Geo-temporal vehicular environmental sensing dataset

TL;DR

This paper introduces a dataset of CO₂ and temperature measurements collected by vehicles in Benevento, Italy, to support urban environmental research.

Contribution

A novel geo-temporal dataset capturing CO₂ and temperature using mobile sensing platforms in an urban setting.

Findings

The dataset includes georeferenced CO₂, temperature, humidity, and speed measurements from three vehicles.

The data enable trajectory reconstruction and analysis of spatio-temporal environmental patterns.

The dataset supports benchmarking and validation of mobile sensing strategies and interpolation algorithms.

Abstract

This dataset contains georeferenced, timestamped environmental measurements of CO₂ concentration and air temperature collected by a fleet of mobile sensing platforms operating in the urban area of Benevento, Italy. The data were acquired using low-power, vehicle-mounted sensor nodes designed to capture fine-grained spatio-temporal variability across the city. Each sensing unit integrates a Sensirion SCD41 nondispersive infrared CO₂ sensor, connected to an embedded computing board responsible for local preprocessing, GPS acquisition, timestamp synchronization, and data transmission.

Measurements were performed using three vehicles equipped with the same sensing hardware. The sensing fleet traversed the urban environment along diverse, randomly selected roads. All vehicles generated continuous timestamped and georeferenced (latitude, longitude, altitude) streams of CO₂, temperature, humidity, and speed measurements.

The dataset includes raw sensor readings, spatial coordinates, temporal metadata, and vehicle identifiers, enabling multi-vehicle trajectory reconstruction and environment-driven analyses.

This dataset is intended to support research on urban mobile sensing, spatio-temporal analysis, environmental monitoring, and vehicular sensor networks. Potential reuse includes benchmarking interpolation algorithms, validating mobile sensing strategies, and studying mobility-aware environmental sampling. The dataset may also be used to investigate sensor behavior in real-world dynamic conditions or to develop methods for robust, resource-aware urban monitoring.

Figures