Temperatura notturna al suolo

Microclimatev00.60 credits/per_km2

Responsible:Niccolò Calandri

Night Land Surface Temperature — RF Downscaling

Overview

The Night LST layer provides high-resolution nighttime land surface temperature maps by downscaling MODIS thermal data (~1 km) to 25 m resolution using Random Forest machine learning. Nighttime LST is particularly informative for urban heat island (UHI) analysis because it reveals heat retention patterns that daytime measurements cannot capture.

The downscaling uses three auxiliary variables:

NDVI from Sentinel-2 (10 m) — proxy for vegetation cooling effects
Land Cover from ESA WorldCover (10 m) — categorical surface type influencing thermal behavior
Elevation from SRTM (30 m) — altitude-driven temperature lapse rates

In addition to the LST map itself, the product generates:

UTFVI (Urban Thermal Field Variance Index) — quantifies urban heat island intensity
SUHI (Surface Urban Heat Island) analysis — temperature anomaly relative to rural surroundings

Calculation Methodology

Random Forest downscaling pipeline:

MODIS LST extraction — Nighttime LST values from MOD11A1 (Terra, ~22:30 local) and MYD11A1 (Aqua, ~01:30 local) are extracted for the analysis area
Quality filtering — Pixels with LST errors > 2 K or excessive cloud contamination are excluded
Feature preparation — NDVI, land cover, and elevation rasters are aligned to the MODIS grid
Model training — Random Forest learns the relationship between MODIS LST and the auxiliary features at 1 km resolution
High-resolution prediction — The trained model predicts LST at the 25 m resolution of the auxiliary features
UTFVI computation — UTFVI = (LST - LST_mean) / LST_mean, where LST_mean is the spatial mean
SUHI analysis — Temperature difference between urban and reference rural pixels

MODIS LST characteristics:

MOD11A1 (Terra): ~22:30 local overpass
MYD11A1 (Aqua): ~01:30 local overpass
Both provide daily LST with ±1 K accuracy under clear-sky conditions

Data Sources

Code	Name	Provider	Type	Coverage	Resolution	Availability
`MODIS/061/MOD11A1`	MODIS Terra LST Daily	NASA LP DAAC	Thermal raster	Global	~1 km	2000 — present
`MODIS/061/MYD11A1`	MODIS Aqua LST Daily	NASA LP DAAC	Thermal raster	Global	~1 km	2002 — present
`COPERNICUS/S2_SR_HARMONIZED`	Sentinel-2 L2A (NDVI)	ESA/Copernicus	Multispectral	Global	10 m	2017 — present
`ESA/WorldCover/v200`	ESA WorldCover 2021	ESA	Land cover	Global	10 m	2021
`USGS/SRTMGL1_003`	SRTM DEM	NASA/USGS	Elevation	60°N–56°S	30 m	2000

Technical Notes

Night LST is only available under clear-sky conditions; cloud-covered periods have no data
The RF downscaling preserves the thermal patterns from MODIS but adds spatial detail from the auxiliary variables
Urban areas typically show 2–8°C higher nighttime temperatures compared to surrounding rural areas (UHI effect)
UTFVI values are relative to the spatial mean within the analysis area, making them sensitive to the area extent
Elevation corrections assume a standard environmental lapse rate (~6.5°C/km)

References

Wan, Z. (2008). "New refinements and validation of the MODIS Land-Surface Temperature/Emissivity products". Remote Sensing of Environment, 112(1), pp. 59–74.
Liu, L., Zhang, Y. (2011). "Urban heat island analysis using the Landsat TM data and ASTER data: A case study in Hong Kong". Remote Sensing, 3(7), pp. 1535–1552.
Hutengs, C., Vohland, M. (2016). "Downscaling land surface temperatures at regional scales with random forest regression". Remote Sensing of Environment, 178, pp. 127–141.

Data Sources & References

Sources

WRD_MAQLT_991d

WRD_MTRLT_991d

Methodology

La Land Surface Temperature notturna viene calcolata attraverso una tecnica di downscaling spaziale che migliora la risoluzione dalla scala MODIS (~1km) a circa 25 metri. I dati di base sono MODIS MOD11A1 (Terra) e MYD11A1 (Aqua), banda 'LST_Night_1km', convertiti in gradi Celsius secondo i fattori di scala ufficiali del prodotto. Il downscaling è ottenuto con un modello di regressione addestrato sulla relazione tra la temperatura osservata e tre variabili ausiliarie disponibili ad alta risoluzione: 1) NDVI da Sentinel-2 (la vegetazione riduce la temperatura tramite evapotraspirazione), 2) Land Cover da ESA WorldCover (materiali diversi hanno diversa capacità termica), 3) Elevazione da SRTM DEM (la temperatura decresce con l'altitudine). Nuvole, valori nodata e corpi idrici vengono mascherati e non contribuiscono al risultato. La configurazione del modello e i dettagli implementativi del downscaling sono metodologia proprietaria 3Bee. Nota sulla risoluzione: i 25 metri sono la risoluzione della griglia di output; l'informazione termica proviene dalle osservazioni MODIS a 1 km e viene disaggregata spazialmente sulla base delle variabili ausiliarie sopra elencate. La temperatura superficiale notturna è un indicatore chiave del comfort termico urbano e dell'effetto isola di calore (UHI): di notte le aree urbane rilasciano il calore accumulato durante il giorno, mantenendo temperature più alte rispetto alle aree rurali circostanti.

Night Land Surface Temperature — RF Downscaling

Overview

The downscaling uses three auxiliary variables:

NDVI from Sentinel-2 (10 m) — proxy for vegetation cooling effects

Land Cover from ESA WorldCover (10 m) — categorical surface type influencing thermal behavior

Elevation from SRTM (30 m) — altitude-driven temperature lapse rates

In addition to the LST map itself, the product generates:

UTFVI (Urban Thermal Field Variance Index) — quantifies urban heat island intensity

SUHI (Surface Urban Heat Island) analysis — temperature anomaly relative to rural surroundings

Calculation Methodology

Random Forest downscaling pipeline:

MODIS LST extraction — Nighttime LST values from MOD11A1 (Terra, ~22:30 local) and MYD11A1 (Aqua, ~01:30 local) are extracted for the analysis area

Quality filtering — Pixels with LST errors > 2 K or excessive cloud contamination are excluded

Feature preparation — NDVI, land cover, and elevation rasters are aligned to the MODIS grid

Model training — Random Forest learns the relationship between MODIS LST and the auxiliary features at 1 km resolution

High-resolution prediction — The trained model predicts LST at the 25 m resolution of the auxiliary features

UTFVI computation — UTFVI = (LST - LST_mean) / LST_mean, where LST_mean is the spatial mean

SUHI analysis — Temperature difference between urban and reference rural pixels

MODIS LST characteristics:

MOD11A1 (Terra): ~22:30 local overpass

MYD11A1 (Aqua): ~01:30 local overpass

Both provide daily LST with ±1 K accuracy under clear-sky conditions

Code

Name

Provider

Type

Coverage

Resolution

Availability

MODIS/061/MOD11A1

MODIS Terra LST Daily

NASA LP DAAC

Thermal raster

Global

~1 km

2000 — present

MODIS/061/MYD11A1

MODIS Aqua LST Daily

NASA LP DAAC

Thermal raster

Global

~1 km

2002 — present

COPERNICUS/S2_SR_HARMONIZED

Sentinel-2 L2A (NDVI)

ESA/Copernicus

Multispectral

Global

10 m

2017 — present

ESA/WorldCover/v200

ESA WorldCover 2021

ESA

Land cover

Global

10 m

2021

USGS/SRTMGL1_003

SRTM DEM

NASA/USGS

Elevation

60°N–56°S

30 m

2000

Data Sources & References

Sources

WRD_MAQLT_991d

WRD_MTRLT_991d