Responses of a widespread pest insect to extreme high temperatures are stage-dependent and divergent among seasonal cohorts

Insect populations in temperate climates can show seasonal differences in demographic rates. Extreme high-temperature events (EHTs) are increasing in frequency across all seasons. These may, in turn, disrupt insects’ season-specific demographic strategies. However, whether insect populations respond to EHTs in a season-specific fashion is not known, but may be critical for forecasting their responses to climate change. We conducted a series of common garden experiments measuring the inherent vital rates of spring, summer, and autumn cohorts of a global nonmigratory pest, the oriental fruit moth (Grapholita molesta), under constant mild temperatures, EHT exposures at different life stages, and short-term heat-hardening followed by EHT exposure at different life stages. We simultaneously conducted a 4-year survey of G. molesta in orchards to link our experimental results to observed population dynamics in the field. We encountered intrinsic variation in vital rates and reproductive traits among seasonal cohorts, such that summer cohorts had higher intrinsic population growth rates but smaller eggs than spring and autumn cohorts. Moreover, these responses varied among life stages to which the EHT was applied. Pre-acclimation via heat-hardening did not strongly affect these results but did slightly increase growth rates when applied to late-stage larvae. These results were qualitatively recovered in natural population time-series, with moderate EHTs enhancing population growth in the spring and reducing it in autumn. Our results underscore the importance of climatic seasonality in shaping season-specific thermal responses to EHTs among ectotherms. Season-specific EHT responses can potentially enhance growth of weak spring cohorts and depress growth of autumn cohorts failing in dormancy, thereby contributing to insects’ annual population persistence. Our study helps clarify how a deeper understanding of season and life-stage-specific heat responses can enable a more informed forecasting and management of multivoltine insect populations under climate change. Here are the raw data of this research.

Methods
Lab experiment: 

We conducted a common garden experiment to assess whether the three seasonal cohorts had different vital rates under benign climate conditions. For the grand-offsprings of each seasonal cohort, we added 40 newly laid eggs of G. molesta into the calyx cavity of a single apple. This was repeated for at least twenty apples per cohort. We then placed the inoculated apples under a mild climate (25±1°C, RH: 60±10%, and 16 Light: 8 Dark). We randomly photographed and measured the length of over 47 newly laid eggs for each seasonal cohort. After hatching, the young larvae fed inside their apple until they became mature larvae (3rd-4th instar) and then left the apple. We counted the larvae emerging from fruits, then transferred each one to its own small plastic tube (0.6 cm in diameter and 5 cm in length) for pupation. Once adults emerged, we randomly paired newly eclosing adults (male and female) into plastic cups (6 cm in diameter and 8 cm in length) fed with 5% honey-water solution in a cotton ball. Mated females could lay eggs on the wall of plastic cups. We renewed all cups every day and counted the number of eggs on the cup until the adults died. At least 20 pairs of adults were tested per seasonal cohort.

To investigate the basal demographic responses to EHTs for seasonal cohorts of G. molesta, we conducted additional common garden experiments where EHTs were applied to a particular life stage (eggs, young larvae, mature larvae emerging from fruits and young adults). We did not choose the pupal stage to be tested in this experiment because a previous study (Zheng et al., 2017) in this species suggested pupae to be resistant to EHTs. For each seasonal cohort, we prepared 20 Fuji apples (replicates) in the test groups of EHT exposure on eggs, young larvae, or mature larvae, and prepared 10 Fuji apples (replicates) in the test group of EHT exposure on new-emerging adults. For all the treatments, we added 40 newly laid eggs of G. molesta in the calyx cavity of each apple, and then placed all inoculated apples under mild rearing conditions (25±1°C, RH: 60±10%, and 16 Light: 8 Dark) in climate-controlled chambers. For each test group, when individuals developed into the first day of their focal life stage, we moved them into a new climate-controlled chamber with a heat-stress condition of 38±1°C, RH 60±10%, and light for 7 hours. We chose this heat-stress condition as an EHT exposure based on results from Zheng et al. (2017) and Liang et al. (2014), and natural high temperatures and durations in China’s temperate apple-growing regions in recent years which match the ‘typical’ EHTs. After this EHT exposure, we returned the individuals to the mild rearing condition and continued to rear them until they died. For the testing group of heating on eggs, young larvae (L1), or mature larvae (L3), we tracked the survival, age stage and sex of each tested individual across their entire life cycle. All resulting demographic rates were calculated as described above in the benign conditions. 

To investigate the effects of heat-hardening on demographic responses, we ran a third set of experiments exactly as previously described, but this time with a brief heat-hardening phase preceding the EHT which consisted of a two-hour pre-treatment at 35±1°C, RH: 60±10%, recovery at 25±1°C for 30 minutes, followed by a 7-hour EHT exposure as previously described. We chose this heat-hardening condition based on Zheng et al. (2017) and our pilot experiments. Demographic data was collected during every life cycle stage and transition during the experiment as described above.

Field survey:

We conducted field surveys for population density of G. molesta during their growing season (March to September) from 2019 to 2021 at Xingcheng, China (40.62°N, 120.74°E) and from 2018 to 2021 at Yingkou, China (40.20°N, 122.13°E). At each site, we selected between 1 apple orchard in Yingkou and 3 apple orchards in Xingcheng separated by at least 1 kilometer. Five G. molesta traps were installed in each orchard, which were spaced at least 200 meters apart. Traps consisted of a 30 cm diameter plastic basin filled 2/3 with water and a pheromone lure fixed above the center of the water. We counted the number of captured adults per device every day from Mar 1st to Sep 30th in each year. We renewed water and removed all trapped adults every morning, and the lure was replaced once a month. We obtained hourly records of temperatures and precipitation of 2018-2021 in each site from the database of CPC Global Unified Temperature and CPC Global Unified Gauge-Based Analysis of Daily Precipitation (https://psl.noaa.gov/data/gridded/index.html) and calculated the corresponding daily average temperatures, daily maximum temperatures, and daily accumulated precipitation.