Spatio-Temporal Jump Model for Urban Thermal Comfort Monitoring

Abstract

Thermal comfort is essential for well-being in urban spaces, especially as cities face increasing heat from urbanization and climate change. Existing thermal comfort models usually overlook temporal dynamics alongside spatial dependencies. We address this problem by introducing a spatio-temporal jump model that clusters data with persistence across both spatial and temporal dimensions. This framework enhances interpretability, minimizes abrupt state changes, and easily handles missing data. We validate our approach through extensive simulations, demonstrating its accuracy in recovering the true underlying partition. When applied to hourly environmental data gathered from a set of weather stations located across the city of Singapore, our proposal identifies meaningful thermal comfort regimes, demonstrating its effectiveness in dynamic urban settings and suitability for real-world monitoring. The comparison of these regimes with feedback on thermal preference indicates the potential of an unsupervised approach to avoid extensive surveys.

Figures (13)

• Figure 1: Average Balanced Accuracy (BAC) computed between true and estimated latent sequences, for $P=10$, and for varying dimensions $M$ and temporal lengths $T$.Each subplot presents BAC results across different values of $\lambda$ (top) and $\gamma$ (bottom), with each curve corresponding to a fixed level of $\gamma$ (or $\lambda$, respectively). Missing data were introduced by randomly dropping 20% of simulated time-series observations, preliminarily adjusting the total number of observations to preserve dimensionality.
• Figure 2: Geographic locations of the 14 weather stations (S1–S14) used for monitoring thermal comfort across Singapore.
• Figure 3: Hourly distribution of comfort regimes (Cool, Neutral, Hot) throughout the day.
• Figure 4: Temporal distribution of comfort regimes (Cool, Neutral, Hot) across locations. The red dashed line indicates the average air temperature, relative humidity, rainfall, and wind speed versus time. The left y-axis shows the proportion of locations classified into each comfort regime, while the right y-axis reflects the corresponding variables.
• Figure 5: Temporal distribution of comfort regimes (Cool, Neutral, Hot) across locations, with the red dashed line representing the UTCI (°C) versus time. The left y-axis depicts the proportion of locations classified into each comfort regime, while the right y-axis indicates the UTCI.