Seasonal phenology of Empoasca fabae (Hemiptera: Cicadellidae) in Québec, Canada

Abstract

Climate change is reshaping insect population dynamics in North America, notably impacting the migratory pest Empoasca fabae (Harris) (Hemiptera: Cicadellidae). While its phenology is well studied in the United States, knowledge gaps exist regarding its dynamics in Eastern Canada, one of its northernmost migration areas. Our study integrates degree-day models, CLIMEX ecological niche modeling, and field-collected data from Québec to assess E. fabae seasonal phenology and monthly climatic suitability. Our results indicate that E. fabae completes one to two generations in Québec, with earlier emergence and higher generational potential in warmer southeastern regions compared to cooler northeastern regions. CLIMEX modeling showed that suitable climatic conditions for E. fabae growth begin in April, peak from May to September, and decline by November. First adult captures occurred from late May to early June, with population peaks in June-July and a decrease by September. Observed adult peaks occurred earlier than predicted by degree-day models, suggesting that additional environmental factors, such as wind patterns and host plant availability, influence early-season population dynamics. This study provides a comprehensive understanding of E. fabae phenology in Québec and highlights the importance of incorporating regional climate to predict population trends. Further research on diapause onset, late-season persistence, and migration patterns is needed to refine predictive models and inform pest management strategies in Québec. Understanding these factors will be essential in mitigating potential economic impacts amid ongoing climate change.

Introduction

Ongoing climate change is reshaping insect population dynamics in North America, particularly through shifts in overwintering ranges and seasonal phenology, with diverse consequences depending on species and life history (Maredia et al. 1998, Baker et al. 2015, Walter et al. 2018, Lawton et al. 2022). Empoasca fabae (Harris) (Hemiptera: Cicadellidae), the potato leafhopper, is a migratory insect pest species native to North America (DeLong, 1938, Ross et al. 1964) whose phenology is likely influenced by these changes. For instance, increased temperatures have led to the earlier arrival of E. fabae in northern areas of the United States (USA) and increased crop damage (Maredia et al. 1998, Baker et al. 2015). Although evidence of effects on overwintering populations is growing, it remains unclear how these changes influence E. fabae migratory population dynamics in northern areas, particularly in Canada, during the growing season and fall, and how this relates to its baseline phenology.

Each spring, E. fabae migrates from its overwintering zone in the southeastern USA to northern areas in the USA and Eastern Canada (DeLong 1938, Medler 1957, Specker et al. 1990, Sidumo et al. 2005) with arrival occurring from early May until mid-June, though some early arrivals have been reported by the end of March in North Central USA states (Maredia et al. 1998, Baker et al. 2015). Then, local generations emerge, and population size increases and peaks from June to August (Medler et al. 1966, Sidumo et al. 2005). The population is then expected to decline by September, likely associated with the return of individuals to the overwintering zone, the reduction of suitable hosts, and the beginning of freezing temperatures impacting individual survival and reproduction (Medler et al. 1966, Hogg and Hoffman 1989, Specker et al. 1990). However, from the end of October to early November, E. fabae can still be found outside the overwintering area (Medler et al. 1966).

Québec, Canada, is one of the final northern frontiers where E. fabae immigrates every spring (Bostanian et al. 2003, Plante et al. 2024). In this province, studies on E. fabae phenology have primarily focused on abundance from mid-May to the end of September (Bostanian et al. 2003, 2006, Saguez et al. 2014, Plante et al. 2024). These studies have also aimed to delineate management practices to reduce pest damage in field crops and vineyards, often assuming similar life history dynamics (eg adult emergence dates and female reproductive diapause) as in northern USA states (Appleton et al. 2003, Bostanian et al. 2006, Shi et al. 2025). Valiquette and Comtois (2022) proposed a seasonal model for E. fabae in Québec, and this model suggests that the first-generation adult emergence starts in mid-June and the second generation by the beginning of August. They also indicated that newborn females enter a reproductive diapause state—associated with the return migration (Taylor et al. 1995)—by mid-August and that non-migratory insects die by October (Valiquette and Comtois 2022). However, this proposed model did not account for regional variations within the province, and those variations, mainly related to minimum temperature, are suggested to impact the leafhopper population dynamics at regional scales (DeLong 1938, Hogg and Hoffman 1989).

Understanding the seasonal phenology of E. fabae in Québec and its variations within the province is crucial to further estimating the ongoing changes caused by climate change on this species in Canada (Taylor and Shields 2018). In this study, we aimed to describe the seasonal phenology of E. fabae in four geographical regions of the province using two approaches: (i) degree day models based on the thermal thresholds developed for each life stage of E. fabae and (ii) ecological niche modeling to estimate monthly areas of suitability.

Materials and Methods

Degree Days Model

We estimated the development time of E. fabae based on the accumulated degree days (ADD, °) for adult emergence (F1, F2), considering the first day of the year, January 1st, as the null model and the first adult capture on the yellow sticky traps as the first capture model for each region sampled (Fig. 1a–b). These two models were performed to estimate the adult emergence date and time and the number of generations based on two scenarios: (i) the hypothetical scenario where E. fabae is considered a local species, considering that relative species from the subtribe Empoascina overwinter in the egg stage, such as Hebata (Empoasca) erigeron (DeLong) (DeLong 1938) (null model), and (ii) the realistic scenario where E. fabae migrates to Canada (first capture) (Plante et al. 2024). Daily temperatures (at 2 m) for each region (see Field Sampling section) were obtained from the NASA Prediction of Worldwide Energy Resources (POWER) database (NASA 2024; version 2.4.3) to calculate degree-days (Fig. 1c). We used this database because no local weather station was available within a 4 km buffer zone (average distance to the nearest station: 7.9 km, range: 4.2–11.84 km). This database also had incomplete daily or monthly data. Considering the limitations of available weather data for the fields, the NASA POWER dataset was validated through a linear regression analysis (α = 0.05) comparing the minimum and maximum daily temperatures from the nearest available weather station to the fields (y-axis) with the corresponding predicted values of NASA POWER (x-axis). All analyses were conducted using R (version 4.4.2; R Core Team, 2024 ). Our analysis demonstrates high consistency between measurements (minimum and maximum values), as evidenced by the agreement index (d = 0.94–0.98) and coefficient of determination (R² = 0.87–0.95) across all regions (Supplementary Fig. S1). While mean absolute (MAE: 1.6–2.1°C) and root mean square (RMSE: 1.7–2.6°C) errors were slightly elevated in one region for maximum temperature (Chaudière-Appalaches: MAE = 2.7°C; RMSE = 3.1°C)—likely due to topographic and microclimatic influences (eg presence of ponds)—these divergences are within acceptable limits for phenological models (White et al. 2008, Barboza et al. 2024).

Fig. 1.

The geographical position(a-b), and temperature conditions of the four geographic regions in Québec, Canada (c). The red dashed line indicates the minimum temperature (-9°C) related to the survival of Empoasca fabae (Decker and Cunningham 1967, Decker and Maddox 1967) and for the delineation of its overwintering zone in southeastern USA (Sidumo et al. 2005). The temperature data were obtained from the NASA POWER dataset at the 2 m level.

The temperature thresholds adopted were 9°C for egg to nymph and 7.6°C for nymph to adult development (Hogg 1985). We calculated the ADD using the degree day package in R (Lyons 2022) for each stage using the simple average degree days method and the ‘zeroes out’ method when the daily temperature was below the threshold temperature (Mcmaster and Wilhelm 1997):

$$\begin{array}{l}\mathrm{D}\mathrm{D}=[(\mathrm{T}\mathrm{m}\mathrm{a}\mathrm{x}.+\mathrm{T}\mathrm{m}\mathrm{i}\mathrm{n}.)÷2]-\mathrm{T}\mathrm{b}\mathrm{a}\mathrm{s}\mathrm{e}\end{array}$$

where T_max is the maximum temperature, Tmin is the minimum temperature, and Tbase is the lower development threshold for each stage. The ADD for each stage was based on the estimations under 18 to 29 °C (Hogg 1985) and 13 to 24 °C (Sher and Shields 1991). In the warmer regime, the number of degree days accumulated from egg to nymph development is 253°, and from nymph to adult, it is 345° (Σ = 598°) (Hogg 1985). In the cool regime, those ADDs are 272° and 453°, respectively (Σ = 725°) (Sher and Shields 1991). Considering the two regimes in both models, we recorded the day as the egg hatch when the ADD reached 253°/272°. From that recorded day, we continued calculating the ADD for nymph development until it reached 345°/453° and noted the day. We recorded this last day as the adult emergence (Σ ADD 598°/725°), and another generation was estimated, assuming an interval of 5 days from the day of adult emergence, as E. fabae oviposition initiates on the 6th day after and continues up to 25 to 50 d after emergence (Sher and Shields 1991). We conducted these phenological estimations, considering the four fields where sampling was performed (see Field sampling section) in 2021 and 2022. Finally, we averaged both model estimations to determine the first and second generations’ emergence dates (Supplementary Fig. S2 and Supplementary Table S1) following the same approach as Sidumo et al. (2005).

Climex Model

We used the CLIMEX algorithm (version 4.0) to develop spatiotemporal climate suitability for E. fabae in Québec (Kriticos et al. 2015). This algorithm generates maps that identify potential areas based on climate conditions favorable for species growth and stress conditions (Table 1). It also includes predictions on how this suitability varies at temporal scales (eg monthly). As a result, it allowed us to identify how suitable areas for E. fabae vary over the year in the province.

Table 1.

Final parameter values used to model the seasonal climate suitability of Empoasca fabae based on the CLIMEX algorithm. The index description is based on Kriticos et al. (2015).

Index	Parameter	Value	Index description
Temperature	DV0 = lower threshold	-9°C	The temperature index models the species’ population growth rate as a function of the daily temperature cycle. Its formulation incorporates four parameters that define the shape and boundaries of the temperature range favorable to growth.
	DV1 = lower optimum temperature	13°C
	DV2 = upper optimum temperature	31°C
	DV3 = upper threshold	38°C
Moisture	SM0 = lower soil moisture threshold	0.1	It assumes soil moisture influences moisture content and microclimatic conditions. The index is scaled: 0 = over dry, 1 = represents 100% of the soil’s water-holding capacity, and > 1 indicates an excess moisture beyond field capacity, leading to runoff.
	SM1 = lower optimum soil moisture	0.4
	SM2 = upper optimum soil moisture	1.0
	SM3 = upper soil moisture threshold	1.5
Diapause	DPD0 = diapause induction day length	10 h	DPD0 is the average weekly day length that induces diapause. DPT0 is the average weekly minimum temperature (for winter diapause) or maximum temperature (for summer diapause) that induces diapause.
	DPT0 = diapause induction temperature	−9°C
Cold Stress	TTCS = cold stress temperature threshold	−9°C	Cold stress represents the accumulation of the difference between a threshold temperature and the average weekly minimum temperature. When the average weekly minimum temperature is below the TTCS, cold stress accumulates at a THCS.
	THCS = stress accumulation rate	−0.00003/week
Heat Stress	TTHS = heat stress temperature threshold	38°C	Heat stress indicates the temperature (TTHS) at which the species fails to survive when exposed. The higher the maximum temperature, the greater the stress each week.
	THHS = stress accumulation rate	0.0001/week
Wet Stress	SMWS = wet stress threshold	1.5#	Wet stress accumulates if soil moisture exceeds the SMWS. The difference between SMWS and the soil moisture level is multiplied by HWS to give the resultant stress.
	HWS = stress accumulation rate	0.002/week
Dry Stress	SMDS = dry stress threshold	0.1#	Dry stress accumulates when the soil moisture level falls below the SMDS. The difference between SMDS and the soil moisture level is multiplied by the HDS to give the resultant stress.
	HDS = stress accumulation rate	−0.007/week
Degree Days	PDD = degree days per generation	725°	PDD denotes the degree-days above DV0 (the lower temperature threshold for population growth) required to complete an entire generation.

The CLIMEX algorithm relies on information on species development thermal requirements and a climate database to fit models and generate Ecoclimatic (EI) and Weekly growth (GIw) index maps. The EI results from growth and stress parameters give each species an overall yearly measure of suitable location conditions. EI values range from 0 to 100, where 0 means the area is inadequate for species establishment; 0 to 30 means that the area has less favorable climate conditions for species growth and development; and > 30 means that the area has highly favorable climate conditions. On the other hand, the GIw index indicates favorable climate conditions over a week on a scale from 0 to 1, where values close to 0 indicate inappropriate conditions for species growth. The conditions are more favorable as GIw values are closer to one (Kriticos et al. 2015).

We used the information in the literature to establish the temperature thresholds for E. fabae development and diapause. At the same time, humidity indices and adjusted values were based on the areas of known occurrence of the species. A description of each model parameter is indicated in Table 1. For the EI model, we used the CliMond 10′ grid climatic data (https://www.climond.org/) at a spatial resolution of 10 km (Kriticos et al. 2012). The climatic data represent long-term values, including monthly averages of total precipitation (Ptotal), maximum (Tmax), and minimum (Tmin) temperatures, as well as relative humidity at 09:00 (RH 09:00) and 15:00 (RH 15:00) for recent historical climate. For the GIw model, a monthly time series from the Climate Research Unit (version 3.26, https://crudata.uea.ac.uk/cru/data/hrg/), covering the period from 1901 to 2019, was also used at 10 km of spatial resolution.

Climex Model Calibration

Temperature Index

We set the following temperature parameters: lower temperature threshold (DV0) as −9°C (Decker and Cunningham 1967, Decker and Maddox 1967), lower optimal temperature (DV1) as 13°C, the upper optimal temperature (DV2) as 31°C, and the upper-temperature threshold (DV3) as 38°C (Table 1) (Kouskolekas and Decker 1966).

Moisture Index

We defined the lower soil moisture threshold (SM0) as 0.1 and the lower ideal soil moisture (SM1) as 0.4 to be consistent with the soil moisture of those regions with high E. fabae occurrence records. The upper ideal soil moisture (SM2) was set at 1, indicating that the soil moisture content is 100% of its capacity. In contrast, the upper soil moisture threshold (SM3) was set at 1.5, representing areas with excessive soil moisture (Kriticos et al. 2015).

Diapause Index

E. fabae reproductive diapause is associated with photoperiod conditions ≤ 10 h at both high and low temperatures (Taylor et al. 1995). In addition, Sidumo et al. (2005) estimated the overwintering temperature range for E. fabae at −9°C. Thus, we defined the diapause induction day length (DPD0) as 10 h and the diapause induction temperature (DPT0) as −9°C.

Cold and Heat Stress

The cold stress temperature threshold (TTCS) was set at −9°C (Sidumo et al. 2005), and the stress accumulation rate (THCS) was set at −0.00003 week^-1. These values were adjusted to suit the cold regions of Canada (Kriticos et al., 2015). We set the heat stress temperature threshold (TTHS) at 38°C (Kouskolekas and Decker 1966), and the stress accumulation rate (THHS) was 0.0001 week^-1 (Kriticos et al. 2015).

Wet and Dry Stress

The wet stress threshold (SMWS) was set at 1.5, while the stress accumulation rate (HWS) was defined as 0.002 week^-1. The dry stress threshold (SMDS) was fixed at 0.1, with the stress accumulation rate (HDS) set at −0.007 week^-1. These values were adjusted for E. fabae to achieve a high Ecoclimatic Index (EI) in the USA and Canada.

Degree Days

As previously indicated, two-degree days models were determined for E. fabae (Hogg 1985, Sher and Shields 1991). The final model used as the degree-days per generation (PDD) was 725° because it provided a more accurate model consistent with the known distribution of E. fabae.

Climex Model Validation

We validated the model by adjusting the parameters (Table 1) to achieve the best association between areas with EI > 0 and the occurrence records of E. fabae. These occurrence data were compiled from the Global Biodiversity Information Facility (GBIF 2024), from which an initial retrieval resulted in 7,459 records. To reduce temporal bias arising from older records potentially not reflecting the current distribution of E. fabae, we filtered the dataset to include only occurrences dated between 1970 and 2025. Another potential bias was the presence of multiple data points within the same climate pixel resolution. To address this, we retained only one record per pixel (~10 km). Following both filtering steps, 704 occurrence records across North America were used and subsequently plotted on the final EI map, categorized by the year of occurrence: 1970 to 2010, 2011 to 2020, and 2021 to 2025.

For the GIw model, we performed a Spearman correlation (α = 0.05) between the number of adults caught (see Field sampling section) and the GIw for the corresponding week. We extracted the GIw for each sampled field from 2010 to 2019 and then used the averaged values to conduct the correlation analysis (Supplementary Table S2). We used this approach because monthly climate data for the field sampling years (2021 to 2022) were unavailable.

Field Sampling

We used a leafhopper collection survey conducted on strawberry farms in Québec during the growing seasons of 2021 and 2022 (Plante et al. 2024) to depict E. fabae seasonal phenology, comparing it with the degree days models and CLIMEX predictions. We selected one farm per geographical region explored by Plante et al. (2024), making a total of 4 farms with distances between farms ranging from 38 to 210 km (Fig. 1a–b). Briefly, each year, from the third week of May until the end of September, surveys were conducted using yellow sticky traps attached to wood pins placed at the level of the strawberry plants (Fig. 2a). The traps were spatially distributed to cover the fields at least 1 m apart from each other and changed weekly. The identification of E. fabae was based on external morphology (Fig. 2b), and male genitalia dissection using the key developed by Beirne (1956). For confirmation, we compared the identified specimens with those from the Canadian National Collection of Insects, Arachnids, and Nematodes in Ottawa (Ontario, Canada). For the first capture date, since the traps were not monitored daily after the first installation week for logistical reasons, we adopted a conservative approach, considering the first capture day as the last one when the trap was in the field when the first E. fabae was captured in each region (Table 2). Finally, we performed a Spearman correlation (α = 0.05) to verify whether the first capture is associated with latitude.

Fig. 2.

Yellow stick trap used to capture leafhoppers in the strawberry fields (a). Empoasca fabae adult with the primary morphological feature for identification based on the six white spots on the top of the head behind the eyes (b).

Table 2.

Sampling areas information related to each geographic region, year, the strawberries cultivars, number of traps per week, first installation and capture of Empoasca fabae in Québec, Canada.

Geographic region	Year	Cultivars	Traps	1st installation	1st capture
Capitale-Nationale	2021	Seascape	15	26-May	1-Jun
Chaudière-Appalaches		Malwina and Seascape	15-20	2-Jun	8-Jun
Mauricie		Malwina and Albion	15-20	24-May	30-May
Montérégie		Yambu/Murano and Seascape	10	24-May	6-Jun
Capitale-Nationale	2022	Clery and Seascape	15	23-May	6-Jun
Chaudière-Appalaches		Malwina and Seascape	12	7-Jun	13-Jun
Mauricie		Yambu and Seascape	10	16-May	21-May
Montérégie		Seascape	10	22-May	29-May

Results

Estimating Adult’s Emergence

By combining the Hogg (1985) and Sher and Shields (1991) degree days estimations, and the null and first capture models, we found that the date of the first-generation adult emergence of E. fabae in Québec ranged (minimum and maximum) from 9 July to 18 August, while the second from 10 September to 25 October (Table 3). Interestingly, these models indicated that the southeast regions of the province (Mauricie and Montérégie) had early emergence dates compared to the northeast ones (Capitale-Nationale and Chaudière-Appalaches). Nevertheless, the first capture model suggests that a second generation of insects would be expected only in Mauricie and Montérégie, as in the other regions, E. fabae did not reach enough ADD for adult emergence (Table 3). The number of days required for the first adult emergence was longer (90 to 124 d) than the second (47 to 90 da) in the null model. In the first capture models, the first (53 to 79 d) and second (49 to 85 d) generations took similar emergence times (Table 3).

Table 3.

Estimated date and number of days for Empoasca fabae adult emergence in four geographic regions in Québec, Canada. The null model considers E. fabae as an overwintering species in Canada, while the first capture is based on the first date of adult capture in yellow stick traps. The values are averaged according to the accumulated degree-days models of Hogg (1985) and Sher and Shields (1991).

Geographic region	Year	Emergence (null model)		Emergence (first capture)
			F1	F2	F1	F2
Capitale-Nationale	2021	06 Aug, 120 d	22 Oct, 77 d	14 Aug, 75 d	-
	2022	07 Aug, 97 d	-	18 Aug, 79 d	-
Chaudière-Appalaches	2021	26 Jul, 124 d	23 Sep, 60 d	12 Aug, 66 d	-
	2022	27 Jul, 124 d	25 Oct, 90 d	16 Aug, 65 d	-
Mauricie	2021	21 Jul, 119 d	25 Sep, 62 d	01 Aug, 64 d	23 Sep, 53 d
	2022	23 Jul, 90 d	21 Sep, 59 d	30 Jul, 71 d	23 Oct, 85 d
Montérégie	2021	09 Jul, 121 d	31 Aug, 47 d	28 Jul, 53 d	21 Sep, 50 d
	2022	13 Jul, 119 d	10 Sep, 54 d	28 Jul, 61 d	16 Sep, 49 d

Climex Models

The CLIMEX model performed well with the final set of parameters for E. fabae (Table 1). When this model was validated with the occurrence records of E. fabae, all occurrences aligned with moderate and high suitability areas (EI > 0; Fig. 3), and we could estimate how this suitability would vary monthly using the GIw model.

Fig. 3.

Potential suitability of North America for Empoasca fabae based on the Ecoclimatic Index [EI = 0 (unsuitable), EI > 0 < 30 (moderate suitability), EI > 30 (high suitability)]. The circles in green, red, and yellow represent the final set of occurrence records (n = 704) used for model calibration.

The seasonal model showed that the suitability of Québec for E. fabae varies monthly with periods of unsuitable and high conditions for species growth (Fig. 4). From January to March, the GIw values were closer to zero. Then, GIw increased in April, with the highest values from June to August (0.5 to 0.6), subsequently decreasing. The GIw values for the four regions showed a similar pattern, with the highest values from May to September and decreasing afterward (Fig. 5). These values also presented a high variability (standard deviation) based on the period evaluated from 2010 to 2019 (Fig. 5). No significant correlation was found between the number of adults and GIw values (r = −0.006, P = 0.94).

Fig. 4.

Seasonal climate suitability based on the weekly Growth Index (GIw) for Empoasca fabae in Québec (Canada). The index varied from 0 to 1, where values > 0.1 indicate growing conditions for the species.

Fig. 5.

Estimated weekly growth index based on the seasonal model using the CLIMEX algorithm for each geographical region in Québec, Canada. The solid line indicates the average, and the shadows show ± standard deviation from 2010 to 2019. The index varied from 0 to 1, where values > 0.1 indicate growing conditions for the species. The dashed line indicates GIw values (> 0.40) where the first capture occurred.

Field Sampling

The first captures of E. fabae occurred at the end of May and the beginning of June (Table 2). Although not significant (P = 0.38), the first capture appeared to be associated with latitude (Fig. 6; r = 0.36). Interestingly, the peak of adults captured varied among regions from the end of June to the middle of July, except for Chaudière-Appalaches in 2021, where it occurred in August (Fig. 7). In all cases, the number of adults captured was lower by September, with no adults detected in the yellow sticky traps by the end of that month. When we compared the peaks of adults captured with the estimated first emergence dates by the two models (null and first capture), the peak was before the predicted dates in both cases (Fig. 7), except in Chaudière-Appalaches (2021) and Mauricie (2022). Considering the GIw values, the first capture occurred only by May when values were > 0.40.

Fig. 6.

Association between latitude and the first capture date (Julian date) of Empoasca fabae in four geographical regions in Québec, Canada.

Fig. 7.

Average number of Empoasca fabae adults collected per trap in each Québec, Canada geographical region. The dotted and solid lines indicate the first adult emergence based on the null degree days and first capture models.

Discussion

Québec is one of the northernmost regions where E. fabae migrates annually from its overwintering zone in the Southeastern USA (Sidumo et al. 2005). While numerous studies have investigated the phenology and population dynamics of this species in the USA (eg Delong 1928, DeLong 1931, 1938, Poss 1932, Delong and Caldwell 1935, Medler 1957, Pienkowski and Medler 1964, Kouskolekas and Decker 1966, Medler et al. 1966, Simonet and Pienkowski 1980, Taylor and Reling 1986, Lamp et al. 1989, Parr and Pass 1989, Carlson et al. 1992, Taylor and Shields 1995, Taylor et al. 1995, Maredia et al. 1998, Sidumo et al. 2005, Baker et al. 2015, Lagos-Kutz et al. 2024), research in Canada has primarily focused on economic aspects related to pest control in field crops (eg Cheng and Roy 1985, Appleton et al. 2003, Shi et al. 2025). In Québec, studies have emphasized the phenology of E. fabae from May to September (Bostanian et al. 2003, 2006, Saguez et al. 2014, Plante et al. 2024), with limited information available for the earlier and later parts of the year, and a seasonal model for the province has been proposed (Valiquette and Comtois 2022). By combining degree-day models and a monthly ecological model using the CLIMEX algorithm, we gained insights into the annual phenology of E. fabae in Québec, which will contribute to a better understanding of its population dynamics.

Across North America, E. fabae generations typically range from one to two, with a partial third to fourth generation possible (DeLong 1938, Sidumo et al. 2005). Our findings suggest the potential for two E. fabae generations in Québec (Table 3), aligning with the trend of decreasing generation numbers with increasing latitude across North America (Sidumo et al. 2005). However, the first capture model revealed regional variations within the province, with southeast regions showing a higher likelihood of supporting two generations. In contrast, northeast regions may only support a single generation.

While humidity and precipitation are generally considered primary factors influencing E. fabae occurrence, it has been suggested that under stable wet conditions, temperature becomes the dominant factor influencing development time and population size (DeLong 1938, Baker et al. 2015). This scenario appears relevant in Québec, as previous reports indicated a lack of significant association between rainfall and Nearctic leafhopper population size (Plante et al. 2024). Conversely, temperature, particularly minimum temperatures during the growing season, likely plays a crucial role in accelerating development, potentially enabling the completion of two generations in warmer regions while limiting development to a single generation in cooler regions (DeLong 1938, Medler et al. 1966).

Earlier adult emergences observed in the southeast regions align with warmer growing season temperatures, whereas lower temperatures are more common in the northeastern regions (Fig. 1). Cold conditions extend E. fabae development time (Hogg 1985, Sher and Shields 1991), inhibit oviposition between 10 and 21°C (Taylor et al. 1995) and below 15.6°C (Kieckhefer and Medler 1964), and increase mortality in immature stages (Simonet and Pienkowski 1980, Specker et al. 1990). Mortality in adults induced by freezing conditions begins at −9°C and becomes complete between −13°C and −14°C (Decker and Cunningham 1967, Decker and Maddox 1967). Consequently, colder conditions in the northeast may extend E. fabae development time, delay the onset of reproduction, and increase mortality rates, ultimately limiting the number of generations completed within the growing season (DeLong 1938, Medler et al. 1966). Furthermore, local microclimatic variations within each region, such as those associated with elevation, vegetation cover, and proximity to water bodies, can further influence temperature and contribute to the observed regional differences in E. fabae phenology. These findings suggest that environmental heterogeneity, primarily driven by minimum temperature variations across the province, could primarily explain the observed regional differences in E. fabae phenology in Québec.

Our seasonal model indicates that suitable conditions (GIw > 0.1) for E. fabae development appear in April, with the highest values from May to September (0.4 to 0.6), aligning with the phenology described during this time frame in the growing season in Québec. In vineyards, the first collections of adults typically occur at the beginning of June, with a peak in July and August, while nymphs appear in June, peaking in July and August, and are not found at the end of September (Bostanian et al. 2003). The first collection in alfalfa fields is similar to that in vineyards but shows multiple population peaks in July in southeastern Québec, whereas only one peak occurs in the northeast (Shi et al. 2025). Our field sampling also observed an early adult peak from June to July (except for Chaudière-Appalaches in 2021), which differs from the later peak in alfalfa and vineyards, suggesting potential differences in E. fabae population dynamics based on host plant availability.

The first capture in our field sampling occurred between late May and early June, coinciding with a GIw value exceeding 0.40, which only appears in May. This aligns with historical records of E. fabae first captures in the Northeastern USA (eg New York), typically recorded from mid-May to early June, with no reports in April (Medler 1957, Maredia et al. 1998, Baker et al. 2015). These findings suggest that the onset of favorable spring conditions likely constrains the arrival of migratory individuals in Québec. Notably, none of our models fully capture the complex factors influencing arrival and establishment timing, such as continental winds in the overwintering zone and host plant phenology.

Although our field data collection ended in late September, our model indicated a significant decline in suitable conditions for E. fabae by November. In new ongoing sampling efforts, we are extending the sampling period in strawberry fields to November, which will allow us to determine the latest presence of adult E. fabae. This effort corresponds with model predictions of suitable conditions in October, highlighting the importance of extended sampling for a more complete understanding of E. fabae phenology in Québec.

Our models primarily focused on temperature and precipitation to describe the population dynamics of E. fabae in Québec. However, it is crucial to acknowledge that other factors significantly influence populations. These factors include (i) continental winds, which are associated with the arrival of migratory individuals and the dispersal of populations (Taylor and Reling 1986, Taylor and Shields 2018); (ii) polyphagy, as the ability of E. fabae to utilize a wide range of host plants impacts their survivorship and fecundity (Lamp et al. 1994); (iii) natural enemies, which can exert top-down control on E. fabae populations, although to a minor extent (Hogg and Hoffman 1989); and (iv) the high degree of generation overlap, which can lead to complex interactions within and between generations due to females’ long reproductive lives, potentially contributing to population outbreaks (Hogg and Hoffman 1989). These factors were not incorporated into our models, highlighting the need for future research to investigate their interactive effects on E. fabae populations in Québec.

Our study aimed to describe the seasonal phenology of E. fabae in Québec, providing a data-driven assessment of its population dynamics. While some findings support Valiquette and Comtois’s (2022) proposed model, some discrepancies emerged. Our estimates indicate that the first-generation emergence occurs in July rather than June, aligning even with the null model (ie E. fabae as a local species), which predicts the first-generation emergence by July. However, our models agree with Valiquette and Comtois (2022) regarding the second generation’s emergence in August-September. Notably, while their model suggests adult E. fabae mortality by mid-to-late October, our findings indicate that unfavorable conditions dominate only in November. Given E. fabae’s cold hardiness, temperatures below −9°C, expected to cause complete mortality (Decker and Maddox 1967, Sidumo et al. 2005), are typically reached in November rather than October. Although field crops may no longer be available at that time, E. fabae is polyphagous and known to feed on pine species (Lamp et al. 1994), which remain present year-round and may serve as late-season hosts, as observed in the overwintering zone (Taylor and Shields 1995, Sidumo et al. 2005). These results highlight the importance of considering regional factors for accurate first-generation predictions while confirming the utility of the Valiquette and Comtois (2022) model for a second one.

The initiation of reproductive diapause in E. fabae in Québec remains an open question. Valiquette and Comtois’s (2022) proposed model suggests diapause onset (ie newly emerged females in a complete diapause state) in the second week of August. However, field studies in New York (1991 to 1992) indicated that females enter a complete reproductive diapause in late August to early September (Taylor et al. 1995). Temperature data from our study support this later timing, as conditions within the temperature range of 10 to 21°C, known to inhibit ovipoition even under photoperiods exceeding 15 h per day (Taylor et al. 1995), occur more frequently in late August. Furthermore, photoperiods below 14 h per day are associated with diapause induction (Taylor et al. 1995), beginning in late August (Supplementary Fig. S3). Thus, females are more likely to enter full reproductive diapause in late August rather than earlier. However, determining the precise timing of diapause onset requires detailed dissections of females throughout the season to assess their reproductive status.

Rising temperatures are reshaping E. fabae phenology, leading to earlier arrivals and increased crop damage in the USA (Maredia et al. 1998, Baker et al. 2015). While climate models predict a northward range expansion of E. fabae and other leafhopper pests in North America (Santos et al. 2024), understanding the current temporal dynamics of this pest within Québec is crucial for developing effective, region-specific pest management strategies. Nevertheless, key questions remain: (i) What is E. fabae spatial distribution within Québec following their arrival? (ii) Does it reach northern areas, including non-agricultural habitats in the Québec subarctic and arctic? (iii) When the adult population peaks, does it indicate multiple migratory events or local population buildup? (iv) When does female reproductive diapause begin? And perhaps most importantly, (v) What happens to E. fabae populations in Québec when environmental conditions become unsuitable?

Author contributions

Abraão Santos (Conceptualization [Lead], Data curation [Lead], Formal analysis [Lead], Investigation [Lead], Software [Supporting], Validation [Lead], Visualization [Lead], Writing - original draft [Lead], Writing - review & editing [Supporting]), Fausto Vieira Araújo (Formal analysis [Equal], Investigation [Supporting], Software [Supporting], Validation [Supporting], Visualization [Supporting], Writing - original draft [Supporting], Writing - review & editing [Supporting]), Nicolas Plante (Data curation [Supporting], Investigation [Supporting], Validation [Supporting], Writing - original draft [Supporting], Writing - review & editing [Supporting]), Ricardo Siqueira (Supervision [Supporting], Writing - original draft [Supporting], Writing - review & editing [Supporting]), and Edel Pérez-López (Conceptualization [Supporting], Funding acquisition [Lead], Resources [Lead], Supervision [Lead], Writing - review & editing [Supporting])

Funding

Fausto Henrique Vieira Araújo is supported by a scholarship (PDSE 88881.933705/2024-01) from the Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES). This work was funded by the Réseau québécois de recherche en agriculture durable (RQRAD), the Ministère de l'Agriculture, des Pêcheries et de l'Alimentation du Québec (MAPAQ), and the Fonds de recherche du Québec–Nature et technologies (FRQNT) through the Programme de recherche en partenariat—Agriculture durable—Volet II—2e concours (application number 337847), as well as by the Natural Sciences and Engineering Research Council of Canada (NSERC) through the Alliance-SARI Program (Grant ALLRP 588519-23).

Conflict of interest. None declared.