Animals and insects rely on olfaction to navigate their environments. One way to understand the strategies they employ and the computational bases of this behaviour is the synthetic approach: as the capability gap between biological and artificial olfaction becomes smaller, it becomes possible to explore a greater range of olfactory-driven behaviours with robotic experiments. But experiments with electronic noses are often performed in laboratory conditions. These are not representative of an animal's natural environment, which can be considerably more challenging due to the uncontrolled influences of humidity, temperature and wind. We set out to collect a dataset that would give us more insights into the olfactory landscapes encountered in the field, exploring whether the data from electronic nose recordings can support navigation bio-inspired sensing strategies such as those based on intermittent odour encounters (bouts). For this purpose, we built a mobile electronic nose that can acquire day-long field recordings from multiple channels of metal-oxide gas sensors, together with environmental conditions like humidity, temperature and atmospheric pressure, geolocation, orientation, and acceleration data. Our system can simultaneously record the responses to odorants and modulate the sensor temperature, an active sensing technique which shows promise for improving the temporal resolution and odorant specificity of the gas sensor signal. This lets us explore a variety of heater modulation techniques and how these should be adapted in real-time to varying environmental conditions. We believe it will prove particularly useful in solving open questions in artificial and biological olfaction. Once embedded on a mobile robot it will also allow these strategies to be validated within a closed sensorimotor loop. DD and MS were funded from EU H2020 Grants 785907 and 945539 (Human Brain Project). MS was funded by MRC grant MR/T046759/1 (NeuroNex: From Odor to Action).