12The decision for a migratory animal to be site faithful in its non--breeding season has 13 profound implications for migratory connectivity, resilience to winter habitat loss 14 and population dynamics through carry--over effects on future breeding success and 15 fitness. Knowledge of the temporal and spatial scale of site fidelity and dispersal is 16 also central to accurate survival estimates. We established the observed spatial and 17 temporal scale of site fidelity and the ability to detect small--scale dispersal within 18 and between years for a wintering long--distance Palearctic migrant, the Whinchat 19 Saxicola rubetra, by comparing predicted and observed detection rates within the 20 study site. Across two years, 54% of birds returned to the study site and all returning 21 birds reoccupied the territories they used in the previous winter. Observed dispersal 22 was very low despite the high probability of detecting any local dispersal, suggesting 23 that return rates are indicative of true between--winter survival rates for this 24 population. In any winter, 50% of returning individuals had a previously occupied but 25 now empty territory that was less than one territory--span away from the centre of 26 their current territory; high site fidelity was therefore very unlikely to be because of 27 limited territory availability. Over--winter residency time (defined by departure 28 month) differed significantly across sites and with age, but did not determine the 29 probability of whether a bird returned in the following year. This suggests the use of 30 more than one wintering site for some individuals, rather than reduced over--winter 31 survival. This study is one of the first to comprehensively document site fidelity at 32 the territory scale in a Palearctic system, although less comprehensive studies or 33 anecdotal evidence suggest that high winter site fidelity may be relatively common. 34Here we provide evidence for the serial residency hypothesis, where selection acts 35 for individual migrants to have generalist habitat requirements, allowing them to 36 survive in and remain site faithful to even relatively low quality, but sufficient and 37 familiar sites. Lower dispersal and higher site fidelity compared to that during 38 breeding suggest that annual survival estimates are more accurate when measured 39 on the wintering grounds. This study supports previous findings that wintering 40 conditions do not limit Whinchat populations.

For migrant birds, which habitats are suitable during the non‐breeding season influences habitat availability, population resilience to habitat loss, and ultimately survival. Consequently, habitat preferences during winter and whether habitat segregation according to age and sex occurs directly influences migration ecology, survival and breeding success. We tested the fine‐scale habitat preferences of a declining Palearctic migrant, the whinchat Saxicola rubetra, on its wintering grounds in west Africa. We explored the influence of habitat at the territory‐scale and whether dominance‐based habitat occupancy occurs by describing the variation in habitat characteristics across wintering territories, the degree of habitat change within territories held throughout winter, and whether habitat characteristics influenced territory size and space‐use within territories or differed with age and sex. Habitat characteristics varied substantially across territories and birds maintained the same territories even though habitat changed significantly throughout winter. We found no evidence of dominance‐based habitat occupancy; instead, territories were smaller if they contained more perching shrubs or maize crops, and areas with more perching shrubs were used more often within territories, likely because perches are important for foraging and territory defence. Our findings suggest that whinchats have non‐specialised habitat requirements within their wintering habitat of open savannah and farmland, and respond to habitat variation by adjusting territory size and space‐use within their territories instead of competing with conspecifics. Whinchats show a tolerance for human‐modified habitats and results support previous findings that some crop types may provide high‐quality wintering habitat by increasing perch density and foraging opportunities. By having non‐specialised requirements within broad winter habitat types, migrants are likely to be flexible to changing wintering conditions in Africa, both within and across winters, so possibly engendering some resilience to the rapid anthropogenic habitat degradation occurring throughout their wintering range.

Data from location logging tags have revolutionised our understanding of migration ecology, but methods of tagging that do not compromise survival need to be identified. We compared resighting rates for 156 geolocator-tagged and 316 colour ringed-only whinchats on their African wintering grounds after migration to and from eastern Europe in two separate years. We experimentally varied both light stalk length (0, 5 and 10 mm) and harness material (elastic or non-elastic nylon braid tied on, leg-loop 'Rappole' harnesses) in the second year using a reasonably balanced design (all tags in the first year used an elastic harness and 10 mm light stalk). Tags weighed 0.63 g (0.01 SE), representing 4.1% of average body mass. There was no overall significant reduction in between-year resighting rate (our proxy for survival) comparing tagged and untagged birds in either year. When comparing within tagged birds, however, using a tied harness significantly reduced resighting rate by 53% on average compared to using an elastic harness (in all models), but stalk length effects were not statistically significant in any model considered. There was no strong evidence that the fit (relative tightness) or added tag mass affected survival, although tied tags were fitted more tightly later in the study, and birds fitted with tied tags later may have had lower survival. Overall, on a precautionary principle, deploying tags with non-elastic tied harnesses should be avoided because the necessary fit, so as not to reduce survival, is time-consuming to achieve and does not necessarily improve with experience. Geolocator tags of the recommended percentage of body mass fitted with elastic leg-loop harnesses and with short light stalks can be used without survival effects in small long-distance migrant birds.

For migratory birds, it is necessary to estimate annual and overwinter survival rates, identify factors that influence survival, and assess whether survival varies with age and sex if we are to understand population dynamics and thus inform conservation. This study is one of the first to document overwinter and annual survival from the wintering grounds of a declining Afro-Palaearctic migrant bird, the Whinchat Saxicola rubetra. We monitored a population of marked individuals for which dispersal was low and detectability was high, allowing accurate estimates of survival. Annual survival was at least 52% and did not differ significantly across demographic groups or with habitat characteristics or residency time in the previous winter. Overwinter survival was very high and monthly survival at least 98% at some sites. Although winter residency varied spatially and with age, lower residency did not correlate with reduced annual survival, suggesting occupancy of multiple wintering sites rather than higher winter mortality of individuals with shorter residency. Our results suggest that mortality occurs primarily outside the wintering period, probably during migration, and that wintering conditions have minimal influence on survival. The similarity between survival rates for all age and sex classes when measured on the wintering grounds implies that any difference in survival with age or sex occurs only during the first migration or during the post-fledging stage, and that selection of wintering habitat, or territory quality, makes little difference to survival in Whinchats. Our findings suggest that the wintering grounds do not limit populations as much as the migratory and breeding stages, with implications for the conservation of declining Afro-Palaearctic migrants more widely.

The spatial scale of non‐breeding areas used by long‐distance migrant animals can vary from specific, relatively small non‐breeding areas for each independent breeding population (high connectivity) to a distribution over a large non‐breeding area with mixing of breeding populations (low connectivity). Measuring variation in the degree of connectivity and how it arises is crucial to predict how migratory animals can respond to global habitat and climate change because low connectivity is likely to be an adaptation to environmental uncertainty. Here, we assess whether use of non‐breeding areas in a long‐distance migrant may be stochastic by measuring the degree of connectivity, and whether it is annually variable. Twenty‐nine wintering Whinchats tagged with geolocators over 2 years within 40 km2 in central Nigeria were found to be breeding over 2.55 million km2 (26% of the land area of Europe), without an asymptote being approached in the relationship between area and sample size. Ranges differed in size between years by 1.51 million km2 and only 15% of the total breeding range across both years overlapped (8% overlap between years when only first‐year birds were considered), well above the range size difference and below the proportion of overlap that would be predicted from two equivalent groups breeding at random locations within the observed range. Mean distance between breeding locations (i.e. migratory spread) differed significantly between years (604 ± 18 km in 2013 and 869 ± 33 km in 2014). The results showed very low and variable connectivity that was reasonably robust to the errors and assumptions inherent in the use of geolocators, but with the caveat of having only ranges of 2 years to compare, and the sensitivity of range to the breeding locations of a small number of individuals. However, if representative, the results suggest the scope for between‐year variation (cohort effects) to determine migrant distribution on a large scale. Furthermore, for species with similarly low connectivity, we would predict breeding population trends to reflect average conditions across large non‐breeding areas: thus, as large areas of Africa become subject to habitat loss, migrant populations throughout Europe will decline.