The first coordinated trans-North African mid-winter waterbird census: The contribution of the International Waterbird Census to the conservation of waterbirds and wetlands at a biogeographical level

Biological Conservation 206 (2017) 11–20

Contents lists available at ScienceDirect

Biological Conservation journal homepage: www.elsevier.com/locate/bioc

The first coordinated trans-North African mid-winter waterbird census: The contribution of the International Waterbird Census to the conservation of waterbirds and wetlands at a biogeographical level Sayoud M.S. a, Salhi H. b, Chalabi B. c, Allali A. d, Dakki M. e, Qninba A. e, El Agbani M.A. e, Azafzaf H. f, Feltrup-Azafzaf C. f, Dlensi H. f, Hamouda N. f, Abdel Latif Ibrahim W. g, Asran H. g, Abu Elnoor A. g, Ibrahim H. g, Etayeb K. h, Bouras E. i, Bashaimam W. j, Berbash A. i, Deschamps C. k, Mondain-Monval J.Y. l, Brochet A.L. k, Véran S. k, Defos du Rau P. l,⁎ a

Centre Cynégétique de Réghaia (C.C.R.), BP 54/02 Réghaia, 16112 Alger, Algeria Direction Générale des Forêts, 11 Chemin Doudou Mokhtar, Ben Aknoun, 16000 Alger, Algeria Université d'El Tarf, BP 73, 36000, El-Tarf, Algeria d Conservation des Forêts de la Wilaya de Naama, 45000, Algeria e Institut Scientifique, Université Mohammed V de Rabat, Av. Ibn Battota, Morocco f Association ‘Les Amis des Oiseaux’ (AAO / BirdLife Tunisie), Immeuble ERIS, Bureau N°4 au 2ème étage, 14 Rue Ibn El Heni, 2080 Ariana, Tunisia g Egyptian Environmental Affairs Agency, 30 Misr/Helwan El Zyrae Road, PO 11728, El Maadi Helwan, Egypt h Zoology Dept., Tripoli University, PO Box: 13227, Tripoli, Libya i Environment General Authority, Ganzor Algheran, PO Box 13793, Tripoli, Libya j Libyan Society for Birds, P.O. Box 81417, Tripoli, Libya k Institut de Recherche de la Tour du Valat, Le Sambuc, 13200 Arles, France l Office National de la Chasse et de la Faune Sauvage, Unité Avifaune Migratrice, Le Sambuc, 13200 Arles, France b c

a r t i c l e

i n f o

Article history: Received 1 June 2016 Received in revised form 23 November 2016 Accepted 3 December 2016 Available online xxxx Keywords: Waterbird North Africa Wetlands Ramsar convention International waterbird census Sampling design

a b s t r a c t The International Waterbird Census (IWC) is one of the most widespread biodiversity monitoring programs, assessing waterbird populations in the framework of several international agreements including the African-Eurasian migratory Waterbirds Agreement and the Ramsar Convention. In 2013, the IWC was coordinated for the first time across the whole of North Africa with the aim of developing recommendations for methodological improvements to current design in North Africa, as well as to update the conservation status of certain waterbird populations and wetlands of international importance. We show that coordinating the IWC across all five North African countries significantly improves knowledge of waterbird population sizes and distribution and confirmed that current North African Ramsar wetlands perform well in conserving waterbirds. Nevertheless, biodiversity conservation could potentially be further enhanced by designating additional Ramsar sites, which this study contributes to identifying. We show that reducing sampling effort by half over the entire region would have been sufficient to cover 100% of the species richness of wintering waterbirds recorded and N 98% of the total abundance. Finally, we show that larger wetlands are insufficiently sampled. These findings call for revised sampling design in a coordinated, region-wide framework. The maintenance, optimization and reinforcement of the IWC program over time on a regional scale, with the collected data made available in a shared database, seems essential in order to make appropriate conservation decisions for waterbirds and wetlands in the future. Adding a temporal dimension to these analyses will be critical to confirm the patterns observed in the 2013 census. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Large-scale datasets on the Earth's biodiversity are today increasingly useful for answering important ecological or conservation questions of global concern (Pereira et al., 2013). For example, two recent studies ⁎ Corresponding author. E-mail address: [email protected] (P. Defos du Rau).

http://dx.doi.org/10.1016/j.biocon.2016.12.005 0006-3207/© 2016 Elsevier Ltd. All rights reserved.

use large-scale biodiversity datasets: one indicates that although the effectiveness of birds as biodiversity surrogates can be improved by using other taxa, birds alone perform fairly well (Larsen et al., 2012). The other shows that increased monitoring efforts, notably of waterbirds, are required in the southern half of the Mediterranean Basin despite being one of the most important biodiversity hotspots in the world (Galewski et al., 2011). Waterbird monitoring is one of the oldest, largest and most internationally widespread environmental surveillance

12

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

systems. The monitoring of waterbirds has become of even greater global concern as the role of waterbirds in wetland ecosystems is increasingly understood. Waterbirds are today recognized as important ecosystem service providers, acting for instance as a biological proxy for the condition of wetlands, as dispersers of seeds or as sentinels of potential epidemics (Amat and Green, 2010; Green and Elmberg, 2014). The long-term monitoring of waterbirds is managed on a global scale by Wetlands International (www.wetlands.org) through the International Waterbird Census (IWC). The IWC was launched in 1967 and today covers over 25,000 sites in N100 countries, making it one of the largest global monitoring schemes based largely on citizen science. The objectives of the IWC are (i) to estimate and monitor the size of waterbird populations, (ii) to describe changes in the number and distribution of waterbirds and (iii) to assess the importance of individual wetlands for waterbirds through synchronized counts (Dodman and Diagana, 2003 for Africa, Wetlands International, 2012 for the whole African-Eurasian region). These results are used to inform international agreements especially the EU Birds Directive, the Convention on Biological Diversity, the Conservation of Arctic Flora and Fauna, the Ramsar Convention on Wetlands and the African-Eurasian migratory Waterbirds Agreement (AEWA) and may bring about new policies and regulatory measures regarding the conservation or use of these populations within AEWA Contracting Parties (AEWA Secretariat, 2013). With a potential 2 million wintering waterbirds belonging to N150 different species (a total reached by adding up historical maximum annual counts for each of the five countries: Wetlands International, 2014) and despite hosting only 1.5% of the total surface of wetlands worldwide (Mediterranean Wetlands Observatory, 2012), North Africa (a region comprising Morocco, Algeria, Tunisia, Libya and Egypt) is of major importance for waterbirds of the African-Eurasian flyways (Meininger and Atta, 1994; Dakki et al., 2002; Isenmann et al., 2005; Samraoui and Samraoui, 2008; EGA - RAC/SPA waterbird census team, 2012). During both autumn and spring migrations, North African wetlands provide a last refuelling stopover for millions of migratory waterbirds before they cross the Sahara or the Mediterranean Sea (Green et al., 2002); both crossings can be energetically demanding. These wetlands also shelter hundreds of thousands of waterbirds during the winter. Conserving and managing these waterbird populations and the North African wetlands hosting them requires among other actions filling in various knowledge gaps on species' population sizes and distribution patterns on a region-wide scale (Samraoui et al., 2011; Galewski et al., 2011). The first IWC in North Africa took place in the 1960s, and the census has continued to develop since then. IWC became more regular from 1983 in Morocco, from 1985 in Algeria and from 2002 in Tunisia but for political and/or financial reasons the spatial coverage of wetlands has been variable, depending on the country (EGA - RAC/SPA waterbird census team, 2012; Etayeb et al., 2015; Wetlands International, 2014). Moreover, censuses were always conducted at a national level without region-wide coordination or joint analysis, although nomadic exchanges among migratory waterbird populations between North African countries are strongly suspected (Samraoui et al., 2011), of course in addition to even larger migratory exchanges between North and South Mediterranean. This has meant that only general overviews and trends could be drawn from results and datasets, restricting knowledge about regional waterbird distribution and impacting decision-making on conservation in North Africa. In 2013, for the first time, waterbird counts were coordinated jointly between all five North African countries. They were conducted in January with extended spatial coverage (with the exception of Libya where coverage remained limited and 10 important wetlands were skipped due to security issues). The results were then analyzed at a trans-North African scale to improve knowledge on wintering population sizes and on the wetlands preferentially used within this region. This article presents the results of this first region-wide coordinated survey along four lines of analysis: (i) Methodological improvement: How

could coordinated streamlining of the current IWC sampling design at a region-wide scale improve IWC cost-effectiveness in a resourceconstrained context? (ii) Assessment of North-African wetlands conservation policies through evaluation of the relative importance of sites for waterbirds: Do sites currently protected under conservation status host more waterbirds than others? (iii) Spatial distribution patterns of waterbirds: Could any environmental predictors help to evaluate and improve monitoring design and to develop hypotheses for future spatial analyses of waterbird distributional ecology? and (iv) Reassessment of population size estimates: how influential coordinated waterbird counts summed over North-Africa could be in updating latest published information on these population sizes from this region? 2. Methods 2.1. The 2013 North African IWC This first coordinated trans-North African waterbird count took place between 1 January and 9 February 2013 at 411 sites (Fig. 1). Because IWC ideally takes place around 15 January, the coordinated effort of all teams of observers resulted in N86% of the 411 wetlands being surveyed between 5 and 25 January, most of the teams generally covering several wetlands in one field session or even one day (1). All teams included experienced observer(s) who had been trained specifically for IWC in their country by national and/or international training programs. All wetlands were then positioned on a geographical information system (Quantum GIS Development Team, 2015). Just over half the sites (215 sites, 52.3%) were located in Algeria, and the others were located in Tunisia (88 sites, 21.4%), Morocco (61 sites, 14.8%), Egypt (31 sites, 7.5%) and Libya (16 sites, 3.9%) (1). Despite important logistical, administrative and, sometimes, security constraints as well as very limited financial and material resources, each site was visited once, generally in one day from land (two wetlands were surveyed by boat and one over two days) by a small team of observers (mean = 3 observers, SE = 2). Each team applied the field protocol for waterbird counting recommended by Wetlands International (2010). The count data were obtained by scanning multi-species flocks of waterbirds with telescopes or binoculars as appropriate and counting individuals of each species, or by estimating species-specific abundance using ‘blocks’ of known size in the appropriate order of magnitude. Only waterbird genera to which the AEWA applies (AEWA Secretariat, 2015) are considered in this article. 2.2. Relative importance of sites for waterbirds The most widespread and commonly accepted international conservation treaty devoted to wetlands is the Ramsar Convention on Wetlands of International Importance. Nine criteria are used to classify a wetland as a site of international importance under the Ramsar Convention, two of which are based on waterbirds (Ramsar Convention Secretariat, 2016): a wetland should qualify for Ramsar designation if it regularly supports (1) N20,000 waterbirds, or (2) N 1% of a biogeographical waterbird population. These 1% thresholds are provided by the 5th Waterbird Population Estimates (WPE5) from the online database (wpe.wetlands.org) which defines a waterbird population as a distinct assemblage of individuals which does not experience significant emigration or immigration (Wetlands International, 2016). We compared waterbird abundance in officially designated Ramsar sites with abundance in non-Ramsar sites (Kleijn et al., 2014). We scored the sites according to the Waterbird Conservation Value (WCV): this index “sums the ratio of each species' abundance to its published 1% threshold across all observed species” (Harebottle and Underhill, 2016). WCV thus provides in a single score an overall measure of the value of a given wetland based on its waterbird abundance and species richness. Waterbird abundance and species richness did not follow a normal (but a rather right-skewed) distribution over the

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

13

Fig. 1. The geographical location of the 411 sites monitored in the 2013 trans-North African IWC. 2-column fitting image.

sampled wetlands. In order to compare relevance of Ramsar criteria and WCV score in rating potential for waterbirds and thus assessing potential conservation status, we compared waterbird abundance and species richness between Ramsar criteria and WCV score using non-parametric Mann-Whitney tests. 2.3. Spatial distribution patterns of the waterbird community Spatial autocorrelation (SA) is a frequently found pattern in species distribution data, i.e. locations close to each other exhibit more similar or dissimilar values than would be predicted by chance alone (Dormann et al., 2007). Wetland use by waterbirds could therefore display some SA (as a wetland is more or less likely to be occupied by waterbirds depending on its habitat types and conditions, as well as on whether neighboring wetlands are themselves more or less occupied by waterbirds). In our study, SA of waterbird abundance was assessed by Moran's I (Rangel et al., 2010) using the coordinates of each sampled wetland. When no clear SA could be detected, we took this to indicate that local habitat and environmental factors are the most important drivers of waterbird distribution across North Africa. Since a primary predictor of abundance and species richness is a habitat's surface area (de Goeij et al., 1992), we controlled for surface area by respectively modeling the log-transformed density of waterbirds (i.e. abundance divided by wetland surface area) and the log-log species–area relationship (Lomolino, 2000; Paracuellos and Tellería, 2004). We modeled the distribution of waterbird density in North Africa using five covariates (country, longitude, altitude, distance from the coast, and distance from the nearest urban agglomeration) in order to provide recommendations for optimizing the sampling design of the IWC as well as to develop hypotheses for future spatial analyses of North African waterbird macro-ecological patterns. These models of waterbird density were developed on the sub-sample of 154 wetlands of known surface area. The following a priori hypotheses governed our choice of covariates: • Country: The field experience of a country's team of ornithologists, site accessibility and/or the political situation in the country could affect both the sampling and the detection processes, and, as a result, the estimates of waterbird abundance on a national level. • Longitude: As the study region mainly extends along an east–west gradient, longitude was used as a proxy for biogeographical and macro-ecological factors (including climate) that affect waterbird abundance.

• Altitude and distance from the coast: Both these covariates were used to compare waterbird abundance between different wetland ecosystems, particularly low-altitude lagoons (mainly coastal) and inner plateau salt marshes (chotts) or reservoirs in mountainous areas. In particular, the combination of low altitude and proximity to the coast was used as a proxy for the threatened ecosystems of Mediterranean lagoons (Ayache et al., 2009; Hüttich et al., 2012). • Distance from the nearest urban agglomeration: This was log-transformed and used as a proxy both for monitoring accessibility and the disturbance and/or pollution of wetlands, hence this covariate could potentially negatively or positively impact the detection of waterbirds. The effects of these large-scale factors on waterbird communities were assessed through Generalized Linear Models using appropriate link function and response probability distribution. The competing models were ranked through a model-selection procedure using Akaike information criteria (AIC). 2.4. Reassessment of population size estimates Species-specific total counts were then analyzed at a trans-North African scale to improve knowledge on flyway-specific wintering population sizes. For each waterbird species, there are generally one to two flyways encompassing biogeographical parts of North-Africa as defined by the Waterbird Population Estimates online database (Wetlands International, 2016) and CSN tools (Wings Over Wetlands Project, 2010) which provide a population estimate for each flyway-specific populational unit. The complete dataset of the 2013 census was provided to Wetlands International in order to update the Waterbird Population Estimates online database (Wetlands International, 2016). To assess whether the 2013 census had improved our knowledge of waterbird population sizes, we compared the fractions represented by the species-specific counts of the 2013 North African census in the estimates of the sizes of the different biogeographical populations published in the Waterbird Population Estimates online database before (5th edition of the Waterbird Population Estimate or WPE5) and after the 2013 census (6th edition of the Conservation Status Report or CSR6). Irrespective of the non-surveyed range of these populations, if this percentage was found to be higher than 30%, this indicated that the North African region was of major importance for this species and that a potentially serious reassessment of its biogeographical population

14

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

size, either upward or downward, could be expected from WPE5 to CSR6 following the present work. Finally, in order to confirm that waterbird distribution at the North African scale had not been exceptionally affected by a continentalscale effect, we checked that weather in northern Mediterranean this winter 2012–2013 was neither particularly warm nor cold (University of Nebraska–Lincoln, 2016). 3. Results 3.1. Methodological improvements A total of 1,074,501 waterbirds were counted in the 2013 North African census; 38.8% (387,730) in Morocco, 29.8% (319,875) in Tunisia, 23.6% (253,380) in Algeria, 10.1% (108,222) in Egypt, and 0.5% (5294) in Libya. The cumulative abundance of birds counted in the different sites, ranked by ascending order, presents a saturation plateau (Fig. 2): the 11 wetland sites with the highest abundance represent alone 50.3% of the total abundance of birds at the 411 sites (Table 1). The 100 wetlands with the highest abundance cumulate 90.4% of the total abundance, and the 200 wetlands with the highest abundance contribute 98.3% of the total abundance. These 200 sites include 40% of the 215 Algerian sites, 26% of the 31 Egyptian sites, 44% of the 16 Libyan sites, 72% of the 61 Moroccan sites, and 64% of the 88 Tunisian sites. Thus the 200 wetlands of highest waterbird abundance surveyed in 2013 are found in all five countries, with the highest proportion in Morocco, Algeria and Tunisia. A total of 115 species were recorded across all sites. The site with the highest species richness was the 40-km section of the Nile River from Aswan to Kom Ombo (Egypt), where 63 species were recorded. On average, 10.6 ± 9.8 species were recorded per site. The cumulative curve of species richness (Fig. 3) shows that the 200 sites of highest waterbird abundance contributed 112 species out of the 115 recorded across the whole region. In light of this, monitoring only 203 sites over the entire region (the top 200 with 112 species, plus the three sites with the remaining three species) would have been sufficient to cover all the species richness recorded across the 411 surveyed wetlands, and N98% of the total abundance of wintering waterbirds.

3.2. Relative importance of sites for waterbirds Of the 411 wetlands monitored in the 2013 trans-North African census, 135 are formally part of designated Ramsar sites, but in this census we found that only 63 potentially fulfilled one or both of the Ramsar criteria concerning waterbirds (i.e. a wetland regularly supporting over 20,000 waterbirds, or N1% of a biogeographical population). In contrast, 42 non-designated Ramsar sites potentially fulfilled one or both of these Ramsar criteria (Table 2, 1). Of the 11 wetlands with the highest waterbird abundance in 2013 (Table 1), nine were Ramsar sites (as well as having other types of conservation status), six were Important Bird and Biodiversity Areas (IBAs: www.birdlife.org/datazone/site), and one had no conservation status. When comparing the distribution of waterbird abundance in the 135 Ramsar sites with the 276 non-Ramsar ones, Ramsar sites have, as would be expected, a distribution of abundance classes skewing towards sites with high abundance (Mann-Whitney test, Z = − 5.793, p b 0.001). Median abundance for Ramsar sites is 1499 waterbirds (inter-quartile range, or IQR = 4117), while for non-Ramsar sites it is only 203 waterbirds (IQR = 758). The distribution of waterbird abundance across the 42 non-Ramsar sites potentially meeting Ramsar waterbird criteria (median = 1633 waterbirds, IQR = 4337) does not significantly differ from the distribution of waterbird abundance in the current Ramsar sites (Fig. 4, Mann-Whitney test, Z = − 1.528 p = 0.13). The present census found that 63 out of the 135 Ramsar sites (47%) met the Ramsar waterbird criteria in 2013 (Table 2). Considering that not all of these 135 Ramsar sites were originally designated according to waterbird criteria, these results suggest that current North African Ramsar sites remain among the wetlands hosting the most waterbirds in the region. In terms of Ramsar site identification, the 105 sites (63 Ramsar sites and 42 non-Ramsar sites) that in our census met the standard Ramsar criteria (median = 3017 waterbirds, IQR = 5934) as well as the 135 sites with the highest WCV scores (median = 2582 waterbirds, IQR = 3931) had higher waterbird abundance than current Ramsar sites. Indeed, 53% of the latter did not meet Ramsar criteria in 2013 and included many sites with a low number of waterbirds (1). Abundance distributions of the 105 sites meeting the Ramsar waterbird criteria and the 135 sites with the highest WCV scores are significantly more skewed towards higher abundance than the abundance distributions of the current Ramsar sites (Mann-Whitney tests: Z = − 4.045

Fig. 2. The cumulative abundance of waterbirds counted (left) and the cumulative proportion of waterbirds counted (right) over the 411 surveyed wetlands (ordered by increasing waterbird abundance). single-column fitting image.

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

15

Table 1 The 11 sites with the highest waterbird abundance based on data from the 2013 trans-North African IWC. Site ID

WIa site code

Site name

Country

Conservation statusb

Waterbird abundance

Proportion of cumulative abundance (%)

1 2 3 4 5 6 7 8 9 10 11

MA00148 TN00132 TN00111 TN00216 EG00013 DZ00079 MA00022 TN00188 EG00185 DZ00093 EG00008

Merja Zerga Lac Ichkeul Îles Kneïss Sebkhet Sejoumi Lake Qarun La Grande Sebkha d'Oran Baie d'Ad-Dakhla Sebkhet Ariana Nile - Aswan to Kom Ombo Lac de Telamine Lake Burullus

Morocco Tunisia Tunisia Tunisia Egypt Algeria Morocco Tunisia Egypt Algeria Egypt

NP/Ramsar NP/Biosphere reserve/Ramsar/IBA NR/IBA/Ramsar Ramsar/IBA Ramsar/IBA Ramsar Ramsar (No status) IBA Ramsar Ramsar/IBA

196,178 66,865 40,189 39,813 38,101 36,834 30,760 26,357 24,281 22,311 18,320

18.3 24.5 28.2 31.9 35.5 38.9 41.8 44.2 46.5 48.6 50.3

a

WI: Wetlands International. NP: National Park, NR: National Reserve, IBA: Important Bird and Biodiversity Area – IBA are discrete areas of high avian biodiversity value, selected according to a set of internationally recognized criteria, partly similar to the Ramsar ones but encompassing all bird species and not only wetland-related taxa. They are generally not protected legally as such but they can provide the information basis for future legal protection (Beresford et al., 2011). b

p b 0.001 and Z = −3.944 p b 0.001 respectively, Fig. 4). These results suggest that many wetlands in North Africa would qualify for Ramsar status, but are not yet designated as such. In terms of identifying wetlands of international importance, using either standard Ramsar criteria or WCV scores resulted in the selection of sites of very similar abundance and species richness distribution (Mann-Whitney tests, Z = 0.692, p = 0.49 and Z = 0.916, p = 0.36 respectively, this time using only the 105 sites with the highest WCV scores in order to compare both criteria with an equal sample size). Although conceptually appealing, the WCV score used on its own did not result in a significant improvement over the standard Ramsar criteria in selecting wetlands of potential interest for waterbirds in the case of the 2013 trans-North African IWC. 3.3. Ecological features of the sites and the spatial distribution patterns of waterbirds Abundance and density showed only very weak SA (Moran's I b 0.04 for both), indicating that these are spatially independent and likely to be driven by local environmental factors, possibly at the wetland or landscape scales. An assessment of the effects of the selected site-specific geographical or ecological factors on log-transformed waterbird density was therefore of interest and was conducted by a GLM with a Poisson distribution and a log link. Following a standard model-selection procedure (Table 3), the most supported model (i.e. with lowest AIC) showed a reasonable absolute fit (ĉ = 1.33) and had the least covariates among models with ΔAIC b 2; its parameters were thus retained for inference (Table 4). Egypt and Tunisia were predicted to be the countries with

the highest waterbird density in the surveyed wetlands, whereas Algeria and Libya were predicted to have the lowest density. The distance of a surveyed wetland from the coast and from the nearest urban agglomeration were negatively linked with the density of waterbirds. Thus, these factors affecting observed densities point at sampling and detection issues: waterbird density seems to vary primarily between national IWC schemes and to be greater in more accessible areas, i.e. closer to urban agglomerations and coasts (Table 4). All these factors suggest an accessibility bias in the sampling of larger and/or more remote wetlands, because of likely limited road networks surrounding and/or reaching them. For the subsample of wetlands of known surface area, the slope of the log-log species–area relationship was not significantly different from 0 (n = 154, r2 = 0.023, slope = 0.063 [− 0.002, 0.129], p = 0.059), indicating a likely downward bias in species richness in larger wetlands. Both these results corroborate the apparent insufficient sampling of larger and thus less accessible North African wetlands. 3.4. Reassessment of population size estimates Of the 115 species observed and counted in the 2013 trans-North African IWC, the eight species most represented proportionally to their published population size estimates remained the same before (WPE5) and after (CSR6) the 2013 census (Table 5, 2). Among those eight species for which North Africa is thus of major importance, two are globally threatened (IUCN, 2015): the marbled teal (Marmaronetta angustirostris) and the white-headed duck (Oxyura leucocephala). The North African fraction of those eight species changed to a large extent (Table 5) between WPE5 and CSR6 because of the reassessment of their published biogeographical population size estimates in the CSR6 (Wetlands International, 2016) on the basis of the present 2013 transNorth African IWC dataset. Among those eight species showing strong evolution in their North African fraction, two gained importance in North Africa, suggesting a decline in their overall population and the six others decreased in proportion relatively to their overall population size estimate which was increased using the present 2013 trans-North African IWC dataset (Table 5). 4. Discussion

Fig. 3. Cumulative species richness of waterbirds in the 411 wetlands surveyed (ordered by increasing waterbird abundance). single-column fitting image.

The IWC scheme aims at a global coordination. Nevertheless this coordinated region-wide IWC was particularly desirable because of the importance of this region's wetlands for migratory waterbirds moving seasonally between sub-Saharan Africa and Eurasia which the present total count of 1,074,501 wintering waterbirds still probably largely underestimates. Our results demonstrate that coordinating the IWC across the five North African countries of Morocco, Algeria, Tunisia, Libya and

16

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

Table 2 Sizes of the different categories of wetlands surveyed for the 2013 trans-North African IWC with respect to their Ramsar designation and their fulfillment of Ramsar waterbird criteria.

Current Ramsar sites Others Total

Meeting Ramsar criteria on waterbirds

Not meeting Ramsar criteria on waterbirds

Total

63 42 105

72 234 306

135 276 411

Egypt would potentially significantly add to our knowledge of waterbird population sizes and distribution for the following reasons:

4.1. Methodological improvement: How can coordination at a region-wide scale improve monitoring?

- it could potentially prove more cost-effective than in the past because concurringly reducing the sampling effort by half over the entire region would have been sufficient to cover the total species richness recorded in the census and N98% of the total abundance of wintering waterbirds - it improved inference power on waterbird population dynamics and distribution by increasing the sample size of monitored wetlands: in the 2013 census, 411 sites were covered thanks to the coordination efforts (out of the 739 sites surveyed at least once since 1967), whereas before this, no more than 200 sites were monitored in the same year (Wetlands International, 2014). Thus, the 2013 IWC was by far the largest waterbird sampling effort ever conducted in one year in North Africa. - it allowed to confidently reevaluate the population size of two globally threatened species thanks to the simultaneity of the coordinated field surveys - thanks to the improved sample size aggregated across the five countries, it allowed to assess that larger and/or remote wetlands are insufficiently inventoried and sampled.

Beyond the mere benefit that regional coordination can result in a synchronous estimation of a number of waterbird populations in a given year, the plateau of cumulative abundance reached with only the 200 sites with the highest waterbird numbers (Fig. 2) shows that coordinating the IWC across North Africa (and possibly across international regions elsewhere e.g. van Roomen et al., 2014; van Roomen et al., 2015) can help to optimize monitoring design through concerted effort of all five countries to decide on a sample of most important wetlands. The 200 wetlands with the highest waterbird counts (out of a total of 411 surveyed wetlands) covered 98.3% of the overall surveyed populations. Thus, the remaining 211 sites contributed b 2% of the overall surveyed populations. The top 11 most important sites (Table 1) contributed to over 50% of the overall surveyed populations. In the resource-limited context of biodiversity monitoring schemes in North Africa, these results show the clear potential of reallocating overall IWC field effort by decreasing the sampling of less important sites and increasing sampling in the most important sites. Obviously, the design of such a reallocation cannot be based solely on one year of data; variability in waterbird abundance in the different sites would need to be assessed in the longer term with further full-scale IWC and historical datasets in order to confirm the sound identification of the most and least important sites. In addition to abundance, the distribution of species richness showed that most of the species (112 out of 115) were observed and counted in these 200 most important sites (Fig. 3). Only three sites of less importance hosted three species not found in any of the major sites. This reinforces

The results also show that the current North African network of Ramsar wetlands performs well in conserving the number of waterbirds of international significance, but that waterbird and wetland conservation could potentially be further enhanced by designating additional Ramsar sites that this study could contribute to identifying.

Fig. 4. Distribution of waterbird abundance classes in the 135 surveyed Ramsar wetlands (black), the 42 wetlands meeting Ramsar waterbird criteria but not currently designated as Ramsar (grey), the 135 wetlands with the highest WCV scores (striped), and the 105 wetlands meeting Ramsar criteria in 2013 (dotted). single-column fitting image.

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

17

Table 3 Model selection for GLM of log-transformed wintering waterbird density from the 2013 trans-North African IWC. Covariates included in the GLM Country Country Country Country Country Country Country Country

Distance to town Distance to town Distance to town Distance to town Distance to coast

Distance to coast Distance to coast Distance to coast

Longitude Longitude

Longitude Altitude

the conclusion that reallocating IWC sampling effort (again, after further full-scale IWC in the medium-term future and with historical datasets) would allow adequate sampling of the waterbird community. Rare species would need to be accounted for when reallocating the monitoring effort, by additionally sampling those sites regularly hosting rare species that may not be adequately sampled in the most important sites. These results lead us to advocate the coordination of future IWC sampling efforts in North Africa in order to reallocate and optimize monitoring; this seems highly desirable and would be readily feasible, assuming further full-scale IWC are first coordinated and analyzed in the medium-term future and long-term IWC datasets are made fully available and shared. However, as a prerequisite to any such optimized IWC design, all wetlands in North Africa would need to be inventoried and their waterbird communities sampled, including through further exhaustive IWC, so as to confidently identify the most important sites in terms of species abundance or richness. We thus strongly recommend that complete wetland inventories and further full-scale IWC (at least 3 to 5) are carried out in North Africa as it is likely that not all major wetlands have yet been identified. For example, Lake Nasser in Egypt was only fully sampled for waterbirds as recently as January 2015, yet it certainly qualifies as one of the most important sites in North Africa, with an estimated 100,000 wintering waterbirds. 4.2. Relative importance of sites for waterbirds: Do sites currently under conservation status host more waterbirds than others? Each of the five North African countries is a contracting party to the Ramsar Convention, thus some of the region's most significant wetland complexes are designated as Ramsar sites. Indeed the Ramsar convention is possibly the most efficient, if not the only, international instrument devoted to wetlands conservation in North Africa. These five countries have committed to taking the necessary steps to maintain their ecosystem components, processes and benefits/services, including by backing Ramsar designation with other protection status or by imposing a ban on hunting and some restriction of disturbance in the great majority of Ramsar sites (Dakki et al., 2011; Karem et al., 2007; Direction générale des forêts, 2004, https://rsis.ramsar.org/ris). Our 2013 census found, as would be expected, that North African Ramsar wetlands hosted a higher density of waterbirds than non-Ramsar wetlands, either because they were designated due to the high abundance in waterbirds or because their conservation status has helped to conserve or restore waterbird populations, as shown in Morocco (Dakki et al., 2011; Kleijn et al., Table 4 Parameter estimates and effects size of the most supported GLM of log-transformed wintering waterbird density from the 2013 trans-North African IWC.

Intercept Tunisia Libya Egypt Algeria Distance to coast Distance to town

Parameter

SE

Wald test

p

1.798 0.067 −0.153 0.216 −0.344 −0.001 −0.123

0.149 0.076 0.148 0.103 0.099 0.001 0.045

145.665 20.965

0.000 0.000

5.757 7.335

0.016 0.007

Altitude

AIC

ΔAIC

514.014 514.575 515.093 518.436 519.240 525.348 526.004 527.342

0 0.561 1.079 4.422 5.226 11.334 11.990 13.328

2014). However, no less than 42 non-Ramsar wetlands met one or both Ramsar criteria concerning waterbirds (hosting over 20,000 waterbirds or over 1% of a biogeographical population) in 2013; these sites could potentially qualify for Ramsar designation; it would therefore be of conservation interest to monitor them in a more targeted way in the near future in order to confirm their potential international importance. In particular, 10 non-Ramsar wetlands in Morocco and 22 in Algeria met Ramsar criteria in 2013, illustrating the importance of both of these countries for waterbirds and their international potential for wetland and waterbird conservation. Thus, although current Ramsar wetlands in North Africa are already doing a good job of hosting waterbirds in numbers of international significance, conservation of North African waterbird populations could be further enhanced. A further improvement would be to assess whether the 72 Ramsar wetlands that did not meet any waterbird criteria in 2013 have lost some of their waterbird biodiversity on a permanent basis or were originally designated as Ramsar sites for other criteria than the waterbird ones (Ramsar Convention Secretariat, 2016). As a site monitoring tool, IWC could thus be useful to monitor and evaluate the state of Ramsar sites but also, more generally of any wetlands of heritage value. This site monitoring dimension of IWC should be promoted further in a trans-North African framework to strengthen or complement the IBA monitoring, of strong relevance in Africa (Beresford et al., 2011). On the basis of our analysis of the 2013 trans-North African IWC, the recently proposed criteria of the WCV score (Harebottle and Underhill, 2016), despite its appealing integrated scope, would not select wetlands richer in waterbirds (in either abundance or species richness) than the standard Ramsar criteria and thus would not improve identification of sites of international importance. However, again, this result is based only on one year of data and will need confirmation in the longer term. 4.3. Ecological features of the sites and the spatial distribution patterns of waterbirds: Could any ecological or environmental predictors help to improve current IWC design? The apparent downward bias in species richness for larger wetlands could be indicative of insufficient sampling of species richness (and thus abundance) in the field: comprehensive waterbird censuses are more difficult or limited in larger wetlands, which are often partly invisible and/or inaccessible. In order to avoid missing any waterbird species, a further recommendation for North African IWC sampling design would be to increase sampling efforts on large wetlands relative to smaller wetlands; North Africa has a number of huge wetlands, including the Merja Zerga in Morocco, the Great Sebkha of Oran and many other vast chotts and sebkhas in Algeria, the Chott El Djerid in Tunisia, the Tawargha Sebkha in Libya, and the Nile River in Egypt. The lack of SA displayed by waterbird density distribution across North African wetlands is counterintuitive at this continental scale for supposedly vagile or dispersive species such as migratory birds (Koenig, 1999). It is likely to be indicative of the relative importance of local, sitespecific environmental factors that affect waterbird density. In other words, waterbird density in a given wetland seems to depend primarily on the condition of the wetland's local habitat more than on any largerscale biotic or abiotic factor. The lack of SA in waterbird density would

18

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

Table 5 The percentage of the estimated biogeographical population (based on the Waterbird Population Estimates from Wetlands International) calculated from counts in the 2013 trans-North African IWC. Only species that represented N30% of their estimated population are presented. Species

Population

Fulica cristata Marmaronetta angustirostris Tadorna ferruginea Oxyura leucocephala Phoenicopterus roseus Aythya nyroca Tadorna tadorna Charadrius hiaticula

Spain & Morocco West Mediterranean/West Medit. & West Africa North-west Africa Algeria & Tunisia West Mediterranean West Mediterranean/North & West Africa Black Sea & Mediterranean Hiaticula, Northern Europe/Europe & North Africa

Number counted

Pop. size estimate

Algeria Egypt Libya Morocco Tunisia total count WPE5 200 4297 818 35,034 881 24,437 720

thus indicate that waterbirds would not preferentially use wetlands neighboring a favored wetland. Due to their local habitat features, some wetlands retain larger numbers of waterbirds; protection of these mostused wetlands is thus a priority as there is nothing that indicates that the waterbirds they host would switch to neighboring wetlands if they were degraded or reclaimed. This underlines the ecological and natural heritage value of the most waterbird-rich wetlands in North Africa, particularly in the context of environmental impact assessments (Weller, 1988; Dolman and Sutherland, 1995) and the mitigation hierarchy, as the reclamation or degradation of one wetland might not be compensated by the protection of neighboring wetlands. The main factors affecting observed densities relate to sampling and detection issues, as waterbird density in North Africa appears to vary primarily between countries and is greater closer to urban agglomerations and coasts (Table 4). All these factors suggest an accessibility bias in the sampling of wetlands, since wetlands near better road networks, i.e. closer to cities or coasts, or in smaller countries, are more easily and better surveyed. Could there be other hypotheses than accessibility to explain the higher waterbird densities in wetlands located closer to cities or coasts? For example, wetlands near cities or coasts might have higher organic productivity and thus provide more food resources for some waterbird families like e.g. larid or ardeid. However, our field knowledge of the difficulties to access large and remote wetlands leads us to favour the sampling bias hypothesis because: a) it is consistent with the observed downward bias in species richness of larger wetlands, which could be due, for example, to the general lack of peripheral access. b) there are also a number of reasons why waterbirds would avoid vicinity of urban and coastal areas: disturbance, hunting, poaching, pollution, reclamation. This result corroborates the apparent insufficient sampling of larger and thus less accessible North African wetlands. These findings, revealing inadequate sampling of waterbird species richness and abundance in North Africa, point to the necessity of revising IWC sampling design so it is carried out in a coordinated region-wide framework. We recommend as a first step a better inventory of larger and/or remote wetlands, especially in countries where they have not yet benefitted from comprehensive and recent exploration: mainly in Algeria (Samraoui and Samraoui, 2008), Libya (EGA - RAC/SPA waterbird census team, 2012), Egypt, and also the large chotts in central Tunisia. As such wetlands are typically huge, barely accessible and do not flood every year, remote sensing techniques might be useful before planning field surveys. Such an inventory of larger and/or remote wetlands should be considered a high priority as it would likely improve regional knowledge of population size of wintering waterbirds in the region and thus potentially modify some of the conclusions of this work. 4.4. Reassessment of population size estimates One straightforward way to illustrate the importance of coordinating a simultaneous IWC at a trans-North African scale is to compare

18

1575

31

1877 3755 1567 3755 801 9215 19,381

2114 74 386 61,276 658 17,810 79

1877 6087 5938 1204 100,065 2340 53,068 20,180

5000 4000 3000 500 132,500 2500 120,000 73,000

2013 North African %

CSR6

%WPE5 %CSR6

938 6750 7000 1600 150,000 6000 150,000 62,100

37.5 152.2 197.9 240.8 77.0 93.6 44.2 27.6

200.2 90.2 84.8 75.3 66.7 39.0 35.4 32.5

two consecutive current estimates of flyway-specific population sizes (as published before 2013 in WPE5 and after 2013 in CSR6: Wetlands International, 2016) for those waterbirds species a large fraction of which winters in North Africa. For eight species (Table 5), major changes occurred in their published biogeographical population size estimates, thus highlighting the added-value of the 2013 trans-North African IWC dataset and the coordination behind it. Four species (marbled teal, ferruginous duck (Aythya nyroca), ruddy shelduck (Tadorna ferruginea), white-headed duck) had as many or more individuals counted in the 2013 trans-North African IWC than the corresponding total WPE5 population estimates. Of these four species, two are globally threatened (IUCN, 2015). Along with two other species (greater flamingo Phoenicopterus roseus and common shelduck Tadorna tadorna), this study contributed to updating their population size and distribution and thus demonstrating improvement of their conservation status. Contrary to those six species whose conservation status improved, two species demonstrated a marked decline in conservation status, since their North African fraction was increased using the present 2013 transNorth African IWC dataset: common ringed plover (Charadrius hiaticula) and red-knobbed coot (Fulica cristata). In particular, the latter species' critical conservation status in Spain (Martínez-Abraín et al., 2007; Martínez-Abraín et al., 2011) contributed to decrease its population size estimate as published in the CSR6 and thus to confer an even greater importance to the North African fraction of this endangered population. In addition, the recommendations previously put forward for improving sampling design would benefit the monitoring of these threatened species. A time-series dataset to investigate the change in count data over time for these species would be necessary in order to disentangle confounding factors to determine whether these populations have increased (or decreased) or, as is equally possible, they were underestimated (or overestimated) in the past. 5. Conclusion This 2013 trans-North African IWC was by far the largest waterbird survey ever conducted in mid-winter in North Africa but its resulting total count of 1,074,501 waterbirds still probably largely underestimates the true importance of North African wetlands for waterbirds. The findings of the 2013 trans-North African IWC highlighted a knowledge gap on waterbird population sizes and distribution across North Africa, and demonstrated how a coordinated, simultaneous, redesigned monitoring program can provide new and valuable information to help fill this gap. Redesigning the current trans-North African IWC sampling would be more cost-effective, as this study shows there is a potential to reduce the sampling effort by half over the entire region if the observed waterbird distribution pattern is beforehand confirmed on the longterm. This work also shows that the current North African network of Ramsar wetlands performs well in conserving waterbirds, but that waterbird and wetland conservation could potentially be further enhanced by designating additional Ramsar sites that this study could contribute

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

to identify. Finally, our analysis allowed to reevaluate the population size of two globally threatened species, and also to determine that larger and/or remote wetlands remain insufficiently inventoried and sampled. The maintenance, optimization and reinforcement of this census on a trans-North African scale, with collected data made available in a shared database, seem essential in guiding conservation management decisions through international cooperation on waterbird populations and wetlands. Adding a time-trend dimension to these analyses (e.g. analyzing population trends over the last 10 or 20 years) will be a key requirement in confirming the patterns observed in 2013 and addressing inter-annual variability in waterbird numbers and distribution. In particular, inter-annual water levels can vary dramatically in North African wetlands due to precipitations, directly and largely affecting the distribution and number of wintering waterbirds. Funding This work, including data collection and analysis for the 2013 transNorth African IWC, was supported by the MAVA Fondation pour la Nature(11/001) and the French Ministry in charge of Environment (Ministère de l'Environnement, de l'Energie et de la Mer) (2101208162 for 2013 & 2101398404 for 2014). The following are the supplementary data related to this article. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.biocon.2016.12.005. Acknowledgements We would like to thank all the field observers who participated in the 2013 trans-North African IWC, and who made this study possible; their names are listed in the supplementary materials (3). We are also grateful to Laura Dami and Marie Suet for their help with data management, Arnaud Béchet and four anonymous reviewers for comments on previous drafts of this paper and Elise Bradbury for editing the English.We would like to thank all the field observers who participated in the 2013 transNorth African IWC, and who made this study possible; their names are listed in the supplementary materials (3). We are also grateful to Laura Dami and Marie Suet for their help with data management, Arnaud Béchet and four anonymous reviewers for comments on previous drafts of this paper and Elise Bradbury for editing the English. References AEWA Secretariat, 2013. Report of the 5th Session of the Meeting of the Parties (MOP5) to the African-Eurasian Migratory Waterbird Agreement (AEWA), 14–18 May 2012, La Rochelle, France. AEWA Secretariat, 2015. Agreement Text and Annexes as amended at the 6th Session of the Meeting of the Parties (MOP6) to the African-Eurasian Migratory Waterbird Agreement (AEWA), 9–14 November 2015, Bonn, Germany. Amat, J.A., Green, A.J., 2010. Waterbirds as bioindicators of environmental conditions. In: Hurford, C., Schneider, M., Cowx, I. (Eds.), Conservation Monitoring in Freshwater Habitats. Springer, Netherlands. Ayache, F., Thompson, J.R., Flower, R.J., Boujarra, A., Rouatbi, F., Makina, H., 2009. Environmental characteristics, landscape history and pressures on three coastal lagoons in the Southern Mediterranean Region: Merja Zerga (Morocco), Ghar El Melh (Tunisia) and Lake Manzala (Egypt). Hydrobiologia 622, 15–43. Beresford, A.E., Buchanan, G.M., Donald, P.F., Butchart, S.H.M., Fishpool, L.D.C., Rondinini, C., 2011. Poor overlap between the distribution of protected areas and globally threatened birds in Africa. Anim. Conserv. 14, 99–107. Dakki, M., Qninba, A., El Agbani, M.A., Benhoussa, A., 2002. Recensement hivernaux d'oiseaux d'eau au Maroc: 1996–2000. Travaux de l'Institut Scientifique, Rabat. Série Zoologie. 45. Dakki, M., El Agbani, M.A., Qninba, A., 2011. Zones humides du Maroc inscrites jusqu'en 2005 sur la liste de la Convention de Ramsar. Trav. Inst. Sci., Rabat, Sér. Générale. 7. de Goeij, P.J., Van der Have, T.M., Keijl, G.O., Van Roomen, M.W.J., Rutters, P.S., 1992. The network of wetlands for waterbird migration in the eastern Mediterranean. In: Finlayson, C.M., Hollis, T., Davis, T. (Eds.), Managing Mediterranean Wetlands and Their Birds. Proceedings of an IWRB International Symposium, Grado, Italy. Vol. 20. IWRB special publication, pp. 70–72. Direction Générale des forêts, 2004. Atlas [IV] des zones humides Algériennes d'importance internationale. République Algérienne démocratique et populaire. Ministère de l'Agriculture et du Développement rural.

19

Dodman, T., Diagana, C.H., 2003. African waterbird census/Les Dénombrements d'Oiseaux d'eau en Afrique 1999, 2000 & 2001. Wetlands International Global Series 16, Wageningen. Dolman, P.M., Sutherland, W.J., 1995. The response of bird populations to habitat loss. Ibis 137, 38–46. Dormann, C.F., McPherson, J.M., Araújo, M.B., Bivand, R., Bolliger, J., Carl, G., Davies, R.G., Hirzel, A., Jetz, W., Kissling, W.D., Kühn, I., Ohlemüller, R., Peres-Neto, P.R., Reineking, B., Schröder, B., Schurr, F.M., Wilson, R., 2007. Methods to account for spatial autocorrelation in the analysis of species distributional data: a review. Ecography 30, 609–628. EGA - RAC/SPA waterbird census team, 2012. Atlas of Wintering Waterbirds of Libya, 2005–2010. Imprimerie COTIM, Tunisia. Etayeb, K.S., Berbash, A., Bashimam, W., Bouzainen, M., Galidan, A., Saied, M., Yahia, J., Bourass, E., 2015. Results of the eighth winter waterbird census in Libya in January 2012. Biovers. J. 6, 253–262. Galewski, T., Collen, B., McRae, L., Loh, J., Grillas, P., Gauthier-Clerc, M., Devictor, V., 2011. Long-term trends in the abundance of Mediterranean wetland vertebrates: from global recovery to localized declines. Biol. Conserv. 144, 1392–1399. Green, A.J., Elmberg, J., 2014. Ecosystem services provided by waterbirds. Biol. Rev. 89, 105–122. Green, A.J., El Hamzaoui, M., El Agbani, M.A., Franchimont, J., 2002. The conservation status of Moroccan wetlands with particular reference to waterbirds and to changes since 1978. Biol. Conserv. 104, 71–82. Harebottle, D.M., Underhill, L.G., 2016. Assessing the value of wetlands to waterbirds: exploring a population-based index at flyway and regional levels. Ostrich 87, 1–15. Hüttich, C., Reschke, J., Keil, M., Dech, S., Weise, K., Beltrame, C., Fitoka, E., Paganini, M., 2012. Using the Landsat Archive for the Monitoring of Mediterranean Coastal Wetlands: Examples from the Glob-Wetland-II Project. (accessed 13.09.15). http:// www.earth-zine.org/2011/12/20/using-the-landsat-archive-for-the-monitoring-ofmediterranean-coastal-wetlands-examples-from-the-globwetland-ii-project/. Isenmann, P., Gaultier, T., El Hili, A., Azafzaf, H., Dlensi, H., Smart, M., 2005. Birds of Tunisia. Société d'études ornithologiques de France, Paris. IUCN, 2015. The IUCN Red List of Threatened Species. Version 2014.3. (accessed 09.04.15). www.iucnredlist.org. Karem, A., Faouzi, M., Fekih Salem, A.R., Smart, M., 2007. Les 20 sites Ramsar de Tunisie: Zones humides d'importance internationale. République Tunisienne. Ministère de l'Agriculture et des Ressources hydraulique, Direction Générale des forêts. Kleijn, D., Cherkaoui, I., Goedhart, P.W., van der Hout, J., Lammertsma, D., 2014. Waterbirds increase more rapidly in Ramsar-designated wetlands than in unprotected wetlands. J. Appl. Ecol. 51, 289–298. Koenig, W.D., 1999. Spatial autocorrelation of ecological phenomena. Trends Ecol. Evol. 14, 22–26. Larsen, F.W., Bladt, J., Balmford, A., Rahbek, C., 2012. Birds as biodiversity surrogates: will supplementing birds with other taxa improve effectiveness? J. Appl. Ecol. 49, 349–356. Lomolino, M.V., 2000. Ecology's most general, yet protean pattern: the species-area relationship. J. Biogeogr. 27, 17–26. Martínez-Abraín, A., Viedma, C., Bartolomé, M.A., Gómez, J.A., Oro, D., 2007. Hunting sites as ecological traps for coots in southern Europe: implications for the conservation of a threatened species. Endanger. Species Res. 3, 69–76. Martínez-Abraín, A., Regan, H.M., Viedma, C., Villuendas, E., Bartolomé, M.A., Gómez, J.A., Oro, D., 2011. Cost-effectiveness of translocation options for a threatened waterbird. Conserv. Biol. 25, 726–735. Mediterranean Wetlands Observatory, 2012. Biodiversity — Status and Trends of Species in Mediterranean Wetlands 1. Tour du Valat, Arles, France. Meininger, P.L., Atta, G.A.M., 1994. Ornithological studies in Egyptian wetlands 1989/ 1990. FORE Report 94-01/WIWO Report 40 (Vlissingen/Zeist, The Netherlands). Paracuellos, M., Tellería, J.L., 2004. Factors affecting the distribution of a waterbird community: the role of habitat configuration and bird abundance. Waterbirds 27, 446–453. Pereira, H.M., Ferrier, S., Walters, M., Geller, G., Jongman, R.H.G., Scholes, R.J., Bruford, M.W., Brummitt, N., Butchart, S.H.M., Cardoso, A.C., Coops, N.C., Dulloo, E., Faith, D., Freyhof, J., Gregory, R., Heip, C., Höft, R., Hurtt, G., Jetz, W., Karp, D., McGeoch, M., Obura, D., Onoda, Y., Pettorelli, N., Reyers, B., Sayre, R., Scharlemann, J., Stuart, S., Turak, E., Walpole, M., Wegmann, M., 2013. Essential biodiversity variables. Science. 339:pp. 277–278. http://dx.doi.org/10.1126/science.12299. Quantum GIS Development Team, 2015. Quantum GIS Geographic Information System. Open Source Geospatial Foundation Project. (accessed 10.06.15). http://qgis.osgeo. org/. Ramsar Convention Secretariat, 2016. An Introduction to the Convention on Wetlands. Ramsar Convention Secretariat, Gland, Switzerland. Rangel, T.F.L.V.B., Diniz-Filho, J.A.F., Bini, L.M., 2010. SAM: a comprehensive application for spatial analysis in macroecology. Ecography 33, 46–50. Samraoui, B., Samraoui, F., 2008. An ornithological survey of Algerian wetlands: important bird areas, Ramsar sites and threatened species. Wildfowl 58, 71–96. Samraoui, F., Alfarhan, A.H., Al-Rasheid, K.A., Samraoui, B., 2011. An appraisal of the status and distribution of waterbirds of Algeria: indicators of global changes? Ardeola 58, 137–163. University of Nebraska–Lincoln, 2016. School of Natural Resources. (accessed 4.09.16). http://snr.unl.edu/lincolnweather/data/Year2012Global-Anomalies.asp. van Roomen, M., van Winden, E., Langendoen, T., 2014. The Assessment of Trends and Population Sizes of a Selection of Waterbird Species and Populations From the Coastal East Atlantic Flyway for Conservation Status Report 6 of the African Eurasian Waterbird Agreement. Wadden Sea Flyway Initiative, Wetlands International & Birdlife International, Wageningen, Ede, The Netherlands.

20

M.S. Sayoud et al. / Biological Conservation 206 (2017) 11–20

van Roomen, M., Nagy, S., Foppen, R., Dodman, T., Citegetse, G., Ndiaye, A., 2015. Status of Coastal Waterbird Populations in the East Atlantic Flyway. With Special Attention to Flyway Populations Making Use of the Wadden Sea. Programme Rich Wadden Sea, Leeuwarden, The Netherlands, Sovon, Nijmegen, The Netherlands, Wetlands International, Wageningen, The Netherlands, BirdLife International, Cambridge, United Kingdom &, Common Wadden Sea Secretariat, Wilhelmshaven, Germany. Weller, M.W., 1988. Issues and approaches in assessing cumulative impacts on waterbird habitat in wetlands. Environ. Manag. 12, 695–701. Wetlands International, 2010. Guidance on Waterbird Monitoring Methodology: Field Protocol for Waterbird Counting. (accessed 5.10.12). wpe.wetlands.org. Wetlands International, 2012. International Waterbird Census Count Totals 2009–2012: African-Eurasian Region. (accessed 4.02.15). www.wetlands.org.

Wetlands International, 2014. IWC Online Database. (accessed 3.09.16). http://iwc. wetlands.org. Wetlands International, 2016. Waterbird Population Estimates. (accessed 3.09.16). wpe. wetlands.org. Wings Over Wetlands Project, 2010. Critical Site Network Tool. UNEP-WCMC, BirdLife International and Wetlands International, Cambridge, UK and Wageningen, The Netherlands (accessed 4.09.16) csntool.wingsoverwetlands.org/csn/default.html#state= home.