No abstract

ABSTRACT. This paper describes an experimental method for simulating an odour plume in the field so that its fine‐scale characteristics may be determined. It was found that the ‘odour’ arrived at a series of fixed points up to 15 m from the source in a series of discrete bursts, which were widely distributed in time but were typically 0.1 s long and 0.5 s apart. The strengths of the bursts were also found to be widely distributed and some contained considerable fluctuations. Thus an odour plume is not continuous, but intermittent, and appears at a fixed point downwind of the odour source as a series of bursts of odour which are variable in strength and duration. The distribution of the length of bursts and the time between them does not greatly vary with distance from source. Although concentration does decay with distance from source, instantaneous measurement would not be a reliable guide to the mean concentration and hence to position relative to source. To obtain reliable positional information from the odour plume an insect would have to average received stimuli over many seconds. More information is therefore required about the dynamics of insects' olfactory responses.

Summary Wind‐borne odour stimuli from a small point‐source of pheromone are intermittent owing to the effects of atmospheric turbulence on the odour plume. The work reported here measures the characteristics of the intermittent stimulus in open fields and forests during typical daytime atmospheric conditions. To model the pheromone plume, negatively charged atmospheric ions were used as a tracer. They were released from an ion generator, and ion detectors measured the fluctuating flux of ions at positions up to 20 m downwind in the open field case and 10 m in the forest. In both the open field and in the forest, ion signals were highly intermittent, with a signal present only 20% of the time. Ion signals recorded in the forest consisted of bursts with gaps between them of at least three‐fold greater duration than those from the open field. In both environments, bursts generally each comprised a series of ‘spikes’, on average three in the field and seven in the forest. To validate the use of ionized air plumes as models of pheromone plumes, the antennae of male Lymantria dispar (gypsy moth) were used as detectors to quantify the plume of synthetic (+)‐disparlure emanating from a 2000 ng point source placed ≈ 10 cm from the ion source. A comparison of ion signals and EAGs (electronantennograms) suggests that the antennae respond to the main spikes within a burst, but no consistent relationship between the strength of the spikes and the magnitude of the EAG response was found. The average strengths of bursts in the ion detector signal decreased systematically as the distance from the ion generator to the ion detector increased. A similar trend, however, was not detected in the EAG response.

Abstract. This article presents the theoretical motivation, implementation approach, and example validation results for a computationally efficient plume simulation model, designed to replicate both the short-term time signature and long-term exposure statistics of a chemical plume evolving in a turbulent flow. Within the resulting plume, the odor concentration is intermittent with rapidly changing spatial gradient. The model includes a wind field defined over the region of interest that is continuous, but which varies with location and time in both magnitude and direction.The plume shape takes a time varying sinuous form that is determined by the integrated effect of the wind field.Simulated and field data are compared. The motivation for the development of such a simulation model was the desire to evaluate various strategies for tracing odor plumes to their source, under identical conditions. The performance of such strategies depends in part on the instantaneous response of target receptors; therefore, the sequence of events is of considerable consequence and individual exemplar plume realizations are required. Due to the high number of required simulations, computational efficiency was critically important.