Ranges of Evening Grosbeak (Hesperiphona vespertina) call types 1 and 2 documented from community science and targeted placement of automated recorders

The Evening Grosbeak (Hesperiphona vespertina) is a species of North American Fringillid finch with at least five populations defined by differences in their flight calls. Groups with different flight calls generally occupy different geographic ranges during the breeding season and are referred to as call types. In Oregon and California, USA, call types 1 and 2 overlap in the Cascade and Sierra Nevada mountain ranges, but the geographical extent of this overlap remains unclear. We used citizen science data, our own ad hoc surveys, and a network of autonomous recording units (ARUs) across a longitudinal range from northern California into Oregon to quantify the distributions of these call type populations. We found a clear north-south transition, with type 1 birds occurring primarily in the northern sections of our study area while type 2 birds were predominantly detected in the Sierra Nevada mountains. Both types overlapped during the breeding season in southern Oregon. Both types also wandered widely after the breeding season and often co-occurred throughout the study area, especially in the Sierra Nevada. Type 2 was always rare north of southern Oregon. Additionally, our ARU data indicate greater occurrence of both call types where Evening Grosbeaks were most often recorded overall, especially after the breeding season. Our results align with Grinnell’s (1917) hypothesis of the presence of a contact zone between two populations of Evening Grosbeaks in Oregon and California and clarify the current extent of this geographic overlap.

1 Introduction

A central challenge in evolutionary biology is understanding the factors that influence the divergence and maintenance of evolutionary lineages (Edwards et al., 2005; Hill, 2017; Uy et al., 2018). In birds, the study of subtle differences in morphology and vocal characteristics across species’ geographic ranges has revealed “cryptic” species and has highlighted substantial infraspecific diversity within many avian taxa (Toews and Irwin, 2008; Chesser et al., 2020; Williamson et al., 2024). Subspecies are commonly designated within avian taxa recognized as species, with such infraspecific diversity largely being based on morphological differences in plumage coloration or morphometrics such as bill size. Some finch lineages present unique levels of evolutionary diversity expressed as variation among groups in their calls (Adkisson, 1981; Groth, 1988, 1993). Distinct vocal signatures, often referred to as call types, can align with geographic differences in breeding ranges, can be used to track migratory or nomadic movements of groups appearing to be nearly identical in morphological characteristics, and potentially allow identification of boundaries between genetic groups (Groth, 1988, 1993; Robinson et al., 2024). The Evening Grosbeak (Hesperiphona vespertina), a nomadic North American finch with striking vocal variation across its range (Sewall et al., 2004), presents an ideal case for exploring how vocal and geographic data together can illuminate patterns of infraspecific diversity.

Studies investigating the systematic status of the Evening Grosbeak have a long history. Over a century ago, Allan Brooks wrote to Joseph Grinnell at the California Museum of Vertebrate Zoology, describing “certain peculiarities” he had observed in Evening Grosbeak specimens from British Columbia (Grinnell, 1917). Intrigued, Grinnell gathered Brooks’ birds along with additional skins from other collectors. Over the next two years, Grinnell measured morphological traits including the dimensions of the bill, the size of the yellow band across the forehead of males, and other subtle plumage variation, to assess geographic variation in the Evening Grosbeak across North America (Grinnell, 1917). His efforts and attention to detail led him to an interesting conclusion: what was once considered a single subspecies, Hesperiphona vespertina montana (the Western Evening Grosbeak), actually comprised four diagnosably distinct groups (Grinnell, 1917). These included distinct Mexican (H. v. montana) and Rocky Mountain (H. v. warreni) forms, as well as a further split between the birds of the Pacific Northwest (H. v. brooksi) and those inhabiting California’s Sierra Nevada (H. v. californica).

Yet, Grinnell’s detailed subspecific distinctions raised new questions. Because his work relied primarily on specimens collected during winter, when many grosbeaks were possibly distant from their breeding areas, the association between morphological traits of subspecies and their breeding ranges was obscured. For example, although H. v. brooksi was documented from British Columbia and Washington, its status in Oregon remained uncertain (Grinnell, 1917). Similarly, while H. v. californica was associated with the Sierra Nevada, its northern extent into Oregon and whether its breeding range overlapped with H. v. brooksi were unclear (Grinnell, 1917).

Decades later, Sewall et al. (2004) reinvigorated interest in these groups by documenting four categorically distinct forms of Evening Grosbeak flight calls and examining their geographic distributions. Flight calls are single-note vocalizations, frequently given while birds are flying but also when perched. The notes are now known to vary distinctively among at least 5, possibly 6, different infraspecific groups (Kirsch et al., 2025). These vocal populations, hereafter referred to as call types, appeared to largely align with Grinnell’s subspecies designations. For instance, Pacific Northwest birds (presumably H. v. brooksi) produce a flight call that differs markedly from those of the Sierra Nevada birds (presumably H. v. californica). Sewall et al. (2004) identified the primary flight call given by Pacific Northwest birds as the type 1 call, whereas the California birds largely produced the type 2 call. In this way, the “certain peculiarities” Brooks observed in his British Columbia specimens extended not only to morphology but to vocal traits as well.

Recently, the breeding ranges and movements of Evening Grosbeak call types have come under additional scrutiny (Duman and Hahn, 2024; Robinson et al., 2024; Kirsch et al., 2025). Evening Grosbeaks present certain challenges given their secretive nature during the breeding season, their tendency to nest and forage high in the forest canopy, and their apparently nomadic wandering (Bekoff and Scott, 1989; Bekoff et al., 1989; Scott and Bekoff, 1991). Recognition that calls differ distinctively and can therefore be used to map ranges and track movements of call types has created an opportunity to understand aspects of grosbeak biology that were previously challenging to resolve (Robinson et al., 2024). Fortunately, technological advances available to the public for recording bird sounds, along with the accessibility of large data repositories such as the Macaulay Library (Sullivan et al., 2009), have generated new data streams useful for studying subtle differences among infraspecific groups of birds. Previously unanswerable questions about Evening Grosbeaks now seem tractable: Does each call type maintain a breeding range exclusive to the other call types? When and where do call types form mixed-call-type groups, if at all? Recent evidence from Wyoming shows consistent overlap in the breeding ranges of call types 1 and 4 (e.g., Duman and Hahn, 2024). Does similar overlap occur in the ranges of other geographically adjacent call types?

We focused our research on a zone of potential overlap in the breeding season ranges of types 1 and 2 Evening Grosbeaks. Type 1 is considered the predominant breeding call type of the Pacific Northwest, extending from British Columbia south through at least southern Oregon. Type 2 has been documented breeding largely in the Sierra Nevada of California, but the northern extent of its range remains unclear. By integrating data from recordings contributed by community scientists, autonomous recording units we placed at targeted locations along a longitudinal transect from Oregon into northern California, and our own field observations, we investigated the spatial and temporal overlap of type 1 and type 2 Evening Grosbeaks from May through August. Our goal was to clarify the extent of spatial overlap in occurrence of the two call types during the breeding season (which is concentrated in late May through early July, [(Gillihan and Byers, 2020)]) while also advancing our understanding of co-occurrence of call types 1 and 2 during pre-breeding migration (early to mid-May) and the early post-breeding period (mid-late July to August).

2 Methods

2.1 Mapping archived records and personal field observations

The Macaulay Library contains bird sound recordings associated with eBird checklists contributed by community scientists (Sullivan et al., 2009). Those checklists contain location data and date of observation. We visually inspected the spectrograms and identified to call type Evening Grosbeak recordings archived in the Macaulay Library from May through August within California, Nevada, and Oregon (south of Bend), primarily spanning the years 2017 through 2024 (n=119). Specifically, this media search was conducted by viewing “Species Maps” on the eBird explore page, using the filter “explore rich media” and searching “Evening Grosbeak”. Each archived recording was evaluated using characteristics described by Sewall et al. (2004); Robinson et al. (2024), and Kirsch et al. (2025) to ensure consistency in call type identification and analysis. That is, we inspected parameters such as frequency (kHz), duration (s), and shape of flight calls, which are highly distinct (Figure 1). We mapped the location of identified Evening Grosbeak call types using ArcGIS Pro Version 3.2.0 (Esri, 2022).

Figure 1

Figure 1. Typical structure of Evening Grosbeak type 1 and type 2 flight calls. The type 1 flight call is a single descending whistle with at most a very brief rising component that is difficult for a human to detect, while the type 2 flight call has a protracted and pronounced early ascending element that is conspicuous to a human, followed by a slightly longer descending element. Each call spans the 3–5 kHz frequency range and is approximately 0.2 seconds in duration, with type 1 calls averaging shorter and type 2 averaging longer. Discriminant function analyses by Robinson et al. (2024) found these flight calls to be reliably identifiable in 99% of cases. Image created by Reya Coats.

In addition to these community-sourced records, we also included observations provided by one of us (TPH [n=98]) for the same region between the years 1997-2025. His field notes were based on by-ear identification of Evening Grosbeaks to call type, and these observations were made opportunistically while conducting other field work in California, Oregon, and Washington. In many instances, audio recordings were also made of the same observations and were used to confirm in-field identification of Evening Grosbeak call types. However, Evening Grosbeak call types are extremely distinctive to experienced observers, and ambiguous calls are extremely unusual (Robinson et al., 2024; Figure 1).

2.2 Autonomous recording unit deployment

We augmented the Macaulay Library data by deploying a network of autonomous recording units (ARUs) across 19 unique sites extending from the Sierra Nevada in central California to the central Cascades in central Oregon (Figure 2). We used SwiftOne (https://www.birds.cornell.edu/ccb/swift-one/) and Wildlife Acoustic Song Meter 4 ARUs (https://www.wildlifeacoustics.com/products/song-meter-sm4). Because Evening Grosbeak flight calls are always below 6 kHz in frequency, an ARU sample rate of 12,000 Hz or higher is sufficient to detect Evening Grosbeaks flight calls. The SwiftOne models recorded with a sampling rate of 32,000 Hz, while the Wildlife Acoustic Song Meter 4 ARUs recorded with a sampling rate of 24,000 Hz. Evening Grosbeak flight calls are loud and distinctive, diminishing concerns that differences in recorder specifications could contribute to sampling issues.

Figure 2

Figure 2. Map of ARU transect. Our ARU transect covered an approximately 775 km latitudinal range from central California (the south-central portion of the map) to central Oregon (the north-central portion of the map). The 20 unique sites (19 sites are visible on the map; both Four Mile Lake locations are represented by one point due to their close proximity) sampled over a span of four years, with densest sampling prioritized in the area that is the primary zone of overlap between type 1 and type 2 Evening Grosbeaks in southern Oregon near the California border.

The ARUs were placed initially in 2021, with additional units added annually from 2022 to 2024. The additional ARUs were positioned to prioritize sampling more densely within areas we previously noted to contain both call types (Table 1). In total, we used data from ARUs placed at 20 different locations, with each location sampled from 1–4 years, for a total of 36 year-location combinations. Sampling effort across dates and locations varied slightly from year to year, due to logistical issues associated with snow-melt dates, fires, and other access limitations (Table 1).

Table 1

Table 1. ARU locations, dates of sampling and elevations for 2021-2024, Oregon and California.

2.3 Data processing and analysis

For each hour of recording, ARUs generated a.WAV file, which we analyzed for the presence of Evening Grosbeak calls; when detected, we identified flight calls to type. We subsampled ARU files for analysis by randomly selecting one file per day starting within the hours of 06:00-08:00 (Swifts) and 06:15-08:30 (Wildlife Acoustics) for analysis. Each selected hour-long file was processed in Raven Pro Version 1.6 using the BirdNET Learning Detector to identify potential Evening Grosbeak vocalizations (Kahl et al., 2021). We set the BirdNET Learning Detector threshold at 0.25. We used this low setting to minimize false negatives; all detections flagged by the learning detector were then manually reviewed to verify accuracy and to identify additional call notes in the surrounding audio context one minute before and after the flagged detection. Using this method, we confidently discarded all false positive detections and manually flagged fainter calls that were missed by the BirdNET detector. Identified calls were categorized into call types using the classification framework outlined by Robinson et al. (2024). This process involved detailed comparisons of acoustic parameters, including frequency, duration, and modulation patterns to ensure accurate call type assignment.

To extract information about possible cross-type flocking, we recorded observations made by an ARU of “simultaneous occurrences” of both types. Simultaneous occurrences were instances in which the presence of both type 1 and 2 flight calls were identified within the same three-second time frame. We use this arbitrary time frame to infer close association and the possibility of call types flocking together. We hypothesize that calls heard further apart in time suggest a higher likelihood of birds acting separately. While related, statements about “sympatry” simply refer to the observation of both types being present at a particular location or set of locations over longer sampling intervals. To quantify the proportion of days a particular Evening Grosbeak type was observed at a particular ARU location, we calculated a metric we called “encounter frequency”. Encounter frequency was calculated based on the number of days Evening Grosbeaks were recorded (e.g., those of a particular type) on a given ARU divided by the number of days the ARU was recording during a particular time frame (e.g., 2023, July 2022, or across all deployment days).

We analyzed the relationship between latitude and proportion of days each call type was recorded with polynomial regression implemented in JMP Pro 18. The “Fit Y by X” platform was used with least-squares estimation. Models of increasing polynomial order (linear, quadratic, cubic) were compared. We compared adjusted R² values to assess fit.

3 Results

We analyzed 1,865 hours of ARU data sampled across 20 unique locations over four years and mapped 119 community-sourced Evening Grosbeak recordings to characterize the geographic distribution of type 1 and type 2 Evening Grosbeaks from May through August (Figures 3–5). We also reviewed 98 records of Evening Grosbeaks observed by TPH during the years 1997 to 2025 to better assess the May to August range limits of both types, as well as the stability of the type 1 and 2 contact zone over recent time. Throughout the study area, we only detected type 1 and 2 Evening Grosbeaks, never any of the other Evening Grosbeak types. With 3 exceptions, recordings we identified were unambiguously attributed to one or the other type (Supplementary Table 1). Type 2 birds dominated detections in the southern part of the transect, from Soda Springs (39.2 N; our southernmost ARU) to Trout Creek (41.4 N). In the middle part of the transect – which included ARUs at Baldy Creek (42.1 N) north to Sevenmile Guardstation (42.7 N) – neither type was clearly dominant, with call type proportions relatively equal or varying significantly between sites (Table 2). In the northern part of the transect, from ARUs at Crater Lake (42.7 N) north to Waldo Lake (43.6 N), detections were dominated by type 1 birds. Low detection rates and the presence of both types at the Camp Sherman ARU (44.4 N; our northernmost site) prevent us from drawing conclusions about call type status there. Overall, the quadratic relationship between latitude and proportion of days each type was recorded was not significant (R² = 0.05, F_2,33 = 2.0, p>0.15). The proportion of days type 2 birds were recorded was less farther north than south, with the quadratic relationship again fitting best (R² = 0.39, F_2,33 = 12.2, p < 0.0001). Thus, overlap of geographic ranges was extensive but type 2 tended to be rare to absent farther north while type 1 was more widespread (Figure 6).

Figure 3

Figure 3. Range of type 1 Evening Grosbeaks in Oregon and California, during May – August based on recordings archived and identified from the Macaulay Library (n=54). The inset shows the typical call shape of type 1.

Figure 4

Figure 4. Range of type 2 Evening Grosbeaks in Oregon and California, during May through August based on recordings archived and identified from the Macaulay Library (n=65). The inset shows the typical call shape of type 2.

Figure 5

Figure 5. Relative proportions of Evening Grosbeak call types in Oregon and California, from May through August ARU data. Blue bars labeled “T1” represent type 1 detection frequencies; red bars labeled “T2” represent type 2 detection frequencies; purple bars labeled “DS” represent the frequency that both call types were detected simultaneously (i.e. within 3 secs of each other). The six sites shown here were selected based on their location and number of total Evening Grosbeak detections; observations from each site were included across all years an ARU was recording and each site histogram depicted is based on ≥ 27 Evening Grosbeak detections.

Table 2

Table 2. Proportions (percentage of time intervals sampled in which birds of that type were detected) of each call type detected at each sampling site from 2021-2024.

Figure 6

Figure 6. As a function of latitude, the proportion of days each Evening Grosbeak call type was detected at each ARU site. Lines of best fit are quadratic with 95% confidence intervals. Call type 1 (blue) and call type 2 (red).

Type 1 birds were detected across almost the entire transect whereas type 2 birds were nearly limited to the central and southern portions of the study region. The main zone of type 1 and 2 sympatry was identified to occur in southern Klamath and Jackson Counties, Oregon, approximately centered between Fish Lake and Lake of the Woods (42.3 N). It extended across a range of about 30 km, where the ratios of type 1 to type 2 individuals were approximately equal (Table 2). We detected both types at each unique ARU location (n=20); however, only at 13 of these locations did we detect simultaneous occurrences (i.e., calls of both type 1 and type 2 birds within the same 3-second recording interval) of type 1 and type 2 birds, and only at 4 of these locations (Trout Creek, Chinquapin Mountain, Four Mile Lake, and Crater Lake) did we detect simultaneous occurrences of both types across multiple years (Table 2). Furthermore, we observed a higher proportion of simultaneous occurrences of both types later in the season. In June, Evening Grosbeaks were detected on 23% of total ARU recording days (n=529), with simultaneous occurrences of both types comprising 11% of those detections. In July, the detection frequency of Evening Grosbeaks was similar at 22% of total July recording days (n=926), but the proportion of those detections with simultaneous occurrences of both types increased to 15%.

We also report striking variation in detection rates across sites, even among those in close proximity (Table 3). For instance, in 2024, the Lake of the Woods ARU had a detection frequency more than twice that of the Keno ARU, approximately 22 km away. Furthermore, the Keno ARU had a detection frequency more than triple that of the Chinquapin Mountain Road ARU, approximately 18 km away (Table 3). Evening Grosbeaks were also most frequently detected in the core contact zone and in the southern portions of the transect, with significantly fewer detections in northern areas, and east in the Warner Mountains. Except for the 2024 Waldo Lake ARU, each of the three northernmost ARU sites had Evening Grosbeak detection frequencies ≤ 0.08 for each season sampled. In the Warner Mountains, the 2024 Bear Camp Flat ARU had a detection frequency of 0.10 and the Drake Springs ARU detection rate was ≤ 0.05 for each of the years it ran from 2021 through 2023. Numbers of detections also fluctuated at the same sites between years (Table 4). At Soda Springs, Domingo Springs, Trout Creek, and Crater Lake, detection frequencies varied by at least a factor of 2 during the month of July (the month for which we have the most consistent recording data) from 2021 through 2023 (Table 4).

Table 3

Table 3. Variation in ARU site detection in the core contact area in 2024.

Table 4

Table 4. Interannual site variability of SwiftOne ARUs recording from 2021 to 2023.

As for defining the May – August range limits of type 1 and type 2 Evening Grosbeaks, TPH reported the presence of type 2 birds in Washington on three occasions: near Mount Adams (46.2 N) on 20 July 2003, at Hugo Lake (46.5 N) on 30 June 2015, and at Bethel Ridge Camp (46.7 N) on 30 June and 1 July 2015 (Supplementary Files). These records represent the northernmost detections of type 2 Evening Grosbeaks during the May–August period. For the southern limits of type 1 Evening Grosbeaks, TPH has consistently observed a northward movement through the Sierra Nevada in May, followed by a general absence in June. However, there are several early June records in California, including one near the Donner Camp Picnic Area (39.4 N) on 4 June 1998 and multiple observations around Tioga Pass (37.9 N) on 1, 2, and 21 June 2025 (Supplementary Files). Additionally, late June records include detections at Northstar (39.3 N) on 29 June 2006 and near Soda Springs (39.3 N) on 25 June 2022 (Supplementary Files).

4 Discussion

We found a north-south transition between type 1 and type 2 Evening Grosbeaks, confirming that type 1 Evening Grosbeaks primarily occupy areas further north in the Pacific Northwest, while type 2 Evening Grosbeaks are mainly concentrated in the Sierra Nevada of California and the southern Cascades of southern Oregon. These results are consistent with our (TPH) ad hoc field observations indicating that type 1 become even more abundant further north and west in Washington, with type 2 being absent or near-absent there. However, we observed widespread co-occurrence of type 1 and type 2 Evening Grosbeaks, with both types detected at every ARU location sampled. Despite this overlap, type 2 birds dominated detections from Soda Springs (our southernmost ARU) to Trout Creek (just south of the CA-OR border). From the Baldy Creek ARU north to the Sevenmile Guardstation ARU, neither type was clearly dominant, with call type proportions relatively equal or varying significantly between sites. From the Crater Lake ARU north to the Waldo Lake ARU, detections were consistently dominated by type 1 birds. Low detection rates and the presence of both types at the Camp Sherman ARU (our northernmost site) prevent us from drawing conclusions about call type status there, but type 1 generally predominates in this area based on our in-person visits to this site. We identified the main area of overlap between type 1 and type 2 Evening Grosbeaks to occur in southern Klamath and Jackson Counties, Oregon, approximately centered between Fish Lake and Lake of the Woods. This primary zone of overlap, where the ratios of type 1 to type 2 detections were approximately equal, apparently extended across a latitudinal range of about 30 km. We consistently observed these call type dynamics in each year sampled, suggesting that the observed dynamics are persistent features of Evening Grosbeak call type 1 and 2 distributions from May through August.

The ecological and evolutionary significance of call type populations, particularly those with stable geographic boundaries during the breeding season, requires further study. Duman and Hahn (2024) studied another Evening Grosbeak call type contact zone in western Wyoming and similarly found that call types 1 and 4 overlapped during the breeding season in and around Jackson Hole. They also observed that these call types shared similar habitat associations and only call types 1 and 4 were present during the breeding season in approximately similar proportions across a 24-year period. Flight calls with characteristics intermediate between the two were exceedingly rare. However, instances of “call switching,” where individuals produced both type 1 and type 4 flight calls in quick succession, were reported. Because our study primarily relied on ARU data, we were unable to confirm the occurrence of call switching within the southern Oregon contact zone. Additional data on the frequency of call switching and mixed-type pairs would allow us to more confidently evaluate the ecological and cultural factors that maintain call type geographic range boundaries. Nevertheless, presently available evidence suggests that formation of mixed-type pairs and call switching are both rare. Additional detailed observational studies are needed.

Flight calls in fringillids likely evolved as a mechanism for communication between distinct cultural and/or ecological groups (i.e., ecotypes). Notably, only certain finch species – such as Evening Grosbeaks (Sewall et al., 2004; Kirsch et al., in review), Red Crossbills (Loxia curvirostra) (Groth, 1988, 1993; Young et al., 2024), and Pine Grosbeaks (Pinicola enucleator) (Adkisson, 1981) – appear to exhibit discrete geographic variation in vocalizations. In contrast, other closely related species, such as Pine Siskins (Spinus pinus), White-winged Crossbills (Loxia leucoptera), and American Goldfinches (Spinus tristis) do not appear to show such clear geographic vocal differentiation. Even among species with regionally or ecologically associated vocal differences, the degree of distinction can vary. For example, Evening Grosbeak call types are highly distinct, confirmed by statistical analysis reliably identifying separate call type groups (Kirsch et al., 2025). Red Crossbill call types, however, appear to vary more gradually, based on extensive personal observations (personal observations by authors MAY, TPH, and WDR), with greater overlap and a higher prevalence of intermediate forms. This pattern may reflect more frequent interactions among different Red Crossbill call types and/or broader geographic overlap during the breeding season. The presence of more ambiguous or intermediate call types may also suggest weaker selection for vocal divergence in Red Crossbills, or possibly an adaptive advantage (or absence of a significant fitness cost) to having intermediate calls and associated social and morphological differences. However, more concrete, quantitative analyses of Red Crossbill vocalizations at both continental and global scales are needed, along with further investigation into the social and ecological forces driving the more pronounced differences in call types seen in Evening Grosbeaks.

One possible explanationfor the evolution and maintenance of such discrete Evening Grosbeak call type groups is that differences in migration patterns between call type populations contributed to divergence, with call types facilitating group cohesion for birds best adapted for a particular migratory route. Similar movement-driven divergence has been the proposed diversifying mechanism for cryptic species of Giant Hummingbirds (Patagonas gigas and P. chaski) in South America, as well as in populations of Eurasian Blackcaps (Sylvia atricapilla) in Europe (Irwin, 2009; Williamson et al., 2024). Given the broad, irruptive movements of type 1 Evening Grosbeaks compared with the more localized distribution of type 2 birds around the southern Cascades and Sierra Nevada (Figure 2), this hypothesis could be explored further using modern tracking technologies and the identification of key wintering areas, major stopover sites, and general migratory routes from such movement data combined with additional collection of sound recordings. Our observations from the ARU network, community-sourced recordings, and fieldwork (TPH) align with a consistent pattern of type 1 Evening Grosbeaks migrating north through parts of California in spring (May), followed by a southward return in the post-breeding season (July to August). Similarly, historical notes and recent standardized counts from the Oregon State University campus show a consistent peak in Evening Grosbeak numbers around late April and early May each year, further supporting the idea of regular, potentially route-specific migration in type 1 individuals (Robinson et al., 2022).

Laboratory work by Groth (1993) and Sewall (2011) has found that young Red Crossbills learn the call type identity of their “social tutor,” be it their biological parent or a foster parent. This learning appears to occur during a critical period early in life. In contrast, adult Red Crossbills are rarely able to perfectly imitate the flight calls of other call types but can gradually modify both their own calls and their responsiveness to other Red Crossbill types with time (Sewall, 2009; Sewall and Hahn, 2009). Such call type flexibility could be a byproduct of social learning in Red Crossbills and/or the product of high social motivation; however, it is unknown the extent to which different Red Crossbill call types can learn the calls or modify their calls to those of other call types. For instance, Sewall (2011) demonstrated that young type 3 and type 4 Red Crossbills – ecotypes with relatively similar bill morphology and foraging preferences on soft-coned conifers – can accurately copy each other’s calls when experimentally cross-fostered. However, it is unknown whether a young type 3 bird could (or would) learn the call of more morphologically distinct types, such as type 5 or type 9, which are adapted to feed on harder-coned pines. These social and ecological factors likely contribute to the maintenance of various call type populations in Red Crossbills. Understanding whether Evening Grosbeaks share similar patterns of vocal development and identifying any associated ecological differences – potentially through studies characterizing the palate structure in different Evening Grosbeak types – could help clarify the mechanisms underlying the persistence of their call type populations (Benkman, 2003). Furthermore, if flight calls influence migratory routes, genetic predispositions (e.g., differences in morphology or migratory orientation) may favor individuals that adopt the same call type identity as their biological parents.

Despite these insights, the evolutionary processes responsible for the formation and maintenance of call type boundaries in Evening Grosbeaks remain unclear. Limited laboratory and field studies make it difficult to draw firm conclusions. Future research should aim to revisit the morphological distinctions between Evening Grosbeak populations identified by Grinnell (1917) and confirm whether morphological variation across the Evening Grosbeak’s range is clearly associated with vocal identity. For example, such a study might confirm that all (≥ 95%) Evening Grosbeaks that have morphological traits aligned with the C.v. brooksi subspecies give type 1 vocalizations. Additionally, comparative studies involving population genomics, vocalizations, and life histories of all Coccothraustes species and subspecies populations may offer a compelling case study on the evolution of vocal structures and the potential for cryptic speciation.

Ultimately, our study provides a modern context for revisiting the early observations of Grinnell (1917). Brooks’ original description of “certain peculiarities” in Evening Grosbeak specimens, later expanded upon by Grinnell’s morphological analyses, hinted at underlying complexities within what was then considered a single subspecies. By using flight calls to identify these otherwise cryptic populations, we were able to map the breeding ranges and describe the core contact zone between call types 1 and 2 in the Pacific Northwest and the Sierra Nevada. Further observational studies of type 1 and type 2 individuals within this contact zone could yield valuable insights into the extent of call switching, the possibility of interbreeding among call types, and potential habitat use and behavioral differences between the types. Likewise, increased sampling in the Warner Mountains, the Coast Range, and the Cascade Mountains of Northern California would help clarify the relative abundance and distribution of Evening Grosbeak call types in these regions, along with their seasonal movement patterns throughout the annual cycle.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Author contributions

WK: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. TH: Conceptualization, Data curation, Investigation, Methodology, Writing – review & editing. MY: Conceptualization, Investigation, Methodology, Resources, Validation, Writing – review & editing. WR: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. Support was provided by the Finch Research Network, College of Agricultural Sciences Student Researcher Program, the ER Jackman Internship Support Program (WK), the Bob and Phyllis Mace Watchable Wildlife Professorship, and Hatch funds (WR).

Acknowledgments

We thank the many birders who upload their recordings of Evening Grosbeaks to eBird. We thank the K. Lisa Yang Center for Conservation Bioacoustics who provided access to Raven Pro Version 1.6. We thank Reya Coats for Figure 1.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author WR declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbirs.2026.1671357/full#supplementary-material

References

Adkisson C. S. (1981). Geographic variation in vocalizations and evolution of North American pine grosbeaks. Condor 83, 277. doi: 10.2307/1367494

Crossref Full Text | Google Scholar

Bekoff M. and Scott A. C. (1989). Aggression, dominance, and social organization in evening grosbeaks. Ethology 83, 177–194. doi: 10.1111/j.1439-0310.1989.tb00528.x

Crossref Full Text | Google Scholar

Bekoff M., Scott A. C., and Conner D. A. (1989). Ecological analyses of nesting success in evening grosbeaks. Oecologia 81, 67–74. doi: 10.1007/BF00377012

PubMed Abstract | Crossref Full Text | Google Scholar

Benkman C. W. (2003). Divergent selection drives the adaptive radiation of crossbills. Evolution 57, 1176–1181. doi: 10.1111/j.0014-3820.2003.tb00326.x

PubMed Abstract | Crossref Full Text | Google Scholar

Chesser R. T., Isler M. L., Cuervo A. M., Cadena C. D., Galen S. C., Bergner L. M., et al. (2020). Conservative plumage masks extraordinary phylogenetic diversity in the Grallaria rufula (Rufous Antpitta) complex of the humid Andes. Auk 137, ukaa009. doi: 10.1093/auk/ukaa009

Crossref Full Text | Google Scholar

Duman K. W. and Hahn T. P. (2024). Sympatric breeding occurrence of two call types of Evening Grosbeak in Western Wyoming. Front. Bird Sci. 3. doi: 10.3389/fbirs.2024.1371155

Crossref Full Text | Google Scholar

Edwards S. V., Kingan S. B., Calkins J. D., Balakrishnan C. N., Jennings W. B., Swanson W. J., et al. (2005). Speciation in birds: Genes, geography, and sexual selection. Proc. Natl. Acad. Sci. 102, 6550–6557. doi: 10.1073/pnas.0501846102

PubMed Abstract | Crossref Full Text | Google Scholar

Esri Inc (2022). ArcGIS Desktop: Release 10.8.2. (Redlands, CA: Environmental Systems Research Institute). Available online at: https://www.esri.com/en-us/arcgis/products/arcgis-pro/overview.

Google Scholar

Gillihan S. W. and Byers B. E. (2020). “Evening Grosbeak (Coccothraustes vespertinus), version 1.0,” in Birds of the World. Eds. Poole A. F. and Gill F. B. (Cornell Lab of Ornithology, Ithaca, New York, USA). doi: 10.2173/bow.evegro.01

Crossref Full Text | Google Scholar

Grinnell J. (1917). The subspecies of hesperiphona vespertina. Condor 19, 17–22. doi: 10.2307/1362452

Crossref Full Text | Google Scholar

Groth J. G. (1988). Resolution of cryptic species in appalachian red crossbills. Condor 90, 745–760. doi: 10.2307/1368832

Crossref Full Text | Google Scholar

Groth J. G. (1993). Evolutionary Differentiation in Morphology, Vocalizations, and Allozymes Among Nomadic Sibling Species in the North American Red Crossbill (Loxia Curvirostra) Complex (Berkeley, California USA: University of California Press).

Google Scholar

Hill G. E. (2017). The mitonuclear compatibility species concept. Auk 134, 393–409. doi: 10.1642/AUK-16-201.1

Crossref Full Text | Google Scholar

Irwin D. E. (2009). Speciation: new migratory direction provides route toward divergence. Curr. Biol. 19, R1111–R1113. doi: 10.1016/j.cub.2009.11.011

PubMed Abstract | Crossref Full Text | Google Scholar

Kahl S., Wood C. M., Eibl M., and Klinck H. (2021). BirdNET: A deep learning solution for avian diversity monitoring. Ecol. Inf. 61, 101236. doi: 10.1016/j.ecoinf.2021.101236

Crossref Full Text | Google Scholar

Kirsch W. M., Centanni C. T., Ngo S. H., Young M. A., Hahn T. P., and Robinson W. D. (2025). Quantitative analysis of the flight calls and trills of Evening Grosbeaks (Coccothraustes vespertinus). Front. Bird Sci. 4. doi: 10.3389/fbirs.2025.1613273

Crossref Full Text | Google Scholar

Robinson W. D., Greer J., Masseloux J., Hallman T. A., and Curtis J. R. (2022). Dramatic declines of evening grosbeak numbers at a spring migration stop-over site. Diversity 14, 496. doi: 10.3390/d14060496

Crossref Full Text | Google Scholar

Robinson W. D., Nanau M., Kirsch W., Centanni C. T., and Clements N. M. (2024). Flight calls and trills of Evening Grosbeaks can be used to map movements and ranges of call types 1 and 2. Front. Bird Sci. 3. doi: 10.3389/fbirs.2024.1340750

Crossref Full Text | Google Scholar

Scott A. C. and Bekoff M. (1991). Breeding behavior of evening grosbeaks. Condor 93, 71–81. doi: 10.2307/1368608

Crossref Full Text | Google Scholar

Sewall K. B. (2009). Limited adult vocal learning maintains call dialects but permits pair-distinctive calls in red crossbills. Anim. Behav. 77, 1303–1311. doi: 10.1016/j.anbehav.2009.01.033

Crossref Full Text | Google Scholar

Sewall K. B. (2011). Early learning of discrete call variants in red crossbills: implications for reliable signaling. Behav. Ecol. Sociobiol. 65, 157–166. doi: 10.1007/s00265-010-1022-0

PubMed Abstract | Crossref Full Text | Google Scholar

Sewall K. B. and Hahn T. P. (2009). Social experience modifies behavioural responsiveness to a preferred vocal signal in red crossbills, Loxia curvirostra. Anim. Behav. 77, 123–128. doi: 10.1016/j.anbehav.2008.09.016

Crossref Full Text | Google Scholar

Sewall K., Kelsey R., and Hahn T. P. (2004). Discrete variants of evening grosbeak flight calls. Condor 106, 161–165. doi: 10.1093/condor/106.1.161

Crossref Full Text | Google Scholar

Sullivan B. L., Wood C. L., Iliff M. J., Bonney R. E., Fink D., and Kelling S. (2009). eBird: A citizen-based bird observation network in the biological sciences. Biol. Conserv. 142, 2282–2292. doi: 10.1016/j.biocon.2009.05.006

Crossref Full Text | Google Scholar

Toews D. P. L. and Irwin D. E. (2008). Cryptic speciation in a Holarctic passerine revealed by genetic and bioacoustic analyses. Mol. Ecol. 17, 2691–2705. doi: 10.1111/j.1365-294X.2008.03769.x

PubMed Abstract | Crossref Full Text | Google Scholar

Uy J. A. C., Irwin D. E., and Webster M. S. (2018). Behavioral isolation and incipient speciation in birds. Annu. Rev. Ecol. Evolution Systematics 49, 1–24. doi: 10.1146/annurev-ecolsys-110617-062646

Crossref Full Text | Google Scholar

Williamson J. L., Gyllenhaal E. F., Bauernfeind S. M., Bautista E., Baumann M. J., Gadek C. R., et al. (2024). Extreme elevational migration spurred cryptic speciation in giant hummingbirds. Proc. Natl. Acad. Sci. 121, e2313599121. doi: 10.1073/pnas.2313599121

PubMed Abstract | Crossref Full Text | Google Scholar

Young M. A., Spahr T. B., McEnaney K., Rhinehart T., Kahl S., Anich N. M., et al. (2024). Detection and identification of a cryptic red crossbill call type in northeastern North America. Front. Bird Sci. 3. doi: 10.3389/fbirs.2024.136399

Crossref Full Text | Google Scholar