We integrate a decade of summer satellite observations with block level census indicators to estimate daytime air temperature across the entire city. Results indicate systematic exposure disparities: older adults, disabled residents and households without a car are disproportionately located in hotter blocks, commonly experiencing daytime conditions around 2 to 3 °C above the citywide mean.

Moving beyond surface mapping, we test whether thermal exposure aligns with the city's everyday movement structure. By combining street-network configuration (Space Syntax) with built-form descriptors (Space Matrix), we show that high-exposure areas also contain highly connected, heavily traversed corridors.

At the pedestrian scale, microclimate simulations clarify the dominant mechanism of discomfort: radiant load from the sun and heated urban surfaces. Simulations show that mean radiant temperature (Tmrt) drives thermal stress during the 10:00 to 14:00 period.

The contribution is a screening workflow that translates evidence into implementable targets. By jointly considering demographic vulnerability, local temperature variation and network centrality, the method identifies priority corridors where interventions are likely to protect the most exposed groups.

The approach avoids costly citywide microclimate modelling, providing a practical tool for local authorities to diagnose risk rapidly and track progress through outcome focused indicators.