Continuous monitoring of personal exposures indoors and outdoors to assess human environmental resilience through sleep quality

Understanding how environmental conditions influence human resilience is critical for designing climate-responsive buildings. Sleep quality represents a key human-centric indicator of environmental resilience, impacting cumulative exposure to different stimuli. This study presents an exploratory analysis of longitudinal monitoring data which examines how environmental exposures, physiological responses, and personal characteristics interact to shape human resilience through sleep quality in real-life settings. Sixteen participants were continuously monitored for ten days using wearable devices to collect physiological data and characterize environmental exposure patterns associated with sleep quality. A composite Predicted Sleep Score (PSS) was developed by regressing sleep stages and efficiency (R 2 = 0.713) and, together with resting heart rate (RHR), was used as complementary indicator s of physiological dimensions of sleep quality. With PSS and RHR, three distinct sleeper profiles were defined using k-means clustering: poor, resilient, and good sleepers. Six parameters differed significantly across clusters: nocturnal levels of temperature, illuminance, total volatile organic compounds and formaldehyde, and the daily mean level of exhaled CO 2 and temperature. The cluster with resilient sleepers was characterized by slightly lower bedroom temperatures (−1.7 ± 0.7 °C), along with 11.6 ± 1.3 more minutes of deep sleep and 2.0 ± 0.3% higher sleep efficiency. A mixed-effects model showed that female participants had 50% probability of being classified in the poor sleepers' cluster, whereas males and participants with PSQI