Introduction
Biological control of pests by their natural enemies represents a valuable regulating ecosystem service (Symondson et al. Reference Symondson, Sunderland and Greenstone2002; Losey and Vaughan Reference Losey and Vaughan2006). The economic importance of insect pests in agricultural fields and the potential biological control by their natural enemies warrant foundational studies for the development of conservation and integrated pest management strategies (Tscharntke et al. Reference Tscharntke, Tylianakis, Wade, Wratten, Bengtsson, Kleijn, Stewart, New and Lewis2007). This has become more pressing with global insect population declines (Hallmann et al. Reference Hallmann, Sorg, Jongejans, Siepel, Hofland and Schwan2017; Leather Reference Leather2018; Goulson Reference Goulson2019; Wagner et al. Reference Wagner, Grames, Forister, Berenbaum and Stopak2021) and the potential, exacerbating effects of climate change on pest issues in agricultural fields (Cannon Reference Cannon1998; Estay et al. Reference Estay, Lima and Labra2009; Tonnang et al. Reference Tonnang, Sokame, Abdel-Rahman and Dubois2022). Alfalfa, Medicago sativa Linnaeus (Fabaceae), is one of the most important forage crops, particularly in temperate regions of the world, and is often grown for animal forage and for its nitrogen-fixing abilities (Burity et al. Reference Burity, Ta, Faris and Coulman1989; Soroka and Otani Reference Soroka, Otani and Floate2011; Edde Reference Edde2021). In Canada, alfalfa dominates forage seed production with over 3 million hectares grown annually, of which 73% is produced in the Canadian Prairie Provinces of Alberta, Saskatchewan, and Manitoba (Statistics Canada 2022).
A variety of insect pests occur in alfalfa fields in the Canadian prairies, including the alfalfa weevil, Hypera postica (Gyllenhal) (Coleoptera: Curculionidae), the pea aphid, Acyrthosiphon pisum (Harris) (Hemiptera: Aphididae), the spotted alfalfa aphid, Therioaphis maculata (Buckton) (Hemiptera: Aphididae), lygus bugs, Lygus spp. Hahn (Hemiptera: Miridae), the alfalfa plant bug, Adelphocoris lineolatus (Goeze) (Hemiptera: Miridae), the alfalfa seed chalcid, Bruchophagus roddi Gussakovsky (Hymenoptera: Eurytomidae), and the alfalfa blotch leafminer, Agromyza frontella (Rondani) (Diptera: Agromyzidae) (Hobbs et al. Reference Hobbs, Nummi and Virostek1959; Harper Reference Harper1988; Schaber and Entz Reference Schaber and Entz1988; Uddin Reference Uddin2005; Sim and Meers Reference Sim and Meers2017). In addition, Sim and Meers (Reference Sim and Meers2017) identify minor pests in forage alfalfa, including a three-species complex of Sitona spp. Germar weevils (Coleoptera: Curculionidae), thrips (Thysanoptera), and the lucerne flea, Sminthurus viridis (Linnaeus) (Symphypleona: Sminthuridae), some of whose populations occasionally can cause significant damage.
Hypera postica is considered one of the most economically important pests of alfalfa in North America (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). It was introduced from Eurasia in the early 1900s (Titus Reference Titus1911) and was first reported in southern Alberta in 1954 (Hobbs et al. Reference Hobbs, Nummi and Virostek1959). Since then, H. postica has undergone a rapid geographical expansion in the Canadian prairies in the past two decades (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). Hypera postica is univoltine in Canada and the northern United States of America (Bereza Reference Bereza1970). Overwintering adults emerge in spring and lay several clusters of approximately 10 eggs on plant stems (Whitford and Quisenberry Reference Whitford and Quisenberry1990). Eggs develop over two weeks, larvae feed and mature over a period of 3–5 weeks, and adults continue to feed briefly before hibernation (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). During both the larval and adult stages, the insect feeds on alfalfa foliage, but larvae peak during the early- and mid-bloom stages and are particularly injurious to the crop during this period (Soroka and Otani Reference Soroka, Otani and Floate2011; Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). Management of H. postica on the Canadian prairies includes the use of monitoring, economic thresholds, insecticide applications, and biological control. Previous studies have investigated biological control of H. postica in alfalfa fields in the Canadian prairies by two introduced parasitoids, Bathyplectes curculionis (Thomson) (Hymenoptera: Ichneumonidae) and Oomyzus incertus (Ratzeburg) (Hymenoptera: Eulophidae) (Soroka Reference Soroka2013; Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). Nonetheless, the expansion of H. postica is relatively recent, and parasitoid population buildup can take several years (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020), warranting monitoring and surveying.
Several generalist predators have been documented in alfalfa fields in the Canadian prairies, including the following: ground beetles (Coleoptera: Carabidae); damsel bugs (Hemiptera: Nabidae); big-eyed bugs (Hemiptera: Geocoridae); lacewings (Neuroptera: Chrysopidae); lady beetles (Coleoptera: Coccinellidae); hoverfly larvae (Diptera: Syrphidae); minute pirate bugs, Orius insidiosus (Say) (Hemiptera: Anthocoridae); Aeolothrips fasciatus (Linnaeus) (Thysanoptera: Aeolothripidae); harvestmen (Opiliones: Phalangiidae); and numerous spider families (Araneae) (Harper Reference Harper1988; Uddin Reference Uddin2005; Sim and Meers Reference Sim and Meers2017). Nonetheless, the seasonality and spatiotemporal relationships between H. postica and the generalist predators occurring in alfalfa fields in the Canadian prairies have been scarcely studied (but see Schaber and Entz Reference Schaber and Entz1988 and Uddin Reference Uddin2005).
The variety of insect pests and natural enemies found in alfalfa agroecosystems and the importance of this crop in the Canadian prairies warrant more research into the seasonality of these taxa for the development of integrated pest management strategies. In the present study, we investigated the seasonal abundance of three major pests in alfalfa seed production fields in southern Alberta, Canada: H. postica, A. lineolatus, and Lygus spp. We also examined the seasonal abundance, richness, and diversity of generalist predators and the seasonality of two H. postica parasitoids, B. curculionis and O. incertus. Finally, we examined correlations between pests and natural enemies and discuss their potential under a biological control framework.
Materials and methods
Field sites and insect sampling
An insect survey targeting a subset of pests and natural enemies was conducted in alfalfa fields in southern Alberta. Sixteen irrigated alfalfa seed production fields in their second or third year of production were sampled during two years (2020: n = 8; 2021: n = 10; two fields were sampled in both years; Fig. 1). Sampled fields were quarter sections (approximately 0.65 km2) separated by a minimum distance of 1.5 km (10.94 ± 11.92, mean ± standard error) and were managed by growers according to standard agronomic practices, including insecticide application when necessary. Insecticide application data were requested, but only a partial dataset was obtained; therefore, it is not included here.
Figure 1. Locations of alfalfa seed production fields surveyed in 2020 and 2021 in southern Alberta. The thick, winding grey line represents the Bow River.
Fields were monitored weekly and sampled at three crop stages, based on Mueller and Teuber (Reference Mueller and Teuber2008): bud (stage 4), start of flowering (stage 6), and full seed (stage 8; Table 1). Insects within the crop canopy were sampled using a standard sweep net (38.1 cm diameter) along a 100-sweep transect. The sweeping transect began 10 m into the field and consisted of 25 sweeps in a 45° angle perpendicular to the field edge, followed by 50 sweeps parallel to the field edge and 25 more sweeps in a 45° angle opposite to the first. Sweep net samples were collected by the same individual within each year for consistency. Fields were sampled between 08:00 and 18:45 hours, local time, when temperatures were above 15 °C. Samples were collected into resealable plastic bags and transported inside an insulated container with ice to the laboratory, where they were stored at –20 °C before identification.
Table 1. Abundance of generalist predators and two H. postica parasitoids collected in alfalfa seed production fields in southern Alberta, 2020 and 2021. Values indicate means of individuals per 100 sweeps. Eight and 10 fields were sampled in 2020 and 2021, respectively. Collection dates in 2020 occurred between 10 and 12 June, between 4 and 6 July, and between 10 and 11 August during the bud, flower, and seed stages, respectively. Collection dates in 2021 occurred on 16 June, 21 July, and 11 August during the bud, flower, and seed stages, respectively
* Individuals were not identified to genus or species and were not used in the species richness assessments.
† Taxon reported by Hobbs et al. (Reference Hobbs, Nummi and Virostek1959).
‡ Taxon reported by Harper (Reference Harper1988).
§ Taxon reported by Uddin (Reference Uddin2005).
|| Taxon reported by Sim and Meers (Reference Sim and Meers2017).
¶ Taxon reported by Soroka et al. (Reference Soroka, Bennett, Kora and Schwarzfeld2020).
Sample identification
Collected arthropods were classified as pests or natural enemies based on Harper (Reference Harper1988) and Soroka and Otani (Reference Soroka, Otani and Floate2011) and were identified to species or genus using taxonomic keys (Marshall Reference Marshall2006; Larson Reference Larson2013; Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020) and reference specimens provided by Julie Soroka (Agriculture and Agri-Food Canada, Saskatoon, Saskatchewan, Canada). Due to the high volume of specimens collected and to time constraints, we limited our observations to three key pests in alfalfa, namely H. postica, A. lineolatus, and Lygus spp., as well as to their potential generalist predators and to two H. postica parasitoids, B. curculionis and O. incertus (Table 1). Parasitoids related to other pests were present in our samples but were not investigated. Voucher specimens of identified taxa were deposited in the Strickland Museum of Entomology, University of Alberta, Edmonton, Alberta, Canada (UASM417149 to UASM417182).
Data analysis
All analyses were conducted in R, version 4.2.1 (R Core Team 2022). Seasonal abundance was examined across three crop stages for H. postica, Lygus spp., A. lineolatus, B. curculionis, and O. incertus. Using the Stats (R Core Team 2018) and PMCMR (Pohlert Reference Pohlert2021) packages, we computed Friedman tests, then Conover multiple comparisons (P-values adjusted using the Benjamini–Hochberg method), because data were neither normally distributed nor independent across time points (Zar Reference Zar2010).
To investigate seasonal richness and diversity of generalist predators across crop stages, Hill numbers were used for richness (q = 0) and Simpson diversity (q = 2) using the package iNEXT (Hsieh et al. Reference Hsieh, Ma and Chao2016). Richness and diversity were calculated at the genus level after sample size–based rarefaction in iNEXT, for which the number of individuals per sample were standardised to 2m, where m is the lowest sample size found per year (Chao et al. Reference Chao, Gotelli, Hsieh, Sander, Ma, Colwell and Ellison2014; Hsieh et al. Reference Hsieh, Ma and Chao2016). Specimens not identified to genus or species (Table 1) were not included in the richness and diversity analyses. Coleomegilla maculata (DeGeer) (Coleoptera: Coccinellidae) was excluded from the analysis because only one individual was caught during the two years of sampling. Richness and diversity of generalist predators were compared between crop stages using Friedman tests, followed by Conover multiple comparison tests (P-values adjusted using the Benjamini–Hochberg method).
Abundance correlations between pests and potential natural enemies were examined using repeated measured correlations with the rmcorr package (Bakdash and Marusich Reference Bakdash and Marusich2017) to account for non-independence between time observations. The P-values were adjusted using the Benjamini–Hochberg method.
Results
Seasonal abundance of Lygus spp. and A. lineolatus
Lygus spp. showed inconsistent abundance trends across crop stages between years. Lygus spp. abundance did not differ between crop stages in 2020 (χ2 = 5.25, df = 2, P > 0.05) but did in 2021 (χ2 = 6.2, df = 2, P = 0.045). In 2021, Lygus spp. abundance was highest during the seed stage (351.4 ± 75.11 (standard error) individuals per 100 sweeps) compared to the bud (96.9 ± 36.74 (standard error) individuals per 100 sweeps) and the flower (175.6 ± 50.6 (standard error) individuals per 100 sweeps) stages and did not differ between the bud and flower stages (Fig. 2).
Figure 2. Lygus spp. and Adelphocoris lineolatus abundances, sampled at three alfalfa crop stages in 2020 and 2021. Values indicate individuals collected per 100 sweeps and include immature and adult insects: A, Lygus spp. abundance in 2020; B, A. lineolatus abundance in 2020; C, Lygus spp. abundance in 2021; and D, A. lineolatus abundance in 2021. Letters indicate statistical differences between values at each crop stage, based on Conover tests, with P-values adjusted with the Benjamini–Hochberg method.
Adelphocoris lineolatus abundance differed between crop stages in 2020 (χ2 = 9.75, df = 2, P = 0.008) and 2021 (χ2 = 6, df = 2, P = 0.049). In 2020, A. lineolatus abundance was highest during the bud (92.88 ± 46.13 (standard error) individuals per 100 sweeps) and seed (43 ± 16 (standard error) individuals per 100 sweeps) stages compared to the flower stage (5.25 ± 1.46 (standard error) individuals per 100 sweeps) and did not differ between the bud and seed stages. In 2021, A. lineolatus abundance was higher during the seed (147 ± 61.69 (standard error) individuals per 100 sweeps) and bud (102.2 ± 44.77 (standard error) individuals per 100 sweeps) stages compared to the flower stage (25.7 ± 13.84 (standard error) individuals per 100 sweeps; Fig. 2).
Predator seasonal abundance, richness, and diversity
In 2020, there were 40.88 ± 2.48 (standard error), 21.88 ± 8.30 (standard error), and 82.88 ± 13.32 (standard error) predators identified to genus or species per 100 sweeps at the bud, flower, and seed stages, respectively (Table 1). In 2021, there were 99.1 ± 21.42 (standard error), 243.4 ± 55.42 (standard error), and 221.9 ± 35.29 (standard error) predators identified to genus or species per 100 sweeps at the bud, flower, and seed stages, respectively (Table 1). Rarefied richness and diversity of predators showed inconsistent trends across years. No differences were observed in predator richness (χ2 = 0.75, df = 2, P > 0.05) or diversity (χ2 = 0.25, df = 2, P > 0.05) between crop stages in 2020 (Fig. 3). In 2021, however, significant differences were identified in predator richness (χ2 = 6.2, df = 2, P = 0.045) and diversity (χ2 = 8, df = 2, P = 0.018) across crop stages. Predator richness in 2021 was higher during the seed stage (3.14 ± 0.17 standard error) compared to the bud stage (2.36 ± 0.17 standard error) but did not differ between the bud and flower stages (3.27 ± 0.31 standard error) or between the flower and seed stages (Fig. 3). Similarly, predator diversity in 2021 was higher during the seed stage (1.70 ± 0.09 standard error) compared to the bud stage (1.25 ± 0.06 standard error) but did not differ between the bud and flower stages (1.79 ± 0.28 standard error) or the flower and seed stages (Fig. 3).
Figure 3. Rarefied richness and diversity (genus level) of generalist predators sampled at three alfalfa crop stages during 2020 and 2021: A, predator richness in 2020; B, predator diversity in 2020; C, predator richness in 2021; and D, predator diversity in 2021. Rarefactions were based on twice the lowest sample size per year (8 and 24 individuals in 2020 and 2021, respectively). Letters indicate statistical differences between values at each crop stage based on Conover tests, with P-values adjusted with the Benjamini–Hochberg method.
Seasonal abundance of H. postica larvae, B. curculionis, and O. incertus
Hypera postica larval abundance trends differed between crop stages inconsistently across 2020 (χ2 = 13, df = 2, P = 0.002) and 2021 (χ2 = 14.6, df = 2, P < 0.001). In 2020, H. postica larval abundance was higher during the flower stage (1703.63 ± 658.78 (standard error) individuals per 100 sweeps) compared to the bud (433.38 ± 128.45 (standard error) individuals per 100 sweeps) and seed (26. 13 ± 9.65 (standard error) individuals per 100 sweeps) stages and was higher during the bud stage compared to the seed stage (Fig. 4). In 2021, H. postica larval abundance was highest during the bud stage (492.7 ± 143.01 (standard error) individuals per 100 sweeps) compared to flower (81.6 ± 26.95 (standard error) individuals per 100 sweeps) and seed (9.9 ± 2.43 (standard error) individuals per 100 sweeps) stages and was higher during the flower stage compared to the seed stage (Fig. 4). Over the two years of this study, the two H. postica parasitoids were found in alfalfa fields at low but varying numbers between crop stages across years.
The abundance of B. curculionis varied significantly across crop stages in 2020 (χ2 = 13, df = 2, P = 0.002) but not in 2021 (χ2 = 4.06, df = 2, P = 0.13). In 2020, B. curculionis abundance decreased with crop stage and was highest during the bud stage (9.34 ± 2.56 (standard error) individuals per 100 sweeps) compared to the flower (3.25 ± 0.59 (standard error) individuals per 100 sweeps) and the seed (0.125 ± 0.125 (standard error) individuals per 100 sweeps) stages, differing also between flower and seed stages (Fig. 4).
The abundance of O. incertus varied between crop stages in 2021 (χ2 = 10.89, df = 2, P = 0.004) but not in 2020 (χ2 = 4.88, df = 2, P > 0.05). It should be noted that in 2020, no O. incertus individuals were collected during the flower stage. In 2021, O. incertus abundance was higher at the flower stage (19.1 ± 9 standard error) individuals per 100 sweeps) compared to the bud (1.5 ± 1.39 (standard error) individuals per 100 sweeps) and seed (0.5 ± 0.31 (standard error) individuals per 100 sweeps) stages (Fig. 4).
Figure 4. Abundance of Hypera postica larvae and its parasitoids, Bathyplectes curculionis and Oomyzus incertus, sampled at three alfalfa crop stages in 2020 and 2021. Values indicate individuals collected per 100 sweeps: A, H. postica larvae in 2020; B, B. curculionis in 2020; C, O. incertus in 2020; D, H. postica larvae in 2021; E, B. curculionis in 2021; and F, O. incertus in 2021. Letters indicate statistical differences between values at each crop stage based on Conover tests, with P-values adjusted with the Benjamini–Hochberg method.
Abundance correlations between pests and natural enemies
We found a number of positive correlations between pests and natural enemy abundances (Table 2). In 2020, Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae) and Coccinellidae were positively correlated with Lygus spp. Also in 2020, Araneae were positively correlated with A. lineolatus. In 2021, Nabis spp. was positively correlated with Lygus spp. We found no correlations between H. postica larvae and either B. curculionis or O. incertus in either year.
Table 2. Repeated measures correlations between pests and natural enemies in alfalfa seed production fields in southern Alberta in 2020 and 2021. The P-values are adjusted with the Benjamini–Hochberg method (*, **, and *** indicate significant correlations at P < 0.05, P < 0.01, and P < 0.001, respectively)
Discussion
Characterising the seasonality of insect pests and their natural enemies, and identifying the correlations occurring between them, can provide important foundational knowledge for the development and implementation of conservation and integrated pest management strategies. In this study, we examined the presence and abundance of a selected guild of insect pests and their natural enemies across six arthropod orders from alfalfa seed fields in southern Alberta and investigated their correlations throughout the growing season.
Seasonal abundance of Lygus spp. and A. lineolatus
In the present study, we observed inconsistent seasonal abundance trends for Lygus spp. across years. Lygus spp. abundance did not differ between crop stages in 2020. However, in 2021, Lygus spp. abundance differed between crop stages and increased continually throughout the season. The peak in Lygus spp. abundance that we observed in 2021 is consistent with the second-generation adult peaks in mid-August that have been reported in Manitoba (Uddin Reference Uddin2005), Alberta (Butts and Lamb Reference Butts and Lamb1991), and Saskatchewan (Braun et al. Reference Braun, Erlandson, Baldwin, Soroka, Mason, Foottit and Hegedus2001), but we did not observe this pattern in 2020. Inconsistent seasonal abundance trends may be attributed partially to differences in climate between years, among other environmental and biotic factors. In 2021, high summer temperatures (discussed below, in the Discussion section) and drought likely allowed Lygus populations to increase locally and may also have resulted in higher migration to irrigated seed alfalfa fields from drought-affected crops. Insecticide applications also may have varied between years, impacting Lygus dynamics. However, our data do not allow us to discuss this further. In contrast to Lygus spp. abundances, we observed somewhat consistent seasonal abundance trends for A. lineolatus in 2020 and 2021, with peaks occurring during the bud and seed stages in both years. This result concurs with observed abundance peaks reported by Uddin (Reference Uddin2005) in Manitoba in mid-June and later again in mid-August. Even though A. lineolatus is mostly considered univoltine north of 51° N, two generations can occasionally occur, depending on temperature (Craig Reference Craig1963). In agreement with Uddin’s (Reference Uddin2005) observations of a partial second generation in Manitoba, the relatively high numbers of A. lineolatus that we observed during the seed stage suggest a second generation and deserve further study. Sim and Meers (Reference Sim and Meers2017) noted that Lygus spp. and A. lineolatus tend to stay under the economic thresholds in alfalfa forage crops in southern Alberta. In accordance with this, over the two years of the present study, Lygus spp. and A. lineolatus abundances mostly remained under the economic threshold (2–3 late-instar larvae or adults per sweep for seed alfalfa; Government of Alberta 2024). Nonetheless, in 2021, adult Lygus spp. densities reached economic thresholds (i.e., > 200 individuals per 100 sweeps) during the seed stage in three of 10 fields (data not shown).
Predator seasonal abundance, richness, and diversity
During the present study, we documented the presence and seasonal abundance of 10 predatory arthropod taxa in alfalfa seed production fields in southern Alberta. The occurrence of these taxa in alfalfa fields concurs with the results of previous studies in alfalfa fields in the Canadian prairies (Table 1; Hobbs et al. Reference Hobbs, Nummi and Virostek1959; Harper Reference Harper1988; Uddin Reference Uddin2005; Sim and Meers Reference Sim and Meers2017; Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). We found that the predator complex was dominated by Orius spp., Nabis spp., Coccinellidae, and Araneae, consistent with other studies conducted in the Canadian prairies (Schaber and Entz Reference Schaber and Entz1988) and the United States of America (Elliott and Kieckhefer Reference Elliott and Kieckhefer1990; Elliott et al. Reference Elliott, Kieckhefer, Michels and Giles2002; Pons et al. Reference Pons, Núñez, Lumbierres and Albajes2005, Reference Pons, Lumbierres and Albajes2009; Rand Reference Rand2017). Despite this, our data suggest considerable variation in seasonal abundance trends for predatory taxa between years. Based on data collected over 13 years, Elliott and Kieckhefer (Reference Elliott and Kieckhefer1990) noted that natural enemy abundances in alfalfa fields in South Dakota, United States of America, fluctuate considerably from year to year and are often unrelated to co-occurring taxa and abundances in previous years. The inconsistent abundance trends we observed for predators across the two years of the present study can be attributed to multiple environmental and biotic factors, as well as to management practices such as insecticide applications. Climatic differences, landscape characteristics, resource availability, biotic interactions, and habitat disturbances can strongly influence natural enemy densities in agricultural fields (Chaplin-Kramer et al. Reference Chaplin-Kramer, O’Rourke, Blitzer and Kremen2011; Lantschner et al. Reference Lantschner, de la Vega and Corley2019). The influence of these factors and particularly the timing and choice of insecticide application should be considered in the development of integrated pest management strategies and natural enemy population predictions.
In the present study, Coccinella septempunctata Linnaeus (Coleoptera: Coccinellidae) strongly dominated the Coccinellidae assemblage. Following the introduction and establishment of C. septempunctata in North America, concern has been raised as to what appears to be a displacement of native Coccinellidae species by C. septempunctata, as evinced by decreasing lady beetle diversity (Alyokhin and Sewell Reference Alyokhin and Sewell2004; Harmon et al. Reference Harmon, Stephens and Losey2007). In an earlier survey, Harper (Reference Harper1988) recorded 22 different species of lady beetles in alfalfa fields in Alberta, all native to North America, whereas Sim and Meers (Reference Sim and Meers2017) recorded 12 species, with C. septempunctata dominating this group. In the present study, we documented six Coccinellidae species over two sampling years, of which five were native. Thus, results from the present study seem to be following a trend of decreasing lady beetle richness in alfalfa. However, differences in sampling effort between the present and earlier studies should be considered, and further research in this area is warranted.
Overall, in the present study, the richness and diversity of sampled predators varied over time, with inconsistent patterns occurring between years. Predator richness stayed constant across crop stages in 2020, but interestingly, in 2021, it increased throughout the season and peaked during the seed stage. Similarly, predator diversity remained constant in 2020 but increased throughout the season in 2021, again peaking during the seed stage. Although natural enemy diversity is generally considered beneficial in terms of biological control, this relationship is a subject of debate in the literature (Jonsson et al. Reference Jonsson, Kaartinen and Straub2017); further research should explore the effects of natural enemy diversity on pest populations in alfalfa fields. In addition, because we did not identify Araneae and juvenile lady beetles to genera, we likely underrepresented the true species richness and diversity of predators in the study fields.
Seasonal abundance of H. postica larvae, B. curculionis, and O. incertus
We observed variable trends in the abundance of H. postica larvae and its parasitoids, B. curculionis and O. incertus, between collection years. Although we observed a peak in H. postica larval abundance at the flower crop stage in 2020, this peak occurred during the bud stage in 2021. Observed peaks in H. postica larval abundance, however inconsistent, seem to fall between those reported by Soroka et al. (Reference Soroka, Bennett, Kora and Schwarzfeld2020) in Saskatchewan as occurring between mid-June and early July. The inconsistent trends in H. postica larvae observed in the present study may be due in part to differences in temperature between years. This factor is critical for H. postica overwintering and spread (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020): survival, longevity, and reproduction of H. postica greatly decrease at temperatures rise above 32 °C (Levi-Mourao et al. Reference Levi-Mourao, Madeira, Meseguer and Pons2022). During the present study, 2021 was much warmer than 2020: June and July 2020 had averages of 19.0 °C and 22.6 °C and high extremes of 25.8 °C and 29.1 °C, respectively, whereas June and July 2021 had averages of 23.7 °C and 25.4 °C and extremes of 35.0 °C and 35.5 °C, respectively (Government of Canada 2022). Extreme temperatures therefore may have depressed H. postica populations in 2021. In addition to between-year temperature differences, crop phenology differed between 2020 and 2021. Although collection dates at the bud stage occurred six calendar days apart in 2020 relative to 2021 (from 10 to 12 June versus 16 June, respectively), and only one day apart at the seed stage (from 10 to 11 August 2020 versus 11 August 2021), collection dates at the flower stage occurred almost two weeks apart in 2020 compared to 2021 (from 4 to 6 July versus 21 July, respectively). We suspect that the higher temperatures seen in 2021 may have accelerated crop development in July, further affecting H. postica larval abundance. We observed the presence of H. postica larvae throughout the entire season during both years of this study. Despite this species being univoltine in Canada (Bereza Reference Bereza1970; Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020), H. postica larvae seem to be present throughout the entire season in the Canadian prairies (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). Nonetheless, larval abundance decreased dramatically by the seed stage, consistent with Soroka et al.’s (Reference Soroka, Bennett, Kora and Schwarzfeld2020) results.
In the present study, we collected B. curculionis throughout the entire growing season in both years. Although B. curculionis is documented to have a partial second generation in some areas of the United States of America (Chamberlin Reference Chamberlin1926; Radcliffe and Flanders Reference Radcliffe and Flanders1998), the species is univoltine in Canada (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020), even though, like its host, it remained present throughout the season. Bathyplectes curculionis peak flight occurs in mid-June, 1–2 weeks before H. postica larval populations peak (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). In the present study, we observed a peak in B. curculionis abundance in mid-June (flower stage) in 2021 but not in 2020.
Oomyzus incertus is documented to have 3–4 generations per year in the United States of America, with peak abundances generally occurring in mid-summer, depending on the region (Radcliffe and Flanders Reference Radcliffe and Flanders1998). In the Canadian prairies, O. incertus is multivoltine and persists until late August (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020); however, the exact number and timing of peak flights are unknown in Alberta because this species is considered recently established in the Canadian prairies (Soroka et al. Reference Soroka, Bennett, Kora and Schwarzfeld2020). In 2020, no individuals of O. incertus were collected during the flower stage, in contrast to 2021, when O. incertus abundance peaked during the flower stage. The varying numbers of O. incertus collected during the two years of this study and the lack of previous information on population peaks in southern Alberta make drawing conclusions on the species’ seasonal abundance difficult. Given the differences in abundance and population trends across years and locations, more research investigating the life cycles and seasonal abundance of B. curculionis and O. incertus over a longer time span is needed to assess the species’ potentials as biocontrol agents for H. postica in southern Alberta.
Abundance correlations between pests and natural enemies
During the two years of our study, we found positive correlations between alfalfa pests and some generalist predators. Although our findings do not reveal trophic interactions between species and any correlations may result from shared responses to environmental factors, such correlations have biological implications and may indicate numerical responses to pest densities. First, we observed a positive correlation between C. carnea and Lygus spp. in 2020. This result is consistent with Schaber and Entz (Reference Schaber and Entz1988), who found a positive correlation between lacewings and Lygus spp. abundances, also in southern Alberta. In addition, molecular gut content analyses show evidence of light predation on Lygus spp. by C. carnea under controlled conditions and in alfalfa as a trap crop in California, United States of America (Mansfield and Hagler Reference Mansfield and Hagler2016; Hagler et al. Reference Hagler, Nieto, Machtley, Spurgeon, Hogg and Swezey2018). Second, we observed a positive correlation between Coccinellidae and Lygus spp. abundances in 2020, concurring again with Schaber and Entz (Reference Schaber and Entz1988) but contrasting with Uddin (Reference Uddin2005), who found a negative relationship between Coccinellidae and Lygus spp. abundances in alfalfa fields in Manitoba. Mansfield and Hagler (Reference Mansfield and Hagler2016) report predation and scavenging on L. hesperus by H. convergens. Nonetheless, C. septempunctata, the dominant lady beetle species collected in our study, did not consume L. lineolaris in controlled feeding experiments by Arnoldi et al. (Reference Arnoldi, Stewart and Boivin1991) and displayed low consumption rates on L. hesperus in a cage experiment by Hagler et al. (Reference Hagler, Casey, Hull and Machtley2020). This indicates that further research is required to assess whether Coccinellidae assemblages in southern Alberta alfalfa feed on Lygus spp. Third, we identified a positive correlation between Araneae and A. lineolatus in 2020. This correlation agrees with correlations noted by Schaber and Entz (Reference Schaber and Entz1988), and Howell and Pienkowski (Reference Howell and Pienkowski1971) observed evidence of predation on Adelphocoris sp. Reuter by spiders in the families Thomisidae, Salticidae, Lycosidae, and Tetragnathidae under controlled conditions. However, we are not aware of field studies that investigate trophic interactions between these taxa. Finally, we observed a positive correlation between Nabis spp. and Lygus spp. in 2021. Previous research by Schaber and Entz (Reference Schaber and Entz1988) supports this finding, and feeding experiments by Arnoldi et al. (Reference Arnoldi, Stewart and Boivin1991) and Clancy and Pierce (Reference Clancy and Pierce1966) also show relatively high incidences of Nabis spp. feeding on L. lineolaris and Lygus hesperus Knight (Hemiptera: Miridae). Moreover, molecular gut content analysis shows that N. alternatus Parshley (Hemiptera: Nabidae) feeds on L. hesperus (Hagler et al. Reference Hagler, Nieto, Machtley, Spurgeon, Hogg and Swezey2018) in alfalfa trap crops in California, and N. americoferus Carayon (Hemiptera: Nabidae) successfully reduced L. lineolaris populations in a release experiment in strawberry (Rosaceae) fields in Québec (Dumont et al. Reference Dumont, Solà, Provost and Lucas2023). Given the correlations we noted between pests and natural enemies in the present study, further research is needed to determine the effects of these generalist predators on pest populations.
We did not observe correlations between Araneae and Lygus spp. abundances in either year, despite evidence from Arnoldi et al. (Reference Arnoldi, Stewart and Boivin1991), Hagler et al. (Reference Hagler, Nieto, Machtley, Spurgeon, Hogg and Swezey2018), and Hagler et al. (Reference Hagler, Casey, Hull and Machtley2020) of predation on Lygus spp. by spiders in the families Agelenidae, Philodromidae, and Thomisidae under controlled and field conditions. Similarly, we did not find correlations between Orius spp. and Lygus spp. in either year, despite evidence by Hagler et al. (Reference Hagler, Nieto, Machtley, Spurgeon, Hogg and Swezey2018) and Dumont et al. (Reference Dumont, Solà, Provost and Lucas2023) of predation on Lygus spp. by Orius tristicolor White (Hemiptera: Anthocoridae) and O. insidiosus in the field.
Even though we did not find significant correlations between H. postica larval abundances and any of the predators we sampled, we noted that relationships between these taxa seem to follow a negative direction. This observation agrees with Schaber and Entz (Reference Schaber and Entz1988), who found only negative correlations between predators and H. postica in southern Alberta. Nonetheless, Rand (Reference Rand2017) documented aggregation responses of Nabidae and Coccinellidae to H. postica in Montana and North Dakota, United States of America. We should note, however, that given the cyclic nature of prey–predator dynamics (Crawley Reference Crawley1975), the direction of correlations between two taxa may potentially differ according to the time of sampling. Interestingly, Evans and England (Reference Evans and England1996), Elliott et al. (Reference Elliott, Kieckhefer, Michels and Giles2002), and Batuecas et al. (Reference Batuecas, Alomar, Castañé, Piñol, Gallardo-Montoya and Agustí2022) provide evidence of predation on Hypera spp. by Orius spp. and Coccinellidae in Spain and Utah, United States of America, respectively, and Ouayogode and Davis (Reference Ouayogode and Davis1981) suggest evidence of predation on H. postica by Coccinellidae, Chrysopa oculata (Say) (Neuroptera: Chrysopidae), Nabis spp., and Xysticus sp. Koch (Araneae: Thomisidae) in laboratory feeding trials. Therefore, the potential trophic interactions between generalist predators and H. postica in alfalfa fields in Alberta should be investigated.
We found no correlations between H. postica larvae and B. curculionis or O. incertus in either 2020 or 2021. In North Dakota and Montana, United States of America, Rand (Reference Rand2013) correlated B. curculionis parasitism rates to alfalfa weevil larval densities, even though the direction of these correlations changed from a positive to negative in a two-year span, which might be attributed to environmental factors.
In the present study, we did not quantify aphids in the field, although aphid abundances likely would influence predator populations: Pons et al. (Reference Pons, Núñez, Lumbierres and Albajes2005, Reference Pons, Lumbierres and Albajes2009) noted evidence of predation and numerical responses to aphids by Coccinellidae, Chrysopidae, and Nabidae. In addition, a positive correlation may exist between aphid abundance and parasitism of H. postica by B. curculionis that could be due to honeydew consumption by the parasitoid (Evans and England Reference Evans and England1996; Rand and Lundgren Reference Rand and Lundgren2019), which deserves attention. Although we requested insecticide application data for the fields we sampled, we obtained only a partial dataset and had insufficient field replication to investigate the effect of insecticide applications on pest and natural enemy populations. Chemical insecticides have been shown to affect both pest and natural enemy populations in alfalfa (Harper Reference Harper1978), and recommendations on insecticide use in alfalfa have been made in relation to crop phenology, pest populations, and some beneficial insects occurring in fields (Soroka et al. Reference Soroka, Goerzen and Murrell2011). Lastly, we focused on only a few of the pests that are present in alfalfa fields. This limitation affected our observations, and future investigations into other pests in alfalfa (e.g., alfalfa seed chalcid, aphids, and cicadellids) should be considered.
Here, we have documented seasonal abundance of three pests and the seasonal abundance, richness, and diversity of generalist predators in alfalfa fields in southern Alberta during two collection years. Although some of our abundance observations concur with other studies, we observed important inconsistencies in trends between years, which need to be addressed further in future studies to improve understanding of the effects of environmental conditions on abundance patterns. Moreover, the life cycles of H. postica parasitoids in alfalfa fields in relation to host abundance patterns in southern Alberta require further investigation. In addition, the potential occurrence of a second-generation of A. lineolatus in southern Alberta requires further research. This information is essential to guide growers towards improved management strategies to conserve natural enemies and reduce dependency on insecticides. The trophic interactions between generalist predators and alfalfa pests under a biological control framework in southern Alberta also present ample opportunity for further research.
Acknowledgements
We thank Dr. Linda Gorim for reviewing an earlier version of this manuscript. We are grateful to the journal reviewers for their valuable revisions and comments provided for manuscript improvement. We thank and acknowledge the alfalfa growers who allowed us to perform this study in their fields. Finally, we thank B. Alexander, J. Narayan, and M. Kohlman for helping with field work during this study.
Funding statement
Funding for this project was provided to B.A.M. and H.A.C. by the Canadian Agricultural Partnership, which is funded by Agriculture and Agri-food Canada, in partnership with the Alberta Alfalfa Seed Commission, the Saskatchewan Alfalfa Seed Producers Association, and the Peace Region Forage Seed Association (ASP-017 AGR-15675). Additional funds were provided by a NSERC Industrial Research Chair (545 088) and partner organisations (Alberta Wheat Commission, Alberta Barley Commission, Alberta Canola Producers Commission, and Alberta Pulse Growers Commission), as well as an NSERC Discovery Grant (2021–02479) to B.A.M.
Competing interests
The authors declare they have no competing interests with respect to the publication of this article.
Open access
