Isole di calore urbano

Risksv00.80 credits/per_km2

Responsible:Niccolò Calandri

UTFVI — Urban Thermal Field Variance Index

Overview

The UTFVI (Urban Thermal Field Variance Index) quantifies the intensity of the urban heat island (UHI) effect by measuring how far each pixel's surface temperature departs from a reference temperature representative of the surrounding non-urbanised areas, normalised over the thermal range of the analysis area. It is derived from the Night LST product at 25 m.

The non-urban reference is identified automatically from the Built-up class of ESA WorldCover, excluding water bodies. The exact formulation of the index and the criteria used to identify the reference are 3Bee proprietary methodology.

Increasing values indicate a stronger heat island effect relative to the non-urbanised context; negative values indicate pixels cooler than the reference itself. The index is dimensionless and is classified into 5 intensity levels:

Class	UTFVI range	Interpretation
1	< 0.2	Minimum intensity (parks, dense vegetation, water bodies)
2	0.2 – 0.4	Low intensity
3	0.4 – 0.6	Moderate intensity
4	0.6 – 0.8	High intensity (dense urban)
5	> 0.8	Maximum intensity (industrial, highly sealed)

Calculation Methodology

UTFVI derivation pipeline:

Night LST generation — 25 m night-time LST obtained by spatially downscaling the MODIS signal (~1 km) using high-resolution auxiliary variables (proprietary model)
Non-urban reference identification — from the Built-up class of ESA WorldCover, excluding water bodies
Index computation — departure of each pixel from the reference, normalised over the thermal range of the area
Classification — UTFVI values are classified into 5 classes
Spatial analysis — comparison of urban vs. non-urban UTFVI distributions for SUHI characterisation

Key characteristics:

UTFVI is a relative index: values depend on the extent of the analysis area
Night-time UTFVI is preferred over daytime because it captures heat retention
Dense urban areas with high surface sealing show the highest UTFVI values
Vegetated areas act as cooling islands with low or negative UTFVI

Data Sources

Same as the Night LST layer: MODIS MOD11A1/MYD11A1, Sentinel-2 (NDVI), ESA WorldCover, SRTM DEM.

Technical Notes

The 25 m figure is the output grid resolution, not an observed resolution: the thermal information comes from MODIS observations at ~1 km and is spatially disaggregated using the auxiliary variables. The achievable detail is therefore bounded by those input layers
UTFVI is highly sensitive to the spatial extent of the area; comparisons between sites of different sizes must be made with caution
The index works best for urban and peri-urban areas with significant thermal gradients
In homogeneous landscapes values cluster near zero with low information content
Clear-sky conditions are required for the underlying LST data

References

Liu, L., Zhang, Y. (2011). "Urban heat island analysis using the Landsat TM data and ASTER data". Remote Sensing, 3(7), pp. 1535–1552. (conceptual reference: the 3Bee formulation differs from the one in the paper)
Zhang, J. et al. (2006). "A C++ program for retrieving LST from Landsat TM/ETM+". Computers & Geosciences, 32(10), pp. 1796–1805.
Oke, T.R. (1982). "The energetic basis of the urban heat island". Q. J. R. Meteorol. Soc., 108(455), pp. 1–24.

Data Sources & References

Sources

WRD_MAQLT_991d

WRD_MTRLT_991d

Methodology

L'UTFVI (Urban Thermal Field Variance Index) è un indice derivato dalla Land Surface Temperature notturna per valutare l'intensità dell'effetto isola di calore urbana (Urban Heat Island - UHI). L'indice misura di quanto la temperatura di ciascun pixel si discosta da una temperatura di riferimento rappresentativa delle aree non urbanizzate del contesto, normalizzata sull'escursione termica dell'area analizzata. Il riferimento non urbano viene individuato automaticamente a partire dalla classe 'Built-up' di ESA WorldCover, escludendo i corpi idrici. La formulazione esatta dell'indice e i criteri di individuazione del riferimento sono metodologia proprietaria 3Bee. Valori crescenti indicano una maggiore intensità dell'effetto isola di calore rispetto al contesto non urbanizzato; valori negativi indicano pixel più freschi del riferimento stesso. L'indice è adimensionale ed è classificato in 5 livelli di intensità, con soglie 0.2 / 0.4 / 0.6 / 0.8: classe 1 (verde) intensità minima, tipica di parchi e aree vegetate; classi intermedie 2, 3 e 4; classe 5 (magenta) intensità massima, tipica di centri urbani densi con scarsa vegetazione. L'analisi produce inoltre metriche SUHI (Surface Urban Heat Island) su anelli concentrici attorno alle aree urbane, salvate come metadati e non come mappa.

UTFVI — Urban Thermal Field Variance Index

Overview

Class	UTFVI range	Interpretation
1	< 0.2	Minimum intensity (parks, dense vegetation, water bodies)
2	0.2 – 0.4	Low intensity
3	0.4 – 0.6	Moderate intensity
4	0.6 – 0.8	High intensity (dense urban)
5	> 0.8	Maximum intensity (industrial, highly sealed)

Calculation Methodology

UTFVI derivation pipeline:

Night LST generation — 25 m night-time LST obtained by spatially downscaling the MODIS signal (~1 km) using high-resolution auxiliary variables (proprietary model)

Non-urban reference identification — from the Built-up class of ESA WorldCover, excluding water bodies

Index computation — departure of each pixel from the reference, normalised over the thermal range of the area

Classification — UTFVI values are classified into 5 classes

Spatial analysis — comparison of urban vs. non-urban UTFVI distributions for SUHI characterisation

Key characteristics:

UTFVI is a relative index: values depend on the extent of the analysis area

Night-time UTFVI is preferred over daytime because it captures heat retention

Dense urban areas with high surface sealing show the highest UTFVI values

Vegetated areas act as cooling islands with low or negative UTFVI

Data Sources & References

Sources

WRD_MAQLT_991d

WRD_MTRLT_991d