Assessment of the Threatened Status of Testudo werneri Perälä, 2001 (Testudines: Testudinidae) for the IUCN Red List
ABSTRACT
Testudo werneri is one of the smallest tortoise species in the Mediterranean region, with a historical distribution in northeastern Egypt, Sinai, and parts of the Negev desert in Israel. Populations in Egypt are already practically extinct due to anthropogenic factors affecting the habitat and collection for the pet trade. Currently, T. werneri has a national Israel Red List status of Endangered (EN, A2cd, B). The species' primary habitat, desert sands, is degrading rapidly because of a multitude of human induced factors. Other threats include unnatural predators attracted by human settlements, and collecting for food and pet trade. According to present estimates, the global population has declined by around 95% in less than three generations. The remaining T. werneri population is essentially restricted to an area of around 700 km2 or less in northwestern Negev, Israel. Ten individuals are known from Zaranik in northern Sinai of Egypt. The present population size is estimated at between 2520 and 3150 individuals depending on parameters used, of which around 1890 to 2360 tortoises would be adults. True figures may be considerably less, however. Testudo werneri is clearly threatened by extinction, potentially within decades, but more data are needed for an accurate estimate. The establishment of new nature reserves in the sands of northwestern Negev in conjunction with effective management would enhance the survival prospects of the species, assuming that most of the known threat factors can be addressed. Testudo werneri qualifies globally for listing as Critically Endangered (CR, A2abcde+3de) under current IUCN Red List Criteria; this was recommended and accepted by the IUCN in 2003.
The Negev tortoise, Testudo werneri Perälä, 2001, had not previously been listed by the IUCN Red List. This report formed the basis for its initial listing (IUCN 2003).
Taxonomy
Order: Testudines. Family: Testudinidae. Scientific name: Testudo werneri. Species authority: Perälä, 2001.
Taxonomic Comments
Previously included in Testudo kleinmanni Lortet, 1883, thus all literature before Perälä (2001), plus Bouskila (2002), apply the names T. kleinmanni Lortet, 1883 or T. leithii Günther, 1869 — a senior objective synonym of the preceding taxon (Mertens and Wermuth 1955; Loveridge and Williams 1957) — for the population east of the Nile Delta through Negev. Current nomenclature is used throughout this report with reference to older literature.
Common Names
Negev tortoise (E); tortue de Negev (F); Negev-Landschildkröte (G); tortuga de Negev (S).
Countries of Occurrence
Israel (Negev desert); Egypt (east of Nile Delta; Sinai).
Current Population Trends
Decreasing.
Major Threats
Habitat degradation (various factors), settlements, off-road traffic, unnatural predators, collecting for food and pet trade.
Current Red List Status
Endangered (A1abcd) as Testudo kleinmanni (sensu lato). Listed nationally in Israel as Endangered (EN, Acd, B) (Bouskila 2002, 2004).
Proposed (2003) Red List Status
Establish as Critically Endangered (CR, A2abcde+3de). Proposed change accepted by IUCN Red List Authority in 2002 and subsequently by IUCN (2003).
Distribution
The historical range of Testudo werneri runs along the Mediterranean coastal strip east of the Nile delta in continental Egypt, further eastwards along the Sinai Peninsula, and entering northwestern and northern Negev in Israel (Perälä 2001). Zoogeographically and by biotope affiliation, as modified from Werner (1987, 1988), the species can be characterized as a Saharo-Arabian stenoecious psammophile testudinid, occurring only or almost only in sandy biotopes. The Saharo-Arabian biogeographical zone (Fig. 1 in Werner 1987) coincides almost perfectly with the distribution of T. werneri in northeastern Africa, Sinai, and Israel.
Citation: Chelonian Conservation and Biology 5, 1; 10.2744/1071-8443(2006)5[57:AOTTSO]2.0.CO;2
All available distribution data for T. werneri, past and present, are presented below with added commentary. The transliteration of Hebrew locality names follow those published in Survey of Israel (1997).
Egypt
Known localities for T. werneri in Egypt are as follows: Suez Canal, Kantarah-il-Khastreh (Al Qanţarah); Damietta (Dumyãt); vic. of Port Said; Bir Gindali; Wadi Digla; Ismailia; El Bardawil, W to Nahal Yam settlement (= 1970s Israeli settlement in Sinai, 70 km SW of El Arish and 60 km from the Suez Canal); El Teloul; SE El Teloul; El Gora; nr. El Gora; 8 km SE El Gora; El Auja; W of El Auja; S of Kharruba; W of El Arish; 20 km SW of El Arish; El Arish; E of El Arish; El Midan; S of El Midan; Katia; Romani; Salmana; Bir El Abd; 10 km SE of Bir El Abd; E of El Nasr; Lahfan (Bir Lahfan); Wadi Hareidhin; Khabra Abu Guzoar; Wadi El Amr; SW of Zaranik; Zaranik Protected Area; E of Zaranik; El Roda islands; N of El Roda; S of El Midan; Um El Rumiat island in western Lake Bardawil (Lortet, 1883; Flower, 1933; Loveridge and Williams, 1957; Buskirk, 1985; Baha El Din, 1992; Baha El Din, 1994; Bringsøe and Buskirk, 1998; Perälä, 2001).
In addition to the above, Iverson (1992) listed T. werneri (as T. kleinmanni) from two localities in southern Sinai. According to Iverson (pers. comm., August 2002), one is based on Siebenrock (1909) who cited the species from “Halbinsel Sinai, Tor” (= El Tor = at-Tūr in SW Sinai), the other locality further east, around Mt. Sinai, is based on Iverson's personal communications with J. Buskirk. Siebenrock's (1909) account is traceable to a specimen in the herpetological collection of the Vienna Natural History Museum: NMW 18667:4, collected during 1895–1896 by “I.R.M. Expedition.” The abbreviation stands for “I. Österreichische Expedition nach dem Rothen Meere” (First Austrian Expedition to the Red Sea), on which excursion Siebenrock participated in person together with the contemporary director of the Vienna museum, Steindachner (Tiedemann et al. 1994). Friedrich Siebenrock and Franz Steindachner were world-famous herpetologists and turtle specialists, and there is no reason to doubt the accuracy of their record. Major sand units exist (not only) around El Tor and further east in southern Sinai, as can be inferred from geological (EGSMA [Egyptian Geological Survey and Mining Authorty] 1995) and aeronautical (DGIA [Defence Geographic and Imagery Intelligence Agency, UK] 2000) maps, as well as a figure in Werner (1987, Fig. 2). The above data, combined with persistent rumors originating among Egyptian animal dealers about southern Sinai (Buskirk 1985) and Mount Sinai (Baha El Din, 1994) as the source of their animals even as recently as the 1980s and early 1990s, strengthen considerably the case about a natural (past) occurrence of T. werneri in the region. Such a scenario challenges the traditional view about a strict Mediterranean affinity in pure geographical terms. The species' range could have been much larger in the past. Although dismissed as extralimital or dubious without further proof (e.g., Buskirk, 1985), remote localities (with at least correct soil parameters) might actually represent the last remnants of a more extensive distribution in historical times, which range could have decreased and fragmented with the natural aridification and expansion of the desert. This was already hinted at by Baha El Din (1994) regarding two western localities, Bir Gindali and Wadi El Natrun.
A relevant addition by Y.L. Werner (pers. comm., August 2002): “Our rediscovery of the species at 14 km south of Be'er Sheva [Israel] in 1963 followed decades of fruitless searching and asking Bedouins who claimed ignorance. We wondered whether our success was due to the unprecedented large search party at the unprecedented correct time, or the then mass-poisoning of foxes incidental to excessive rodent poisoning by neighbouring Qibuzim. The lesson is that, to assess the reports from southern Sinai one would have to go into the probability of encounter based on statistics of search effort (and gathering these data is a major and non-promising enterprise). Therefore, to me its occurrence there [in southern Sinai] seems more enigmatic than unlikely.” Major sand units are found running southwards from northern Sinai, and in Africa, along and near the Gulf of Suez, and fragmentation of the belt is obvious in many areas (Werner 1987; EGSMA 1995; DGIA 2000).
Today, T. werneri is known in Egypt from only 10 individuals living in Zaranik Protected Area, northern Sinai (Baha El Din 2002). The major extirpation process of the species in Egypt took only a few decades between the late 1960s to early 1970s, and the early 1990s (Baha El Din 1994). Data on trade, unnatural predators and recent search efforts, derived from the above literature alone, relative to the fact that the main range of the species lay in Egypt, support a Red List status of CR A2ade under current Red List Categories and Criteria (IUCN 2001).
Israel
Testudo werneri was first discovered (as T. kleinmanni) in Israel in 1963 (Werner 1982). Older records (Lortet 1883; Tristram 1884) from the same region before the erection of the State of Israel were dismissed by Flower (1933) and most subsequent authors until Bringsøe and Buskirk (1998).
The map in Bringsøe and Buskirk (1998) covers roughly the general distribution of T. werneri in Israel, the northwestern Negev and a disjunct range further east. However, the species' actual occurrence is more restricted, encompassing several more or less interconnected sands, according to information provided by B. Shacham (pers. comm., August 2002), and Bouskila (2002, 2004). These are: the 
olot 'Agur sands, olot alu 
a sands, 
e'elim sands, Ni ana (= Bir Malaga) sands, Shunera sands, Mash'abbim (= Bir Asluj) sands, and the Sekher River sands near Ramat ovav—all south to west of Be'er Sheva in the geographic division of northwestern Negev (Fig. 1). In the Central Negev geographic division southeast of Dimona, the species is additionally reported from the isolated Mishor Rotem and Mishor Yamin plains (e.g., Werner 1987). Geographic divisions above are based on classification adopted by Werner (1987).)
More detailed localities include the following: sand dunes 14 km S of Be'er Sheva; 16 km S of Be'er Sheva; Be'er Sheva (this locality could indicate the seasonal watercourse [wadi] of Nahal Be'er Sheva rather than the city); nr. Ashalim; 3 km N of Revivim; Gevulot; 1 km E of Re 
ovot; “Be'er Chagil” (Nahal Be'er ayil); 15 km N of Be'er Milka; ca. 6 km NE of Ni ana; “Tel-Yerucham” (Yeroham); Kadesh Barne'a; and Har Qeren sands (Geffen and Mendelssohn, 1988; Bringsøe and Buskirk, 1998; Perälä, 2001; B. Shacham, pers. comm., August 2002).
Bouskila (2002, 2004) remarks that records from the Yeroham sands represent only lone individuals and that it is unclear whether a population exists there. There is no population in that area, according to B. Shacham (pers. comm., August 2002). Lortet (1883) mentioned T. werneri (as T. kleinmanni) from the vicinity of Hebron and from Be'er Sheva. His first locality is doubtful or it is based on a stray animal because the region lacks the correct habitat. Tristram's (1884) locality “between Hebron and Be'er Sheva” seems to be a combination based on Lortet (1883) whereas his reference to “Arabah” (Ha'Arava) might represent a badly formulated indication of the Mishor Rotem and/or Mishor Yamin plains, as suggested by Bringsøe and Buskirk (1998). The locality Tel Aviv for specimen MNHN 1984–500 from the Natural History Museum in Paris is extralimital as no suitable habitats exist that far north. The record is probably based on a captive specimen from the Tel Aviv University Expo Zoo (Perälä 2001). A further extralimital record, from Lebanon (Iverson 1992), is likely to be based on originally misidentified specimens of T. graeca (sensu lato) in the Chicago Field Museum, which I have examined.
The general distribution of T. werneri in Israel has not changed since its discovery in 1963, with one possible reservation. Although the area is cited as belonging to the range of this species as reviewed recently by Bringsøe and Buskirk (1998) and Bouskila (2002, 2004), B. Shacham (pers. comm., August 2002) is not confident that a population exists in the Mishor Rotem and Mishor Yamin plains at all: “We [Israeli herpetologists] consider the sands of the Rotem plains to be too far and detached from the western Negev/Sinai sands to have a werneri population.” These sands have eroded relatively recently in situ from sandstone (Werner 1987), and are therefore not historically connected to the northwestern Negev sands. Records from the area might represent lone individuals.
Ecology
The Negev tortoise T. werneri is one of the smallest Mediterranean tortoises, only attaining a carapace length of up to 131 mm (Perälä 2001). The species is found basically in sand dune habitats in regions with an annual precipitation of roughly between 100 to 200 mm (Mendelssohn 1982; Baha El Din 1994). The ecology of T. werneri has been studied relatively well in Israel (e.g., Mendelssohn 1982; Geffen and Mendelssohn 1988, 1989, 1991). In northwestern Negev, mean annual temperature is 20°C, with average maxima and minima of 30°C and 12°C in July and January, respectively; fluctuations of 10°–15°C are common during the day (Geffen and Mendelssohn 1989). Extreme temperatures reach 40°C in the summer when the soil surface may warm up to 60°C, and subzero temperatures are known to occur in the winter, although these are rare (Mendelssohn and Geffen 1995). Testudo werneri has its main activity period in winter (Mendelssohn 1982). The tortoises tend to estivate the summer in rodent burrows under bushes (Geffen and Mendelssohn 1989). Individuals have well-defined home ranges with great overlap between those of different males and both sexes (Geffen and Mendelssohn 1988). The species feeds mainly on annual plants but leaves of perennial shrubs are possibly used to compliment the diet (Mendelssohn 1982).
Age estimates of 7–9 years for mature wild T. werneri females published in Geffen and Mendelssohn (1991) are not considered reliable; these figures most probably represent underestimates. Although age determination in chelonians by scute growth ring counts is a commonly used practice (and the technique applied in the study quoted above) this method is based on several assumptions to be accurate. It becomes increasingly difficult to recognize annual growth marks after maturity has been reached when growth has slowed or ceased altogether, and annuli are commonly lost due to abrasion (Zug 1991). Maturity is probably reached at about 10–20 years in the wild, and females take longer to mature than males, which is also the case in several other testudinids adapted to life in semiarid or arid environments (Woodbury and Hardy 1948; Brushko 1977; Kuzmin 2002, among others). Accordingly, the generation length (where the term is defined as the average age of parents of the current cohort [IUCN 2001]) of the Negev tortoise would be at least around 20 years, and probably more.
The mating period of wild T. werneri is limited to the spring whereas nesting season lasts from March to the end of June; females lay 2 to 3 clutches of 1–3 eggs per year (Geffen and Mendelssohn 1991).
The Negev tortoise does not have many natural predators. According to Mendelssohn and Geffen (1995), the main predator is the desert monitor Varanus griseus, but as this lizard is active in the summer it can encounter tortoises only during late spring when activity of both species overlap. Except for foxes, endothermic predators are rare in the Negev under natural conditions (Mendelssohn and Geffen 1995).
Habitat Preferences and Current Habitat Status
Habitat Preferences
Although the species is also known to occur in coastal areas adjacent to salt marshes near Lake Bardawil (Baha El Din 1994), Testudo werneri prefers almost entirely sandy soils, dunes, and solidified sands with some plant cover of small bushes, shrubs, and short lived annual vegetation (Mendelssohn 1982; Geffen and Mendelssohn 1988, 1989, 1991; Baha El Din 1994; Mendelssohn and Geffen 1995). Bouskila (2002, 2004) adds that Israeli herpetologists consider only the sand dunes without loess soils as valid potential habitat for this species.
The vegetative cover of the northwestern Negev sands, which are partially connected to sands in the Sinai and thus loosely connected to those of the Sahara, consists mainly of Stipagrostis scoparia, Artemisia monosperma, and Retama raetam plant associations, depending on local soil factors and precipitation (Werner 1987; Mendelssohn and Geffen 1995). In the Mishor Rotem and Mishor Yamin plains dominant bushes are Anabasis articulata, R. raetam, Thymelaea hirsuta, and A. monosperma in that order of abundance (Werner 1987).
Vegetation of the inland dunes in Sinai show domination by A. monosperma and S. scoparia, whereas halophytes Halocnemum strobilaceum and succulents such as Zygophyllum album dominate near the coast (Baha El Din 1994). Plant cover in intact habitat in the Negev can be as much as 30%–40% (Mendelssohn 1982), or even 70% in Sinai (Baha El Din 1994).
Using the modified version of the Global Land Cover Characterization (GLCC) classification according to the Red List Criteria, Annex C (IUCN 2001), the primary habitat preference of T. werneri can be characterized as sand desert (GLCC category 8.1.3).
Current Habitat Status
Habitat in African Egypt east of the Nile is increasingly arid, severely grazed, and degraded (Baha El Din 1994). In Sinai, vast areas of habitat were almost completely devoid of vegetation already over 20 years ago; hence uninhabitable by tortoises (Mendelssohn 1982). Unprotected (= most) areas in Sinai have less than 5% plant cover, and the coastal strip east of El Arish is completely destroyed up to 10–35 km inland, and marginally suitable habitats elsewhere on the peninsula are rapidly disappearing (Baha El Din 1994). Baha El Din (1994) concludes that the best habitat patches in Sinai are found at Zaranik Protected Area.
Although plant cover averages (or at least averaged) about 20%–30% in a tiny 3 km2 study area at olot 'Agur where ecological work on T. werneri was conducted in the past (Geffen and Mendelssohn 1988, 1989, 1991), vegetation cover is generally very low in the Negev, only 5%–10% in most areas occupied by the species (Mendelssohn 1982).
According to recent information, all areas within the extent of occurrence of T. werneri in the Negev, other than the protected Be'er Mash'abbim Sands Reserve (ca. 12 km2), are in clear present danger of becoming a wasteland (B. Shacham, pers. comm., August 2002). Northern parts even within the Mash'abbim Reserve itself seem to be overgrazed, although this is denied by the head scientist of the Israel Nature and National Parks Protection Authority who claims that Bedouin herders are part of the Authority's vegetation management scheme in that area (B. Shacham, pers. comm., August 2002).
Judging by vegetation cover indices of 40–70% for intact habitat relative to figures on current levels at less than 5%–10% (Baha El Din 1994; Mendelssohn 1982), the decline in plant cover within less than three generations is estimated at between 75% and > 92%. The higher estimate (which applies to most parts of the distribution) meets the Red List Criterion CR A2c for a Critically Endangered status (IUCN 2001), assuming that there is a correlation between population size and quality of habitat. This assumption is viable (IUCN 2001).
Causal factors influencing habitat degradation and destruction are reviewed in the Major Threats section below.
Extent of Occurrence
The extent of occurrence of T. werneri less than three generations (≥ 60 years) ago—using the minimum convex polygon method (IUCN 2001)—is estimated at around 21,000 km2 excluding, and around 40,000 km2 including, southern Sinai. These estimates are based on information published in Loveridge and Williams (1957), Buskirk (1985), Iverson (1992), Baha El Din (1992, 1994), Bringsøe and Buskirk (1998), Perälä (2001), and Bouskila (2002, 2004).
Today the species' range is effectively restricted to the Negev (Perälä 2001). Additionally, 10 individuals are known from Zaranik Protected Area at the eastern end of Lake Bardawil in Sinai (Baha El Din 2002). Data presented above in the Distribution section are used for the following estimate. The present estimated extent of occurrence of T. werneri is around 3,625 km2 assuming that no subpopulations exist in the Mishor Rotem and Mishor Yamin plains, as hypothesized by B. Shacham (pers. comm, August 2002). Zaranik in Sinai was taken into account in the calculations.
According to these estimates, the extent of occurrence of T. werneri has decreased by around 82%−90% in less than three generations, depending on whether the potential occurrence in southern Sinai is included or not. This estimated reduction meets the Red List Criterion CR A2c for a Critically Endangered status (IUCN 2001).
Area of Occupancy
Based on data on primary habitat (major sand units) in northeastern Egypt, Sinai, and Israel (Werner 1987; EGSMA 1995; DGIA 2000), the past area of occupancy of T. werneri is estimated at around 10,000 km2 or 13,950 km2, depending on whether the potential distribution in southern Sinai is excluded or not, respectively. These estimates are applicable to the situation less than 3 generations ago based on field and museum data, and interviews, as published in Loveridge and Williams (1957), Buskirk (1985), Iverson (1992), Baha El Din (1992, 1994), Bringsøe and Buskirk (1998), Perälä (2001), and Bouskila (2002, 2004).
Judging by potentially suitable habitat (sand dunes) in the northwestern Negev, as estimated by G. Gabai (assistant manager of the Southern Province with the Israel Nature and National Parks Protection Authority), the current area of occupancy of the Negev tortoise is around 700 km2 (B. Shacham, pers. comm, August 2002). The Zaranik reserve in Sinai covers an area of 250 km2 of which 32% or 80 km2 are sand dunes and most of the area (68%) covers Mediterranean waters. Thus, the total area of occupancy of T. werneri is around 780 km2.
Based on the above estimates, the area of occupancy of T. werneri has decreased by around 92%−94% in less than three generations, depending on whether the potential occurrence in southern Sinai is included or not. This estimated reduction meets the Red List Criterion CR A2c for a Critically Endangered status (IUCN 2001).
Population Size and Density
Although a small study area (3 km2) of Geffen supported a relatively dense population of tortoises in the early 1980s, estimated at 27 individuals per km2 (reviewed in Bouskila; 2002, 2004), the average population density for T. werneri in Israel was estimated at only 4–5 individuals per km2, of which ca. 25% were immature (Mendelssohn 1982).
(a) Using Mendelssohn's (1982) population density estimate (calculated as 4.5 tortoises per km2) relative to the estimated areas of occupancy, the past population size less than three generations ago would have been around 45,000 individuals (ca. 33,750 mature) not taking the potential distribution in southern Sinai into account, and around 62,775 tortoises (ca. 47,080 mature) including southern Sinai.
(b) Using the same criteria, but omitting 80 km2 of sand dunes at Zaranik that support only 10 individuals to avoid unjustified overestimates, the current world population of Testudo werneri would be estimated at around 3150 individuals of which ca. 2360 are adults.
(c) Mendelssohn's (1982) density estimate is 20 years old. The species' range in Israel is shrinking, and local densities have supposedly decreased, although exact data are lacking (Bouskila 2002, 2004). When Zaranik is not taken into account on viable grounds as described above, and assuming that 10% of the sand dunes in the Negev are already degraded to the extent that these represent unsuitable habitat for T. werneri (leaving 630 km2 of occupied habitat), and that densities have additionally declined from 4.5 to 4 individuals per km2 because of other threat factors (as listed under Major Threats below), then the present global population of the Negev tortoise would be estimated at around 2520 animals of which 1890 are adults. The assumption is that around 75% of individuals are mature (Mendelssohn 1982). Because no systematic population and habitat inventory has been conducted in two decades, and because several well-known threat factors continue to affect the population (Bouskila; 2002, 2004), even these “corrected” figures might represent overestimates.
Population Decline
Based on the preceding simplistic inferences on past (a), and present (b) or (c), population sizes, the population of Testudo werneri is estimated to have declined by: (ab) ca. 93% assuming no subpopulations existed in southern Sinai, or ca. 95% assuming the species occurred in southern Sinai, or (ac) ca. 94% assuming no subpopulations existed in southern Sinai, or ca. 96% assuming the species occurred in southern Sinai.
These estimates meet with Red List criterion CR A2b (Critically Endangered) of observed, estimated, inferred or suspected population size reduction of ≥ 80% over the last 10 years or three generations, whichever is longer, where the reduction or its causes may not have ceased or may not be understood or may not be reversible (IUCN 2001).
Major Threats
Detailed explanations of the factors leading to the near to complete extirpation of T. werneri from Egypt in only a few decades are presented in Baha El Din (1994), and to some extent Buskirk (1985). It is noteworthy, however, that in addition to other factors, trade played a major role in the extinction process in Egypt (Baha El Din 1994). Tortoises have a low annual biomass production resulting in a high degree of sensitivity to population disturbance and consequent poor recovery abilities from such activities as trade collection (Iverson 1982).
Several threat and disturbance factors have been identified in Israel where the remaining world population of T. werneri effectively occurs. Mendelssohn and Geffen (1995) pointed out that anthropogenic factors change the habitat and the assemblage of naturally coexisting animal and plant species in a drastic fashion. These authors exemplified a chain reaction caused by overgrazing and overbrowsing by Bedouin herds (camels, goats, sheep) in conjunction with the building of human settlements: the reduction in shrub density and amount of annuals due to overgrazing leads to less cover and food for tortoises, and it destroys rodent burrows dug under bushes, which burrows are also used for shelter and estivation by the tortoises. Eventually seed production decreases, leading to the disappearance of the rodents and their borrows, and leaving the tortoises without shelter with the effect that they make relatively easy prey for unnatural predators such as ravens and crows, as well as dogs, attracted by human settlements (Mendelssohn and Geffen 1995).
In his recent reviews, Bouskila (2002, 2004) lists the following threats: habitat destruction by agricultural development; sand mining and reservoir construction; off-road vehicular usage causing damage to both tortoises and their habitat (tortoises are killed whether active or in burrows, especially in the Nahal Sekher and e'elim sands); overgrazing and over-trampling of sand habitats by livestock, especially in the Nahal Sekher sands; increase in predation threat due to growing raven and stray dog populations, as well as spread of jackals near human settlements, military camps, and garbage dumps (juveniles are at highest risk); illegal collection for the pet trade; and threats associated with the release of confiscated T. kleinmanni in Sinai.).
B. Shacham (pers. comm., August 2002) confirms the above and adds that the other main threats for natural inhabitants of the northwestern sands (especially tortoises) are the development of roads, new settlements, industry, and military facilities. Also, the use of military 4-wheel-drive vehicles is important in addition to recreational cruising. Shacham further stresses not to underestimate the threat posed by Bedouin herders because illegal settlers are sprawled over many hectares with their sheep and goats in many areas adjacent to sand dunes, often grazing within the dunes themselves. His general conclusions are that, apart from protected sands within the tiny Be'er Mash'abbim Reserve, the rest of the habitat is in clear current danger of becoming a wasteland. Most parts of the land are either used as military training zones, or under various levels of physical use; in other parts there are settlements.
Recently, yet another threat has arisen. Apart from illegal collection for the pet trade, B. Shacham (pers. comm., August 2002) pinpoints an acute problem: T. werneri are collected for food by Thai migrant workers living in the agricultural areas of Kemehin and Kadesh Barne'a. This information comes from rangers in the area but no data on the scale of exploitation are available. Even in an unrealistic scenario with the assumption that no other threat factors exist, and if one individual on average were to be removed from the population every week for food, or the pet trade, T. werneri would be collected to extinction in ca. 50–60 years, based on present population size estimates. Taking all known parameters threatening the remaining population into account, as listed in this paper, including the effects of introduced taxa (unnatural predators and T. kleinmanni), it is reasonable to predict that the global population of the Negev tortoise is under severe threat of extinction in the near future, possibly in only a few decades, and hence meets Red List criteria CR A3de. Werner (1988) listed T. werneri (as T. kleinmanni) as being under threat of imminent local extinction in Israel with reference to habitat factors presented by Mendelssohn (1982). This “local” extinction threat has now become a global issue because most of the world population of T. werneri occurs essentially in a small region in Israel (except for 10 individuals in Sinai).
Major threats affecting T. werneri are listed in Table 1 in accordance with standard classification (Major Threats Authority File) as required by IUCN Red List Criteria, Annex 3 (IUCN 2001). These factors have been compiled from Mendelssohn (1982), Iverson (1982), Geffen and Mendelssohn (1995), Baha El Din (1994), Baha El Din (2002), Bouskila (in press, a,b), Perälä (2001), and B. Shacham (pers. comm., 2002). Past, present, and future trends are given in conjunction with each specific threat category. Each of these time frames represents three T. werneri generations. Only those categories that are met are listed. Numbers indicate IUCN standards (IUCN 2001).
Conservation Measures
Conservation measures both present and those needed, and which are realistically achievable within the next five years, are listed in Table 2 on a standard Conservation Measures Authority File as required by Red List Criteria, Annex 3 (IUCN 2001). Selection of a higher level action, for example 1.2. Legislation, does not mean that all the actions below this, e.g., 1.2.1 Development and 1.2.2. Implementation, are indicated. It simply indicates that legislation is either in place or is needed as part of a policy-based action for the taxon concerned. Selection of any action lower down the hierarchy automatically implies that the higher levels are indicated (IUCN 2001). As some of the points do not apply to conservation measures in both Egypt and Israel, a brief commentary of categories, as indicated with numbers (two last columns), follows. In such cases only the “Needed” category is used.
Note (1)
Completing the declaration of the olot 'Agur and Shunera Sands Nature Reserves in Israel is essential for the survival of T. werneri (Bouskila; in press, a,b); the establishment of a nature reserve in olot 'Agur has been long campaigned for (Mendelssohn 1982; Geffen 1990). The olot 'Agur Reserve would cover an area of approximately 290 km2 (B. Shacham, pers. comm., August 2002), around 40% of the area potentially occupied by T. werneri. It is noteworthy that the Rotem plains near Dimona are a closed military area and off-limits to the public; thus effectively preserved—although it is not clear whether tortoises exist there anymore (B. Shacham, pers. comm., August 2002). An understanding should be reached with the Israel Defence Forces to reduce off-road traffic in army training areas [in the Negev sands] (Bouskila 2002, 2004).
Note (2)
The species is protected by law in Egypt (Baha El Din 1994; Anonymous 2000), and in Israel (Mendelssohn 1982). However, the law is not always implemented in Egypt (Baha El Din 1994; Anonymous 2000). Despite full legal protection in Israel, the law is useless because of factors influencing the species' population existence (Geffen and Mendelssohn 1988). Laws that forbid off-road vehicular traffic in protected sandy areas must be enforced in Israel (Bouskila 2002, 2004).
Note (3)
New field studies and subsequent monitoring are needed to accurately estimate the current tortoise population size, demography, and population trends, as well as status of the habitat in northwestern Negev plus the Mishor Rotem and Yamin plains in Israel. Up-to-date estimates would enhance the quality of a possible population viability analysis (inclusive of other relevant parameters). Such data would be helpful regarding management plans in the future. Bouskila (2002, 2004) suggests the postponing of ostrich reintroduction into natural sandy habitats in the western Negev until clarification of their potential impact on the soil crust and the tortoise population. More research also is needed in Egypt. Systematic searches are encouraged to identify habitat pockets where tortoises might still exist.
Note (4)
Any statistics on contemporary levels of exploitation (collecting for the pet trade and for food) would be welcome. Bouskila (2002, 2004) calls for increased supervision and enforcement by the Israel Nature and National Parks Protection Authority to prevent tortoise collection.
Note (5)
Bouskila (2002, 2004) lists the reduction of raven, jackal, and stray dog populations, as well as the removal of raven nesting sites, as a necessary step for species preservation.
Note (6)
In association with the local tortoise conservation program, Bedouins are engaged in protecting wild T. werneri and their habitat at Lake Bardawil in Sinai (Baha El Din 2002).
Note (7)
Before the recent description of T. werneri, TortoiseCare (The Egyptian Tortoise Conservation Program) ran a reintroduction and breeding program in northern Sinai using confiscated T. kleinmanni of Libyan origin as founder stock (Perälä 2001, 2002). Recently, 10 individuals of wild T. werneri were identified within this area (Zaranik Protected Area) (S. Baha El Din, in litt., 2001; Baha El Din 2002), as feared from a conservation point of view (genetic contamination; pathogens; a multitude of ecological risks) (Perälä, 2001, 2002). This potential threat has been acknowledged in Israel (Bouskila 2002, 2004). As a result of the discovery of wild T. werneri in northern Sinai, TortoiseCare has also shifted its focus to conservation of this wild population, and released T. kleinmanni in the same area have been collected and returned to enclosures in Cairo (Baha El Din 2002). This is welcome news. Nevertheless, there is a chance that mixing (reproduction) between the two species, or exchange of potential pathogens (all of which can never be screened for because of limited resources or because these are as yet undiscovered) has already taken place. Thus both stocks, the wild T. werneri living in situ at Zaranik, as well as the originally released T. kleinmanni from this area, should be considered potential carriers of introduced disease and/or hybrid offspring, and both groups should be kept isolated from any other tortoises or wild populations. Sperm storage is a well known phenomenon in chelonians, and delayed fertilization can occur years after copulation (e.g., Zug et al. 2001). Israeli authorities are not pursuing reintroduction or reinforcement of T. werneri but the captive populations or individuals in Israel are due to be surveyed by zoo personnel from the Biblical Zoo in Jerusalem as a first step to planning a captive breeding colony (Y.L. Werner, pers. comm., August 2002).
The use of captive or confiscated stock of unknown origins or parentage in conservation programs should be discouraged. However, such animals have educational value. Potential captive breeding or reintroduction schemes in the future—if found to be legitimate after analysis, and under IUCN guidelines and basic principles in conservation biology as a scientific discipline—should be based on wild stock of exactly known geographical affinities to avoid genetic mixing and incalculable ecological as well as pathological risks. On the other hand, in situ head-start programs could turn out to be effective and are comparatively safe. Nevertheless, to save the species, habitat conservation and management as well as control of collecting, and off-road vehicular use, must have priority over anything else.
Conclusions
Today, T. werneri is found almost entirely in a small region in northwestern Negev in Israel. Ten individuals survive in a protected area in northern Sinai (Egypt), where trade played a notable role in the extirpation of the species on top of habitat destruction. Testudo werneri was previously listed by CITES on Appendix I as T. kleinmanni (sensu lato), but has recently been listed independently. Habitat has largely disappeared in Egypt, and is degrading fast due to anthropogenic factors in Israel. The decrease in the population size of T. werneri is estimated at around 95% within less than three generations. Focus must be placed on basic research, including population and habitat based studies throughout the distribution, as well as conservation and management of the species and its habitat, including eradication or control of unnatural predators. Education of the general public to incease awareness to the plight of T. werneri is not to be underestimated. Planned nature reserves should be established swiftly and off-road traffic in northwestern Negev effectively controlled. More data are needed on illegal collection of tortoises for food and pet trade, and this exploitation of the remaining population must be stopped. There is a real possibility that the global population of Testudo werneri faces extinction in only a few decades, but more data are needed for an accurate population viability analysis. Concerted national and international conservation efforts could save the species from extinction. Such efforts must be based strictly on scientific criteria as opposed to animal welfare philosophy. This latter trend is apparent in many recent programs regarding chelonians with potentially catastrophic effects on wild populations (Pieh 2001; Perälä 2002).
Based on evidence presented above, and the IUCN Red List Categories and Criteria Version 3.1 (IUCN 2001), T. werneri is established as CR A2abcde+3de − Critically Endangered. This was reviewed and accepted by the Red List Authority and subsequently by the IUCN (2003).
The remaining world population of Testudo werneri is restricted to around 700 km2 of potential habitat, desert sands (shaded) in northwestern Negev (Israel), and 80 km2 of dunes at Zaranik Protected Area in the northern Sinai of Egypt supporting only a few individuals. Major roads near the sands are drawn in. The possible isolated occurrence of the species southeast of Dimona (Israel) is indicated with a question mark.
