Introduction

Whiteflies cause significant agricultural losses worldwide, both through direct damage by feeding on phloem sap, and indirect damage by excreting honeydew and vectoring viruses (Li et al., 2021; Fiallo-Olivé et al., 2020; Stansly & Naranjo, 2010; Byrne & Bellows Jr, 1991). Trialeurodes vaporariorum (Westwood) and Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) are the most economically damaging species, known for their invasions and associated virus outbreaks (Milenovic et al., 2019; Nasruddin & Mound, 2016; Boykin, 2014; Cavalieri et al., 2014; Legg et al., 2014; Liu et al., 2012). Whiteflies are globally distributed, with Central Europe being on the northern edge of their range. Luxembourg is a good representative country for the climate of Central Europe, as it is located at its latitudinal midpoint.

Biological control is an indispensable tool in integrated pest management of whiteflies (Park et al., 2021; Horowitz et al., 2020; Rodríguez et al., 2019; Wang et al., 2018; Xie et al., 2018). In protected cropping systems, biocontrol relies mainly on augmentative release of parasitoid wasps and generalist predators, with Encarsia formosa (Gahan) and Eretmocerus eremicus (Rose and Zolnerowich) (Hymenoptera: Aphelinidae) currently being the most widely used and commercially available parasitoids. Biocontrol by parasitoids is strongly dictated by the climatic conditions, which are currently changing (Arias et al., 2021). Regional climate projections have advanced sufficiently to support climate change adaptation strategies in agriculture, especially when multi-model ensemble approach is applied (IPCC, 2021; Semenov & Stratonovitch, 2010; Tebaldi & Knutti, 2007). Today, a larger source of uncertainty in predicting the response of organisms to the changing climate comes from the lack of knowledge on the responses of the involved organisms (Tylianakis et al., 2008). Predictions of insect development in the future are based on experiments under constant conditions, but in nature environmental factors vary. A recent study used a climate chamber simulation to show that climate change will shorten the development time of the whitefly B. tabaci by almost half, potentially leading to faster population growth in the spring (Milenovic et al., 2023). Fitness parameters of its parasitoids are also directly influenced by abiotic factors (Zandi-Sohani & Shishehbor, 2011; Asplen et al., 2009; Qiu et al., 2006; McCutcheon & Simmons, 2001).

To understand the effectiveness of biocontrol methods in the future, simulations of future climate that consider daily changes in environmental factors are needed. Under these climate scenarios, several fitness parameters are trivial to describe the impact on these insects. In this preliminary study, we started with the longevity of adult E. eremicus parasitoids, one of the most important life-history traits for biocontrol agents (Plouvier & Wajnberg, 2018). This study aims to start closing this knowledge gap by using physical climatic chamber simulation to assess the longevity of the whitefly parasitoid E. eremicus in present and projected future climates.

Materials and methods

Two Bronson Incrementum 1400 and 1500 climatic chambers (Bronson Climate BV, The Netherlands) were used to simulate the present and the projected mid-term future (2061–2070) climate of Luxembourg, under RCP 8.5 (Supplementary Material S1), as described by Milenovic et al. (2023). Briefly, the climatic chambers were equipped with Valoya NS12 luminaries, set to deliver 480 µmol/m2s photosynthetic photon flux density (PPFD) at 20 cm distance constantly during 12 daylight hours. The CO2 concentration was maintained at 410 ppm and 700 ppm for the present and future conditions, respectively. The mean daily temperature for the present climate (2006–2015) was 19.8 °C, while the one for the future climate (2061–2070) was 23.4 °C. The mean daily relative humidity for the present climate was 69.2%, while the one for the future climate was 67.5% (Supplementary Material S1).

Parasitized T. vaporariorum nymphs were purchased (Koppert, Belgium), unpacked, and kept in glass petri dishes at room temperature (23 ± 1 °C) to facilitate the emergence. Adult parasitoids were kept for at least 2 h and no more than 24 h before being used for experiments. Approximately 50 unsexed adults and one 8 ± 3 mm cotton ball soaked in 300 µl diluted honey (50% Robinia pseudoacacia L. honey, 50% tap water) were introduced in mesh-capped glass vials (8 × 3 Ø cm) and placed inside the climatic chamber. An additional control vial was included in both climate experiments, containing non-soaked pressed cotton ball (no-food). A total of nine vials with honey and one no-food control vial were prepared per climate. The vials were placed in the two climatic chambers and dead individuals in each vial were counted daily on weekdays until no living individuals remained.

All statistical analyses were conducted using R software v 4.1.2 employing methodology as described by Ripamonti et al. (2022) with slight modifications. Individuals that died in the first 24 h of the experiment were excluded from analysis as this mortality is likely a result of their manipulation. E. eremicus longevity was estimated through Kaplan-Meier estimates. Generalized Additive Cox Model was applied with Peto’s correction for ties and experimental replicates stratified. Covariate (climate and diet) effects were graphically represented by Aalen’s Additive Regression Model (package ‘survival’, function ‘aareg’; Supplementary Material S2) (Therneau, 2022). Vial identity was added to the model as a random effect. GAM results were subjected to EMMs comparisons (Lenth, 2022), with Tukey’s p-value adjustment and all-pairwise comparisons (Hothorn et al., 2008). Summary statistics table was produced and paired with the results from all pairwise comparisons. The data that support the findings of this study are openly available in OSF at https://osf.io/qwsza/ (DOI: https://doi.org/10.17605/OSF.IO/QWSZA), while the complete R code for data analysis is publicly available at https://github.com/matteo-rpm/papers.

Results

Survival time analysis showed significant differences between E. eremicus adult longevity in the two climatic conditions. Future climate significantly shortened E. eremicus survival: the median survival time of the adults was 13 days, e.g. almost half of the 22 days survival observed in present climate (Table 1; Fig. 1).

Table 1 Summary statistics for Eretmocerus eremicus survival under present (2006–2015) and future climate conditions (2061–2070), measured as days between adult emergence and death
Full size table
Fig. 1
figure 1

Survival curves (Kaplan-Meier estimates) for adult Eretmocerus eremicus under present (2006–2015) and future climate conditions (2061–2070). Groups are divided based on climate (present vs. future) and diet (honey vs. no-food control). Risk table is reported, with the number of residual alive adults and percentage of total adults in parentheses

Full size image

A percentage of 75% of the adult parasitoids under future climate conditions died earlier (Q3, 14 days) than the 25% most short living adults under present climate conditions (Q1, 17 days) (Table 1). The most long-living specimen survived 18 days in the future climate, while the two most long living parasitoids survived 35 days in the present climate conditions. The individuals in the no-food control in both climate conditions had the shortest lifespan: the median survival time was only 2 days in future climate and 3 days in present climate (Table 1).

Discussion

The impact of climate change on survival of whitefly parasitoid E. eremicus is appraised here for the first time. The present study assesses the longevity of adults emerged from purchased parasitized nymphs, as representative tool of augmentative biocontrol practices. The results show severely reduced longevity in future climatic condition of Luxembourg (RCP 8.5) and likely in the one of all Central Europe. Three quarters of parasitoid individuals reared in future climate conditions lived less than 14 days after emergence. In comparison, under present climate conditions, the same mortality (75%) was reached only after 30 days, with an average longevity being more than twice longer than in the future condition experiment. At 14 days after emergence, when the future climate-reared insects reached 75% of mortality, only 25% of the present climate-reared parasitoids were dead. An overall accelerated development is to be expected: Qiu et al. (2004) and Gerling (1966) reported longer survival of the parasitoid at lower temperatures both in the presence and absence of the host. Although the mentioned authors employed a method of fixed constant temperatures, unspecified relative humidity, and uncontrolled CO2 concentration, some comparison between their results and the present findings can still be made.

In the study of Qiu et al. (2004), E. eremicus adults lived for a mean of 38.4 days at 15 °C, 33.8 days at 20 °C, and 18.9 days at 25 °C, in absence of the whitefly host. In the work of Gerling (1966), E. eremicus lived for 40.5 days at 15.5 °C, and 8.6 days at 26.7 °C. In the present study we observed a mean longevity of 22.1 days in the present climate conditions (with 19.8 °C mean temperature), and 11.2 days at the future climate conditions (with 23.4 °C mean temperature). In comparison with Qiu et al. (2004) and Gerling (1966), a discrepancy can be observed, with E. eremicus living shorter in the present study. This suggests a generally shorter lifespan in the present study compared to the life table parameters obtained at constant conditions.

Furthermore, according to the work of Qiu et al. (2004), E. eremicus longevity in the presence of its host is 37–51% shorter at the same environmental conditions. This indicates that the longevity of this biocontrol agent, when in presence of its host, would be likely even shorter in the future than predicted by the present experiment.

Overall, the results show that future climate conditions will significantly affect E. eremicus lifespan, consequently reducing the timespan for parasitization, and possibly limiting its biocontrol capacity. Previous works (Qiu et al., 2004; Headrick et al., 1999; Powell and Bellows Jr, 1992) described E. eremicus lifetime parasitism in relation to age-specific fecundity and adult age as dependent on the temperature. Considering this, future climate may enhance the parasitization of whitefly nymphs (Aregbesola et al., 2019), but contemporaneously limit E. eremicus survival. The temporal limitation caused by the warmer climate may affect the capacity of controlling whiteflies in the future. Additionally, the development of the whitefly host will be shortened (Milenovic et al., 2023), resulting in faster population growth and a shorter release window for the parasitoid. This may require more frequent releases, increasing agricultural production cost. As the present results are focused solely on E. eremicus longevity without its host, further research is needed to predict the effect of the future climate in more detail, which includes parasitization rate and the development of preimaginal stages.