1 2 3 4 5 COMMENTARY 6 7 8 EXPANDING THE TRADITIONAL DEFINITION OF MOLT-MIGRATION 9 Christopher M. Tonra1* and Matthew W. Reudink2 10 1 School of Environment and Natural Resources, The Ohio State University, 2021 Coffey Road, 210 Kottman Hall, Columbus, OH 43210, USA. 11 12 13 2 Department of Biological Sciences, Thompson Rivers University, 805 TRU Way, Kamloops, BC V2C 0C8, Canada 14 15 16 17 18 19 20 21 22 23 *Corresponding author: tonra.1@osu.edu 1 1 ABSTRACT 2 The occurrence of molt during migration, known as “molt-migration,” has increasingly received 3 attention across many avian taxa, since first being described in waterfowl in the 1960’s. 4 However, despite the many different types of molt stages and strategies, most, if not all, uses of 5 the term “molt-migration” apply to the definitive prebasic molt of flight feathers in post-breeding 6 adults, whereas fewer studies address migration for body-feather molts. Here, we argue that the 7 current definition of molt-migration, as applied, is limited in focus relative to the diverse ways in 8 which it can manifest in avian populations. We suggest a new, broader definition of molt- 9 migration and highlight examples of molt-migration as traditionally defined, and the many 10 examples that have not been defined as such. We propose a new, two-tiered typology for 11 defining different forms of molt-migration, based on 1) its progression relative to stationary 12 portions of the annual cycle and 2) the stage of molt involved. In order to advance our 13 understanding of the ecology and evolution of this increasingly documented phenomenon and 14 apply this knowledge to conservation and management, avian researchers must begin to utilize a 15 common framework for describing molt-migration in its’ various forms. 16 Keywords: migration, molt, molt-migration, seasonal interactions, stopover 17 18 19 20 21 22 23 2 1 OVERVIEW 2 Migratory birds must balance three energetically expensive events during the annual cycle: 3 breeding, migration, and molt. Although breeding and migration have received an enormous 4 amount of attention, relatively little work has been dedicated to understanding the evolution of 5 molt strategies (Leu and Thompson 2002). This is an important knowledge gap, given the 6 energetic costs associated with molt (Dietz et al. 1992, Vézina et al. 2009), and the vast variation 7 in when, where, and which feathers birds molt (e.g., Howell et al. 2003, Pyle et al. 2009, 8 Lourenço and Piersma 2015, Wiegardt et al. 2017b). Indeed, molt appears to be highly labile, 9 with different molt strategies arising within and among species independently over relatively 10 short evolutionary time (Pyle et al. 2009, Pyle 2013a). 11 Many birds balance energetically expensive events through temporal separation, for 12 example by performing the definitive prebasic molt prior to fall migration (Pyle 1997, Leu and 13 Thompson 2002, Froehlich et al. 2005). However, this strategy does not hold for all species, with 14 some species actively molting feathers while migrating and others interrupting migration to molt 15 at stopover locations. The nature of this type of molt strategy can take many forms, even within a 16 single family (e.g., Leu and Thompson 2002). Further complicating matters, there is substantial 17 variation in which feathers are molted (e.g., complete or partial molt; contour feathers or flight- 18 feathers), as well as intraspecific variation among (Nordell et al. 2016), and even within (Tonra 19 et al. 2015), individuals. This variation is often captured by the single term, “molt-migration” 20 (for flight-feather molts), or lacks any classification at all (e.g., for partial molts). As a result, the 21 term “molt-migration” is lacking in specificity and is in need of an updated definition that better 22 captures its various forms. The impetus behind this commentary was to shed light on the 23 numerous molt strategies used by migratory birds and to disentangle and simplify the language 3 1 used to describe these strategies. We also hope to draw more attention to the complexity of molt- 2 a chronically understudied, but critical aspect of avian life history. 3 4 DEFINING MOLT MIGRATION 5 Since the term “molt-migration” was first introduced, the definition of this phenomenon has 6 evolved in ways that slightly alter what types of movements are included. Salomonsen (1968; 7 summarized by Jehl 1990) defined molt-migration as: “birds moving from the breeding grounds 8 to a special molting area where they can rapidly replace their flight feathers at a low predation 9 risk before resuming their migration to the winter quarters.” This definition confines molt- 10 migration to post-breeding flight-feather molts, and to the use of “special moulting areas”. Thus, 11 instances where molting sites also serve as refueling locations, or molt occurs during active 12 migrating are seemingly excluded. Per Salomensen’s definition, molt-migration is distinct from 13 migration to overwintering sites, as opposed to occurring during/overlapping with migration. Leu 14 and Thompson (2002) updated the definition to describe when “bird species interrupt the annual 15 fall migration at specific locations to molt their flight feathers.” Unlike the previous definition, 16 Leu and Thompson more explicitly imply molt-migration occurs within the larger migration life 17 history stage and they included both elevational and latitudinal movements in their review of 18 molt-migration. However, not captured in this definition are birds that molt continuously during 19 active migration (though they include such examples in their literature review). Further, it 20 excludes spring migration and continues to refer to molt as the primary function of the area 21 utilized, excluding birds that molt at staging/stopover sites also utilized for refueling. Pyle et al. 22 (2009) largely used the same definition as Leu and Thompson (2002) in describing the “monsoon 4 1 migrant” systems of western North America, but are more general in saying birds “stop and 2 molt”, broadening the definition beyond flight-feather molt, as in the two previous definitions. 3 In previous definitions (e.g. Salomonsen 1968), molt-migration is treated primarily as a 4 form of movement relative to molt, and separate from migration. However, there are many 5 examples of instances where migratory movement and molt of feathers overlap, such that the 6 destinations are not solely visited for molting, and serve as refueling sites as well (e.g. Lourenço 7 and Piersma 2015; see below for more examples). Thus, we feel that a more comprehensive 8 definition is required to capture the full breath of strategies whereby these two life history stages 9 overlap and interact. Ramenofsky and Wingfield (2005) reviewed and conceptualized such 10 overlap in life history stages, in the context of the transition leading into breeding for migratory 11 birds. Later, Wingfield (2008) described the organization of annual cycles in terms of a “finite 12 state machine,” where annual cycles have a finite number of distinct stages that exist on a 13 continuum, with the development of one stage often overlapping completion of another. We feel 14 such an annual cycle approach is germane to the instances of transitioning between molt and 15 migration. Therefore, we define molt-migration as “temporal overlap in the molt and migration 16 life history stages.” This definition includes all instances where scheduled feather replacement 17 occurs at “special molting areas” (sensu Salomonsen 1968), refueling sites, or during active 18 migration. Further, it can be applied where molts that do not include flight feathers occur during 19 migration. With this broader definition that better represents the way in which molt-migration is 20 referred to in the literature, we capture a wide variety of systems, thus requiring a typology to 21 more specifically classify each one. In order to develop such a system, we first review some 22 examples of how molt-migration is manifest in birds. 23 5 1 EXAMPLES OF MOLT-MIGRATION STRATEGIES, AS TRADITIONALLY DEFINED 2 Flight-feather molt occurring entirely on migratory stopover 3 Perhaps the first use of the term “molt-migration” was by Salomonsen (1968) in describing 4 waterfowl post-breeding movements. Specifically, the term was applied to the movements of 5 many ducks, and later other Anseriformes (e.g., geese; Ogilve 1978), to secluded marshes in 6 order to complete flight-feather replacement during their definitive prebasic molt, including a 7 period of flightlessness. This phenomenon, which has been detailed in many species of 8 waterfowl since Salomonsen (e.g., Hohman et al. 1992, Robert et al. 2002, Dickson et al. 2012), 9 revealed a critical component of the habitat needs specific to molt in waterfowl, separate from 10 wintering and breeding. In addition, several species of grebe (Family Podicipedidae) have a 11 similar molt-migration as waterfowl (e.g., Storer and Jehl 1985, Piersma 1988). For example, 12 Eared Grebes (Podiceps nigricollis) halt migration at hypersaline lakes in the western U.S. to 13 complete flight-feather molt (e.g., Mono Lake, Great Salt Lake; Storer and Jehl 1985). These 14 grebes also endure a flightless period during this stopover as they molt all of their flight feathers. 15 Some shorebirds migrate to staging areas during fall migration and replace flight feathers, such 16 as Wilson’s Phalaropes (Phalaropus tricolor), which partially replace their primaries while 17 stopping over at many of the same lakes utilized by Eared Grebes (Jehl 1987). 18 In the arid regions of western North America, temperatures soar and resources become 19 increasingly limited by late summer, at which time birds are faced with the prospect of 20 completing their definitive prebasic molt with limited food resources in a harsh, demanding 21 environment. Several western migrants have developed an effective strategy for dealing with 22 limited resources for molt during the post breeding period (summarized in Rohwer et al. 2005, 23 Pyle et al. 2009). By flying south post-breeding and stopping en-route (to wintering grounds) to 6 1 molt, these species are able to temporally separate energetically demanding events. As an 2 example, at the end of the breeding season, Bullock’s Orioles (Icterus bullockii) migrate to 3 northwest Mexico/southwest United States (Pillar et al. 2016), where they take advantage of the 4 high insect and fruit abundance that coincides with late summer rainstorms in this region 5 (Rohwer and Manning 1990). As the current increase in tracking studies using geolocators and 6 miniaturized GPS devices continues (e.g., Black-headed Grosbeak P. melanocephalus; Siegel et 7 al. 2016), it is quite likely we will discover additional species using this strategy. Though molt in 8 the Mexican monsoon region is strongly biased towards western migratory birds, geolocators 9 revealed that Painted Buntings breeding in Oklahoma, USA travel westward several hundred 10 kilometers to the Mexican monsoon region to molt prior to traveling southeast to overwinter in 11 southern Mexico (Contina et al. 2013). A similar system appears to occur in Trans-Saharan 12 migrants, whereby flight-feather molt is delayed until arrival at stopover sights south of the 13 Saharra, and birds arrive in freshly molted plumage at overwintering sites (e.g., European reed 14 warbler Acrocephalus scirpaceus, Dowsett-Lemaire and Dowsett 1986; great reed warbler 15 Acrocephalus arundinaceus, Hedenstrӧm et al. 1993). 16 As noted above, food limitation likely plays a critical role in determining when and 17 where molt occurs (Jenni and Winkler 1994). Thus, in addition to moving latitudinally to a 18 stopover site to molt, movement may also involve changing elevations post-breeding, most 19 commonly in the form of upslope movements to, cooler, moister habitat to complete molt. This 20 appears to be the case for migrants in western North America, including Orange-crowned 21 Warbler (Vermivora celata; Steele and McCormick 1995), Townsend’s Warbler (Setophaga 22 townsendi; Leu and Thompson 2002), Hermit Warbler (S. occidentalis; Pearson 1997), Cassin’s 23 Vireo (Vireo cassinii; Rohwer et al. 2008), and Wilson’s Warbler (Cardellina pusilla; Wiegardt 7 1 et al. 2017a). Recent work suggests that this strategy may be much more common and complex 2 than previously appreciated (Wiegardt et al. 2017b). Consistent with Leu and Thompson (2002), 3 we consider these movements to be migratory, and thus the time spent at the molting location a 4 “stopover” or “staging” event, given that birds are moving to meet an energetic challenge prior to 5 further movement (Warnock 2010). 6 7 Flight-feather molt bridging post-breeding and migration 8 Many waterfowl are rendered flightless during molt and must rely on stopover/staging areas, 9 bearing the risks associated with flightlessness. By contrast, in other taxa (e.g., passerines), there 10 is no flightless period, but the molt period is generally associated with secretive behavior and 11 reduced activity to minimize energy expenditure and predation risk (Newton 1966). Yet, perhaps 12 due to energetic costs of molt, some species appear to avoid overlap of active migratory 13 movement with molt by suspending molt begun near the breeding grounds until arrival at 14 stopover sites or the wintering grounds (e.g., Loggerhead Shrike Lanius ludovicianus; Pérez and 15 Hobson 2006). In addition, migration with gaps in the wing due to remex moult may have 16 negative impacts on flight performance (e.g. Hedenström and Sunada 1999). Yet, despite the 17 apparent risks/energetic costs of continuing to molt while actively migrating, some species of 18 Neotropic-Nearctic passerines (Northern Rough-winged Swallow Stelgidopteryx serripennis, 19 Purple Martin Progne subis, Tree Swallow Tachycineta bicolor, Swainson’s Thrush C. ustulatus, 20 Red-eyed Vireo Vireo olivaceus, Yellow Warbler, Rose-breasted Grosbeak Pheucticus 21 ludovicianus, American Redsart S.ruticilla [but see Reudink et al. 2009]; Table 2 in Leu and 22 Thompson 2002) may molt flight feathers during active migration, combining two highly 23 energy-intensive events. If food resources are limited post-breeding, but flight feather 8 1 replacement is critical for flight performance during migration (e.g., crossing the Gulf of Mexico 2 or long-distance flights over the Atlantic Ocean), selection may favor a strategy whereby 3 individuals molt throughout migration. Furthermore, extremely protracted molts may 4 necessitate/facilitate molting during active migration. Such appears to be the case in Families 5 Accipitridae and Falconidae, where flight-feather molt can take as long as 4-8 months (e.g., 6 Peregrine Falcon Falco peregrinus White et al 2002; Sharp-shinned Hawk Accipiter striatus 7 Bildstein and Meyer 2000). 8 We wish to highlight here that the distinction between suspended and continuous molt 9 bridging migration is a difficult one to document in many cases. For instance, Swainson’s Hawk 10 (Buteo swainsoni) had been assumed to molt flight feathers continuously during migration (e.g., 11 Palmer 1988). However, surveys of large capacity roost sites on the migration route failed to find 12 evidence of this in the form of dropped feathers (Smith 1980; Bechard and Weidensaul 2005), 13 and thus there appears to be support for a suspended molt (Bechard et al. 2010). Tree swallows 14 appear to begin molt following departure from breeding areas and complete molt late in 15 migration, but are assumed to molt continuously, as opposed suspending molt until arrival 16 stopover sites, without direct evidence (Stutchbury and Rowher 1990). Further, although stable 17 isotopes present a valuable technique to examine the distinction between continuous and 18 stopover moult, the coarse nature of isoscapes is problematic. For instance, in Loggerhead Shrike 19 intermediate stable-hydrogen isotope ratios provided evidence that some individuals continued to 20 molt on migration. Yet, for those shrikes that suspended molt, the isotope values for feathers 21 molted south of breeding areas overlapped both possible stopover sites and wintering areas 22 (Perez and Hobson 2006). 23 9 1 EXAMPLES OF MOLT-MIGRATION, NOT PREVIOUSLY DEFINED AS SUCH 2 Although focus on molt-migration as a process has increased over the last decades (Figure 1), 3 research utilizing the term has generally only dealt with one portion of the molt cycle: post- 4 breeding flight-feather (remex) molt (prebasic molt; Howell et al. 2003, Pyle 2005). The reason 5 for this is likely that flight-feather molt is easier to document than body-feather molt and is of 6 great importance, as it directly impacts flight performance (e.g., Tucker 1991, Swaddle et al. 7 1996). However, molt of body feathers is also critical to avian life history, as body feathers play 8 important roles in communication (e.g., Hill 2006; Senar 2006) and thermoregulation (e.g., 9 Vézina et al. 2009). Although the term molt-migration is not applied, there are several examples 10 of body-feather molts (Humphrey and Parks 1959; Howell et al. 2003) occurring during the 11 migration stage. In this section we review several types of molt that can overlap with migration 12 which have not traditionally been considered in previous definitions of molt-migration, but 13 would fit our revised definition. 14 15 Prealternate and staged prebasic molts 16 There are extensive examples of shorebirds completing their prealternate molt during migratory 17 stopover (Lourenço and Piersma 2015). For instance, after spending a relatively brief time in 18 basic plumage, adults of the subspecies of Red Knot Calidris canutus rufa begin prealternate 19 molt in February, just prior to departing wintering sites in extreme southern South America 20 (Buehler and Piersma 2008). They then complete this molt into their breeding plumage while 21 staging on the mid-Atlantic coast of the United States, prior to migrating to breeding locations in 22 the Canadian arctic (Buehler and Piersma 2008). In reviewing this phenomenon across shorebird 23 species, Lourenço and Piersma (2015) suggest that this may be a byproduct of migration distance 10 1 and time, such that birds minimize feather age and wear during mate acquisition on arrival to the 2 breeding grounds. Furthermore, in addition to the long recognized molt-migration during the 3 prebasic remex molt in waterfowl (Salomonsen 1968), many species are known to molt contour 4 feathers during migration in both spring and fall (Pyle 2005). This includes both the autumnal 5 portion of the prebasic molt into colorful nuptial plumages, and the prealternate molt of some 6 females into cryptic breeding season plumage (based on updated terminology; Pyle 2005). For 7 instance, Northern Shoveler (Anas clypeata) males initiate the contour feather portion of their 8 prebasic molt while at post-breeding molting grounds, but either continue this complete contour 9 feather molt during fall migration or suspend it until after migration is completed, with some 10 birds still molting as late as November (DuBowy 1980 & 1986). Northern Pintail (A. acuta) 11 follow a similar pattern, with most of the contour feather molt delayed until after flight-feather 12 replacement peaking in October and continuing into the winter in some cases (Clark et al. 2014) 13 and Long-tailed Ducks (Clangula hyemalis) also appear to continuously molt during migration 14 (Payne et al. 2015). 15 Recently, there are indications that this phenomenon occurs in other families as well. 16 Most studies of passerine prealternate molt have documented, or assume, that this molt occurs on 17 the wintering grounds, prior to departure for spring migration (e.g., Boone et al. 2010, Mowbray 18 1997, Mazerolle et al. 2005, Bulluck et al. 2016). However, few studies have directly addressed 19 questions about the spatial variation in prealternate songbird molt. At least one recent study 20 documented an obligate partial prealternate molt, completed during spring stopover, in Rusty 21 Blackbirds (Euphagus carolinus; Wright et al. 2018). In that case, much like many shorebirds, 22 individuals begin prealternate molt prior to leaving wintering grounds (Mettke-Hoffman et al. 23 2010), but molt peaks in the middle of the stopover period. Molt was negatively associated with 11 1 fat score in this study, potentially indicating it is antagonistic with migratory fattening and a limit 2 on migration phenology (Wright et al. 2018). Furthermore, another recent study has found 3 evidence of a definitive prealternate molt during migratory stopover in Rufous Hummingbird 4 (Selasphorus rufus; Sieburth and Pyle 2018). At this time, it is unclear how many other taxa 5 follow such a pattern, as there is scant treatment in the primary literature. For instance, in 6 songbirds, Indigo Bunting (Passerina cyanea) first prealternate molt appears to begin on the 7 wintering grounds, but often complete on breeding sites (Pyle 1997), but it is not clear that 8 definitive prealternate molt follows the same pattern. There is a great need for further 9 documentation and quantification of such examples to determine how widespread prealternate 10 molt-migration is in most families. 11 12 Presupplemental molts 13 Several shorebird species complete presupplemental molts during migratory stopover/staging. In 14 these cases, feathers already replaced on wintering sites in a prealternate molt are replaced again 15 (Humphrey and Parks 1959; Howell et al. 2003) during migratory stopover. Such appears to be 16 the case in shorebird species that stage in east Asia, such as the Great Knot (C. tenuirostris; 17 Battley et al. 2006) and Ruff (Philomachus pugnax; Jukema & Piersma 2000), and the Bar-tailed 18 Godwit (Limosa lapponica) which stages in western Europe. In the case of the godwit, for 19 example, birds in better condition re-molt contour feathers in the Netherlands, and appear to 20 enhance the quality of their plumage prior to arrival at breeding sites (Piersma and Jukema 1993, 21 Piersma et al. 2001). In addition, there are some indications that Anas ducks have inserted 22 presupplemental molts during their spring migration (Pyle 2013b). 23 12 1 Preformative molt 2 In addition to the definitive molts discussed above, juvenile birds can undergo partial 3 preformative molts during their first fall migration. This has especially been observed in 4 songbirds stopping over in the Mexican monsoon region (e.g., Butler et al. 2002, Pyle et al. 5 2009). These molts can include eccentric flight-feather molts, such as those in Western 6 Kingbirds (Tyrannus verticalis), where juveniles will replace some primaries, but delay this molt 7 until stopover (Barry et al. 2009). Many of these molts however are contour feather only molts, 8 such as those completed by first-year Warbling Vireos (Vireo gilvus), Western Tanagers 9 (Piranga ludoviciana), and other species (Pyle et al. 2009). Preformative molts on stopover have 10 also been observed in Europe in the European Starling (Sturnus vulgaris), where juveniles, but 11 not adults, delay molt until migration (Svardson 1953, Kosarev 1999). 12 13 A CALL FOR STANDARDIZING THE CLASSIFICATION OF MOLT-MIGRATION 14 ACROSS ORNITHOLOGY 15 We argue that the diversity of systems in which molt occurs during migration discussed above 16 should all be classified as “molt-migration”. However, in addition to our broadened definition of 17 the overall phenomenon, this diversity in the timing and extent of the molts involved requires a 18 more specific system of terminology to describe each case that is broadly applicable across 19 systems. We could utilize an entirely spatial system, based on where molt begins and where it 20 ends, relative to stationary phases of the annual cycle. However this would produce a complex 21 system with numerous permutations, even without also including further classification levels for 22 describing which feather tracts are involved. Further, a system based on the type of migration 23 system may be useful (e.g., boreal, austral, altitudinal), however we sought to generate a system 13 1 that would be broadly applicable across all types of migration. Thus, we propose a relatively 2 simple, two-tiered system that classifies 1) when/where molt commences relative to migration, 3 and 2) what type of molt is involved. We expect that although not every single one of the diverse 4 molt strategies globally will fit these definitions, this typology can be applied to the vast majority 5 of variants on the theme of molt-migration in birds. For this classification, we consider the 6 migration stage to have begun once an individual leaves their breeding or stationary non- 7 breeding site, moving to a new landscape (i.e. change in latitude, longitude, altitude). However, 8 we exclude post-breeding movements within the same landscape as the stationary life history 9 stage (e.g., post-fledging movements into adjacent habitats; Vitz and Rodewald 2010). In the 10 first tier of our typology, we classify the following two categories based on when/where molt 11 commences relative to migration: 12 a. Continuous molt-migration - Molt that is initiated on the breeding or stationary non- 13 breeding grounds then continues during migration until stopover or arrival at breeding or 14 stationary non-breeding grounds. 15 b. Suspended molt-migration - Molt that is initiated on the breeding or stationary non- 16 breeding grounds then is interrupted following migratory departure until stopover or 17 arrival at breeding/stationary non-breeding grounds. 18 c. Stopover molt-migration – Molt that is delayed until arrival at specific molting grounds 19 (e.g., high elevation or stopover site, Salomonsen 1968; Leu and Thompson 2002) and 20 completed prior to further progression of migration or at the migration endpoint. 21 In the second tier of our typology, we classify the following three categories based on the type of 22 molt: 14 1 a. Prebasic molt-migration - Definitive molt, involving sequential, simultaneous, or staged 2 replacement of all primaries and contour feathers, occurring in an area distinct in latitude, 3 longitude, and/or elevation from the breeding or stationary non-breeding site. 4 b. Prealternate or presupplemental molt-migration - Definitive partial molts, involving 5 contour feathers, prior to or during (e.g., male waterfowl) the breeding season occurring in 6 an area distinct in latitude, longitude, and/or elevation from the breeding or stationary non- 7 breeding site. These may include replacement of basic or formative feathers (prealternate 8 molt), or replacement of alternate feathers (presupplemental molt). 9 c. Preformative molt-migration - Molts that may be complete, contour feather only, or 10 include eccentric flight-feather molt, occurring post-juvenile dispersal in an area distinct 11 in latitude, longitude and/or elevation from the natal site. 12 13 In Table 1, we provide multiple examples of the application of this typology. It should be noted 14 that, in some cases, a researcher may not have enough information to classify a system at both 15 levels. For instance, one may know a molt observed at a stopover site is “preformative molt- 16 migration”, but may not have clear evidence for whether the molt began prior to departure from a 17 breeding site, or initiated during migration. In these cases we advocate for a partial application of 18 our typology (i.e. simply, preformative molt-migration). 19 20 IMPLICATIONS OF MOLT-MIGRATION AND FUTURE DIRECTIONS 21 Although not exhaustive, the above examples highlight the prevalence of overlap between the 22 molt and migratory life history stages (Ramenofsky and Wingfield 2006), and thus the need to 23 synthesize different systems to elucidate evolutionary and ecological implications. A clear 15 1 understanding of the ecology of migratory birds is dependent on a full annual cycle approach 2 (Marra et al. 2015), which is currently limited by a lack of knowledge about the spatiotemporal 3 aspects of many stages. Molt is primary among these life history stages, though the studies 4 highlighted above exhibit an increasing appreciation for variation among species, and 5 individuals, in where and when molt is completed. As recognized in previous reviews and 6 syntheses on molt-migration (e.g., Jehl 1990, Leu and Thompson 2002, Pyle et al. 2009), without 7 a clear understanding of where and when molt is completed we cannot understand how this 8 critical life history stage is limited. For instance, in terms of flight performance, determining 9 what resources limit flight feather growth rate and feather quality (e.g., de la Hera et al. 2009). 10 Here, we have sought to expand this critical point to include all stages and types of molt in order 11 to move towards a comprehensive understanding of the ecological and evolutionary implications 12 of overlap with the migratory life history stage. This includes determining, in terms of contour 13 feather molts, where critical pigments in intraspecific interactions (e.g., Sparrow et al. 2017) and 14 ectoparasite resistance (e.g., Gunderson et al. 2008) are acquired. With a more comprehensive 15 typology, which can be utilized across all avian taxa, researchers can now classify molt- 16 migration strategies in a common language. The vast array of different molt (Howell et al. 2003) 17 and migration (Salewski and Bruderer 2007) strategies in birds appear to have evolved many 18 times independently, suggesting that common ecological or life history characteristics may drive 19 the evolution of molt strategies, including molt-migration. Phylogenies for birds (e.g., Prum et al. 20 2015) will be instrumental for conducting large-scale phylogenetic reconstructions of molt 21 strategies and phylogenetically-controlled analyses aimed at understanding which ecological, 22 behavioral, or life history traits promote the evolution of different molt strategies. In a 23 conservation sense, whereas molting and staging areas have long received attention as critical 16 1 habitat (reviewed in Jehl 1990, Leu and Thompson 2002), increasing documentation of molt- 2 migration in its myriad forms will require a similar recognition in other migratory taxa. 3 In conclusion, the focus on molt-migration is likely to continue growing, and we hope to 4 see many exciting avenues of research explored to understand these systems. Important 5 questions, recognized by other researchers on this topic (e.g., Leu and Thompson 2002), still 6 remain and are critical to unraveling how molt-migration arises and is maintained as a strategy. 7 For instance, although energetics is likely a driver of many strategies, the nutritional advantages 8 of molting on migration or stopover are not well described for most species. This is likely of 9 great importance, particularly in understanding individual variation in the saptiotemporal aspects 10 of molt (Piersma et al. 2001, Tonra et al. 2015, Nordell et al. 2016). Equally important, is 11 understanding the proximal physiological mechanisms regulating the overlap in molt and 12 migration stages. This is especially true given apparent physiological conflicts between these two 13 energetically expensive and physically challenging states that involve substantial physiological 14 changes (Williams 2012). In order to reach the increasingly prevalent goal of unravelling the full 15 annual cycle ecology of species (Marra et al. 2015), we must continue to explore phenomena 16 such as molt-migration, and other seasonal interactions (Marra et al. 1998, Harrison et al. 2011). 17 This will require a comprehensive focus on the stages of the annual cycle involved and 18 describing them in the same terms across avian systems. 19 20 ACKNOWLEDGEMENTS 21 We are extremely grateful to Peter Pyle for providing feedback and comments, clarification of 22 terminology, and support in the preparation of this manuscript. In addition, we wish to thank 23 three anonymous reviewers whose comments greatly improved the manuscript. Elizabeth Ames, 17 1 Alicia Brunner, Kristie Stein, and Jay Wright also provided valuable feedback on the issues 2 discussed in this commentary and the typology. 3 Funding statement: This research was made possible by funding from the Ohio Agricultural 4 Research and Developmental Center to Tonra and a Natural Sciences and Engineering Research 5 Council of Canada Discovery Grant to Reudink. 6 Author contributions: Tonra: initially proposed the idea of this commentary and contributed to 7 writing, conceptualization of the typology, and editing. Reudink: contributed to writing, 8 conceptualization of the typology, and editing. 9 10 LITERATURE CITED 11 Battley, P.F., D.I. Rogers, and C.J. Hassell (2006). Prebreeding moult, plumage and evidence for 12 a presupplemental moult in the Great Knot Calidris tenuirostris. Ibis 148:27-38. 13 Barry, J.H., L.K.Butler, S. Rohwer, and V.G. Rohwer (2009). Documenting molt-migration in 14 Western Kingbird (Tyrannus verticalis) using two measures of collecting effort. The Auk 15 126:260-267. 16 Bechard, M.J. and Weidensaul, C.S., 2005. Feather molt by Swainson’s hawks (Buteo swainsoni) 17 on the Austral grounds of Argentina. Ornitología Neotropical, 16, pp.267-270. 18 Bechard, M. J., C. S. Houston, J. H. Saransola and A. Sidney-England. 2010. Swainson's Hawk 19 (Buteo swainsoni). In The Birds of North America (P. G. Rodewald, Ed.). Cornell Lab of 20 Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bna.265 21 Bildstein, K. L. and K. D. Meyer (2000). Sharp-shinned Hawk (Accipiter striatus). In The Birds 22 of North America (P. G. Rodewald, Ed.). Cornell Lab of Ornithology, Ithaca, New York, 23 USA. https://doi.org/10.2173/bna.482 18 1 Boone, A. T., P. G. Rodewald, and L. W. DeGroote (2010). Neotropical winter habitat of the 2 magnolia warbler: effects on molt, energetic condition, migration timing, and 3 hematozoan infection during spring migration. The Condor 112:115-122. 4 Bulluck, L. P., M. J. Foster, S. Kay, D. E. Cox, C. Viverette, and S. Huber (2016). Feather 5 carotenoid content is correlated with reproductive success and provisioning rate in female 6 Prothonotary Warblers. The Auk 134:229-239. 7 Butler, L.K., M.G. Donahue, and S. Rohwer (2002). Molt-migration in Western Tanagers 8 (Piranga ludoviciana): Age effects, aerodynamics, and conservation implications. The 9 Auk 119:1010-1023. 10 Buehler, D.M. and T. Piersma (2008). Travelling on a budget: predictions and ecological 11 evidence for bottlenecks in the annual cycle of long-distance migrants. Philosophical 12 Transactions of the Royal Society of London B: Biological Sciences 363:247-266. 13 Clark, R.G., J.P. Fleskes, K.L. Guyn, D. A. Haukos, J. E. Austin and M.R. Miller (2014). 14 Northern Pintail (Anas acuta). In The Birds of North America (P. G. Rodewald, Ed.). 15 Cornell Lab of Ornithology, Ithaca, NY, USA. doi: 10.2173/bna.163 16 Contina, A., E.S. Bridge, N.E. Seavy, J.M. Duckles, and J.F. Kelly (2013). Using geologgers to 17 investigate bimodal isotope patterns in Painted Buntings (Passerina ciris). The Auk 18 130:265-272. 19 de la Hera, I., J. Perez-Tris, and J.L. Telleria (2009). Migratory behaviour affects the trade-off 20 between feather growth rate and feather quality in a passerine bird. Biological Journal of 21 the Linnean Society 97:98-105. 22 23 Dietz, M. W., S. Daan, and D. Masman (1992). Energy requirements for molt in the kestrel Falco tinnunculus. Physiological Zoology 65:1217-1235. 19 1 Dickson, R.D., D. Esler, J.W. Hupp, E.M. Anderson, J.R. Evenson, and J. Barrett (2012). 2 Phenology and duration of remigial molt in Surf Scoters (Melanitta perspicillata) and 3 White-winged Scoters (Melanitta fusca) on the Pacific coast of North America. Canadian 4 Journal of Zoology 90:932-944. 5 6 7 Dowsett-Lemaire, F. and Dowsett, R.J. (1987). European Reed and Marsh Warblers in Africa: migration patterns, moult and habitat. Ostrich, 58:65-85. DuBowy, P. J. (1980). Optimal foraging and adaptive strategies of postbreeding male Blue- 8 winged Teal and Northern Shovelers. Master's Thesis, Univ. of N. Dakota, Grand Forks. 9 Dubowy, Paul J. (1996). Northern Shoveler (Anas clypeata). In The Birds of North America (P. 10 G. Rodewald, Ed.). Cornell Lab of Ornithology, Ithaca, NY, USA. doi: 10.2173/bna.217 11 Froehlich, D.R., S. Rohwer, and B. J. Stutchbury (2005). Spring molt constraints versus winter 12 territoriality. Pages 321-335 in Birds of Two Worlds: The Ecology and Evolution of 13 Migration (R. Greenberg and P.P. Marra, Eds.). Johns Hopkins University Press, 14 Baltimore, Maryland. 15 Gunderson, A. R., A. M. Frame, J. P. Swaddle, and M. H. Forsyth (2008). Resistance of 16 melanized feathers to bacterial degradation: is it really so black and white? Journal of 17 Avian Biology 39:539-545. 18 Harrison, X.A., J.D. Blount, R. Inger, D.R. Norris, and S. Bearhop (2011). Carry-over effects as 19 drivers of fitness differences in animals. Journal of Animal Ecology 80:4-18. 20 Hedenström, A., S. Bensch, D. Hasselquist, M. Lockwood, and U. Ottosson (1993). Migration, 21 stopover and moult of the great reed warbler Acrocephalus arundinaceus in Ghana, West 22 Africa. Ibis, 135:177-180. 20 1 2 Hedenström, A., and Sunada (1999). On the aerodynamics of moult gaps in birds. Journal of Experimental Biology 202:67-76. 3 Hill, G. E. (2006). Female mate choice for ornamental coloration in birds. In: Bird Coloration: 4 Function and Evolution. Hill, G.E. and K. J. McGraw (eds). Harvard University Press 5 University Press, Cambridge, MA. 6 Hohman, W.L., C.D. Ankney, and D.H. Gordon (1992). Ecology and management of 7 postbreeding waterfowl. Pages 128-189 in Ecology and Management of Breeding 8 Waterfowl (B.D.J. Batt, et al., Eds.). University of Minnesota Press, Minneapolis, MN. 9 Howell, S.N.G., C. Corben, P. Pyle, and D. I. Rogers. (2003). The first basic problem: a review 10 11 12 13 14 15 16 17 18 19 of molt and plumage homologies. The Condor 105:635-653. Humphrey, P. S., and K. C. Parkes. (1959). An approach to the study of molts and plumages. The Auk 76:1-31. Jehl Jr., J. R. (1987). Moult and moult migration in a transequatorially migrating shorebird: Wilson's Phalarope. Ornis Scandinavica:173-178. Jehl Jr.,, J. R., (1990). Aspects of the molt migration. Pages 102-113 in Bird Migration. Springer, Berlin, Heidelberg. Jenni, L. and W. Winkler (1994). Moult and ageing European Passerines. Academic Press, London, UK. Jukema, J. and T. Piersma (2000). Contour feather moult of Ruffs Philomachus pugnax during 20 northward migration, with notes on homology of nuptial plumages in scolopacid waders. 21 Ibis 142:289-296. 22 23 Kosarev, V. (1999). Summer movements of starlings Sturnus vulgaris on the Courish Spit of the Baltic Sea. Avian Ecology and Behaviour 30:99-109. 21 1 Leu, M. and C.W. Thompson (2002). The potential importance of migratory stopover sites as 2 flight-feather molt staging areas: a review for Neotropical migrants. Biological 3 Conservation 106:45-56. 4 Lourenço, P.M. and T. Piersma (2015). Migration distance and breeding latitude correlate with 5 the scheduling of pre-alternate body moult: a comparison among migratory waders. 6 Journal of Ornithology 156:657-665. 7 8 9 10 Marra, P.P., Hobson, K.A. and Holmes, R.T. (1998). Linking winter and summer events in a migratory bird by using stable-carbon isotopes. Science 282:1884-1886. Marra, P.P., E. B. Cohen, S. R. Loss, J. E. Rutter and C. M. Tonra (2015). A call for full annual cycle research in animal ecology. Biology Letters 11:20150552. 11 Mazerolle, D. F., K. A. Hobson, and L. I. Wassenaar (2005). Stable isotope and band-encounter 12 analyses delineate migratory patterns and catchment areas of white-throated sparrows at a 13 migration monitoring station. Oecologia 144:541-549. 14 Mettke-Hofmann, C., P. B. Hamel, G. Hofmann, T. J. Zenzal Jr., A. Pellegrini, J. Malpass, M. 15 Garfinkel, N. Schiff, and R. Greenberg. (2015) Competition and habitat quality influence 16 age and sex distribution in wintering rusty blackbirds. PloS One 10: e0123775. 17 Mowbray, T. B.(1997). Swamp Sparrow (Melospiza georgiana), In The Birds of North America 18 (P. G. Rodewald, Ed.). Cornell Lab of Ornithology, Ithica, NY, USA. doi: 19 10.2173/bna.279 20 Newton, I. (1966). The moult of the Bullfinch Pyrrhula pyrrhula. Ibis 108: 41-67. 21 Nordell, C.J., S. Haché, E.M. Bayne, P. Sólymos, K.R. Foster, C.M. Godwin, R. Krikun, P.P. 22 Pyle and K.A. Hobson (2016). Within-site variation in feather stable hydrogen isotope 22 1 (δ2hf) values of boreal songbirds: implications for assignment to molt origin. PloS One 2 11:e0163957. 3 Ogilvie, L.W. (1978) Wild Geese. Poyster, Berkhamsted, UK 4 Palmer, R. S. (1988). Handbook of North American birds. Volume 5: diurnal raptors (Part 2). 5 6 Yale Univ. Press, New Haven, Connecticut. Payne, A. M., M. L. Schummer, and S. A. Petrie (2015). Patterns of molt in Long-tailed Ducks 7 (Clangula hyemalis) during autumn and winter in the Great Lakes Region, Canada. 8 Waterbirds, 38:195-200. 9 Pearson, S.F. (2013). Hermit Warbler (Setophaga occidentalis). In The Birds of North America 10 (P. G. Rodewald, Ed.). Cornell Lab of Ornithology, Ithaca, NY, USA. doi: 11 10.2173/bna.303 12 Pérez, G.E. and K. A. Hobson (2006). Isotopic evaluation of interrupted molt in northern 13 breeding populations of the Loggerhead Shrike. The Condor, 108:877-886. 14 Piersma, T. (1988). The annual moult cycle of Great Crested Grebes. Ardea 76:82-95. 15 Piersma, T., and J. Jukema (1993). Red breasts as honest signals of migratory quality in a long- 16 distance migrant, the Bar-tailed Godwit. The Condor 95:163-177. 17 Piersma, T., L. Mendes, J. Hennekens, S. Ratiarison, S. Groenewold, and J. Jukema (2001). 18 Breeding plumage honestly signals likelihood of tapeworm infestation in females of a 19 long-distance migrating shorebird, the bar-tailed godwit. Zoology 104:41-48. 20 Pillar, A.G., P. P. Marra, N. J. Flood, and M. W. Reudink (2016). Moult migration in Bullock’s 21 Orioles (Icterus bullockii) confirmed by geolocators and stable isotope analysis. Journal 22 of Ornithology 157:265-275. 23 1 Prum, R.O., J.S. Berv, A. Dornburg, D.J. Field, J.P. Townsend, E.M. Lemmon and A.R. 2 Lemmon (2015). A comprehensive phylogeny of birds (Aves) using targeted next- 3 generation DNA sequencing. Nature 526:569-573. 4 5 Pyle, P. (1997). Identification Guide to North American Birds, Part I. Slate Creek Press, Point Reyes Station, California. 6 Pyle, P. (2005). Molts and plumages of ducks. Waterbirds 28: 208-219. 7 Pyle, P. (2013a). Evolutionary implications of synapomorphic wing-molt sequences among 8 9 10 falcons (Falconidae) and parrots (Psittaciformes). Condor 115:593-602. Pyle, P. (2013b). Molt homologies in ducks and other birds: A response to Hawkins (2011) and further thoughts on molt terminology in ducks. Waterbirds, 36:77-81. 11 Pyle, P., W. A. Leitner, L. Lozano-Angulo, F. Avilez-Teran, H. Swanson, E. G. Limón and M. 12 K. Chambers (2009). Temporal, spatial, and annual variation in the occurrence of molt- 13 migrant passerines in the Mexican monsoon region. The Condor 111:583-590. 14 Ramenofsky, M. and J.C. Wingfield (2006). Behavioral and physiological conflicts in migrants: 15 16 the transition between migration and breeding. Journal of Ornithology, 147:135-145. Reudink, M.W., P.P. Marra, K.M. Langin, C.E. Studds, T.K. Kyser, and L.M. Ratcliffe (2008). 17 Molt-migration in the American Redstart (Setophaga ruticilla) revisited: explaining 18 variation in feather δD signatures. The Auk 125:744-748. 19 Robert, M., R. Benoit, and J.P.L. Savard (2002). Relationship among breeding, molting, and 20 wintering areas of male Barrow's Goldeneyes (Bucephala islandica) in eastern North 21 America. The Auk 119:676-684. 24 1 Rohwer, S. and J. Manning (1990). Differences in timing and number of molts for Baltimore and 2 Bullock's orioles: Implications to hybrid fitness and theories of delayed plumage 3 maturation. The Condor 92:125-140. 4 Rohwer, S., L. K. Butler, D. Froehlich (2005). Ecology and demography of east–west differences 5 in molt scheduling of neotropical migrant passerines" Pages 87-105 in Birds of Two 6 Worlds: The Ecology and Evolution of Migration (R. Greenberg and P.P. Marra, Eds.). 7 Johns Hopkins University Press, Baltimore, Maryland. 8 9 10 11 Rohwer, V.G., S. Rohwer, and J.H. Barry (2008). Molt scheduling of western Neotropical migrants and up-slope movement of Cassin's Vireo. The Condor 110:365-370. Salewski, V., and B. Bruderer (2007). The evolution of bird migration—a synthesis. Naturwissenschaften, 94:268-279. 12 Salomonsen, F. (1968). The moult migration. Wildfowl 19:5-24. 13 Senar, J.C. (2006). Bird colors as intrasexual signals of aggression and dominance. In: Bird 14 Coloration: Function and Evolution. Hill, G.E. and K. J. McGraw (eds). Harvard 15 University Press University Press, Cambridge, MA. 16 Sieburth, D., and P. Pyle. 2018. Evidence for a prealternate molt-migration in the Rufous 17 Hummingbird and its implications for the evolution of molts in Apodiformes. The Auk: 18 Ornithological Advances 135:495-505. 19 Siegel, R.B., R. Taylor, J. F. Saracco, L. Helton, and S. Stock (2016). GPS tracking reveals non- 20 breeding locations and apparent molt migration of a Black-headed Grosbeak. Journal of 21 Field Ornithology 87:196-203. 25 1 Smith, N. G. (1980). Hawk and vulture migrations in the Neotropics. Pages 51–65 in Migrant 2 birds in the Neotropics: ecology, behavior, distribution, and conservation (A. Keast and 3 E. S. Morton, Eds.). Smithsonian Institution Press, Washington, DC. 4 Sparrow, K.L., K.K. Donkor, N.J. Flood, P.P. Marra, A.G. Pillar and M.W. Reudink (2017). 5 Conditions on the Mexican moulting grounds influence feather colour and carotenoids in 6 Bullock's orioles (Icterus bullockii). Ecology and Evolution 7:2643-2651. 7 Steele, J. and McCormick, J. (1995). Partitioning of the summer grounds by Orange-crowned 8 Warblers into a breeding grounds, adult molting grounds and juvenile staging areas. 9 North American Bird Bander, 20:152. 10 11 12 13 Storer, R.W. and J.R. Jehl Jr. (1985). Moult patterns and moult migration in the Black-necked Grebe Podiceps nigricollis. Ornis Scandinavica 16:253-260. Stutchbury, B.J. and S. Rohwer (1990). Molt patterns in the Tree Swallow (Tachycineta bicolor). Canadian Journal of Zoology, 68:1468-1472 14 Svardson, G. (1953) Visible migration within Fenno-Scandia. Ibis 95:181-211. 15 Swaddle, J.P., M.S. Witter, I.C. Cuthill, A. Budden, and P. McCowen (1996). Plumage condition 16 affects flight performance in common starlings: implications for developmental 17 homeostasis, abrasion and moult. Journal of Avian Biology 27:103-111. 18 Tonra, C.M., C. Both and P.P. Marra (2015). Incorporating site and year-specific deuterium 19 ratios (δ2H) from precipitation into geographic assignments of a migratory bird. Journal 20 of Avian Biology 46:266-274. 21 22 Tucker, V. A. (1991). The effect of molting on the gliding performance of a Harris' Hawk (Parabuteo unicinctus). The Auk 108:108-113. 26 1 Vézina, F., A. Gustowska, K. M. Jalvingh, O. Chastel, and T. Piersma (2009). Hormonal 2 correlates and thermoregulatory consequences of molting on metabolic rate in a northerly 3 wintering shorebird. Physiological and Biochemical Zoology, 82:129-142. 4 Vitz, A.C. and A. D. Rodewald (2010). Movements of fledgling Ovenbirds (Seiurus aurocapilla) 5 and Worm-eating Warblers (Helmitheros vermivorum) within and beyond the natal home 6 range. The Auk 127:364-371. 7 8 9 Warnock, N. (2010). Stopping vs. staging: the difference between a hop and a jump. Journal of Avian Biology 41:621–626. White, Clayton M., N. J. Clum, T. J. Cade and W. Grainger-Hunt (2002). Peregrine Falcon 10 (Falco peregrinus), In the Birds of North America (P. G. Rodewald, Ed.). Cornell Lab of 11 Ornithology, Ithica, NY, USA. https://doi.org/10.2173/bna.660 12 Wiegardt, A.K., D.C. Barton, and J.D. Wolfe (2017a). Post-breeding population dynamics 13 indicate upslope molt-migration by Wilson's Warblers. Journal of Field Ornithology 14 88:47-52. 15 Wiegardt, A., J. Wolfe, C.J. Ralph, J.L. Stephens, and J. Alexander (2017b). Postbreeding 16 elevational movements of western songbirds in Northern California and Southern 17 Oregon. Ecology and Evolution doi:10.1002/ece3.3326 18 Williams, T. D. (2012). Hormones, life-history, and phenotypic variation: opportunities in 19 evolutionary avian endocrinology. General and Comparative Endocrinology, 176:286- 20 295. 21 Wingfield, J.C. (2008). Organization of vertebrate annual cycles: implications for control 22 mechanisms. Philisophical Transactions of the Royal Society: Biological Sciences 23 363:425-441. 27 1 Wright, J. R., C. M. Tonra, and L. L. Powell (2018). Prealternate molt-migration in Rusty 2 Blackbirds and its implications for stopover biology. Condor: Ornithological 3 Applications 120:507-516. 4 5 Yuri, T. and S. Rohwer (1997). Molt and migration in the northern rough-winged swallow. The Auk 114:249–262. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 28 TABLES Table 1. Examples and applications of proposed typology to classify forms of molt migration across avian taxa. Note, that as in the case of Red Knot, the typology is still useful even when data are not available to classify in both tiers. classification adult - continuous prebasic molt-migration first-year - continuous preformative molt migration adult - stopover prebasic molt-migration first-year - stopover preformative molt migration continuous or suspended prealternate moltmigration suspended prebasic molt-migration species example molt-migration description Northern Rough-winged Swallow (Stelgidopteryx ruficollis) Eastern populations initiate complete prebasic molt and partial preformative molts on breeding grounds, stopover on the northern Gulf of Mexico and complete molt before continuing across the Gulf. However, some juveniles appear to continue molting flight feathers during trans-Gulf migration (Yuri and Rohwer 1997). Western Kingbird (Tyrannus verticalis) Adults undergo a complete prebasic molt during stopover in Mexican monsoon region/montane Southwestern U.S.A. Juveniles also appear to delay both their preformative body feather molt and eccentric flight feather molt until they partially complete their fall migration, and prior to arrival at wintering areas (Barry et al. 2009). Rufa Red Knot (Calidris canutus rufa) Birds begin molting into alternate plumage on wintering grounds in Tiera del Fuego in February, continue to molt during migration, completing molt during stopover in eastern U.S.A. (Bueler and Piersma 2008). It is not currently clear if molts are suspended or continuous. Swainson’s Hawk (Buteo swainsoni) Birds begin molting flight-feathers at breeding sites, apparently pause molt during migratiuon and complete molt at stationary non-breeding sites (Smith 1980; Bechard and Weidensaul 2005) 1 1 FIGURE CAPTIONS 2 Figure 1. Total number of publications since 1978, by 10 year intervals, using the terms “molt- 3 migration” or “moult-migration”, and “bird”. Numbers based on keyword search for terms in 4 Web of Science, March 2018. 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1