Interpolation, Satellite‑Based Machine Learning, or Meteorological Simulation? A Comparison Analysis for Spatio‑temporal Mapping of Mesoscale Urban Air Temperature

Fine-resolution spatio-temporal maps of near-surface urban air temperature (Ta) provide crucial data inputs for sustainable
urban decision-making, personal heat exposure, and climate-relevant epidemiological studies. The recent availability of
IoT weather station data allows for high-resolution urban Ta mapping using approaches such as interpolation techniques
or machine learning (ML). This study is aimed at executing these approaches and traditional numerical modeling within a
practical and operational framework and evaluate their practicality and efficiency in cases where data availability, compu-
tational constraints, or specialized expertise pose challenges. We employ Netatmo crowd-sourced weather station data and
three geospatial mapping approaches: (1) Ordinary Kriging, (2) statistical ML model (using predictors primarily derived
from Earth Observation Data), and (3) weather research and forecasting model (WRF) to predict/map daily Ta at nearly
1-km spatial resolution in Warsaw (Poland) for June–September and compare the predictions against observations from 5
meteorological reference stations. The results reveal that ML can serve as a viable alternative approach to traditional krig-
ing and numerical simulation, characterized by reduced complexity and higher computational speeds within the domain of
urban meteorological studies (overall RMSE = 1.06 °C and R2 = 0.94, compared to ground-based meteorological stations).
The results have implications for identifying the urban regions vulnerable to overheating and evidence-based urban manage-
ment in response to climate change. Due to the open-sourced nature of the applied predictors and input parsimony, the ML
method can be easily replicated for other EU cities.


Dostęp do artykułu:


Cytowanie: Hassani, A., Santos, G.S., Schneider, P. et al. Interpolation, Satellite-Based Machine Learning, or Meteorological Simulation? A Comparison Analysis for Spatio-temporal Mapping of Mesoscale Urban Air Temperature. Environ Model Assess (2023).