A Scale-Invariance-Based Algorithm Application For Land Surface Temperature Downscaling In Denmark

--

https://doi.org/10.3390/rs18132263 <-- shared paper

--

https://zenodo.org/records/20863040 <-- shared open data for downscaled LST dataset for Copenhagen

--

H/T @CLIM4cities

“💡 This study from… #CLIM4cities has focused zooms in on Land Surface Temperature (LST) — the heat stored in surfaces like asphalt, roofs and concrete.

Why is that interesting?

Because LST is directly linked to how cities absorb and release heat it is an instrument for urban planning and climate adaptation identifying the most effective heat mitigation measures.

Therefore, it is important to identify where these large LST values occur so that they may be counteracted with appropriate measures:

· Increasing tree cover and urban vegetation.

· Expanding parks and green spaces.

· Installing green roofs and walls.

· Reducing impervious surfaces where possible.

· Incorporating water features and blue infrastructure.

👉 To identify these hotspots, cities need LST maps at much finer spatial resolution than satellite observations typically provide.

💡 [Their] downscaling model transforms LST observations from ESA’s Sentinel-3 satellites into high-resolution maps suitable for urban-scale analyses.

🔍 Key highlights from the paper:

· Machine learning multi-timestamp models, which do not require re-training and re-tuning for every single timestamp, can perform reasonably well if appropriate standardisation is considered.

· Simpler isn’t always worse: a linear regression singe-timestamp model outperformed the more complex machine learning multi-timestamp models at the target resolution (as the latter tend to break the assumed scale-invariance hypothesis).

· The proposed linear regression approach achieved the lowest downscaling errors.

· Its simplicity also makes it attractive for operational applications.

Heat waves are becoming more frequent and intense as the climate changes. Tools like this can help cities identify urban heat hotspots and support evidence-based adaptation measures -contributing to practical urban planning and climate resilience…”

--

“With an ever-growing recognition of Land Surface Temperature (LST) as a key Essential Climate Variable (ECV), it becomes utmost important to have such a variable at the fine spatial and temporal scales of urban spaces and dynamics. Sentinel-3 provides coarse LST (1 km, daily) based on thermal imagery acquired by its Sea and Land Surface Temperature Radiometer (SLSTR) as well as fine Spectral Directional Reflectances (SDRs, 300 metres, every two days) synergically inferred from both SLSTR and the Ocean and Land Colour Instrument (OLCI), which gives the opportunity for using the latter as a predictor in the downscaling of the former. Herein, two scale-invariance-based architectures were developed: a single-timestamp (STS) model, trained with coarse data of the timestamp whose fine target it infers; and a multi-timestamp (MTS) one, trained with multiple timestamps. Note that while several Machine Learning (ML) models besides Linear Regression (LR) were considered for the MTS architecture, only LR was used for the STS one due to the limited amount of available data which the former require for hyperparameter tuning. The models were developed over four Danish Functional Urban Areas (FUAs) using SRD-derived indices and seasonal and geospatial predictors and validated against Landsat data. While Gradient Boosting (GB) achieved the best coarse-scale performance it corresponded to the worst fine-scale performer together with Random Forest (RF), indicating scale invariance breakdown. Tree-based models performed poorly due to extrapolation limitations, whereas Neural Net (NN) and LR proved more robust. After residual correction, single-timestamp LR achieved the best fine-scale performance, making it the most reliable and recommended architecture for operations. The overall results showed that, although ML models may better predict the target at their training scale, their performance may not significantly generalise at others, therefore revealing scale specificity. Furthermore, the results suggested that usage of the more general multi-timestamp architecture instead of the single one may deteriorate performance…”

#urbanclimate #downscaling #landsurfacetemperature #LST #AI #machinelearning #scaleinvariance #residualcorrection #Sentinel #Landsat #satellite #remotesensing #earthobservation #CLIM4cities #UrbanClimate #ClimateServices #MachineLearning #ClimateAdaptation #heatwave #temperature #ontheground #Copenhagen #Denmark #impervioussurface #asphalt #roof #concrete #albedo #heatabsorption #mitigation #urban #urbancentre #treecover #vegetation #urbanheatisland #planning #design #hotspots #monitoring #spatialanalysis #spatiotemporal #model #modeling #usecase #operational #climatechange #extremeweather #evidencebased #adapation #sustainability #urbanplanning #climateresilience #EssentialClimateVariable #ECV

@+ATLANTIC | @Danish Meteorological Institute | @ESA Φ-lab Collaborative Innovation Network | @CLIM4cities