A computational analysis of two decades of Swiss insect records found that the weeks with the highest probability of detecting species richness advanced by an average of 0.9 weeks between two 20-year periods. The shift was observed across three surveyed insect groups and three altitudinal levels, indicating that the seasonal timing of effective alpine surveys has changed over time.
The study examined Lepidoptera, restricted to butterflies and day-flying moths; Odonata, comprising dragonflies and damselflies; and Orthoptera, comprising crickets and grasshoppers. The researchers used observations held by info fauna, the Swiss biological records centre, for 2003–2022 in the main analysis. They then repeated the optimization for 1983–2022, enabling a comparison between the first and second 20-year periods.
For each taxon, the records were organized by species, year, altitudinal level and calendar week, with the observations of adult individuals distributed among 52 weekly columns. Detectability for a species in a particular week and altitudinal level was modeled as P = 1 − e^(−N), where N was the number of observations of that species during the week. Under this assumption, a more abundant species had a higher probability of being detected during a survey.
The expected number of detectable species was calculated for each week, year and altitudinal level. An optimization algorithm then compared combinations of survey weeks, testing scenarios with one through five surveys per year. The objective was to maximize cumulative detection probability while keeping the number of annual visits as low as possible.
The resulting schedules differed among taxa and elevations. For butterflies in the lowlands, a single survey had its best window between weeks 26 and 30, corresponding to the first half of July. Two surveys were optimized around week 23, in early June, and week 28, in mid-July. For dragonflies at the subalpine level, the median best week for one survey was the last week of July, with annual variability placing the possible window between mid-July and the end of August; the optimized two-visit schedule used mid-July and early August.
Optimal windows generally moved later as elevation increased, with an estimated difference of about two weeks between lowland and subalpine levels. Orthoptera tended to become detectable later in the year than Odonata and Rhopalocera. The authors attributed this pattern to the larger spring and early-summer species pools of the latter two groups, rather than to a uniform seasonal schedule shared by all insects.
When the algorithm was run across 1983–2022, all three groups showed an average advance of 0.9 weeks in their optimal survey windows between the two 20-year periods. Rhopalocera and Orthoptera each advanced by 1.1 weeks, compared with 0.6 weeks for Odonata. The shift also varied by elevation: it was 1.6 weeks at the subalpine level, 0.9 weeks at lowland level and 1.2 weeks at mountain level. The standard deviation around the best windows showed no significant change for any taxon or altitudinal level.
The analysis shows that fixed survey dates may not remain optimal as insect activity periods shift. The authors frame the method as active adaptive monitoring, in which historical observations are used to improve subsequent survey design. They note that climate change can alter both habitat suitability and the phenological periods of insects and other cold-blooded species, making regular revision of existing survey schedules important. The model could also account in future for changes in voltinism or shifts in altitudinal ranges, although those factors were not included in the reported analysis.
The study is an observational analysis and model-based optimization rather than a field experiment. Its schedules are specific to the Swiss records, the three focal taxa, the modeled altitudinal levels and the assumed relationship between abundance and detection probability. It does not directly demonstrate that a particular visit schedule produces a fixed number of species in the field. Its practical result is instead a set of estimated windows intended to improve the likelihood of observing richness while reducing the number of surveys required, potentially strengthening the balance between monitoring costs and conservation work.
Subject of Research: Model-based optimization of seasonal survey timing for alpine insects using Swiss biological records
Article Title: Planning insect surveys in alpine ecosystems
Article References: Pellet, J. (2023). Planning insect surveys in alpine ecosystems. Alpine Entomology, 7, 201-204. https://doi.org/10.3897/alpento.7.110958
Image Credits: AI Generated
Keywords: alpine insects, species richness, survey timing, phenology, Lepidoptera, Odonata, Orthoptera, Switzerland, adaptive monitoring, detection probability, Planning, insect
