roosting habitats. Studies addressing this hypothesis are scarce, but there is some
supportive evidence. For instance, a study showed that the displacement effects of
wind farms on pink-footed geese Anser brachyrhynchus occurred over relatively
short distances (<150 m), and that the negative effects declined markedly in the 10
years after wind farm construction (Madsen and Boertmann 2008). Also,
Donaldson et al. (2007) found that waterbirds were more tolerant of human presence in developed than in undeveloped sites, probably reflecting some level of
behavioural adaptation to potential anthropogenic disturbance (see also Lowry et al.
2013). In contrast to these studies, however, in Chap. 12, Múrias and colleagues, in
a study akin to a BACI design, reported significantly lower shorebird abundances in
saltpans close to a new railway that had recently begun to operate. Another study by
Burton et al. (2002) reported reduced wader abundances in relation to footpaths,
roads and railroads, although this greatly varied across species and human structures. Together, these studies suggest that the exclusion effects of railways on
wetland birds may be small in some cases but not in others, and thus drawing
generalities would require a larger number of studies covering a wide range of
socio-ecological contexts.
Although our study did not find obvious negative effects, the densities of some
species were sometimes higher far from (or close to) the railway, although this was
not statistically significant and there were inconsistencies across species, seasons,
and stages of the tidal cycle. Reasons for this are not certain, but they may be a
consequence of spatial and temporal variations in habitat requirements and quality
for foraging and roosting, rather than reflecting negative (or positive) effects of the
railway. It is well known that waders and other aquatic birds undertake daily
movements between foraging and roosting habitats, which are jointly affected by
the circadian and tidal cycles, and which induce great variations in bird spatial
distributions in estuaries and other wetlands (Dias et al. 2006; Granadeiro et al.
2006). Also, there are changes in the use of habitats through the annual cycle, due
for instance, to differences in habitat selection between wintering and migrating
birds (Martins et al. 2016), or temporal changes in habitat quality (Lourenço et al.
2009). These factors may underlie, for instance, the observation of avocets concentrating at high densities in intertidal mudflats close to the railway at low tide in
autumn, but not in winter, while at high tide they were mainly found in a single salt
pan far from the railway in both autumn and winter. Conversely, dunlins were
found concentrated in a single salt pan close to the railway in winter at high tide,
while at low tide during the same season they were mostly counted in intertidal
areas close to the railway and in a salt pan far from the railway. Overall, these
results suggest that care should be taken when evaluating the effects of railways on
wetland birds, due to the confounding effects resulting from species-specific habitat
requirements and preferences, the availability and quality of habitats, and the
changes in these factors over the seasonal, circadian, and tidal cycles.
Acknowledgements We would like to thank to Pedro Salgueiro and Luís Gomes for their support
in field work. This study was supported by Infraestruturas de Portugal with the contribution of the
Portuguese Science Foundation through the Doctoral Grant SFRH/BD/81602/2011 (Carlos
Godinho) doctoral grant.
188
C. Godinho et al.
supportive evidence. For instance, a study showed that the displacement effects of
wind farms on pink-footed geese Anser brachyrhynchus occurred over relatively
short distances (<150 m), and that the negative effects declined markedly in the 10
years after wind farm construction (Madsen and Boertmann 2008). Also,
Donaldson et al. (2007) found that waterbirds were more tolerant of human presence in developed than in undeveloped sites, probably reflecting some level of
behavioural adaptation to potential anthropogenic disturbance (see also Lowry et al.
2013). In contrast to these studies, however, in Chap. 12, Múrias and colleagues, in
a study akin to a BACI design, reported significantly lower shorebird abundances in
saltpans close to a new railway that had recently begun to operate. Another study by
Burton et al. (2002) reported reduced wader abundances in relation to footpaths,
roads and railroads, although this greatly varied across species and human structures. Together, these studies suggest that the exclusion effects of railways on
wetland birds may be small in some cases but not in others, and thus drawing
generalities would require a larger number of studies covering a wide range of
socio-ecological contexts.
Although our study did not find obvious negative effects, the densities of some
species were sometimes higher far from (or close to) the railway, although this was
not statistically significant and there were inconsistencies across species, seasons,
and stages of the tidal cycle. Reasons for this are not certain, but they may be a
consequence of spatial and temporal variations in habitat requirements and quality
for foraging and roosting, rather than reflecting negative (or positive) effects of the
railway. It is well known that waders and other aquatic birds undertake daily
movements between foraging and roosting habitats, which are jointly affected by
the circadian and tidal cycles, and which induce great variations in bird spatial
distributions in estuaries and other wetlands (Dias et al. 2006; Granadeiro et al.
2006). Also, there are changes in the use of habitats through the annual cycle, due
for instance, to differences in habitat selection between wintering and migrating
birds (Martins et al. 2016), or temporal changes in habitat quality (Lourenço et al.
2009). These factors may underlie, for instance, the observation of avocets concentrating at high densities in intertidal mudflats close to the railway at low tide in
autumn, but not in winter, while at high tide they were mainly found in a single salt
pan far from the railway in both autumn and winter. Conversely, dunlins were
found concentrated in a single salt pan close to the railway in winter at high tide,
while at low tide during the same season they were mostly counted in intertidal
areas close to the railway and in a salt pan far from the railway. Overall, these
results suggest that care should be taken when evaluating the effects of railways on
wetland birds, due to the confounding effects resulting from species-specific habitat
requirements and preferences, the availability and quality of habitats, and the
changes in these factors over the seasonal, circadian, and tidal cycles.
Acknowledgements We would like to thank to Pedro Salgueiro and Luís Gomes for their support
in field work. This study was supported by Infraestruturas de Portugal with the contribution of the
Portuguese Science Foundation through the Doctoral Grant SFRH/BD/81602/2011 (Carlos
Godinho) doctoral grant.
188
C. Godinho et al.