| Abstract |
This study was focused on the distribution and abundance of Cretan bats.
In total, 767 records from 283 localities along the island were collected:
432 own records and 335 resulting for the research efforts of other
scientists (published or communicated to the author). The recent
discovery of Nyctalus leisleri by Czech researchers and the author in the
prefectures of Chania and Rethymno increases the number of Cretan
species to 17. This species, along with Pipistrellus nathusii, Myotis
mystacinus/aurascens, Plecotus kolombatovici and Plecotus macrobullaris
are rarest and less abundant species of Crete, being located in small
numbers in less than 15 localities each. The commonest and more
abundant species in underground shelters are Rhinolophus ferrumequinum,
Myotis blythii/oxygnathus and Miniopterus schreibersii. On the contrary, in
various feeding grounds Hypsugo savii, Pipistrellus hanaki, Pipistrellus kuhlii
and Tadarida teniotis are by far more common and abundant. Similarly
with other Mediterranean areas, some caves on Crete house maternity
colonies of the species Rhinolophus blasii, R. ferrumequinum, R.
hipposideros, Myotis blythii/oxygnathus, M. capaccinii, M. emarginatus and
Miniopterus schreibersii. Furthermore, low to mid‐elevation caves are
being used by Rhinolophus blasii, R. ferrumequinum, R. hipposideros and
Miniopterus schreibersii, which emerge to forage during warm evenings
with no rain. M. blythii/oxygnathus is the only species found to hibernate
on Crete, in caves situated above 1450 m.a.s.l.
The echolocation calls of the Cretan bats were also described here, in order
to facilitate the study of their distribution and ecology. For this purpose
570 bats from 16 species (excluding Pipistrellus nathusii) were recorded
after their capture or during free flight while exiting from their roosts or
foraging. The echolocation calls of Pipistrellus hanaki were studied
thoroughly for first time since its discovery. Investigation of the spectral
and temporal parameters of the recorded calls showed that the recording
conditions (habitat structure) affected the call shape in some species.
Comparison of the calls of Cretan populations with those from northeast
Greece and other European areas revealed significant differences in several
species. These results verify the already proposed existence of geographic
variation in bat echolocation calls and demonstrate the necessity of their
detailed description before conducting acoustic surveys in a certain area.
The collected echolocation calls from seven species (H. savii, Myotis
blythii/oxygnathus, M. capaccinii, M. emarginatus, Miniopterus
schreibersii, P. hanaki and P. kuhlii) were used to develop Quadratic
Discriminant Function Analyses, in order to identify free flight calls of
unknown origin. Evaluation of the method’s results with cross‐validation
showed it has a satisfactorily high performance, with correct classification
rates for different species ranging from 83% to 100%. When Discriminant
Function Analysis was applied for FM and QCF calls separately, it generally
gave higher classification results. The high performance of the method may
be attributed to the small number of the Cretan bat species and the
exclusion of Pipistrellus nathusii and Myotis mystacinus/aurascens (due to
the lack of enough reference samples) which may result in some
misclassifications in acoustic surveys.
An additional aim of this thesis was the study of altitudinal differentiation
of bat feeding activity in shrublands, the most widespread semi‐natural
habitat type on Crete. Spring and autumn recordings were made in 15
shrublands of three elevation zones (500, 1000 and 1500 m.a.s.l.), along 2
Km transects in Idi (Psiloritis) Mts. The Discriminant Function Analysis
developed earlier was used to identify the recorded echolocation calls to
species or species group level and substantiated the presence of at least 13
bat species in the Cretan shrublands. The most common and abundant
species in this habitat type were Hypsugo savii, Pipistrellus kuhlii and
Tadarida teniotis. Elevation had no significant effect on the activity of these
species or the overall activity of bats. Additionally, more species were
recorded in the 500 m zone, but no significant elevational trend was
detected. The lack of a clear altitudinal pattern in bat activity contrasts the
results of other studies reporting low or mid elevation peaks. A possible
explanation for this differentiation is that in this study only one habitat
type was investigated. Since several bat species have specialized foraging
habits, their activity in a certain mountain range shall be affected by the
succession of the main rock and (correspondingly) habitat types across the
elevational gradient. Roosting sites availability is expected to be of
particular importance, since in many species most of the activity takes
place around their refugia in a radius of few kilometers. Uneven
distribution of bat roosts is probably related to the high heterogeneity of
bat activity in each elevation zone which was observed here.
All the available records related to the presence of bats on Idi Mts and
Crete in general (from feeding grounds and roosts) were analysed, in order
to assess the altitudinal distribution on bats. The number of species was
highest in mid to low elevation areas (below 1000 m in Idi, below 600 m in
Crete) and decreased with elevation. This trend is probably related to the
decrease of habitat variability as elevation increases, although sampling
effort in higher areas was limited. The ability of many bat species to travel
very long distances (several kilometers) in a few hours and switch roosts
seasonally obviously inflates their altitudinal distribution, a fact that has
been largely neglected in the relevant studies. Unfortunately, it was not
possible to inspect all possible hibernation and swarming roosts, due to
time restrictions in this study and practical difficulties, since these sites are
expected to be located in high elevation areas, surrounded by snow during
winter.
The foraging habitat preferences and winter activity of bats in Greece and
the Mediterranean in general have been poorly studied so far. In this study
the feeding activity of Cretan bats was investigated in five different habitat
types (olive groves, oak forests, shrublands, villages and rivers).
Echolocation calls of bats were recorded during summers and winters in
five sites in each habitat type, along 2 Km transects. Recorded calls were
identified to species or species group level with the use of Discriminant
function Analyses. Oak forests and rivers support a high number of species
(at least 12 species each), when at least 10 species were recorded in olive
groves. Oak forests and rivers support higher numbers of bats, mainly due
to the high abundance of P. kuhlii (in rivers) and P. hanaki and H. savii (in
forests). These species, together with Tadarida teniotis are the more
common and abundant in the examined habitat types.
The overall activity of bats (all species in all habitat types) was significantly
reduced and a smaller number of species was recorded during winters, in
comparison with summers. The activity of P. hanaki, P. kuhlii and H. savii
was also significantly reduced in winters, while no significant difference
was detected in the activity of Tadarida teniotis. The greatest seasonal
fluctuation in foraging and commuting activity was recorded in oak forests
and shrublands, presumably due to the high elevation of most sites of
these habitat types. Similarly, in the shrublands of Idi Mts the activity of
bats was lower in autumns and less species were recorded in comparison
with springs (in the 500 m zone and in the sum of sites sampled). Since no
significant differences in the ambient temperature of shrublands in springs
and autumns were recorded, we can assume than the annual cycle in the
activity of Cretan bats is synchronized not only by climate, but also by other
factors, not identified so far. The results of this study demonstrate that in
Crete the foraging activity of some bat species (namely P. hanaki, P. kuhlii
and H. savii) exhibit a clear seasonal pattern with reduced activity in
autumns and winters, although ambient temperatures in these seasons
rarely fell below 10 0C. Seasonality in foraging activity was less prominent
in T. teniotis, but also in the three rhinolophids and M. schreibersii. The
latter four species are cave dwellers and the relatively high temperatures
of low elevation caves probably allow them to maintain in a state of
homeothermy at a reduced cost. On the contrary, the only known winter
colonies of Myotis blythii/oxygnathus, another cave dweller, are situated in
high elevation caves with low temperature. As already mentioned, this is
the only species that truly hibernates on Crete, according to our best
knowledge.
In conclusion, the findings of this study not only provide some insight to
the distribution, the echolocation calls and the ecology of Cretan bats, but
can be used as a starting point for their conservation through the
protection of their roosts and the effective management of their key
foraging habitats.
|
Search
