This dataset contains environmental sensor data collected from eight strategically selected manholes in the urbanization of Paraíso Arenal, Córdoba, Spain, between March and June 2022. The data was used to develop and validate a machine learning model for predicting the activity levels of the American cockroach (Periplaneta americana) based on key environmental parameters: carbon dioxide (CO₂) concentration, temperature, and relative humidity. The dataset includes 904 observations, with each row representing a daily measurement. The target variable, "Activity," is a categorical variable with three levels: None (0): No individuals observed. Low (1): Countable number of individuals. High (2): Uncountable number of individuals.