Thermal Comfort • Human-biometeorology

UTCI – Universal Thermal Climate Index

A comprehensive thermal comfort index that combines air temperature, wind speed, humidity, and mean radiant temperature to express how the outdoor environment “feels” for humans.

1. Scientific Definition of UTCI

The Universal Thermal Climate Index (UTCI) is a human-biometeorological index that quantifies outdoor thermal stress and perceived temperature. It is based on an advanced thermophysiological model of the human body and the environment.

Conceptually, UTCI is a function of four main meteorological variables:

UTCI = f(Ta, Tmrt, va, RH)

Where:

  • Ta: air temperature at 2 m (°C)
  • Tmrt: mean radiant temperature (°C)
  • va: wind speed at 10 m (m/s)
  • RH: relative humidity (%) or water vapour pressure

Typical Interpretation of UTCI (°C)

UTCI Range (°C) Thermal Stress Category
< −27 Extreme cold stress
−27 to −13 Very strong cold stress
−13 to 0 Strong to moderate cold stress
0 to 9 Slight cold stress
9 to 26 No thermal stress (comfortable)
26 to 32 Moderate heat stress
32 to 38 Strong heat stress
> 38 Very strong / extreme heat stress

Main Applications

  • Urban heat stress and outdoor thermal comfort mapping
  • Climate risk assessment and heatwave early warning
  • Urban morphology vs. microclimate (street canyons, greenery, shade)
  • Public health, labour safety, and outdoor activity planning

2. Data & Inputs for UTCI

Official UTCI calculation uses a complex polynomial based on a human thermoregulation model. In GIS & RS workflows, we typically approximate UTCI (or a UTCI-like index) using reanalysis data and/or meteorological stations.

Typical Data Sources

  • ERA5 / ERA5-Land (ECMWF reanalysis) – available in Google Earth Engine
    • Air temperature at 2 m
    • Dew point or humidity
    • Wind components (u, v) at 10 m
    • Radiation fluxes for Tmrt approximation
  • Local meteorological stations – for validation and calibration
  • Remote sensing products (e.g. Landsat / Sentinel-3 LST) combined with meteo data for urban heat mapping

Good Practice

  • Work at an appropriate temporal scale (hourly, daily max, or daytime period).
  • Avoid using UTCI as “exact” in RS-only workflows; document it as an approximation.
  • Mask out oceans / lakes if focusing on urban thermal comfort.
  • Always report the data source and time of day for the index.

Typical Resolution

Depends on the data source:

  • ERA5-Land: ~9 km (global, hourly)
  • Downscaled or fused products: 1 km or finer over cities
Unit: °C (apparent temperature) Type: Thermal comfort index Domain: Urban & microclimate

3. Google Earth Engine Code – UTCI-like Index (Educational)

This is a simplified, educational example of a UTCI-like index using ERA5-Land data. It does not implement the full official UTCI polynomial, but provides a useful proxy for mapping relative thermal stress over your AOI.
Steps: open code.earthengine.google.com → New Script → paste the code → draw your AOI as geometry on the map → click Run and export.

// Simplified UTCI-like index for any AOI using ERA5-Land
// -------------------------------------------------------
// 1) Go to: https://code.earthengine.google.com
// 2) Click "New Script" and paste this code.
// 3) On the map: draw your AOI (Polygon/Rectangle).
//    It will appear as a variable named 'geometry' in the left panel.
// 4) Click "Run" to display the UTCI-like index.
// 5) In the Tasks tab, click "Run" to export it to Google Drive.

// -------------------------------------------------------
// 1. Define Area of Interest (AOI)
// -------------------------------------------------------
var roi = geometry;  // Make sure a 'geometry' object exists in the left panel

// Center the map on the AOI
Map.centerObject(roi, 8);

// -------------------------------------------------------
// 2. Define time range
//    You can set a single hot day, a heatwave period, or a month.
// -------------------------------------------------------
var startDate = '2023-07-01';
var endDate   = '2023-07-07';

// -------------------------------------------------------
// 3. Load ERA5-Land hourly reanalysis
//    Dataset: ECMWF/ERA5_LAND/HOURLY
// -------------------------------------------------------
var era5 = ee.ImageCollection('ECMWF/ERA5_LAND/HOURLY')
  .filterBounds(roi)
  .filterDate(startDate, endDate);

// For demonstration: compute daily mean of relevant variables
var daily = era5
  .map(function(img) {
    // Add a date string (YYYY-MM-DD) for grouping
    var date = img.date().format('YYYY-MM-dd');
    return img.set('date', date);
  })
  .select([
    'temperature_2m',            // K
    'dewpoint_temperature_2m',   // K
    'u_component_of_wind_10m',   // m/s
    'v_component_of_wind_10m'    // m/s
  ])
  .reduceColumns({
    reducer: ee.Reducer.mean().repeat(4),
    selectors: ['temperature_2m','dewpoint_temperature_2m','u_component_of_wind_10m','v_component_of_wind_10m']
  });

// NOTE: The above column reduction is just an illustration of how to explore statistics.
// For a proper daily UTCI-like map, we aggregate by date via .toBands() / .reduce().
// Below we compute a single-period mean image over the selected time.

// -------------------------------------------------------
// 4. Compute mean image over the period
// -------------------------------------------------------
var meanImg = era5.mean().clip(roi);

// Extract variables
var TaK = meanImg.select('temperature_2m');          // Kelvin
var TdK = meanImg.select('dewpoint_temperature_2m'); // Kelvin
var u10 = meanImg.select('u_component_of_wind_10m');
var v10 = meanImg.select('v_component_of_wind_10m');

// Convert to Celsius
var Ta = TaK.subtract(273.15);  // air temperature (°C)
var Td = TdK.subtract(273.15);  // dew point (°C)

// Compute wind speed at 10 m
var windSpeed = u10.pow(2).add(v10.pow(2)).sqrt();

// Approximate relative humidity from Ta and Td
// RH ≈ 100 * exp((17.625*Td)/(243.04+Td)) / exp((17.625*Ta)/(243.04+Ta))
var es_Td = Td.multiply(17.625).divide(Td.add(243.04)).exp();
var es_Ta = Ta.multiply(17.625).divide(Ta.add(243.04)).exp();
var RH = es_Td.divide(es_Ta).multiply(100).clamp(0, 100);

// -------------------------------------------------------
// 5. Build a simple UTCI-like index
//    This is NOT the official UTCI formula.
//    It is a pedagogical proxy based on Ta, wind, and humidity.
// -------------------------------------------------------

// Very simple "feels-like" approximation:
// utci_like = Ta - 0.5 * windSpeed + 0.05 * (RH - 50)
var utci_like = Ta
  .subtract(windSpeed.multiply(0.5))
  .add(RH.subtract(50).multiply(0.05))
  .rename('UTCI_like');

// -------------------------------------------------------
// 6. Visualize
// -------------------------------------------------------
var visParams = {
  min: -10,
  max: 45,
  palette: [
    '#313695', '#4575b4', '#74add1', '#abd9e9',
    '#e0f3f8', '#ffffbf', '#fee090', '#f46d43',
    '#d73027', '#a50026'
  ]
};

Map.addLayer(utci_like, visParams, 'UTCI-like (°C)', true);

// Add an outline of the AOI
Map.addLayer(roi, {color: 'white'}, 'AOI', false);

// -------------------------------------------------------
// 7. Export to Google Drive
// -------------------------------------------------------
Export.image.toDrive({
  image: utci_like,
  description: 'UTCI_like_Export',
  fileNamePrefix: 'UTCI_like_Export',
  region: roi,
  scale: 1000,   // 1 km resolution (adjust as needed)
  crs: 'EPSG:4326',
  maxPixels: 1e13
});

// -------------------------------------------------------
// Notes:
// - This script uses a simplified UTCI-like formula for training/demo.
// - For scientific work, use dedicated UTCI implementations and
//   carefully documented workflows.
// -------------------------------------------------------