulence, exchanges of energy and water, and the urban heat island. International Journal of Climatology, 23(1), 1–26. DOI 10.1002/joc.859. Link
Oke, T. R. (1973). City size and the urban heat island. Atmospheric Environment, 7(8), 769–779. DOI 10.1016/0004-6981(73)90140-6. (Crossref90140-6))
Santamouris, M. (2014). Cooling the cities: a review of reflective and green roof mitigation technologies to fight heat island and improve comfort in urban environments. Solar Energy, 103, 682–703. DOI 10.1016/j.solener.2012.07.003. Link
Oke, T. R. (1988). Street design and urban canopy layer climate. Energy and Buildings, 11(1–3), 103–113. DOI 10.1016/0378-7788(88)90026-6. (Crossref90026-6))
Oke, T. R., Mills, G., Christen, A., & Voogt, J. A. (2017). Urban Climates. Cambridge: Cambridge University Press. DOI 10.1017/9781139016476. in nine years, the clearest new standard text. Link
Oke, T. R. (1982). The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society, 108(455), 1–24. DOI 10.1002/qj.49710845502. Link
Masson, V. (2000). A physically-based scheme for the urban energy budget in atmospheric models. Boundary-Layer Meteorology, 94(3), 357–397. DOI 10.1023/A:1002463829265. Link
Bruse, M., & Fleer, H. (1998). Simulating surface–plant–air interactions inside urban environments with a three dimensional numerical model. Environmental Modelling & Software, 13(3–4), 373–384. DOI 10.1016/S1364-8152(98)00042-5. (Crossref00042-5))
Oke, T. R. (1981). Canyon geometry and the nocturnal urban heat island: comparison of scale model and field observations. Journal of Climatology, 1(3), 237–254. DOI 10.1002/joc.3370010304. Link
Ali-Toudert, F., & Mayer, H. (2006). Numerical study on the effects of aspect ratio and orientation of an urban street canyon on outdoor thermal comfort in hot and dry climate. Building and Environment, 41(2), 94–108. DOI 10.1016/j.buildenv.2005.01.013. Link
Gonçalves, J. C. S. (2010). The Environmental Performance of Tall Buildings. London: Earthscan. Link Taught at: Melbourne ABPL90086 2026 (prescribed).
Bokalders, V., & Block, M. (2010). The Whole Building Handbook: How to Design Healthy, Efficient and Sustainable Buildings. London: Earthscan. Link Taught at: UCL BENV0055.
Incropera, F. P., DeWitt, D. P., Bergman, T. L., & Lavine, A. S. (2006). Introduction to Heat Transfer (5th ed.). Hoboken, NJ: Wiley. Link Taught at: MIT 4.42J 2010 (further reading).
van der Linden, A. C. Building Physics (2nd English ed.). Link Taught at: TU Delft AR1B022 (chapters 1, 2, 4 to 7 and 10, as prior knowledge). Free: course material on Klimapedia.
Künzel, H. M. (1995). Simultaneous Heat and Moisture Transport in Building Components: One- and Two-Dimensional Calculation Using Simple Parameters. Stuttgart: Fraunhofer IRB Verlag. ISBN 978-3-8167-4103-9. Link
Moe, K. (2010). Thermally Active Surfaces in Architecture. New York: Princeton Architectural Press. Link Taught at: Miami 2021 (secondary).
Grondzik, W. T., & Kwok, A. G. (2019). Mechanical and Electrical Equipment for Buildings (13th ed.). Hoboken, NJ: Wiley. ISBN 978-1-119-46308-5. Link Taught at: NJIT 2025 (required), Miami 2021 (11th ed.), ASU 2019, Calgary 2012 (11th ed.).
Kreider, J. F., & Rabl, A. (1994). Heating and Cooling of Buildings: Design for Efficiency. New York: McGraw-Hill; later editions with Curtiss. Link Taught at: MIT 4.42J 2010 (further reading), NJIT 2025 (recommended, Kreider, Curtiss and Rabl).
Hensen, J. L. M., & Lamberts, R. (Eds.) (2011). Building Performance Simulation for Design and Operation. London: Spon. DOI 10.4324/9780203891612; 2nd ed. 2019, Routledge, DOI 10.1201/9780429402296. Link Taught at: MIT 4.430 2012 (Reinhart's chapter, pp. 235–276).
Crawley, D. B., Hand, J. W., Kummert, M., & Griffith, B. T. (2008). Contrasting the capabilities of building energy performance simulation programs. Building and Environment, 43(4), 661–673. DOI 10.1016/j.buildenv.2006.10.027. Link
ANSI/ASHRAE/IES Standard 90.1-2025, Energy Standard for Sites and Buildings Except Low-Rise Residential Buildings. Atlanta: ASHRAE. Link
ISO 52016-1:2017, Energy performance of buildings. Energy needs for heating and cooling, internal temperatures and sensible and latent heat loads. Part 1: Calculation procedures. Geneva: ISO. Link Free: REHVA Journal explainer.
Pelsmakers, S., et al. (2022). Designing for the Climate Emergency: A Guide for Architecture Students. Abingdon: Taylor & Francis (subtitle (detail unconfirmed)). Link Taught at: UCL BENV0055 2026/27, Nottingham pre-arrival (printed as "Designing For Climate Change").
Košir, M. (2019). Climate Adaptability of Buildings: Bioclimatic Design in the Light of Climate Change. Cham: Springer (Green Energy and Technology). Link Taught at: SPA Vijayawada 2025/26.
Commission Delegated Regulation (EU) 2021/2139 of 4 June 2021 (Climate Delegated Act), Appendix A, Generic criteria for DNSH to climate change adaptation, Section II classification of climate-related hazards. Link Free: free on EUR-Lex.
IPCC (2021). Summary for Policymakers. In V. Masson-Delmotte et al. (Eds.), Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report (pp. 3–32). Cambridge: Cambridge University Press. DOI 10.1017/9781009157896.001 (Crossref dates the print volume 2023; editors (detail unconfirmed)); with WG II, Climate Change 2022: Impacts, Adaptation and Vulnerability. Link Free: free at ipcc.ch.
CIBSE (2014). TM49: Design Summer Years for London. London: CIBSE. Link Free: free to CIBSE members.
CIBSE (2026). TM59: Overheating Risk in Dwellings: A Design Stage Methodology (2nd ed.). London: CIBSE. Link
Belcher, S. E., Hacker, J. N., & Powell, D. S. (2005). Constructing design weather data for future climates. Building Services Engineering Research and Technology, 26(1), 49–61. DOI 10.1191/0143624405bt112oa.; the morphing method. Link
Heschong, L. (1979). Thermal Delight in Architecture. Cambridge, MA: MIT Press. Link Taught at: MIT 4.430 2012 (recommended), UCL BENV0027 2026/27, Miami 2021.
Tanizaki, J. (1977). In Praise of Shadows. New Haven, CT: Leete's Island Books. Link Taught at: MIT 4.430 2012, Miami 2021.
55-2023 remains current, Addendum a Aug 2026 (ASHRAE)
ISO 13790
2008, withdrawn
ISO 52016-1:2017; any 2025 amendment (unconfirmed) (REHVA)
widely taught 5R1C
EN 15251
2007
EN 16798-1:2019; a split into 16798-1-1 to -1-5 was planned, publication not confirmed
BRE Digest 127 (driving rain)
withdrawn
DD 93:1984, then BS 8104:1992; later transfer to BS EN ISO 15927-3 (unconfirmed) (BRE)
BRE BR 209
2011
3rd ed. 2022, EN 17037-aligned tests for new homes
ISO 15469 / CIE S 011
2004/2003
joint ISO/CIE 15469 listed for 2026-06; status to check (ISO)
EN 17037
2018
2018+A1:2021
Szokolay
2008, 2nd ed.
3rd ed. 2014, Routledge
Melbourne 2026, IIT Tirupati
Lechner
4th ed. 2015
5th ed. 2021, with Andrasik, new subtitle
Miami 2021
Grondzik and Kwok, MEEB
11th ed.
13th ed. 2019
Miami 2021, Calgary 2012
Duffie and Beckman
3rd ed. 2006
5th ed. 2020, with Blair, adds PV and wind
MIT 4.430 2012
Hensen and Lamberts
2011
2nd ed. 2019
MIT 4.430 2012
Others
Olgyay expanded ed. 2015; Sun, Wind and Light 3rd ed. 2014 (author order reversed); Hens 2023 ed.; Architectuur als klimaatmachine 2nd ed. 2025; a 2026 Routledge record of Adaptive Thermal Comfort, DOI 10.1201/9781315645070, status (unconfirmed)
Texts under CC BY 4.0, as the rest of the tool's texts. Cite the tool as: Jakica, N. (2026).
Bioclimatic Explorer: climate-responsive design from a weather file [software]. https://climate-chart.pages.dev