Project Summary Extreme ambient temperature (Ta), which is becoming more common due to climate change, has been associated with the increased incidence of cardiometabolic disease and mortality. Dehydration induced by high Ta stimulates gluconeogenesis in the liver and promotes insulin resistance. Increased blood flow to the skin during heatwaves increases cardiac demand, resulting in cardiac ischemia, infarction, and stroke in individuals with pre-existing cardiovascular disease (CVD). Thus, individuals with diabetes are particularly vulnerable to heat waves due to impaired physiological responses to heat and CVD comorbidities. Glucose variability (GV), measured by continuous glucose monitoring (CGM), is fluctuations in blood glucose levels over a given interval of time (intra-day or inter-day) and is an important emerging parameter of dynamic glycemic control. Glycated hemoglobin (HbA1c), a marker of the average blood glucose levels over the previous 2 to 3 months, is the primary metric used to predict diabetic complications. However, in-depth analyses from large scale, randomized trials, and epidemiological studies have indicated that HbA1c alone may not sufficiently predict the risk of diabetes complications. To date, climate change and health (CCH) research has been done at the population- level with aggregated data and HbA1c levels. Although the previous studies provided valuable insight into climate change and diabetes care, the mean HbA1c level at the population level with aggregated data in different climate zones makes it difficult to understand the real impact of Ta on individual patients. There is an urgent need to address gaps in CCH research studying lifestyle behavioral responses to Ta under climate change, social-built environment, and diabetes outcomes with high spatiotemporal granularity. We will collect time-stamped lifestyle data using ecological momentary assessment (EMA) and actigraphy and perform 24-hour dietary recalls. CGM, a real-time, unobtrusive glucose measure will be blinded from participants during the 14-day observation period. We will collect hourly and daily Ta from weather stations, daily Ta estimates from the Daily Surface Weather Data on a 1-km Grid for North America, and develop a high resolution (1 km2) spatiotemporally resolved hourly Ta model in Connecticut using satellite data with a machine learning approach. Participants will also be asked to wear a personal Ta sensor. Leveraging data in GV and multi-level personal (demographic, comorbidities), lifestyle, and social-built environmental factors that our parent study provides, and linking them to Ta data, we propose the following specific aims: (1) identify the risk groups who are vulnerable to high or low Ta by examining demographics, comorbidities, lifestyle (physical activity, sleep, eating behaviors/diet), social-built environment and GV among multi-ethnic adults with T2D for 14 days each in 2 seasons, (2) examine the influence of Ta on time-varying li...