Wintering waterbird assemblages in the central part of the Gulf of Gabès in southern Tunisia

Abstract

Although the Gulf of Gabès is recognised as an Important Bird Area, several aspects of the ecology of waterbirds inhabiting this area still need to be investigated. We observed how waterbird foraging guilds varied among habitats. In total, 49 species belonging to 16 families were recorded. Winter visitors accounted for 73% of counts. The avifauna was dominated by shorebirds (52% of records), followed by large wading birds (25%), open-water birds (18%) and waterfowl (5%). However, the structure of local waterbird communities and their composition in terms of foraging guilds varied according to habitat type. The bird communities of sandy beaches were dominated by open-water birds, whereas large mudflats were dominated by shorebirds, and wadi estuaries showed a more even representation of foraging guilds. Locally rare species had narrow distributions, whereas locally abundant species were found widely within the gulf. For four species the 1% population level criterion of the Ramsar Convention was exceeded. Overall, our results show that the Gulf of Gabès hosts important numbers of waterbirds with different ecological requirements, which confirms the importance of this gulf as a wintering area for Palearctic waterbirds.

Full text

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=tost20
Download by: [197.3.216.143] Date: 05 October 2016, At: 21:25
Ostrich
Journal of African Ornithology
ISSN: 0030-6525 (Print) 1727-947X (Online) Journal homepage: http://www.tandfonline.com/loi/tost20
Wintering waterbird assemblages in the central
part of the Gulf of Gabès in southern Tunisia
Foued Hamza, Abdessalem Hammouda & Slaheddine Selmi
To cite this article: Foued Hamza, Abdessalem Hammouda & Slaheddine Selmi (2016):
Wintering waterbird assemblages in the central part of the Gulf of Gabès in southern Tunisia,
Ostrich, DOI: 10.2989/00306525.2016.1207721
To link to this article: http://dx.doi.org/10.2989/00306525.2016.1207721
Published online: 05 Oct 2016.
Submit your article to this journal
View related articles
View Crossmark data

Ostrich 2016: 1–7
Printed in South Africa — All rights reserved
Copyright © NISC (Pty) Ltd
OSTRICH
ISSN 0030-6525 EISSN 1727-947X
http://dx.doi.org/10.2989/00306525.2016.1207721
Ostrich is co-published by NISC (Pty) Ltd and Taylor & Francis
This is the nal version of the article that is
published ahead of the print and online issue
Wintering waterbird assemblages in the central part of the Gulf of Gabès in
southern Tunisia
Foued Hamza*, Abdessalem Hammouda and Slaheddine Selmi
Département des Sciences de la Vie, Faculté des Sciences de Gabès, Université de Gabès, Gabès, Tunisia
* Corresponding author, email: Fouedhamza2010@gmail.com
Although the Gulf of Gabès is recognised as an Important Bird Area, several aspects of the ecology of waterbirds
inhabiting this area still need to be investigated. We observed how waterbird foraging guilds varied among habitats.
In total, 49 species belonging to 16 families were recorded. Winter visitors accounted for 73% of counts. The
avifauna was dominated by shorebirds (52% of records), followed by large wading birds (25%), open-water birds
(18%) and waterfowl (5%). However, the structure of local waterbird communities and their composition in terms
of foraging guilds varied according to habitat type. The bird communities of sandy beaches were dominated by
open-water birds, whereas large mudflats were dominated by shorebirds, and wadi estuaries showed a more even
representation of foraging guilds. Locally rare species had narrow distributions, whereas locally abundant species
were found widely within the gulf. For four species the 1% population level criterion of the Ramsar Convention was
exceeded. Overall, our results show that the Gulf of Gabès hosts important numbers of waterbirds with different
ecological requirements, which confirms the importance of this gulf as a wintering area for Palearctic waterbirds.
Rassemblements hivernaux chez les oiseaux aquatiques de la zone centrale du Golfe de Gabes en
Tunisie méridionale
Bien que le Golfe de Gabes soit reconnu comme une zone importante pour la conservation des oiseaux, de
nombreux aspects de l’écologie des oiseaux d’eau qui y vivent restent peu connus. Ainsi, l’objectif de ce travail
était de décrire la diversité de l’avifaune hivernante dans ce golfe et d’étudier la variation de sa composition, en
termes de guildes alimentaires, entre les différents types d’habitat qu’abrite cette région. Au total, 49 espèces
appartenant à 16 familles ont été recensées. Les hivernants représentent 73 % des effectifs enregistrés. Cette
avifaune est dominée par les limicoles (52 % des effectifs enregistrés) suivis par les grands échassiers (25 %),
les laridés (18 %) et les canards (5 %). Cependant, la structure des communautés locales et leurs compositions
en termes de guildes varient en fonction de d’habitat. C’est ainsi que les avifaunes des plages sableuses sont
dominées par les laridés, celles des vasières par les limicoles, alors que les estuaires des oueds abritent une
avifaune plus équilibrée. Les espèces localement rares ont une répartition restreinte, tandis que les espèces
localement abondantes ont une large distribution dans le golfe. D’autre part, pour quatre des espèces recensées, le
critère de 1 % de la Convention de Ramsar est atteint. Globalement, nos résultats montrent que le Golfe de Gabes
abrite un effectif important d’oiseaux d’eau ayant des affinités écologiques variées, ce qui confirme l’importance de
ce golfe pour l’avifaune Paléarctique.
Keywords: abundance, foraging guilds, Gulf of Gabès, Tunisia, waterbirds
Understanding the ecological factors affecting waterbird
diversity is a major question in avian ecology, as well as
for conservation purposes (Danufsky and Colwell 2003;
Granadeiro et al. 2007; Liordos 2010). In recent years there
has been a notable decline in many waterbird populations
around the world (Stroud et al. 2004; BirdLife International
2008; Wetlands International 2012). This decline is particu-
larly severe in the Black Sea/Mediterranean flyway, given
that approximately 65% of shorebirds populations have
decreased in this flyway (International Wader Study Group
2003). Initiation of conservation planning in these areas
requires an understanding of the relationships between
waterbird diversity and habitat features, especially in the
most important areas, such as the Gulf of Gabès, in south-
eastern Tunisia (van Dijk et al. 1986; Isenmann et al. 2005).
The Gulf of Gabès offers a mosaic of wetlands and
coastal habitats, including small sandy beaches, large
mudflats and wadi estuaries. Several previous studies
have shown the ornithological significance of this gulf
(e.g. van Dijk et al. 1986; Spiekman et al. 1993; Bos et al.
2001; Hamdi et al. 2008). However, most investigations
focused on monitoring waterbirds inhabiting the largest
mudflats, whereas less attention has been paid to waterbird
assemblages in the other types of coastal habitats. In
recent related works, we have shown that habitat features
and human presence played important roles in shaping
Introduction

Hamza, Hammouda and Selmi2
waterbird richness and the abundances of individual
waterbird species (Hamza and Selmi 2015; Hamza et al.
2015). However, for a better knowledge of this avifauna
there is a need to investigate further aspects of its diversity.
This paper aims first to provide more information on the
diversity of this avifauna, by describing its composition in
terms of families, migratory status, conservation status and
foraging guilds. Second, we investigate how guild composi-
tion (functional diversity) of waterbird assemblages varies
among the three main gulf habitats.
Material and methods
Study area
Located in south-eastern Tunisia, the Gulf of Gabès
extends from Ras Kaboudia to Elbibane lagoon. It has
the largest tides in the Mediterranean, with tidal amplitude
ranging between 0.8 and 2 m in neap and spring tides,
respectively (Pérez-Domingo et al. 2008). Salinity ranges
from 37.5 to 39.3, and temperature from 13.2 to 26.5 °C
(Ktari-Chakroun and Azouz 1971; Bradaï and Capapé
2001). This gulf is considered one of the most productive
ecosystems in the Mediterranean region (Papaconstantinou
and Farrugio 2000). It is also classified as an Important Bird
Area (Fishpool and Evans 2001) and includes 10 Ramsar
sites, of which the Kneïs islands is a natural reserve (https://
rsis.ramsar.org).
Three main coastal habitats can be distinguished in this
area: (1) sandy beaches with narrow and steep intertidal
zones, (2) large and slightly steep mudflats crossed by
shallow tidal channels and (3) wadi estuaries including one
or more streams crossing saltmarshes and mudflats.
Our work was conducted in the central part of the Gulf
         
          
the south, covering a total of 125 km of coastline (Figure 1).
Fifty study sites, approximately 2.5 km apart, covering the
three habitat types were selected for waterbird surveys
(Figure 1). The mean site area (2 ± 0.4 km²) was calculated
using limits of each sampled site defined by landmarks,
such as tidal channels, sandbanks and saltmarshes.
Data collection
Waterbirds were surveyed from 10 December 2012 to
31 January 2013. This period corresponds with the peak
abundance of wintering waterbirds. Five surveys were
carried out at different dates, so that each site was visited
five times at approximately one-week intervals. All survey
visits were carried out in the morning, during low tide, and
only under suitable meteorological conditions. During each
visit, the observer (F Hamza) was positioned at an elevated
post close to the shoreline and conducted repeated counts
of birds during 30 min, by using binoculars and occasion-
ally a telescope. The highest recorded number of birds
of each detected species was retained. Each survey
represented a counting time of 25 h (30 min × 50 sites) so
that a total sampling effort of 125 h (25 h × five surveys)
was conducted.
The recorded species were classified according to
their migratory status in three categories based on our
prior field experience and on relevant information in the
literature, notably those in Isenmann et al. (2005): resident
(R), regular wintering (RW) and irregular wintering (IW).
A species was considered resident if it could be seen
throughout the year, with evidence of breeding. Regular
wintering species are migratory species that over-winter
and leave at the beginning of the spring. Irregular wintering
species are uncommon species seen occasionally during
winter in the study area.
We also grouped the recorded species into four foraging
guilds according to feeding niche, foraging behaviour
and habitat preferences (Perrins and Ogilvie 1998):
shorebirds, wading birds, waterfowl and open-water birds.
The shorebird guild included tactile surface-foraging
waders (Haematopodidae, Recurvirostridae, Charadriidae,
Scolopacidae and Rallidae, except Eurasian Coot
Fulica atra). Large wading birds comprised long-legged
species that forage on fish and invertebrates in shallow
water (Ardeidae, Threskiornithidae, Ciconiidae and
Phoenicopteridae). The waterfowl guild included swimming
birds (dabblers and divers) that generally feed in wetland
basins (Anatidae, Podicipedidae, Phalacrocoracidae and
Eurasian Coot). Open-water birds are fish-eating and
scavenging seabirds that often catch fish at sea (Laridae,
Sternidae and Procellariidae).
Tunisia
AFRICA
LIBYA
ALGERIA
TUNISIA
Tunis
Sfax
065 130 km
36° N
34° N
32° N
8° E
10° E
GULF
OF GAB ÈS
GULF
OF GABÈ S
S1
S50
Kneïs Islands
Djerba
Island
Boughrara
lagoon
MEDITERRANEAN SEA
Figure 1: Map of the Gulf of Gabès showing the location of study sites

Ostrich 2016: 1–7 3
Finally, the conservation status of each recorded species
was determined according to BirdLife International (2014).
Data analyses
For each recorded species, we determined the rate of site
occupancy as the number of sites where the species was
recorded divided by the total number of sampled sites
(n = 50 sites). We also determined the abundance of each
species in each site as the mean number of individuals
recorded at that site over the five survey visits. The mean
wintering population of each species in the entire study area
was then calculated as the sum of mean local abundances
in each of the 50 sites. We also converted species local
abundances into density values by dividing them by the
areas of the corresponding sites (birds km²). The relation-
ship between average local density and occupancy was then
investigated using Spearman’s correlation coefficient test.
For each site we determined the density of each
foraging guild by summing the local densities of species
belonging to that guild. We also measured the diversity
of the corresponding local waterbird assemblage using
the Shannon index: H =Σ[(ni/Σni)*log2(ni/Σni)], where
ni is the density of guild i in the considered site. The
obtained data were then used to investigate how waterbird
assemblages varied among the three habitats by means of
the non-parametric Kruskal–Wallis test. The Mann–Whitney
test was also used as a post-hoc test for pair-wise compari-
sons between habitats. We also used the c2 test to compare
the relative abundances of the four studied guilds among
the three sampled habitats. In the latter test, data from sites
belonging to the same habitat were pooled together.
All statistical tests and analyses were conducted using
SAS version 8 software (SAS Institute 1999). Means ± one
standard error are reported throughout the manuscript.
Results
Waterbird diversity
A total of 49 species belonging to 16 families were recorded
(Table 1). The families with the highest numbers of species
were Scolopacidae (25% of species), Anatidae (16%)
and Laridae (12%). The most abundant families were
Scolopacidae (41% of birds), Phoenicopteridae (21%)
and Laridae (16%). The most abundant species were the
Dunlin Calidris alpina and Greater Flamingo Phoenicopterus
roseus, which accounted together for 48% of birds recorded.
The great majority of species (47%) had narrow distribu-
tions, as they occurred in less than 20% of sites. The most
widespread species were the Slender-billed Gull Larus genei
and Kentish Plover Charadrius alexandrinus, which occurred
in more than 80% of sites. Moreover, locally abundant
species tended to occur widely while locally rare species
tended to be more localised, as shown by the positive
correlation between the average local density (calculated
across occupied sites) and occupancy (Figure 2).
Concerning migratory status, 11 species were recognised
as resident, 35 were regular wintering species and the three
remaining species were irregular wintering species and
accounted for negligible numbers.
With regard to conservation status, the studied avifauna
was mainly composed of species of Least Concern (47 of
    
species, the Eurasian Curlew Numenius arquata, which
occurred in a large number of sites (68% of sampled sites)
with relatively high numbers (1 054 individuals, 3% of birds).
We also recorded one Vulnerable species, the Marbled Teal
Marmaronetta angustirostris, which was observed in only
one site during a single visit, with a low number (two pairs).
Moreover, for four species, the 1% level criterion of the
Ramsar Convention was exceeded for relevant populations:
Kentish Plover Charadrius alexandrines (3.17%), Eurasian
Spoonbill Platalea leucorodia (4.16%), Greater Flamingo
Phoenicopterus roseus (5.24%) and Slender-billed Gull
Larus genei (1.36%) (Table 1).
Variation in guild composition among habitats
Considering all sampled sites together, we found that the
studied avifauna was dominated by shorebirds, in terms
of both species number (43% of species) and abundance
(52% of birds). Large wading birds were second in
abundance (25%), although they had the lowest number
of species (12%). Open-water birds ranked third in terms
of both species number (18%) and abundance (18%),
whereas the waterfowl guild ranked second in species
number (27%) but last in terms of abundance (5%).
Using the Kruskal–Wallis test, we found that the strength
and direction of the relationship between bird density and
habitat varied among guilds. There were significant habitat
effects on open-water birds, wading birds and waterfowl, but
no significant effect was recorded on shorebirds (respec-
tively: H2 = 20.92, P < 0.0001; H2 = 11.94, P = 0.003;
H2 = 14.58, P = 0.0007; H2 = 0.69, P = 0.71; Figure 3).
Results of the Mann–Whitney test showed that open-water
birds occurred more abundantly in sandy beaches
compared to large mudflats and estuaries, whereas the
difference between the latter habitats was non-significant
(Figure 3a). Waterfowl were significantly more abundant
in wadi estuaries compared to large mudflats and sandy
beaches, with a non-significant difference between the latter
habitats (Figure 3b). For wading birds, significant differences
were found between pairs of habitats, with wadi estuaries
ranked first, followed by large mudflats (Figure 3c).
Waterbird communities of sandy beaches were
dominated by open-water birds, whereas communities
in large mudflats and wadi estuaries were dominated by
shorebirds, with wadi estuaries having the most even distri-
bution among foraging guilds. This finding is also supported
by the results of the comparisons of waterbird diversity,
measured by the Shannon index, among habitats (Kruskal–
Wallis test: H2 = 17.80, P = 0.0001; Figure 4). The highest
waterbird diversity was recorded in wadi estuaries and the
lowest waterbird diversity in sandy beaches (Figure 4).
Discussion
The main goal of this work was to describe the diversity of
waterbirds wintering in the central part of the Gulf of Gabès,
with a particular focus on the variation in guild composition
among habitats. Our results first show that the studied area
hosts an important wintering waterbird community, in terms
of both species richness and abundance. This community
is clearly dominated by winter visitors. Even though our

Hamza, Hammouda and Selmi4
Family Common name Scientific name Migratory
status
Conservation
status
Occupancy
(%)
Estimated
population
size
Mean density
(birds km²)
Shorebirds
Haematopodidae Eurasian Oystercatcher Haematopus ostralegus W LC 48 738 10.423 ± 1.973
Recurvirostridae Pied Avocet Recurvirostra avosetta W LC 2 1 0.002 ± 0.002
Black-winged Stilt Himantopus himantopus R LC 4 5 0.119 ± 0.071
Charadriidae Common Ringed Plover Charadrius hiaticula W LC 42 86 1.457 ± 0.482
Grey Plover Pluvialis squatarola W LC 68 833 9.653 ± 2.015
Kentish Plover Charadrius alexandrinus* R LC 88 2094 26.099 ± 3.872
 Vanellus vanellus W LC 2 1 0.006 ± 0.006
Scolopacidae Dunlin Calidris alpina W LC 66 8739 71.550 ± 24.376
Sanderling Calidris alba W LC 44 411 33.450 ± 9.970
Little Stint Calidris minuta W LC 22 101 2.122 ± 0.590
Bar-tailed Godwit Limosa lapponica W LC 2 9 0.084 ± 0.081
Marsh Sandpiper Tringa stagnatilis W LC 58 122 1.534 ± 0.330
Spotted Redshank Tringa erythropus W LC 30 332 2.602 ± 1.202
Common Greenshank Tringa nebularia W LC 58 112 1.551 ± 0.349
Common Redshank Tringa totanus R LC 30 2211 14.840 ± 6.672
Terek Sandpiper Xenus cinereus W LC 2 1 0.003 ± 0.003
Eurasian Curlew Numenius arquata W 68 1054 10.753 ± 2.526
Ruddy Turnstone Arenaria interpres W LC 18 23 0.247 ± 0.076
Common Snipe Gallinago gallinago W LC 4 1 0.032 ± 0.019
Rallidae Common Moorhen Gallinula chloropus R LC 4 7 0.243 ± 0.152
Water Rail Rallus aquaticus W LC 2 1 0.011 ± 0.011
Waterfowl
Rallidae Eurasian Coot Fulica atra W LC 4 72 3.665 ± 2.627
Podicipedidae Great Crested Grebe Podiceps cristatus W LC 48 60 2.247 ± 0.852
Little Grebe Tachybaptus ruficollis R LC 8 20 0.992 ± 0.632
Black-necked Grebe Podiceps nigricollis W LC 30 86 2.069 ± 0.854
Phalacrocoracidae Great Cormorant Phalacrocorax carbo W LC 72 497 13.330 ± 4.509
Anatidae Eurasian Wigeon Anas penelope W LC 16 598 7.515 ± 4.152
 Anas acuta W LC 16 166 1.672 ± 0.722
 Anas clypeata W LC 10 57 1.253 ± 0.690
Gadwall Anas strepera W LC 4 12 0.649 ± 0.472
Mallard Anas platyrhynchos R LC 6 43 0.652 ± 0.428
Common Shelduck Tadorna tadorna R LC 14 37 0.371 ± 0.220
Marbled Teal Marmaronetta
angustirostris
W VU 2 1 0.046 ± 0.031
Red-crested Pochard Netta rufina IW LC 2 1 0.009 ± 0.009
Wading birds
Ardeidae Little Egret Egretta garzetta R LC 46 196 2.433 ± 0.752
Grey Heron Ardea cinerea W LC 72 379 5.153 ± 1.055
Great Egret Egretta alba W LC 34 178 0.948 ± 0.290
Ciconiidae Black Stork Ciconia nigra IW LC 2 1 0.009 ± 0.006
Threskiornithidae Eurasian Spoonbill Platalea leucorodia* W LC 58 499 3.741 ± 0.818
Phoenicopteridae Greater Flamingo Phoenicopterus roseus* W LC 56 6815 48.167 ± 8.454
Open-water birds
Laridae Slender-billed Gull Larus genei* R LC 84 2338 97.005 ± 27.584
Yellow-legged Gull Larus michahellis R LC 76 1404 68.732 ± 43.429
Herring Gull Larus argentatus IW LC 2 1 0.008 ± 0.008
Lesser Black-backed
Gull
Larus fuscus W LC 58 798 38.613 ± 10.599
Black-headed Gull Larus ridibundus W LC 34 596 47.647 ± 24.988
Mediterranean Gull Larus melanocephalus W LC 8 128 7.894 ± 5.687
Sternidae Caspian Tern Sterna caspia W LC 60 116 3.568 ± 0.955
Sandwich Tern Thalasseus sandvicensis R LC 28 272 10.490 ± 4.869
Procellariidae Cory’s Shearwater Calonectris diomedea IW LC 4 1 0.142 ± 0.101
* Species exceeding the 1% level criterion of the Ramsar Convention
Table 1: List of recorded species classified by foraging guild and family, with information on their migratory status, conservation status, site
occupancy rate, estimated population size (calculated as the mean number of recorded birds in the entire study area over the five surveys),
and mean density (±SE). Migratory status: IW = irregular wintering, R = resident, W = wintering; conservation status: LC = Least Concern,
== Vulnerable

Ostrich 2016: 1–7 5
sampling did not cover the entire gulf, we found that for
four species, namely the Eurasian Spoonbill, Greater
Flamingo, Kentish Plover and Slender-billed Gull, the 1%
level criterion of the Ramsar Convention was exceeded
for relevant populations. Moreover, the Eurasian Curlew,
      
area with relatively high numbers. These findings confirm
the status of the Gulf of Gabès as an important wintering
area for Palearctic waterbirds (van Dijk et al. 1986; Hamdi
et al. 2008), as well as an important area for the resident
Slender-billed Gull.
With regard to the patterns of species occurrence, our
results show that the majority of species had narrow distri-
butions within the gulf. They also show that locally abundant
species tended to range widely, whereas locally rare
species had narrow distributions. This trend is consistent
with one of the most recognised rules in community
ecology, namely the positive interspecific abundance–distri-
bution relationship (Hanski 1999; Gaston et al. 2000; Webb
et al. 2007).
The comparison of our survey results with historical
data on wintering waterbirds in the Gulf of Gabès, notably
those published by van Dijk et al. (1986) who also surveyed
waterbirds during the winter period, were confounded by the
differing extents of the surveys. Our surveys were carried
out in the central part of the Gulf of Gabès, whereas van
Dijk et al. (1986) also monitored peripheral areas, which
permitted them to include more large mudflats in their site
sample but did not take into account small sandy beaches.
The relationship between habitat type and bird density
showed guild-specific habitat effects, which is consis-
tent with recent findings on the determinants of wintering
waterbird richness in the same area (Hamza and Selmi
2015; Hamza et al. 2015). Waterfowl occurred more
abundantly in wadi estuaries compared to large mudflats
and sandy beaches, which can be explained by the
LOG (OCCUPANCY)
LOG (MEAN DENSITY)
−2
−1
0
1
2
3
4
5
123 4
y = 0.8432x − 0.3324
r2 = 38%, P < 0.0001
Figure 2: Relationship between mean density (expressed as
birds km² and calculated using only data from the sites where the
species was recorded) and occupancy rate (%). Both variables
were log-transformed to meet the normality condition. The
regression line and equation are provided
Z = 1.94, P = 0.02
Z = 3.77, P < 0.0001
Z = 1.51, P = 0.07
Sandy beaches Large mudflats Estuaries Sandy beaches Large mudflats Estuaries
DENSITY (birds km−2)
Z = 1.09, P = 0.27
Z = 4.24, P < 0.0001
Z = −3.54, P = 0.0002
Z = 0.28, P = 0.39
Z = 0.93, P = 0.17
Z = 0.05, P = 0.48
Z = 2.92, P = 0.001
Z = −2.46, P = 0.006
Z = 3.78, P < 0.0001
1 000
1 500
500
50
100
150
200
100
200
300
100
200
300
400
0 0
0 0
(a) (b)
(c) (d)
HABITAT
Figure 3: Densities of open-water birds (a), waterfowl (b), wading birds (c) and shorebirds (d) in the three sampled habitats

Hamza, Hammouda and Selmi6
presence of large and deep tidal channels in this habitat
providing attractive roosting and feeding sites. Further
features of wadi estuaries seem to make them more attrac-
tive to waterfowl, in particular the presence of freshwater
for preening and drinking. Moreover, in estuarine habitats
channels were vegetated with dense beds of seagrass,
which seems to be attractive to some waterfowl species,
especially the Eurasian Wigeon, the most abundant species
in this guild and known to feed on seagrass exposed during

Wadi estuaries also provided the most attractive habitat
for wading birds, which also occurred abundantly in large
mudflats. Like the wadi estuaries, large mudflats are
crossed by tidal channels and contain many tidal pools that
retain a thin layer of water at low tide (Hamza et al. 2014).
These tidal channels seem to provide a refuge to juvenile
fishes from predator fishes (Shenker and Dean 1979; Reis
and Dean 1981; Boesch and Turner 1984; Paterson and
Whitfield 2000). During low tides these small fishes provide
highly concentrated feeding opportunities for wading
birds. By contrast, sandy beaches are characterised by
their narrow and steeply inclined intertidal zones. In these
habitats, only a limited intertidal area is available to use as
a feeding ground. Moreover, tidal pools and channels are
absent and the sandy sediment does not host important
invertebrate biomass (Zwarts 1988; Yates et al. 1993;
Hamza et al. 2014), which makes this habitat less attrac-
tive to waterfowl and wading birds (Hamza and Selmi 2015;
Hamza et al. 2015).
Our results also showed that open-water birds occurred
more abundantly along sandy beaches compared to the
other habitats (Hamza and Selmi 2015). This habitat is
commonly used by local artisanal fishermen (Romdhane
1998; FAO 2011), and it seems that open-water birds
have learnt to profit from this artisanal fishing activity. For
instance, terns were commonly seen catching small fishes
stranded in the nets. Moreover, after fishing, fishermen
generally discard large quantities of small fish, crabs and
fish parts that are attractive to large numbers of gulls.
We found that local density of shorebirds did not vary
significantly among habitats. Some of the most abundant
species in this guild, such as Kentish Plover and Eurasian
Oystercatcher, are habitat generalists (Zwarts 1988;
Granadeiro et al. 2007; Liordos 2010). Moreover, this guild
included some species with different habitat requirements
and showing different abundance patterns (Hamza et al.
2015). For instance, unlike the other shorebird species, the
Sanderling occurred more abundantly on sandy beaches
(Hamza et al. 2015), where it feeds on small invertebrates
by pursuing receding waves and snatching items washed in
by tide along wave-break zones (Evans et al. 1980).
Our results also showed that waterbird diversity was
higher in wadi estuaries compared to mudflats and sandy
beaches. This is likely related to the higher diversity and
availability of food resources and micro-habitats offered
by estuaries. They also offer large and permanent water
basins, providing suitable feeding and roosting sites. They
also provided a mosaic of muddy and sandy sediments,
allowing high numbers of shorebirds with different habitat
requirements to coexist. Investigations of factors affecting
the abundance of individual species have also led to the
same conclusion (Hamza et al. 2015).
In summary, our results confirm the importance of the
Gulf of Gabès as a wintering area for Palearctic waterbirds.
They also show that small sandy beaches mainly hosted
open-water birds, whereas waterbird assemblages in wadi
estuaries were more evenly distributed across guilds.
Further investigations of intra-guild habitat partitioning and
interactions between species, such as competition and
commensalism, are needed to better understand the spatial
organization of this avifauna. It would also be necessary to
investigate the responses of waterbird species to human
activity, particularly traditional clam harvesting and fishing,
and the effect of these activities on waterbird diversity. Such
information may be of great relevance for drawing a clearer
picture about the dynamics of waterbirds wintering in the
Gulf of Gabès, as well as for conservation purposes.
Acknowledgements          
comments and suggestions about issues related to this work. We
also thank the Associate Editor, as well as Tim Dodman and one
anonymous reviewer for valuable comments on earlier versions of
the manuscript.
References
BirdLife International (ed.). 2008. Threatened birds of the world.
CD-ROM. Cambridge: BirdLife International.

at http://www.birdlife.org [accessed 14 May 2014].
Boesch DF, Turner RE. 1984. Dependence of fishery species on
salt marshes: the role of food and refuge. Estuaries 7: 460–468.

Ayache F (eds). 2001. Waterbirds in the Gulf of Gabès and other
wetlands in Tunisia – autumn 1999. Report 74 of the Working
Group on International Waterbird and Wetland Research.
Zeist: Working Group on International Waterbird and Wetland
Research.
          Mobula
mobular, dans le golfe de Gabès (Tunisie méridionale,
Méditerranée centrale). Cybium 25: 389–391.
Danufsky T, Colwell MA. 2003. Winter shorebirds communities and
tidal flat characteristics at Humbolt Bay, California. The Condor
105: 117–129.
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Sandy beaches Large mudflats Estuaries
SHANNON DIVERSITY INDEX
Z = 3.78, P < 0.0001
Z = 2.92, P < 0.002
Z = −2.35, P < 0.009
HABITAT
Figure 4: Waterbird diversity, measured by the Shannon diversity
index, in the three sampled habitats

Ostrich 2016: 1–7 7
Evans PR, Breary DM, Goodyer LR. 1980. Studies on Sanderling
    Wader Study Group Bulletin 30:
18–20
        
(ed.). 2011. Guide des engins de pêche utilisés à Ghannouch et
Akarit (golfe de Gabès, Tunisie). Malaga: FAO.
Fishpool LDC, Evans MI (eds). 2001. Important Bird Areas in
Africa and associated islands: priority for conservation. BirdLife
Conservation Series     
Cambridge: BirdLife International.
Gaston KJ, Blackburn TM, Greenwood JJD, Gregory RD, Quinn
RM, Lawton JH. 2000. Abundance–occupancy relationships.
Journal of Applied Ecology 37: 39–59.
Granadeiro JP, Santos CD, Dias MP, Palmeirim JM. 2007.
Environmental factors drive habitat partitioning in birds feeding in
intertidal flats: implications for conservation. Hydrobiologia 587:
291–302.
       Le Peuplement
des oiseaux aquatiques hivernant du Golfe de Gabès (Tunisie).
Bulletin de la Société Zoologique de France 133: 267–275.
Hamza F, Hammouda A, Chokri MA, Bechet A, Selmi S. 2014.
Distribution et abondance du flamant rose Phoenicopterus
roseus hivernant dans la zone centrale du golfe de Gabès,
Tunisie. Alauda 82: 135–142.
Hamza F, Hammouda A, Selmi S. 2015. Species richness patterns
of waterbirds wintering in the gulf of Gabès in relation to habitat
features and human presence. Estuarine, Coastal and Shelf
Science 165: 254–260.
Hamza F, Selmi S. 2015. Habitat features and human presence
as predictors of the abundance of shorebirds and wading birds
wintering in the gulf of Gabès, Tunisia. Marine Ecology Progress
Series 540: 251–258.
Hanski I (ed.). 1999. Metapopulation ecology. Oxford: Oxford
University Press.
International Wader Study Group 2003. Waders are declining
worldwide. Conclusions from the 2003 International Wader Study
Group Conference, Cádiz, Spain. Wader Study Group Bulletin
101/102: 8–12.
Isenmann P, Gaultier T, E-Hili A, Azafzaf H, Dlensi H, Smart
M (eds). 2005. Oiseaux de Tunisie. Paris: Société d’Etudes
     

Ktari-Chakroun F, Azouz A. 1971. Les fonds chalutables de
la région sud-est de la Tunisie (Golfe of Gabés). Bulletin de
l’Institut National Scientifique et Technique d’Océanographie et
de Pêche de Salammbô 2: 5–47.
Liordos V. 2010. Foraging guilds of waterbirds wintering in a
Mediterranean coastal wetland. Zoological Studies 49: 311–323.
  
seagrass beds in the northern Wadden Sea, Helgolander.
Marine Research 54: 87–94.
Papaconstantinou C, Farrugio H. 2000. Fisheries in the
Mediterranean. Mediterranean Marine Science 1: 5–18.
Paterson AW, Whitfield AK. 2000. Do shallow-water habitats
function as refugia for juvenile fishes? Estuarine Coastal and
Shelf Science 51: 359–364.
Pérez-Domingo S, Castellanos C, Junoy J. 2008. The sandy beach
macrofauna of Gulf of Gabès (Tunisia). Marine Ecology 29:
51–59.
Perrins CM, Ogilvie MA (eds). 1998. The complete birds of the
Western Palearctic CD-ROM, version 1.0. Oxford: Oxford
University Press.
Reis RR, Dean JM. 1981. Temporal variation in the utilization of an
intertidal creek by the Bay Anchovy (Anchoa mitchilli). Estuaries
4: 16–23.
Romdhane MS. 1998. La pêche artisanale en Tunisie. Évolution
des techniques ancestrales. Mélanges de l’Ecole française de
Rome, Antiquité 110: 61–80.
SAS Institute. 1999. SAS/STAT user’s guide, version 8. Cary: SAS
Institute.
Shenker JM, Dean JM. 1979. The utilization of an intertidal salt
marsh creek by larval and juvenile fishes: abundance, diversity
and temporal variation. Estuaries 2: 154–163.
Spiekman HW, Keijl GO, Ruiters PS (eds). 1993. Waterbirds in the
Kneiss area and other wetlands, Tunisia, eastern Mediterranean,
spring 1990. Report 38 of the Working Group on International
Waterbird and Wetland Research. Zeist: Working Group on
International Waterbird and Wetland Research.
          
O, Ganter B, Delany S. 2004. Status of migratory wader
populations in Africa and Eurasia in the 1990s. International
Wader Studies 15: 1–259.
van Dijk AJ, van Dijk K, Dijksen LJ, van Spanje TM, Wymenga
E (eds). 1986. Wintering waders and waterfowl in the Gulf of
Gabès, Tunisia, January–March 1984. Report 11 of the Working
Group on International Waterbird and Wetland Research.
Zeist: Working Group on International Waterbird and Wetland
Research.
       

        
Journal of Animal Ecology 76: 123–134.
Wetlands International. 2012. Waterbird population estimates:
summary report (5th edn). Wageningen: Wetlands International.
Yates MG, Goss-Custard JD, McGrorty S, Lakhani KH, Le V, Dit
Durell SEA, Clarke RT, Rispin WE, Moy I, Yates T, Plant RA,
Frost AJ. 1993. Sediment characteristics, invertebrate densities
and shorebird densities on the inner banks of the Wash. Journal
of Applied Ecology 30: 599–614.
         
Guinee-Bissau. Ardea 76: 42–55.

Associate Editor: Michel Louette