Abstract: To study gas dispersion, several statisticalgasdistributionmodelling approaches have
been proposed recently. A crucial assumption in these approaches is that gasdistribution models are
learned from measurements that are generated by a time-invariant random process which can capture
certain fluctuations in the gasdistribution. More accurate models can be obtained by modelling
changes in the random process over time. In this work we propose a time-scale parameter that relates
the age of measurements to their validity to build the gasdistribution model in a recency function.
The parameters of the recency function define a time-scale and can be learned. The time-scale
represents a compromise between two conflicting requirements to obtain accurate gasdistribution
models: using as many measurements as possible and using only very recent measurements. We
have studied several recency functions in a time-dependent extension of the Kernel DM+V. Based
on real-world experiments and simulations of gas dispersal (presented in this paper) we demonstrate
that TD Kernel DM+V improves the obtained gasdistribution models in dynamic situations. This
represents an important step towards statisticalmodelling of evolving gasdistributions.