Population status and conservation of Morelet's crocodile, Crocodylus moreletii, in northern Belize

Transcription

1 Biological Conservation 96 (2000) 21±29 Population status and conservation of Morelet's crocodile, Crocodylus moreletii, in northern Belize Steven G. Platt, John B. Thorbjarnarson * Wildlife Conservation Society, 2300 Southern Blvd., Bronx, New York, , USA Received 21 January 1999; received in revised form 12 January 2000; accepted 4 March 2000 Abstract Spotlight surveys were conducted from 1992 to 1997 to determine the population status of Morelet's crocodile, Crocodylus moreletii, in northern Belize. A total of 754 crocodiles were observed and km surveyed (1.55/km). Encounter rates were highest in non-alluvial (8.20/km) and alluvial (6.11/km) lagoons, and considerably lower in most rivers and creeks (0.95/km), and coastal mangrove habitats (0.24/km). The population sex ratio was signi cantly male biased (1 female:5.3 males), although the reasons for this are unclear. Comparisons with survey data from 1979 to 1980 suggest substantial population recovery has occurred following legal protection in At present there are no immediate threats to the continued survival of Morelet's crocodile in Belize. Conservation actions should address the creation of additional protected areas in anticipation of future regional development. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Morelet's crocodile; Belize; Conservation; Endangered species; Survey 1. Introduction Morelet's crocodile, Crocodylus moreletii, is found in the Atlantic and Caribbean lowlands of Mexico, Guatemala, and Belize (Ross, 1998). It is recognized as endangered under the United States Endangered Species Act (Code of Federal Regulations, 1977), and is listed on Appendix I of the Convention on International Trade in Endangered Species of Flora and Fauna (CITES) (Jenkins, 1987; Ross, 1998). In a recent revision of IUCN Red List categories, based largely on survey data from Mexico, C. moreletii was listed as a species with a low risk of extinction, but ``conservation dependent'' as continued survival is contingent on the success of current e orts to establish conservation programs (Ross, 1996). Morelet's crocodile is one of two species of Crocodylus which occur in Belize (Groombridge, 1987). C. moreletii is found primarily in freshwater wetlands, while the American crocodile, C. acutus, is restricted to o shore islands and coastal mainland habitats (Platt, * Corresponding author. Tel.: ; fax; address: jthorbjarnarson@wcs.org (J.B. Thorbjarnarson). 1996). Early writers commented on the abundance of Morelet's crocodile in Belize (formerly British Honduras). Schmidt (1924) counted the eyeshines of ``several hundred'' small crocodiles in roadside ditches near Belize City. Likewise, Neill and Allen (1959) noted that crocodiles were common during a visit in 1937, and Sanderson (1941) found young crocodiles to be ``very numerous.'' Quantitative survey data from this period are unavailable (Frost, 1974). High quality leather can be prepared from C. moreletii skins (Groombridge, 1987) and over-harvesting depleted populations throughout its range (Powell, 1972, 1973; Alvarez del Toro, 1974). Commercial crocodile hunting began in Belize during the late 1930s and 1940s (Hope and Abercrombie, 1986). Both professional and part-time hunters engaged in commercial hunting, most of which occurred when low water levels concentrated crocodiles during the dry season (Frost, 1974). The major dealers in Belize purchased up to 12,000 skins annually (Frost, 1974), and an unknown number were exported illegally (Charnock-Wilson, 1970). No size limits existed, and much of the catch consisted of juveniles with a total length (TL) <60 cm (Frost, 1974). Population declines were rst noted in the 1950s, and by the late 1960s C. moreletii was nearly extirpated from Belize (Frost, 1974). Neill and Allen (1959, 1961) noted /00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S (00)

2 22 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 crocodiles were rare and could nd no adults, leading Neill (1971) to conclude that C. moreletii was ``nearly exterminated'' in Belize. Frost (1974) concurred and stated crocodiles in Belize were ``practically non-existent...[and] so rare...[as to be] a matter of conversation''. Charnock-Wilson (1970) attributed population declines to excessive hunting. Abercrombie et al. (1980, 1982) provided the only quantitative survey data for northern Belize. Surveys conducted during the late 1970s and early 1980s found smaller C. moreletii common in remote areas, but the scarcity of large adults suggested populations had been over-exploited. Dealers noted steady declines in skin purchases, which they attributed to over-harvesting. Morelet's crocodile was a orded legal protection under the Wildlife Protection Act of 1981 and the export of skins was banned (Marin, 1981). Comprehensive survey data for C. moreletii in Belize are lacking and populations are considered vulnerable (Thorbjarnarson, 1992). Surveys have been accorded priority by the IUCN Crocodile Specialist Group in all countries where C. moreletii occurs (Thorbjarnarson, 1992; Ross, 1998). The objectives of our study were to obtain baseline data and determine the population status of Morelet's crocodile in northern Belize, assess the status of crocodile populations in protected areas, and make comparisons with survey data available from the late 1970s and early 1980s to determine if population recovery has occurred. 2. Materials and methods Northern Belize is dominated by alluvial oodplains and inter uvial swampy depressions and sinkholes (Johnson, 1983), and wetlands are estimated to occupy up to 40% of the region (Alcala-Herrera et al., 1994). Annual precipitation varies from 1300 to 2000 mm with a pronounced wet season from June through November (Hartshorn et al., 1984). We used nocturnal spotlight surveys to census crocodile populations (Messel et al., 1981; Magnusson, 1982; Bayliss, 1987; O'Brien, 1990; King et al., 1994). Survey localities were selected on the basis of accessibility to vehicles and boats, and landowner permission. Single spotlight counts were conducted at most locations. When spotlight counts are replicated, a high degree of precision (low standard error) is generally achieved with a low number of samples (Messel et al., 1981). Standard error decreases dramatically in response to increasing the number of routes, while reduction in standard error due to replication is insigni cant. Optimal allocation of sampling e ort should therefore maximize the number of survey routes in order to increase precision (O'Brien, 1990). Surveys of lagoons, and most creeks and rivers were conducted using an aluminum canoe, while a motorized ski was used to survey mangrove habitats, and the mouth of the Sibun and New Rivers. Crocodiles were located with 12-volt headlights during canoe surveys, while a 400,000 candlepower Q-beam spotlight was used with the ski. Survey routes were traversed before dark, habitat and navigational hazards noted, and surveys began 15 to 30 min after nightfall. Time of night has not been demonstrated to bias spotlight counts (Messel et al., 1981). Inclement weather and periods of elevated water levels were avoided (Woodward and Marion, 1978). Beginning and end points of each survey were recorded to allow replication by future investigators. For comparative purposes wetlands were classi ed as follows: 1. Alluvial lagoons: Lagoons and oxbows located in oodplains of rivers and creeks, and subject to overbank and backwater ooding during periods of high water, especially during the wet season; often lled with mats of oating vegetation (Salvinia spp. and Panicum spp.). 2. Non-alluvial lagoons: Inter uvial wetlands not subject to overbank ooding from rivers and creeks; often formed by dissolution and collapse of underlying karstic substrate; subsurface drainage and springs maintain water levels and prevent drastic uctuations (Johnson, 1983); often heavily vegetated with sawgrass (Cladium jamaicense) and spikerush (Eleocharis spp.). 3. Rivers and creeks: Contain owing water throughout most of the year, although some creeks are reduced to a series of isolated pools by the late dry season; extreme water level uctuations characteristic of wet season; usually little aquatic vegetation. 4. Mangrove: Coastal wetlands dominated by mangrove swamp (Rhizophora mangle, Avicennia germinans) with water salinities ranging from 0 to 35 ppt depending on location, time of year, and amount of recent rainfall; signi cant in ux of freshwater occurs during wet season (McField et al., 1996). For comparison with previous surveys, rivers and creeks, and lagoons (both alluvial and non-alluvial) were categorized as easily accessible, moderately accessible, and inaccessible according to criteria of Abercrombie et al. (1982). Assignment to a category was based on a subjective assessment of accessibility to boats and vehicles. Sites where access was strictly controlled by landowners, even if readily accessible to boats and vehicles, were classi ed as inaccessible. Crocodiles were approached as closely as possible and classi ed by TL as juveniles (TL<100 cm), subadults (TL=100 to 150 cm), or adults (TL>150 cm). When TL could not be determined, crocodiles were classi ed as ``eyeshine-only''. Counts and size estimates were collected by the same observer (SGP) on all surveys. A chisquare analysis was used to compare size-class distributions

3 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 23 within each habitat against a null model of equal distribution. It was assumed that all size classes were equally detectable (Messel et al., 1981). The coordinates of the beginning and endpoint of each survey were determined with a handheld Global Positioning System (GPS). Distance traveled in each survey was determined using 1:50,000 topographical maps obtained from the Department of Lands and Surveys, Belmopan, Belize. Survey routes were calculated as midstream length in linear habitats such as rivers and creeks, or shoreline distance in lagoons (King et al., 1990). A complete description of each route is presented in Platt (1996) and Platt and Thorbjarnarson (1997), copies of which have been deposited in the Campbell Museum (Clemson University, SC), Belize National Archives (Belmopan, Belize), and Wildlife Conservation Society (Bronx, New York). Encounter rates were calculated as the number of crocodiles observed per kilometer of survey route (Bayliss, 1987). Encounter rates provide an index of relative density because not all crocodiles present are observed during a survey (Bayliss, 1987). However, the relationship between the encounter rate and actual population size is assumed to remain constant over time, and any change in the encounter rate should re ect a proportionate change in the total population (Bayliss, 1987). Relative indices are powerful when survey techniques are standardized (Bayliss, 1987). The observed distribution of crocodiles in each habitat was tested against the null model of random distribution using a chi-square goodness of t test and a Z-test (Thorbjarnarson, 1988). Crocodiles were captured whenever possible to con- rm size estimates and determine sex. With the exception of hatchlings and small juveniles, crocodiles were sexed by cloacal examination (Brazaitis, 1968). Observed sex ratios were tested against a null hypothesis of a 1:1 sex ratio using chi-square analyses (Caughley, 1977). 3. Results Spotlight surveys were conducted in 12 alluvial lagoons, 17 non-alluvial lagoons, 20 mangrove wetlands, and 18 rivers and creeks from 1992 to 1997 (Table 1, Fig. 1 ). Surveys were conducted at ve nature reserves (Burdon Canal Nature Reserve, Crooked Tree Wildlife Sanctuary, Freshwater Creek Forest Reserve, Rio Bravo Conservation Area, and Shipstern Nature Reserve), and three private ranches (Banana Banks Ranch, Gallon Jug, and Gold Button Ranch). A total of 754 crocodiles were observed along km of survey route (1.56 crocodiles/km). Of these, 330 (43.7%) were classi ed as ``eyeshine-only'', and 424 (56.3%) were approached closely enough to estimate size: 210 (49.5%) juveniles, 106 (25.0%) subadults, and 108 (25.4%) adults. Encounter rates were highest in non-alluvial and alluvial lagoons, and considerably lower in river/creeks and mangrove habitat (Table 2). A chi-square goodness of t test indicated crocodiles were not randomly distributed between habitat types (w 2 =925.74; df=3; P<0.001). Crocodiles were signi cantly more abundant in non-alluvial and alluvial lagoons, and less abundant in rivers/creeks and mangrove habitat than would be expected in a null model of random distribution (Table 2). Encounter rates for rivers/creeks and lagoons (both alluvial and non-alluvial) categorized according to accessibility criteria were lowest in the most accessible categories and increased in more remote categories (Table 3). When encounter rates for both rivers and lagoons are compared to rates reported by Abercrombie et al. (1982), increases are noted in all categories, with the greatest increase in inaccessible rivers and moderately accessible lagoons. Size-class distributions in mangroves habitats, and rivers and creeks did not di er signi cantly from the null model of equal distribution, but signi cant di erences were found in alluvial and non-alluvial lagoons (Table 4). In the latter two habitats, juveniles were more abundant than expected, and contributed most to the overall chi-square value. A total of 276 crocodiles were captured during this study, and sex was determined for 140 individuals. One hundred and thirty-six hatchlings and small juveniles could not be reliably sexed. The overall sex ratio was strongly male biased (1:5.3), and signi cantly di erent from parity (w 2 =65.8; df=1; P<0.001). Our size estimates of captured crocodiles were generally within 10 cm of the measured TL. 4. Discussion Substantial increases in C. moreletii populations have occurred following legal protection in Little direct crocodile management has been undertaken in Belize, and we largely attribute population recovery to the ban on skin exports. This eliminated the incentive for hunters to penetrate remote areas and systematically harvest crocodiles, and allowed populations to recover from previous exploitation. A comparison of encounter rates for rivers and lagoons in 1978 to 1980 (Abercrombie et al., 1982) and 1992 to 1997 indicate increases in all categories (Table 3). Abercrombie et al. (1982) found no crocodiles in the vicinity of populated areas, while during this survey crocodiles were observed in canals and ditches within the municipal limits of Belize City and Orange Walk, and in wetlands easily accessible from many villages. Nuisance crocodile complaints to the Forest Department have also increased steadily since 1990 (Raphael Manzanero, Forest Department, pers. comm.). Therefore,

4 24 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 Table 1 Morelet's crocodiles observed during spotlight surveys in northern Belize (1992 to 1997) a Location Number observed Distance (km) Encounter rate (crocodiles/km) Location Number observed Distance (km) Encounter rate (crocodiles/km) Alluvial lagoons 1. Black Lagoon (I) Cox Lagoon (M) Crooked Tree Lagoon (E) b Easler Lagoon (I) Jones Lagoon (M) b Mexico Lagoon (M) b Mucklehenny Lagoon (I) O'Neal Lagoon (E) Orange Stump Lahgoon (I) Tonace Lagoon (M) Trash House Lagoon (E) Yellow-stick Lagoon (E) Non-alluvial lagoons 13. Banana Banks (M) c Button Lagoon (M) b Caledonia Lagoon (E) Carmelita Lagoon (E) Chiwa Lagoon (I) Corn eld Lagon (M) Douloon Bank Lagoon (E) b Gold Button Lagoon (I) c Habenero Lagoon (I) b Honey Camp Lagoon (E) b Laguna Seca (I) c Laguna Verde (I) c Reinland Lagon (E) Sallys Pond (E) Salt Creek Lagoon (E) Sapote Lagoon (E) Yo Creek Lagoon (M) Mangrove 30. Almond Hill Barracouta Pond Burdon Canal b Coastline (Midwinter's Lagoon to Potts Lagoon) Coastline (Ladyvill) Coastline (Vista del Mar) Fabers Lagoon Four-mile (San Roque) Lagoon 38. Four-mile Lagoon Haulover Creek Jones Lagoon Katie Pond Laguna Seca (Copper Bank) La Isle Cenote Northern Lagoon Northern Lagoon Petrojam Lagoon Potts Creek/Lagoon Shipstern Lagoon a Southern Lagoon Rivers and creeks 50. Black Creek (E) Blue Creek (E) Chan Chich Creek (I) b Dawson Creek (M) Gold Button Creek (I) b Irish Creek (I) Lemonal Creek (M) Mussell Creek (E) New River (mouth; E) New River (upstream; E) New River Lagoon (E) Northern River (E) Ramgoat Creek (I) Rio Bravo (M) Rio Hondo (mouth; E) Rio Hondo (upstream; E) Sibun River (mouth; M) Sibiun River (upstream; E) Total a Numbers correspond to map in Fig. 1. Rivers and creeks, and lagoons (alluvial and non-alluvial) followed by letter in parentheses denoting accessibility classi cation according to Abercrombie et al. (1982) where E=easily accessible, M=moderately accessible, and I=inaccessible. Place names are in accordance with Ordnance Survey topographical maps. b Nature Reserves (Zisman, 1996). c Private ranches and farms. plans to reintroduce captive-bred C. moreletii to augment impacted populations (Behler, 1989) are now considered unwarranted. Size class distributions also suggest C. moreletii populations are recovering from past over-exploitation. Abercrombie et al. (1980) stated only 5 to 10% of all crocodiles sighted were adults. Because hunters speci cally target larger crocodiles, size class distributions of over-exploited populations are strongly biased towards juveniles and subadults (Cott, 1961; Webb and Messel, 1978a; Glastra, 1983; Rebelo and Magnusson, 1983). However, in our study 25.4% of all crocodiles for which size could be reliably estimated were classi ed as adults, indicating a substantial increase in the number of large animals has occurred. Moreover, adults may comprise a larger segment of the population than our results suggest, because they are often more wary and di cult to approach than juveniles and subadults, and therefore more likely escape detection during spotlight counts (Webb and Messel, 1978b). Changes in size distributions are expected as populations continue to recover from previous hunting. Crocodile populations are resilient to over-exploitation and generally respond well to protection (Bayliss, 1987). Recovery of C. moreletii is attributable to several factors. First, female C. moreletii are capable of reproducing at an

5 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 25 Fig. 1. Map of Belize showing approximate location of Morelet's crocodile surveys (1992 to 1997). Numbers correspond to Table 1. early age, and while growth data are lacking, sexual maturity is probably attained in 7 to 8 years (Platt, 1996). Second, the apparent scarcity of adult crocodiles noted by Abercrombie et al. (1980, 1982) may re ect increased wariness of larger animals rather than actual rarity. Wariness increased among adult C. porosus in response to harassment, and large animals proved di cult to approach (Webb and Messel, 1978b). According to Woodward et al. (1995) wariness resulted in adult alligators being perceived as rare following intense exploitation during the s, while in fact adults were more common than surveys indicated. Finally, much crocodile habitat in northern Belize consists of virtually inaccessible, heavily vegetated marshes and swamps located in roadless areas. Even if these sites could be reached, navigable channels are lacking, precluding the use of boats and spotlights for hunting. Additionally, impassable shoals and rapids on many rivers prevented access, and hunting in these rivers was minimal (Abercrombie et al., 1982). Inaccessible habitats likely served as de facto refuges during the period of intense exploitation, and following protection, dispersing crocodiles bolstered impacted populations. C. moreletii now occurs throughout northern Belize, with the greatest densities in lagoons. Crocodiles are most abundant in non-alluvial lagoons, and the encounter rate we found is believed conservative. An important assumption of spotlight counts is that the

6 26 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 Table 2 Comparisons of encounter rates of Morelet's crocodile in various habitats of northern Belize Habitat Number of crocodiles Kilometers surveyed Encounter rate (crocodiles/km) Z-score Alluvial lagoons ** Non-alluvial ** lagoons Mangrove ** Rivers and creeks ** P< Table 3 Comparison between encounter rates of Morelet's crocodiles in rivers and lagoons prior to protection (1979±1980) and during the present survey (1992±1997) a Encounter rate (crocodiles/km) Habitat Accessibility 1979±1980 b 1992±1997 Change Rivers Easily accessible Moderately accessible Inaccessible Lagoons Easily accessible Moderately accessible Inaccessible a Habitats categorized according to accessibility by boats and vehicles (see text). b Source: Abercrombie et al. (1992). Table 4 Crocodile size classes compared by habitat a Habitat Juveniles Subadults Adults 2 Alluvial lagoons 70(76.0) 17(18.4) 5(5.4) ** Non-alluvial lagoons 78(49.1) 30(18.8) 51(32.0) ** Mangroves 12(25.5) 17(36.1) 18(38.2) 1.31 NS Rivers and creeks 50(39.6) 42(33.3) 34(26.9) 3.02 NS a Total number in each size class followed by frequency within habitat in parentheses. Crocodiles classi ed based on total length (TL) as juveniles (TL<100 cm), subadults (TL=100 to 150 cm), or adults (TL>150 cm). P< NS Not signi cant. proportion of animals observed is constant between habitats (Bayliss, 1987). However, in non-alluvial lagoons this assumption is probably violated because dense vegetation limits visibility to several meters, and undoubtedly decreases the probability of sighting crocodiles. Similarly, the larger boat and more powerful spotlight we used in mangrove and some riverine habitats may have increased the probability of sighting crocodiles, thereby potentially biasing our results. However, encounter rates were generally low in these habitats, and any sampling bias is unlikely to change our results. Indeed, crocodiles may be even less abundant in mangrove habitats than suggested by our study. Size-class distributions also indicate that signi cant recruitment of juveniles occurs in lagoons. Alluvial lagoons had the highest proportion of juveniles of any habitat, although nesting success in this habitat is generally lower than in non-alluvial lagoons (Platt, 1996). Alluvial lagoons may serve as ``rearing stockyards'' where juveniles which hatch elsewhere remain until reaching sexual maturity and moving to breeding habitat (Messel and Vorlicek, 1989). With several exceptions, encounter rates were relatively lower in rivers and creeks. The New River, associated tributary streams, and New River Lagoon, appear to support relatively high densities of C. moreletii. Logwood swamps and sawgrass marshes border the main channel, and in places extend 5 to 10 km on either side of the river. These wetlands are not accessible to boats, and because past hunting pressure was probably minimal, are believed to support high densities of crocodiles. Encounter rates were relatively high in remote rivers and creeks (e.g. Irish, Chan Chich, and Ramgoat Creeks, and sections of the Rio Bravo and Rio Hondo), although shoals, rapids, and fallen trees made these habitats exceptionally di cult to survey. The lowest encounter rates occurred in coastal mangrove wetlands. Similarly, Meerman (1992) and Ouboter (1992) reported few C. moreletii in mangrove wetlands of Shipstern Nature Reserve, and attribute this to low prey densities. C. moreletii and C. acutus occur sympatrically in coastal wetlands of Mexico and Belize (Platt and Thorbjarnarson, 1997) and ecological relationships have not been studied (Thorbjarnarson, 1992). C. moreletii is generally considered a freshwater species (Taplin, 1988), but occurred in salinities as high as 22.0 ppt. Conversely, C. acutus is extremely rare in mainland coastal habitats and largely restricted to o shore islands in Belize (Platt and Thorbjarnarson, 1997). These results contrast with Neill (1971) and Campbell (1972), who stated that C. moreletii does not occur in saline habitats, and when both are present in coastal regions, C. moreletii is con ned to freshwater, and C. acutus to brackish and saline habitats. Patterns of C. moreletii and C. acutus abundance in coastal Belize could re ect past exploitation and subsequent niche expansion by the former. Both species were intensely hunted and following protection C. moreletii may have expanded into habitat previously occupied by C. acutus. Similar models have been proposed to explain patterns of abundance between other species of sympatric crocodilians (Thorbjarnarson, 1989; Thorbjarnarson and Hernandez, 1992; Sigler, 1996). It is also possible that mainland coastal habitats have always represented a zone of contact between the two species with C. moreletii most common in inland wetlands and C. acutus on o shore islands. Encounter rates in nature reserves and private ranches were variable. The highest encounter rate in Belize

7 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 27 occurred at Gold Button Lagoon, a 142 ha lagoon located on the 10,526 ha privately owned Gold Button Ranch. This site was included in a nation-wide inventory of signi cant natural areas (Miller and Miller, 1995). Crocodiles are also abundant in wetlands at Gallon Jug (Laguna Seca and Verde), and Rio Bravo Conservation Area (Habanero Lagoon and Irish Creek). At Crooked Tree Wildlife Sanctuary no crocodiles were found in Crooked Tree Lagoon, and encounter rates were low in Mexico and Jones Lagoons, and Black Creek. Four lagoons (Button, Chiwa, Doubloon Bank, and Honey Camp), were surveyed in the Freshwater Creek Forest Reserve. Of these, only Chiwa Lagoon is completely encompassed by reserve boundaries. Encounter rates were moderately high at all sites except Honey Camp Lagoon, where extensive shoreline clearing and illegal killing have severely impacted crocodile populations. The low encounter rate in Burdon Canal Nature Reserve is similar to other mangrove wetlands. Protected status has been proposed for Cox Lagoon (Zisman, 1996), and the encounter rate was among the highest in Belize. This study constitutes the only reported sex ratio for any wild population of C. moreletii (see review in Thorbjarnarson, 1997), and the reasons for the observed male-bias are unclear. Sampling bias may account for observed deviations from 1:1 sex ratios among reptiles with temperature dependent sex determination (Mrosovsky and Provancha, 1992), but we are unaware of any obvious bias in our study and believe sampling was essentially random. Hatchlings and juveniles, which are di cult to sex (Thorbjarnarson, 1997) were excluded from our analysis, and captures occurred throughout the year and in a variety of habitats. Despite legal protection, laws are rarely enforced and some illegal killing still occurs. Crocodiles are generally perceived as a threat to humans and livestock, and although attacks on humans are extremely rare (Marlin et al., 1995), large individuals are occasionally killed near habitations. Additionally, a limited number of crocodiles drown in sh and turtle nets each year. Combined mortality form these sources is di cult to quantify, but appears minimal at present. We found no evidence of commercial poaching for skins or meat. Habitat destruction has contributed to the decline of C. moreletii in Mexico (Sigler, 1995), but habitat remains widespread in Belize (Hammond, 1993). In comparison to other central American countries, Belize has a relatively small population (200,000 in 1990; Hammond, 1993) and deforestation rates are among the lowest in the region (Myers, 1993). Even where large tracts in northern Belize have been converted to sugarcane production, wetlands generally remain intact. Indeed, relatively high encounter rates were noted at several wetlands (e.g. Caledonia and Sapote Lagoons) surrounded by sugarcane elds. Furthermore, Belize has an extensive system of nature reserves, and many include signi cant areas of crocodile habitat (Zisman, 1996). Although C. moreletii populations seemingly face no immediate threats, exposure to environmental contaminants, especially pesticides, could pose a subtle, long-term threat in agricultural areas. High levels of organochlorine pesticide residues have been found in C. moreletii eggs from northern Belize (Rainwater et al., 1998; Wu et al., 2000), although the e ects of exposure remain unknown. However, exposure to similar chemical contaminants has led to signi cant declines of American alligator (Alligator mississippiensis) populations in Florida, USA (Woodward et al., 1993; Guillette et al., 1994). In conclusion, Morelet's crocodile populations in northern Belize appear to have recovered from past over-harvesting, and at present there exists no immediate threat to their continued survival in northern Belize. Adequate populations are now found in most protected areas, and while illegal killing occurs, the extent of this take appears minimal and sustainable. Conservation actions should address creation of additional protected areas in anticipation of future development Acknowledgements This project was funded by the Wildlife Conservation Society, United Nations Development Program Ð Global Environmental Facility, and Clemson University. Scienti c research permits were issued by Raphael Manzanero, Conservation Division, Ministry of Natural Resources, Belmopan, Belize. Considerable logistic support was provided by Colin, Mark, and Monique Howells, Lamanai Field Research Center, Indian Church, and Robert Noonan, Gold Button Ranch. Additional assistance was provided by Dorian Barrow, K. Mustafa Toure and the University College of Belize, Janet Gibson and the Coastal Zone Management Unit, and Matt Miller of Monkey Bay Wildlife Sanctuary. The following are also thanked for their contributions: Clarence Abercrombie, Tienke Boomsma, Barry Bowen, Steven Brewer, Mary Bunch, John Carr, John Paul Cantareul, Bruce Cullerton, Victor Cruz and family, Rachael Emmer, Lance Fontenot, Richard and Carol Foster, Bill Hasse, Howard Hunt, Tony Garel, William Karesh, Hong Liu, Jacob and Kelly Marlin, Sharon Matola, Scott and Cindy McMurry, Jan Meerman, Bruce and Carolyn Miller, Richard Montanucci, Steven Nichols, Tommy Rhott, Michael Sabal, Graham Sampson and the many students of Corozal Community College, John Scavo, Anita, Mike, and Christine Tupper, Paul and Zoe Walker, and Peter Walsh. Special thanks go to Thomas Rainwater and Travis Crabtree

8 28 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 for eld assistance throughout the project. Comments from Thomas Lacher, Richard Montanucci, Edward Pivorun, David Tonkyn, and two anonymous reviewers greatly improved this manuscript. References Abercrombie, C.L., Davidson, D., Hope, C.A., Scott, D.E., Status of Morelet's crocodile (Crocodylus moreletii) in Belize. Biological Conservation 17, 103±113. Abercrombie, C.L., Davidson, D., Hope, C.A., Scott, D.E., Lane, J. E., Investigations into the status of Morelet's crocodile (Crocodylus moreletii) in Belize, In: Crocodiles: Proceedings 5th Working Meeting, Crocodile Specialist Group. IUCN Ð The World Conser. Union, Morges, Switzerland, pp. 11±30. Alcala-Herrera, J.A., Jacob, J.S., Castillo, M.L.M., Neck, R.W., Holocene palaeosalinity in a Maya wetland, Belize, inferred from the microfaunal assemblage. Quaternary Research 41, 121±130. Alvarez del Toro, M., Los Crocodylia de Mexico, Instituto Mexicano des Recursos Naturales Renovables, Mexico. Bayliss, P., Survey methods and monitoring within crocodile management programmes. In: Webb, G.J.W., Manolis, S.C., Whitehead, P.J. (Eds.), Wildlife Management: Crocodiles and Alligators. Surrey Beatty and Sons Pty. Ltd, Sydney, pp. 157±175. Behler, J.L., The status of wild crocodilians and their culture. Proceedings American Association Zoo Veterinarians 1989, 127± 138. Brazaitis, P.J., The determination of sex in living crocodilians. British Journal of Herpetology 4, 54±58. Campbell, H.W., Preliminary report: status of investigations of Morelet's crocodile in Mexico. Zoologica 57, 135±136. Caughley, G., Analysis of Vertebrate Populations. John Wiley & Sons, London. Charnock-Wilson, J., Manatees and crocodiles. Oryx 10, 236± 238. Code of Federal Regulations, 50, Wildlife and Fisheries., O ce of the Federal Register, National Archives and Records Service, General Service Administration, Washington, DC. Cott, H.B., Scienti c results of an inquiry into the ecology and economic status of the Nile crocodile (Crocodylus niloticus) in Uganda and northern Rhodesia. Transactions Zoological Society London 29, 211±357. Frost, M.D., A Biogeographical Analysis of Some Relationships between Man, Land, and Wildlife in Belize (British Honduras). PhD Dissertation, Oregon State University, Corvallis. Glastra, R., Notes on a population of Caiman crocodilus crocodilus depleted by hide hunting in Surinam. Biological Conservation 26, 149±162. Groombridge, B., The distribution and status of world crocodilians. In: Webb, G.J.W., Manolis, S.C., Whitehead, P.J. (Eds.), Wildlife Management: Crocodiles and Alligators. Surrey Beatty and Sons Pty. Ltd, Sydney, pp. 9±21. Guillette Jr, L.J., Gross, T.S., Masson, G.R., Matter, J.M., Percival, H.F., Woodward, A.R., Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environmental Health Perspectives 102, 68±688. Hammond, A., Environmental Almanac. Houghton Mi in, New York. Hartshorn, G., Nicolait, L., Hartshorn, L., Bevier, G., Brightman, R., Cal, J., Cawich, A., Davidson, W., DuBois, R., Dyer, C., Gibson, J., Hawley, W., Leonard, J., Nicolait, R., Weyer, D., White, H., Wright, C., Belize Country Evironmental Pro le: A Field Study. USAID and Robert Nicolait and Associates, Ltd, Belize City, Belize. Hope, C.A., Abercrombie, C.L., Hunters, hides, dollars and dependency: Economics of wildlife exploitation in Belize. Crocodiles: Proceedings 7th Working Meeting of Crocodile Specialist Group,, IUCN Ð The World Conservation Union Gland, Switzerland, pp. 143±152. Jenkins, R.W.G., The world conservation strategy and CITES: principles for the management of crocodilians. In: Webb, G.J.W., Manolis, S.C., Whitehead, P.J. (Eds.), Wildlife Management: Crocodiles and Alligators. Surrey Beatty and Sons Pty. Ltd, Sydney, pp. 27±31. Johnson, W.C., The physical setting: Northern Belize and Pulltrouser Swamp. In: Turner, B.L., Harrison, P.D. (Eds.), Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Settlement in Northern Belize. University of Texas Press, Austin, pp. 8±20. King, F.W., Espinal, M., Cerrato, L.C.A., Distribution and status of the crocodilians of Honduras. Crocodiles: Proceedings 10th Working Meeting of Crocodile Specialist Group., IUCN Ð The World Conservation Union, Gland, Switzerland. pp. 313±354. King, F.W., Ross, J.P., Gutierrez, L.D., Survey of the status of the crocodilians of Nicaragua. Crocodiles: Proceedings 12th Working Meeting of Crocodile Specialist Group., IUCN Ð The World Conservation Union Gland, Switzerland, pp. 121±161. Magnusson, W.E., Techniques of surveying for crocodiles. In: Crocodiles: Proceedings 5th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union Morges, Switzerland, pp. 389±403. Marin, F., Wildlife Protection Act of No. 4. Ministry of Natural Resources. Government Printing O ce, Belmopan, Belize. Marlin, J., Marlin, K., Platt, S.G., A documented case of an attack by Morelet's crocodile (Crocodylus moreletii) on man. Bulletin Chicago Herpetological Society 30, 165±167. McField, M., Wells, S., Gibson, J., State of the Coastal Zone report, Belize, Coastal Zone Management Programme, United Nations Development Programme and Global Environmental Facility. Project No. Bze/92/G31. Government Printing O ce, Belmopan, Belize. Meerman, J.C., The status of crocodiles in the eastern Corozal District. Occasional Papers of the Belize Natural History Society 1, 1±5. Messel, H., Vorlicek, G.C., A model for the population dynamics of Crocodylus porosus in northern Australia. In: Hall, P.M. (Ed.), Crocodiles: Their Ecology, Management, and Conservation. IUCN Ð The World Conservation Union Publications Gland, Switzerland, pp. 209±227. Messel, H., Vorlicek, G.C., Wells, A.G., Green, W.J., Surveys of tidal river systems in the Northern Territory of Australia and their crocodile populations. Monograph No. 1. Pergamon Press, Sydney. Miller, B.W., Miller, C.M., National protected areas management plan zoological report: Faunal and site analysis. Report to Wildlife Conservation Society and Ministry of Natural Resources, NARMAP Project, Government of Belize. Mrosovsky, N., Provancha, J., Sex ratio of hatchling loggerhead sea turtles: Data and estimates from a 5-year study. Canadian Journal of Zoology 70, 530±538. Myers, N., Tropical forests: The main deforestation fronts. Environmental Conservation 20, 9±16. Neill, W.T., The Last of the Ruling Reptiles: Alligators, Crocodiles, and Their Kin. Columbia University Press, New York. Neill, W.T., Allen, R., Studies on the amphibians and reptiles of British Honduras. Contribution No. 1. Research Division, Ross Allen Reptile Institute. pp. 1±76. Neill, W.T., Allen, R., Further studies on the herpetology of British Honduras. Herpetologica 17, 37±51. O'Brien, T.G., A comparison of 3 survey methods for estimating relative abundance of rare crocodilians. In: Crocodiles: Proceedings 10th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Gland, Switzerland, pp. 91±108.

9 S.G. Platt, J.B. Thorbjarnarson / Biological Conservation 96 (2000) 21±29 29 Ouboter, P.E., Status and conservation of crocodilians in northeastern Belize. In: Crocodiles: Proceedings 11th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Gland, Switzerland, pp. 18±29. Platt, S.G., The Ecology and Status of Morelet's Crocodile in Belize. PhD dissertation, Clemson University, Clemson, SC. Platt, S.G., Thorbjarnarson, J.B., Status and life history of the American crocodile in Belize. Report to United Nations Development Programme, Global Environmental Facility, Belmopan, Belize. Belize Coastal Zone Management Project BZE/92/G31. Powell, J., The Morelet's crocodile: an unknown quantity. Animal Kingdom 1972, 21±26. Powell, J., Crocodilians of Central America, including Mexico and the West Indies: Developments since In: Crocodiles: Proceedings 2nd Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Morges, Switzerland, pp. 27±31. Rainwater, T.R., Platt, S.G., McMurry, S.T., A population study of Morelet's crocodile (Crocodylus moreletii) in the New River Watershed of northern Belize. In: Crocodiles: Proceedings 14th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Gland, Switzerland, pp. 206±220. Rebelo, G.H., Magnusson, W.E., An analysis of the e ect of hunting on Caiman Crocodilus and Melanosuchus niger based on the sizes of con scated skins. Biological Conservation 26, 95±104. Ross, J.P., Application of the new IUCN Criteria to crocodilian status evaluation. In: Crocodiles: Proceedings 13th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Gland, Switzerland, pp. 499±504. Ross, J.P., Crocodiles: Status Survey and Conservation Action Plan. IUCN/SSC Crocodile Specialist Group, 2nd Edition. Gland, Switzerland. Sanderson, I.T., Living Treasure. Viking Press, New York. Schmidt, K.P., Notes on Central American crocodiles. Fieldiana 12, 79±92. Sigler, L., The Crocodylus moreletii recovery in northwest of Chiapas State, Mexico. Crocodile Specialist Group Newsletter 14, 17±18. Sigler, L The caiman situation in Chiapas, Mexico. In: Crocodiles: Proceedings 13th Working Meeting of Crocodile Specialist Group. IUCN Ð The World Conservation Union, Gland, Switzerland, pp. 152±152. Taplin, L.E., Osmoregulation in crocodilians. Biological Review 63, 333±377. Thorbjarnarson, J.B., The status and ecology of the American crocodile in Haiti. Bulletin Florida State Museum, Biological Sciences 33, 1±86. Thorbjarnarson, J., Ecology of the American crocodile (Crocodylus acutus). In: Hall, P.M. (Ed.), Crocodiles: Their Ecology, Management, and Conservation. IUCN-The World Conservation Union Publications. Gland, Switzerland, pp. 228±258. Thorbjarnarson, J Crocodiles: An action plan for their conservation. Messel, H., King, F.W., Ross, J.P. (Eds.). IUCN/SSC Crocodile Specialist Group, Gland, Switzerland. Thorbjarnarson, J., Are crocodilian sex ratios female biased The data are equivocal? Copeia 1997, 451±455. Thorbjarnarson, J., Hernandez, G., Recent investigations into the status and distribution of the Orinoco crocodile Crocodylus intermedius in Venezuela. Biological Conservation 62, 179±188. Webb, G.J.W., Messel, H., 1978a. Morphometric analysis of Crocodylus porosus from the north coast of Arnhem Land, northern Australia. Australian Journal of Zoology 26, 1±27. Webb, G.J.W., Messel, H., 1978b. Wariness in Crocodylus porosus (Reptilia: Crocodilidae). Australian Wildlife Research 6, 227±234. Woodward, A.R., Marion, W.R., An evaluation of factors a ecting night-light counts of alligators. Proceedings Annual Conference Southeastern Association Fish and Wildlife Agencies 32, 291±302. Woodward, A.R., Percival, H.F., Jennings, M.L., Moore, C.T., Low clutch viability of American alligators on Lake Apopka. Florida Scientist 56, 52±63. Woodward, A.R., White, J.H., Linda, S.B., Maximum size of the alligator (Alligator mississippiensis). Journal of Herpetology 29, 507±513. Wu, T.H., Rainwater, T.R., Platt, S.G., McMurry, S.T., Anderson, T.A., Organochlorine contaminants in Morelet's crocodile eggs from Belize. Chemosphere 40, 671±678. Zisman, S., The directory of Belizean protected areas and sites of nature conservation interest. Government Printing O ce, Belmopan, Belize.