More specifically:

(i) Explaining the spatiotemporal variations of air temperature within cities through statistical modelling, using EO-based covariates and sparse temperature measurements recorded by fixed sensors.

(ii) Assessing the feasibility of replacing the sparse fixed temperature measurements with measurements from low-cost thermometers carried by community members (citizens) walking along (flexible) transects, and how this affects the model accuracy.

(iii) Exploring the high-resolution spatial patterns of deprivation variation within DUAs through AI models, using EO, open geospatial data and micro-surveys.

(iv) Exploring the high-resolution spatial patterns of population within SSA cities through the development of statistical and machine learning methods.

(v) Deriving the divergent level of heat exposure of different populations in terms of their deprivation level, by combining (i), (iii) and (iv) and actively involving citizens (in building models and reflecting on results).

(vi) Assessing the feasibility of low-cost transferability of our models to other cities (mainly Lagos, with also a limited experiment in Argentina).