Phenological drivers of ungulate migration in South America: Characterizing the movement and seasonal habitat use of guanacos

doi:10.21203/rs.3.rs-1646244/v1

Download PDF

Research Article

Phenological drivers of ungulate migration in South America: Characterizing the movement and seasonal habitat use of guanacos

https://doi.org/10.21203/rs.3.rs-1646244/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Migration is a widespread strategy among ungulates to cope with seasonality. Phenology, especially on seasonally snow-covered landscapes featuring “white waves” of snow accumulation and “green waves” of plant green-up, is a phenomenon that many migratory ungulates navigate. Guanacos (Lama guanicoe) are native camelids to South America and are one of the last ungulates in South America that migrates. However, a detailed description of this species’ migratory attributes, including whether guanacos surf or jump phenological waves is lacking.

Methods

We quantified the migratory movements of 21 adult guanacos over three years in Patagonia, Argentina. We analyzed annual movement patterns using net squared displacement (NSD) and home range overlap and quantified snow and vegetation phenology via remotely sensed products.

Results

We found that 74% of the individual guanacos exhibited short altitudinal migration. Additionally, we observed fidelity of migratory ranges and residence time, but flexibility around migration propensity, timing and duration of migration. Interestingly, we found that the scarce vegetation and arid conditions within our study area prevented guanacos from surfing green waves; instead, guanacos appeared to avoid white waves.

Conclusions

Our study demonstrates that guanaco elevational migration is driven by a combination of vegetation availability and snow cover presence, reveals behavioral plasticity of their migration, and highlights the importance of snow phenology as a driver of ungulate migrations.

phenology

plasticity

green waves

snow cover

Lama guanicoe

GPS data

Seasonality, the temporal periodic fluctuations in climatic conditions, plays a major role in how animals interact with their environment (1). In general, seasonal habitats consist of alternating favorable and unfavorable periods; winter being a particularly challenging period characterized by low temperatures, snow cover, resource limitation and energetic deficits (2,3). In response to the constraints of winter, many species have evolved strategies to avoid winter conditions via seasonal migration (4). Migration encompasses a variety of periodical round trip movements between discrete, non-overlapping home ranges (5) and arises in systems where seasonality is predictable (6) and the benefits of alternating between ranges exceeds the energetic costs and risks associated with long distance movement (7,8).

Among ungulates, migration is a particularly important strategy to track food availability and reduce competition and predation (9,10). While some species have high fidelity in their migration habits (11,12) including well-defined corridors, ranges that are regularly returned to and relatively fixed departure dates, other species exhibit high levels of plasticity in their migratory behavior. These differences occur even within species and populations (7,13). Some species for example, are partially migratory (14), where some individuals remain resident while others migrate, or where the same individual can alternate between these two strategies. Although originally understood as a particularity of some species, partial migration is now seen as the rule among ungulates (15), it is often attributed to extrinsic factors such as precipitation and resource availability, and demographic factors such as animal density and individual age (16), and is seen as a beneficial strategy when dealing with unpredictable seasonality and heterogeneous, complex landscapes (17).

There is also growing recognition around plasticity on other migration characteristics (e.g., distance, timing, and duration) as a strategy to deal with a changing environment (15). Migration distances can range from < 15 km (18), especially common in mountainous species that exhibit short altitudinal movements (7,8,19), to > 1600 km, in species that inhabit low elevation plains and follow precipitation gradients and high quality vegetation (9,20,21). Timing and duration in migration are also flexible as ungulates respond to interannual and spatial variation in green-up, snow melt and rainfall (9,10,22). Indeed, tracking vegetation green up to maximize intake of high quality forage, or “surfing the green wave” (23), is recognized as the predominant driver among ungulate migration (24,25). Other species, however, have been found to jump green waves by accelerating their migration to arrive at summer ranges before vegetation greenness peaks, which could be a product of non-suitable habitat along the way or due to short migrations that do not allow surfing (26).

While the phenology of vegetation is a recognized driver of ungulate migrations, the importance of snow as a driver is less understood. However, several studies have found snow presence and snow characteristics to influence migration parameters. Severe winter weather and snow cover and depth have been found to be related to the onset of both fall and spring migrations (27,28). The timing of fall departure can be influenced by the trade-off between extending access to summer range foraging grounds and the risk of encountering deeper snow along the way which leads to higher energy demands related to locomotion as well as the struggles of accessing snow covered vegetation (29). Spring migration can also be affected by snow depth and snow melt dates (27), which is unsurprising given the correlation of snow presence and vegetation productivity, particularly in dry landscapes where precipitation occurs mainly as snow (30). Species that inhabit more extreme latitudes that feature longer snow seasons, typically spend more time in their winter than summer ranges and exhibit delayed migration starting dates compared to southern species (31).

The guanaco is a camelid native to South America (32). A monomorphic and social species, guanacos inhabit a wide range of habitat types (33), occupying low altitude open plains during the growing season and higher altitude, rocky areas in the non-growing season (34). Guanacos are the only known South American ungulate with migratory populations (35,36). However, previous studies of guanaco migration were based on seasonal changes in guanaco abundances or radiotracking of a small number of individuals. Thus, a detailed description of the attributes of guanaco migration is lacking. Moreover, whether migration is a plastic trait among guanacos is unknown because previous studies last ≤ 1 year. Further, the drivers of guanaco migration are also unknown, and it is unclear whether guanacos surf or jump green waves or respond to other climatic factors or resource pulses.

To discern the movement strategies and ultimate factors driving guanaco migration we studied a wild population during three migratory cycles in southern South America. We examined the spatial and temporal features that characterize this behavior, as well as the variation around them. To better understand their seasonal habitat use, we analyzed location and size of their summer and winter home ranges, as well as their movement dynamics within them. Given the elevational gradient and seasonality characteristic of our study area, we hypothesized that guanacos would be migratory and generally track favorable seasonal conditions of plant availability and snow presence. More specifically, due to the topological and phenological complexity of the area and lack of competition with domestic cattle, we predicted the population to be partially migratory along an altitudinal gradient, with summer home ranges at higher altitudes than winter ones. Given the reduced productivity during winter as well as larger social groups during this season (36), we expected guanacos to move more during this season while foraging, leading to larger winter home ranges compared to summer ones. Finally, given the scarce vegetation typical of the study area, which is composed of a combination of grasses that show vegetative growth during winter and early spring, and shrubs that concentrate phenological activity during the summer (37), we predicted green-up phenology would not be a smooth wave-like progression and prevent green-wave surfing. Consequently, we predicted that the phenology of snow cover, would become more relevant in mediating migratory characteristics.

Study Area

The study area was located in the Patagonia region of Argentina, it encompassed 131,100 ha, and included part of the recently created Patagonia National Park and private lands (Fig. 1A). The climate is cold with mean temperatures of 5°C, and strong winds that prevail during the summer. Precipitation ranges from 100 to 250 mm falling mainly in winter and spring primarily as snow, and in the form of dew that occur year-round (38). The landscape is a grass-dominated steppe with high elevated plateaus that create a west-east steep elevational gradient. To the north-west the area is characterized by higher altitudes represented by the Buenos Aires plateau (1200–1600 m.), where most of our captures occurred. These plateaus are intersected by several tributaries of the Ecker River and have numerous permanent and ephemeral ponds. To the south-east we find lower altitude areas (200–600 m.) intersected by canyons with ephemeral vegas. The vegetation throughout the study area rarely exceeds 0.5 m in height (39) and is composed mainly by graminoids (e.g., Stipa spp., Festuca spp., Poa ligularis) and shrubs (e.g., Berberis heterophylla, Junelia tridens).

Animal Capture and Data Collection

From January 2019 through February 2021, we darted 25 free-ranging guanacos (9 females and 16 males), and fit them with GPS collars (LiteTrack Iridium 420 collar; weight: 0.6 kg; Lotek, Ontario, Canada). Only one adult guanaco per social group was collared. We used a CO₂ rifle and 1.5 ml darts, with a mean dose of 5 mg (0.05 mg/kg dose) of fentanyl oxalate (Thianil, Wildlife Pharmaceuticals) as an immobilization agent, and a dose of 10 mg of naltrexone hydrochloride (Trexonil, Wildlife Pharmaceuticals) for every mg of Thianil as an antagonist agent. All applicable institutional and/or national guidelines for the care and use of animals were followed (IACUC protocol ID: A006515). We programmed the GPS collars to collect a fix every two hours. Four individuals died before completing a migration period and were excluded from the analysis. For the remaining individuals we eliminated locations that had a dilution of precision (DOP) > 10 and locations with altitudes that were unfeasible. Data management and statistical analysis were conducted using R (40).

Migration Classification and Parameters

To establish the occurrence of migration we combined two different methods. First, we used the elevation variant of the Net Squared Displacement (NSD) analysis which measures the altitudinal distances between the starting location and the subsequent relocations for the movement path of a given individual rather than the straight line distances, and then fits the data using a priori models representing migratory, disperser and resident behaviors (31,41,42). We randomly selected one location a day to thin the data, and visually inspected each individual’s relocations to determine the most appropriate starting season and date for each migratory cycle (Fig. 1B). We then fit NSD a priori models and classified individuals to migratory, resident or disperser strategies based on model ranking (Akaike’s information criterion [AICc])( Figure S1, S2). For migratory animals, we determined the timing, duration and distance parameters using the package MigrateR (42). We estimated the altitudinal differences between seasonal ranges (δ), which corresponds to the asymptote of the double sigmoidal curve from the migratory NSD model for each guanaco. We also obtained temporal parameters of migration timing (θ; the midpoint of the departing movement represented by the inflection point of the curve), migration duration (ρ; the period of time the individual occupied each seasonal range), and the duration of the migratory movement per se (φ; the time required to complete ½ to ¾ of the migration).

For our second method, we estimated overlap of seasonal ranges (43). We first classified each guanaco’s location to its summer range, winter range or to periods of active migration, based on the migration timing parameters from the NSD analysis, using mean migration date values for those individuals classified as residents or dispersers by NSD that did not have estimated parameters. We calculated the 95% Kernel Density Estimate (KDE) for each seasonal range in the package adehabitat (Fig. 1C)(44) and evaluated the Bhattacharyya's affinity index overlap (BA) of each individual’s ranges (43,45). If an individual had two consecutive seasonal ranges (summer-winter) with a BA > 0.15, we classified it as resident for the corresponding period. Furthermore, if alternating seasons had a BA < 0.15 but the overlap between subsequent same season ranges (summer-summer/winter-winter) was < 0.5 we classified that individual as disperser. Otherwise (alternating seasons BA ≤ 0.15, same season BA ≥ 0.5) the individual was considered migratory for that year (Figure S3, S4). If a guanaco was classified as resident or disperser by overlap but as migratory by NSD, we constrained migration distance parameter in NSD to only consider an individual as migratory if its summer and winter ranges had an altitudinal difference > 221.5 m (altitudinal distance [δ] 1st quartile for the overall population). If an individual was reclassified as resident when applying this constraint, it was assigned as such.

To confirm that collared guanacos possessed home ranges for each season as well as to inspect movements within their migratory ranges, we evaluated individual semi-variograms for distances between locations as a function of time. We then estimated seasonal home ranges using the 95% Autocorrelated Kernel Density Estimate (AKDE) estimator as well as overlap of seasonal ranges and average speed of movement within each range in the package ctmm (46). To determine how space use varied throughout the year we tested how home range size was influenced by season (summer, winter) with a generalized linear mixed model in the lm4 package, where individual guanaco and year were random effects (47). We log transformed our data to assure residuals normality and homoscedasticity and verified model assumptions using the DHARMA package. After testing for differences between sexes and finding no significance, individuals were pooled regardless of sex for all analyses.

Environmental Drivers of Migration – Vegetation Green-up and Snow

Using Google Earth Engine, we created time series of 16-day composite measurements of the normalized difference vegetation index (NDVI) and the normalized difference snow index (NDSI) for the study area. We used a combination of Landsat and Sentinel 2 satellite imagery at 30-m resolution. If Sentinel and Landsat had data corresponding to the same day, we selected the highest value for the composite; if data from multiple days was available for a particular composite, we used the average. We floored values for both indices at 0 and replaced missing values using linear interpolation. We only used pixels with a quality assessment (QA) bit = 0, and used a threshold of ≥ 10 to determine snow presence (48). We then created altitudinal ranges of 200 m covering the elevation gradient occupied by guanacos and obtained NDVI and NDSI curves spanning the duration of our study (summer 2019- summer 2022) to visually inspect vegetation and snow cover trends and correlation.

Using the same NDVI composites, we obtained measurements for every relocation of each year of data, and created a space-time-time matrix for each individual guanaco year (26), in which each row represented a spatial location the individual occupied at a certain time (day), and each column was the NDVI value corresponding to that location for every day during the period being analyzed. We calculated the cumulative NDVI value for each location, averaging values by total amount of days to account for differences in individual-year duration. We clustered locations into summer and winter seasonal ranges, and summer and winter migrations according to NSD timing parameters. We calculated the mean NDVI value for each seasonal range to determine what the guanaco would have experienced in terms of vegetation biomass if it would have remained resident in either one of those ranges. Finally, we calculated the values each individual actually experienced by adding up the diagonal values of the matrix and compared the experienced NDVI values with the summer and winter range ones using a generalized linear mixed model analysis with individual guanaco as a random effect. We repeated this procedure with snow cover presence data after using our threshold value to transform the NDSI into a binary index where NDSI < 10 = 0 designated lack of snow cover and NDSI ≥ 10 = 1 indicated presence of snow cover.

Migration Classification and Parameters

We analyzed data from 21 individuals (8 females, 13 males), 9 of which had two or more years of data that we analyzed independently to account for the possibility of an individual switching between strategies (Table 1S), totaling 35 individual years. We classified a total of 26 years, consisting of 12 males and 6 females, as migratory (74.3%). Seventeen of those years were identified as migratory by both methods, whereas 9 were classified as migratory by the NSD and as either resident or disperser by overlap but were ultimately classified as migratory. Nine years (25.7%), 3 males and 4 females, were classified as resident. Three of those were originally classified as non-migratory by both methods while the other 6 were classified as migratory by the NSD and were assigned to be resident after further inspection. Five of the 9 individuals tracked for two years (4 migrants and 1 resident) maintained their strategies throughout the duration of this study, while the other 4 alternated between strategies.

Spatial and temporal characteristics of migration derived from our NSD analysis exhibited high individual variability as well as seasonal trends (Table 1S). Mean migration timing from summer to winter ranges was March 21 in 2019 (range: Feb 8- Apr 22), April 3 in 2020 (range: Jan 7- Jul 27) and March 22 in 2021 (range: Feb 20- May 16). Mean migration timing from winter to summer ranges was October 10 in 2019 (range: Sep 12- Nov 18), October 1 in 2020 (range: Aug 9- Nov 2) and October 11 in 2021 (range: Sep 1- Nov 24). Individuals that were tracked for multiple migration cycles had similar migration timing dates (Fig. 2).

Migration distance ranged from altitudinal differences of 295 m to 1106 m with all summer ranges occurring at higher altitudes than winter ones, which translated to straight line distances that ranged from 12.4 km to 38.7 km. We found the intra-individual variation for guanacos with multiple years of data was smaller than the inter-individual variation for all migratory guanacos for all migratory characteristics (Table S2).

We observed considerable variation in the seasonal spatial use by guanacos (Figure S1, S2). Some individuals had defined seasonal home ranges between which they alternated, while others had less constricted ranges. Indeed, for 9 of the migratory individuals this resulted in winter home ranges partially overlapping with their summer ones, leading to different behavioral classifications by our two methods (Figure S3, S4). Despite this variation, home range size differed by season (χ2 = 27.31, p < 0.001) with winter home ranges (\(\stackrel{-}{x }\)= 354.6 km²; Median = 273.1; SD = 517.6) nearly 3× larger than summer ones (\(\stackrel{-}{x }\)= 120.0 km²; Median = 27.1; SD = 261.6) (Fig. 2). Home range fidelity was high both for winter (\(\stackrel{-}{x }\) = 0.80; SD = 0.2) and summer ranges (\(\stackrel{-}{x }\)= 0.67; SD = 0.2). Guanacos moved at lower speeds within their winter ranges (\(\stackrel{-}{x }\)= 9.1 km/day; SD = 2) compared to summer (\(\stackrel{-}{x }\)= 10.3 km/day; SD = 3; χ2 = 5.2, p = 0.02), but they moved more during winter (\(\stackrel{-}{x }\)= 4.3 km/day; SD = 1.0) than in summer (\(\stackrel{-}{x }\)= 3.9 km/day; SD = 1.3; χ2 = 3.9, p = 0.04). However, we did not detect differences in overall distances moved between migratory and resident guanacos (t = -0.78, df = 43, p = 0.4). We also found that at least 14 individuals had repetitive daily time-lag-dependent behaviors within some of their seasonal home ranges (Fig. 3). This occurred primarily in the summer (79%) for both migratory and resident individuals. Most guanacos that exhibited this behavior, did so for multiple seasons.

Environmental drivers of migration – Vegetation green-up and Snow

Phenology of the study area was highly correlated with altitude. Altitudes > 1000 m presented low and high NDVI values in winter and summer respectively, with summer’s up to 3.2× higher than in winter (\(\stackrel{-}{x }\)_summer = 0.13, SD = 0.02; \(\stackrel{-}{x }\)_winter= 0.04, SD = 0.06 for the 1200-1600m altitudinal range). However, elevations below 600 m presented an opposite pattern with NDVI peaking in winter, with winter mean up to 1.5× higher than summer mean (\(\stackrel{-}{x }\)_summer = 0.11, SD = 0.02; \(\stackrel{-}{x }\)_winter = 0.17, SD = 0.16 for the 200-600m altitudinal range) (Fig. 4A). The lack of a uniform wave like progression of plant phenology through the study area prevented any sort of potential green wave surfing. However, guanacos were present in their summer ranges when NDVI peaked at higher altitudes, and migrated towards lower elevations during the winter, when the winter green-up occurred there (Fig. 4A). Additionally, migratory guanacos experienced higher levels of green-up compared to what they would have experienced if they had remained resident in either one of their seasonal ranges (χ2 = 26.2, df = 3, p < 0.001), with 18 migratory individuals’ paths (69%) having NDVI values that were higher than resident range values (Figs. 4B, 4C). Specifically, migratory guanacos cumulative NDVI was 8.4% higher than residents in their summer ranges, and 15.7% higher than residents in their winter ranges.

NDSI followed the same curve-shaped pattern throughout the altitudinal gradient with higher values during winter, but snow cover was present more consistently and for longer periods of time at higher elevations (Fig. 5A). Guanaco summer ranges coincided with the time of the year in which snow cover was absent in those higher altitudes, while spending winter at lower elevations, with shorter and more sporadic periods of snow cover (Fig. 5A). When comparing experienced versus theoretical snow cover presence values, we found migration paths also typically avoided areas with higher presence of snow (χ2 = 95.24 Df = 2, p < 0.001), since 17 migratory guanacos (65.4%) had lower NDSI values than they would have had if they had remained resident in either of their seasonal ranges (Figs. 5B, 5C). Migratory guanacos cumulative NDSI was 38.2% less than summer ranges and 14.5% less than winter ones.

The majority – nearly three-quarters – of the guanacos in the sampled population were migratory. These migrations were typified by relatively short distances along a steep elevational gradient. Similar to other ungulate species in mountainous landscapes (49), and previous observations for guanacos elsewhere (35,50), guanacos in our study site spent the summer at elevations up 295–1105 m above their winter ranges. These elevational movements appeared to be influenced by both vegetation and snow phenology of the study area. Generally migratory guanacos avoided snow-covered areas by occupying snow-free areas at lower altitudes during the winter. This strategy also provided access to higher-quality forage as guanaco’s use of seasonal ranges coincided with the peak of vegetation green-up in both winter and summer ranges.

Guanacos exhibited a combination of fidelity and plasticity regarding migratory characteristics. Even though our data on multiple migration periods was limited, we found individuals that displayed fidelity to migration (for up to three migratory cycles) as well as individuals that switched strategies from one year to the next. Guanacos with multiple years of data that did not switch between strategies, tended to revisit the same summer and winter ranges, and exhibited similar residency times in their migratory ranges. With one exception, all guanacos that switched between strategies had summer ranges that, while still at higher altitudes than their winter ones, were not located in the high elevation plateau most guanacos migrated to, but were rather neighboring their winter ranges. This change between strategies could be due to a combination of the fact that guanacos often presented exploratory movements outside of their home ranges, coupled with lower NDVI years that could force resident guanacos out of their usual boundaries for foraging purposes. We also observed important behavioral plasticity around the timing and duration of migration, both within and between individuals, with interindividual variation being highest. Indeed, migration departure dates differed by > 100 days, and duration of migration ranged from 1 to 21 days. Changes in migration timing could be related to annual variations in climatic conditions, as we found a trend for a later summer migration departure date in 2020, which registered a longer period of snow cover compared to 2019 and 2021, leading to a later snowmelt date.

Variation in migratory characteristics has also been described for a number of other ungulate species (51). Typically, differences in migration timing, distance and duration have been associated with individual attributes, like age, sex, and reproductive status (16,31,52). Although we did not detect notable differences between male and female guanacos, we did not account for guanacos age, nutritional or reproductive status. Given the tight social nature of guanacos that tend to migrate in family groups, we would not expect major behavioral differences driven by individual attributes.

While most collared guanacos were migratory, we observed variation in migration propensity, with nine individual-years, corresponding to 7 guanacos, that remained resident in the winter ranges. This could be related to density-dependent resource release that follows departure of migratory individuals (49,52), as well as to the absence of domestic cattle in the winter range that has been shown to compete with guanacos for resources (53). Partial migration appears to be common for guanacos across their distributional range (50,53,54). Moreover, partial migration seems to be the predominant strategy among ungulates, with > 27 species exhibiting a mix of migratory and resident individuals within the same population (15,52). This strategy has been related to increased fitness and resilience, given the capacity of the population to respond to environmental changing conditions as well as to fluctuations in population density and predation risk (55). Because pumas (Puma concolor), guanacos main predator, have been found to not follow migratory individuals (50) and occur throughout the study area, and resident guanacos move as much as migratory ones within their home ranges, we do not believe risk of predation or energetic costs associated with long distance movement are influencing guanaco’s decisions to migrate or not. Instead, migratory guanacos likely benefit from accessing newly emerged vegetation, and resident guanacos may benefit from reduced competition during mating and breeding season, while avoiding risk related to encountering roads and fences during migration.

Multiple studies have shown that ungulate migrations follow certain patterns of plant phenology (22,56). For instance, the progression across the landscape of the instantaneous rate of green-up, a proxy of “springness” (26) can be compared to individual locations to evaluate whether animals are maximizing the use of newly emerged vegetation by surfing or jumping green waves (56,56–58). However, we found that these wave-like patterns are not representative of the phenology in our study area. The Patagonian dry shrub and grass steppes exhibit a complex relationship between precipitation, temperature and annual variations in NDVI, with lower elevation areas presenting a lagged vegetative green-up that occurs in early winter (59,60). This corresponds with the pattern we detected in our study area, where higher elevations had wave like patterns with summer peaks while lower elevations had more unpredictable patterns with late fall and winter peaks. The particularities of the phenology along with the short migratory distances, therefore, likely hamper the ability of guanacos to surf green waves. We did, however, detect higher primary productivity levels in the greenscape experienced by migratory guanacos compared to the available greenscapes corresponding to their summer and winter ranges had they remained resident. Thus, although guanacos did not appear to surf green-waves, migratory guanacos did track general patterns of vegetation green-up.

Although less common than vegetation phenology, snow is also a major driver of ungulates migrations (58,61–63). Ungulates tend to avoid snow because increasing snow depth reduces mobility, exacts additional energetic costs, and hinders the search for forage (31). As predicted, guanacos’ migration was associated to snow cover as they appeared to avoid “white waves”: migratory individuals occupied high altitude areas in spring and summer when snow was absent, but moved into low altitude areas featuring less now, during late fall and winter. Indeed, migratory individuals experienced less snow compared to what they would have experienced in their summer range. Having remained in their winter ranges in some cases however, would have meant even less exposure to snow, which could indicate that a synergy of both vegetation and snow phenology are driving guanacos’ migrations. In fact, given the strong correlation between these two variables, in which snow cover mediates vegetation green-up dynamics (64,65), it is difficult to disentangle their effects at such spatial and temporal resolutions. This was the case in our study area, particularly so for high altitudes where NDSI reached higher values and snow cover was present for extended periods of time (Figure S5).

Snow has also been related to smaller winter home ranges given the constraints it can have on animal movement (62,66). While we found that guanacos moved slower during the winter season, they had winter ranges that were larger than summer ones. This is uncommon among migratory ungulates and could be related to snow depth in winter ranges being insufficient to constrain movement, or to a combination of factors including the fact that winter ranges are shared between migratory and resident individuals, as well as the fact that guanacos form larger social groups during this season (36) which could limit vegetation availability and force individuals to forage in larger areas. Reduced forage due to snow cover could also be a factor influencing winter home range size, as 2020 was both the year with higher snow presence cover and larger winter home ranges. Additionally, displaying of male guanacos during the mating season might constrain movements and prevent access to certain areas leading to a reduction of the space effectively available in summer months (53).

While migration occurs at a landscape and seasonal level, driven by broad scale variables, such as snow and vegetation phenology, within seasonal ranges, habitat selection occurs at finer scales, driven by the daily need to maximize foraging while reducing predation risk and minimizing competition (2,67). We found that more than half of the collared guanacos exhibited daily micro-migrations nested within their seasonal ones. Such behaviors were more common during summer, and among residents (6/7 residents vs 12/18 migrants) and resulted in small home ranges. These daily movements are possibly driven by local trade-offs between resource availability and perceived risk, as it has been found previously among vicuñas (68) that tended to visit highly productive but also high risk foraging sites during the day, while moving into lower productivity yet safer sites at night.

Guanacos exhibit a mixture of both fidelity and plasticity regarding migration, which might prove to be critical for the continuity of this strategy in the face of rapid environmental change. Climate change could negatively impact migration by creating a mismatch between climatic conditions, vegetation green-up and migratory movements. This in turn could decrease the benefits of migratory behaviors, which has been observed in other species of migratory ungulates (69). Similarly, land use change and habitat fragmentation, such as those caused by roads and fences, may impact migratory movements by altering connectivity between seasonal ranges. Individual variations in temporal and spatial migration characteristics, including the ones we described for this guanaco population, could provide the necessary plasticity for this species to adapt to changing environmental conditions.

Understanding of where, when, and why animals migrate is important to assess potential adaptations to future scenarios and to inform conservation policies to protect summer and winter ranges as well as corridors that connect them. We found guanacos have altitudinal migrations that track vegetation and snow phenology, displaying a combination of fidelity and flexibility in the temporal and spatial characteristics of their movements. Provided that habitat and connectivity are maintained, and climate change does not outpace the capacity of this species to adapt, the high degree of plasticity in guanaco migration has the potential to buffer these effects. Altering migration timing or location of their seasonal ranges can ensure guanacos continue to match vegetation green-up and snow patterns to their seasonal habitat use, allowing this species the capacity to respond to change without losing their ancestral seasonal migrations.

AKDE: Autocorrelated Kernel Density Estimate

AICc: Akaike’s information criterion

BA: Bhattacharyya's affinity index overlap

KDE: Kernel Density Estimate

NDSI: Normalized difference snow index

NDVI: Normalized difference vegetation index

NSD: Net Squared Displacement

Ethics approval:

All animal captures were approved by the University of Wisconsin (IACUC protocol ID: A006515).

Consent to participate:

Not applicable.

Consent for publication:

Not applicable.

Availability of data and materials:

The datasets supporting the conclusions of this article are included within the article and its supplementary information files.

Code availability:

All code used is available from the corresponding author upon request.

Competing Interests:

The authors declare that they have no competing interests.

Funding:

University of Wisconsin-Madison; American society of Mammalogy Latin American Student Field Research Award; National Geographic (WW-216C-17).

Authors’ contributions:

ED and JNP conceived and secured funding for the research. MC and ED collected data. MC conducted the statistical analysis with input from JNP. MC and JNP wrote the manuscript with input from ED.

Acknowledgments:

We thank Rewilding Argentina and their staff at Parque Patagonia for the help and accommodations provided. We also thank American Society of Mammalogy (ASM) and National Geographic for the funding provided and National Park rangers from Parque Patagonia as well as the technicians that made this possible. Finally, we want to thank Volker Radeloff, David Gudex-Cross and Paula Perrig for their valuable comments.

Boyce MS. Seasonality and patterns of natural selection for life histories. The American Naturalist. 1979 Oct 1;114:569–83.
Anderson KJ, Jetz W. The broad-scale ecology of energy expenditure of endotherms. Ecology Letters. 2005;8:310–8.
Pauli JN, Zuckerberg B, Whiteman JP, Porter W. The subnivium: a deteriorating seasonal refugium. Frontiers in Ecology and the Environment. 2013;11:206–67.
Dingle H, Drake A. What Is migration? BioScience. 2007;57:113–21.
Berger J. The last mile: How to sustain long-distance migration in mammals. Conservation Biology. 2004;18:320–31.
Sinclair ARE. The function of distance movements in vertebrates. The Ecology of Animal Movements. 1983;248–58.
Avgar T, Street G, Fryxell JM. On the adaptive benefits of mammal migration. Canadian Journal of Zoology. 2014;92:481–90.
Middleton AD, Cook JG, Nelson AA, McWhirter DE, Klaver RW, Kauffman MJ. Animal migration amid shifting patterns of phenology and predation: lessons from a Yellowstone elk herd. Ecological Society of America. 2013;94:1245–56.
Fryxell JM, Sinclair ARE. Causes and consequences of migration by large herbivores. Trends in Ecology and Evolution. 1988;3:237–41.
Rickbeil GJM, Merkle JA, Anderson G, Atwood MP, Beckmann JP, Cole EK, et al. Plasticity in elk migration timing is a response to changing environmental conditions. Global Change Biology. 2019;25:2368–81.
Sawyer H M, Kauffman MJ, Nielson R M, Horne JS. Identifying and prioritizing ungulate migration routes for landscape-level conservation. Ecological Applications. 2009;19:2016–25.
Sawyer H, Merkle JA, Middleton AD, Dwinnell S PH, Monteith KL. Migratory plasticity is not ubiquitous among large herbivores. Journal of Animal Ecology. 2019;88:450–60.
Peters W, Hebblewhite M, Mysterud A, Eacker D, Hewison AJM, Linnell JDC, et al. Large herbivore migration plasticity along environmental gradients in Europe: life-history traits modulate forage effects. Oikos. 2019;128:416–29.
Chapman BB, Brönmark C, Jan-Åke N, Lars-Anders H. The ecology and evolution of partial migration. Oikos. 2011;120:1764–75.
Xu W, Barker K, Shawler A, Van Scoyoc A, Smith JA, Mueller T, et al. The plasticity of ungulate migration in a changing world. Ecology. 2021;102.
Eggeman S L, Hebblewhite M, Bohm H, Whittington J, Merrill EH. Behavioural flexibility in migratory behaviour in a long-lived large herbivore. Journal of Animal Ecology. 2016;85:785–97.
Kaitala A, Kaitala V, Lundberg P. A theory of partial migration. The American Naturalist. 1993;142:59–81.
Fuller TK, Keith LB. Woodland caribou population dynamics in northeastern Alberta. The Journal of Wildlife Management. 1981;45:197–213.
Hebblewhite M, Merrill EH. Demographic balancing of migrant and resident elk in a partially migratory population through forage-predation tradeoffs. Oikos. 2011;120:1860–70.
Singh NJ, Grachev IA, Bekenov AB, Milner-Gulland EJ. Tracking greenery across a latitudinal gradient in central Asia - the migration of the saiga antelope. Diversity and Distributions. 2010;16:663–75.
Teitelbaum CS, Fagan WF, Fleming CH, Dressler G, Calabrese JM, Leimgruber P, et al. How far to go? Determinants of migration distance in land mammals. Ecology Letters. 2015;18:545–52.
Merkle JA, Monteith KL, Aikens EO, Hayes MM, Hersey KR, Middleton AD, et al. Large herbivores surf waves of green-up during spring. Proceedings of The Royal Society. 2016;283:1–8.
Van der Graaf AJ, Stahl J, Klimkowska A, Bakker JP, Drent RH. Surfing on a green wave-how plant growth drives spring migration in the Barnacle Goose Branta leucopsis. Ardea. 2006;94:567–77.
Fryxell JM, Greever J, Sinclair ARE. Why are migratory ungulates so abundant? The American Naturalist. 1988;131:781–98.
Hebblewhite M, Merrill EH, McDermid G. A multi-scale test of the forage maturation hypothesis in a partially migratory ungulate population. Ecological Monographs. 2008;78:141–66.
Bischof R, Loe LE, Meisingset E L, Zimmermann B, van Moorter B, Mysterud A. A Migratory Northern Ungulate in the Pursuit of Spring: Jumping or Surfing the Green Wave? The American Naturalist. 2012;180(4):407–24.
Monteith KL, Bleich VC, Stephenson TR, Pierce BM, Conner MM, Klaver RW, et al. Timing of seasonal migration in mule deer: effects of climate, plant phenology, and life-history characteristics. Ecosphere. 2011 Apr 1;2:art47.
Brinkman TJ, Deperno CS, Jenks JA, Haroldson BS, Osborn RG. Movement of female white-tailed deer: effects of climate and intensive row-crop agriculture. Journal of Wildlife Management. 2005 Jul 1;69:1099–111.
Rivrud IM, Bischof R, Meisingset EL, Zimmermann B, Loe LE, Mysterud A. Leave before it’s too late: anthropogenic and environmental triggers of autumn migration in a hunted ungulate population. Ecology. 2016 Apr 1;97:1058–68.
Peng S, Piao S, Ciais P, Fang J, Wang X. Change in winter snow depth and its impacts on vegetation in China. Global Change Biology. 2010 Nov 1;16:3004–13.
Singh NJ, Börger L, Dettki H, Bunnefeld N, Ericsson G. From migration to nomadism: Movement variability in a northern ungulate across its latitudinal range. Ecological Applications. 2012;22:2007–20.
Franklin WL. Biology, ecology, and relationship to man of the South American camelids. Mammalian Biology in South America. 1982;6:457–89.
Wurstten A, Novaro AJ, Walker RS. Habitat use and preference by guanacos, vicuñas, and livestock in an altitudinal gradient in northwest Argentina. European Journal of Wildlife Research. 2014;60:35–43.
Acebes P, Traba J, Malo JE, Ovejero R, Borghi CE. Density and habitat use at different spatial scales of a guanaco population (Lama guanicoe) in the Monte desert of Argentina. Mammalia. 2010;74:57–62.
Mueller T, Olson K A, Dressler G, Leimgrube P, Fuller T K, Nicolson C, et al. How landscape dynamics link individual to population-level movement patterns: a multispecies comparison of ungulate relocation data. Global Ecology and Biogeography. 2011;20:638–94.
Ortega IM, Franklin WL. Social organization, distribution and movements of a migratory guanaco population in the Chilean Patagonia. Revista Chilena de Historia Natural. 1995;68:489–500.
Bertiller MB, Beeskow AM, Coronato F. Seasonal environmental variation and plant phenology in arid Patagonia (Argentina). Journal of Arid Environments. 1991 Jul 1;21:1–11.
Cabrera AL. Fitogeografía de la República Argentina. Boletín de la Sociedad Argentina de Botánica. 1971;14:1–42.
Movia CP, Soriano A, Leon RJC. La vegetación de la cuenca del río Santa Cruz (provincia de Santa Cruz, Argentina). Instituto de Botánica Darwinion; 1987. 9–78 p. (Darwininana; vols. 1–4).
R Core Team. A language and environment for statistical computing [Internet]. Vienna, Austria; 2016. Available from: https://www.R-project.org/
Bunnefeld N, Börger L, van Moorter B, Rolandsen C, Dettki H, Solberg EJ, et al. A model-driven approach to quantify migration patterns: individual, regional and yearly differences. Journal of Animal Ecology. 2011;80:466–76.
Spitz D B, Hebblewhite M, Stephenson T R. ‘MigrateR’: extending model-driven methods for classifying and quantifying animal movement behavior. Ecography. 2017;40:788–99.
Cagnacci F, Focardi S, Ghisla A, van Moorter B, Merrill EH, Gurarie E, et al. How many routes lead to migration? Comparison of methods to assess and characterize migratory movements. Journal of Animal Ecology. 2016;85:54–68.
Calenge C. The package “adehabitat” for the R software: A tool for the analysis of space and habitat use by animals. Ecological Modelling. 2006;197(3):516–9.
Worton BJ. Kernel methods for estimating the utilization distribution in home-range studies. Ecology. 1989;70(1):164–8.
Calabrese JM, Fleming CH, Gurarie E. ctmm: an R package for analyzing animal relocation data as continuous-time stochastic process. Methods in Ecology and Evolution. 2016;7:1124–32.
Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. Journal of statistical software. 2015;67:1–48.
Gudex-Cross D, Keyser S R, Zuckerberg B, Fink D, Zhu L, Pauli JN, et al. Winter Habitat Indices (WHIs) for the contiguous US and their relationship with winter bird diversity. Remote Sensing of Environment. 2021;255:112309.
Mysterud A. Seasonal migration pattern and home range of roe deer (Capreolus capreolus) in an altitudinal gradient in southern Norway. Journal of Zoology. 1999;247:479–86.
Gelin M L, Branch L C, Thornton D H, Novaro AJ, Gould M J, Caragiulo A. Response of pumas (Puma concolor) to migration of their primary prey in Patagonia. PLoS ONE. 2017;12.
Guan TP, Ge BM, McShea W J, Li S, Song YL, Stewart C M. Seasonal migration by a large forest ungulate: a study on takin (Budorcas taxicolor) in Sichuan province, China. European Journal of Wildlife Research. 2013;59:81–91.
Berg JE, Hebblewhite M, St. Clair CC, Merrill EH. Prevalence and mechanisms of partial migration in ungulates. Frontiers in Ecology and Evolution. 2019;7.
Moraga CA, Funes MC, Pizarro JC, Briceño C, Novaro AJ. Effects of livestock on guanaco Lama guanicoe density, movements and habitat selection in a forest–grassland mosaic in Tierra del Fuego, Chile. Oryx. 2015;49:30–41.
Schroeder N M, Matteucci S D, Moreno P G, Gregorio P, Ovejero R, Taraborelli P, et al. Spatial and seasonal dynamic of abundance and distribution of guanaco and livestock: Insights from using density surface and null models. PLoS ONE. 2014;9.
Mysterud A, Loe LE, Zimmermann B, Bischof R, Veiberg V, Meisingset E. Partial migration in expanding red deer populations at northern latitudes – a role for density dependence? Oikos. 2011 Dec 1;120:1817–25.
Aikens EO, Mysterud A, Merkle JA, Cagnacci F, Rivrud IM, Hebblewhite M, et al. Wave-like Patterns of Plant Phenology Determine Ungulate Movement Tactics. Current Biology. 2020;30:3444–9.
Sawyer H, Kauffman MJ. Stopover ecology of a migratory ungulate. Journal of Animal Ecology. 2011;80:1078–87.
Laforge M P, Bonar M, Vander Wal E. Tracking snowmelt to jump the green wave: phenological drivers of migration in a northern ungulate. Ecology. 2021;102.
Bianchi E, Villalba R, Solarte A. NDVI spatio-temporal patterns and climatic controls over northern Patagonia. Ecosystems. 2020;23:84–97.
Bruzzone O, Easdale MH. Archetypal temporal dynamics of arid and semi-arid rangelands. Remote Sensing of Environment. 2021;254.
Ball JP, Nordengren C, Wallin K. Partial migration by large ungulates: characteristics of seasonal moose Alces alces ranges in northern Sweden. Wildlife Biology. 2001;7:39–47.
Luccarini S, Mauri L, Ciuti S, Lamberti P, Apollonio M. Red deer (Cervus elaphus) spatial use in the Italian Alps: home range patterns, seasonal migrations, and effects of snow and winter feeding. Ethology Ecology & Evolution. 2006;18:127–45.
Severson J P, Johnson H E, Arthur SM, Leacock W B, Suitor M J. Spring phenology drives range shifts in a migratory Arctic ungulate with key implications for the future. Global Change Biology. 2021;27:1–18.
Wang X, Wu C, Peng D, Gonsamo A, Liu Z. Snow cover phenology affects alpine vegetation growth dynamics on the Tibetan Plateau: Satellite observed evidence, impacts of different biomes, and climate drivers. Agricultural and Forest Meteorology. 2018 Jun 15;256:61–74.
Xiong T, Zhang H, Zhao J, Zhang Z, Guo X, Zhu Z, et al. Diverse responses of vegetation dynamics to snow cover phenology over the boreal region. Forests. 2019;10.
Shakeri YN, White KS, Waite JN. Staying close to home: Ecological constraints on space use and range fidelity in a mountain ungulate. Ecology and Evolution. 2021 Aug 1;11:11051–64.
Bose S, Forrester TD, Casady DS, Wittmer HU. Effect of activity states on habitat selection by black-tailed deer. Journal of Wildlife Management. 2018;82:1711–24.
Smith JA, Donadio E, Pauli JN, Sheriff MJ, Middleton AD. Integrating temporal refugia into landscapes of fear: prey exploit predator downtimes to forage in risky places. Oecologia. 2019;189:883–90.
Aikens EO, Monteith KL, Merkle JA, Fralick GL, Kauffman MJ. Drought reshuffles plant phenology and reduces the foraging benefit of green-wave surfing for a migratory ungulate. Global Change Biology. 2020;26:4215–25.

No competing interests reported.

Download PDF

Editorial decision: Major revision
09 Jun, 2022
Reviews received at journal
26 May, 2022
Reviewers agreed at journal
20 May, 2022
Reviewers invited by journal
20 May, 2022
Editor assigned by journal
12 May, 2022
Submission checks completed at journal
12 May, 2022
First submitted to journal
11 May, 2022

You are reading this latest preprint version

Phenological drivers of ungulate migration in South America: Characterizing the movement and seasonal habitat use of guanacos

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Study Area

Animal Capture and Data Collection

Migration Classification and Parameters

Environmental Drivers of Migration – Vegetation Green-up and Snow

Results

Migration Classification and Parameters

Environmental drivers of migration – Vegetation green-up and Snow

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1