The feeding ecology of cape vultures Gyps coprotheres in a stock-farming area

The feeding ecology of cape vultures Gyps coprotheres in a stock-farming area

Biological Conservation 35 (1986) 63-86 The Feeding Ecology of Cape Vultures Gyps coprotheres in a Stock-Farming Area A. S, Robertson* Department of ...

1MB Sizes 1 Downloads 23 Views

Biological Conservation 35 (1986) 63-86

The Feeding Ecology of Cape Vultures Gyps coprotheres in a Stock-Farming Area A. S, Robertson* Department of Zoology, University of the Witwatersrand, Johannesburg, South Africa

& A. F. Boshofl The Lakes Nature Conservation Station, Private Bag 6546, George, South Africa

ABSTRACT Cape vultures Gyps coprotheres in the southwestern Cape Province jeed exclusively on sheep carcasses, within a limitedjoraging area. The size and shape of the Joraging range was determined by means of a postal survey and confirmed by a radio-tracking study. The quantity ojJood available within the range, while seasonally variable, was estimated to exceed the colony's requirements. Data pertaining to daily jeeding jorays o! individuals, monthly Joraging patterns oJ the colony and the growth ~] nestlings indicated no seasonal shortages in the amount ojJood obtained. The colony remains susceptible to the eJJects of poisons used in the area; levels oJ contaminants recorded in most eggs are considered low.

INTRODUCTION Along with six other Gyps species distributed throughout the Old World (listed in Brown & A m a d o n , 1968), Cape vultures G. coprotheres are described as specialist feeders on the carcasses of migratory ungulates * Present address: Department of Zoology, University of Natal, Pietermaritzburg 3200, South Africa. 63 Biol. Conserv. 0006-3207/86/$03.50 ~,~'~ElsevierApplied Science Publishers Ltd, England, 1986. Printed in Great Britain

64

A. S. Robertson, A. F. Boshoff

(Houston, 1974). The evolution of this exclusive scavenging behaviour is suggested to have occurred under conditions which probably no longer exist throughout any griffon species' range in Africa (Houston, 1979). Houston's (1972) study of the feeding and breeding ecology of two griffon species inhabiting a relatively natural environment in East Africa is then especially relevant when considering the congeners that range over areas of intensive agricultural development in southern Africa (see Mundy, 1982). In particular, his study of Ruppell's griffon vulture G. rueppellii (Houston, 1974, 1975, 1976), that area's ecological counterpart to G. coprotheres (Mundy, 1982), provides numerous points of potential comparison. Throughout their southern African range, Cape vultures have undergone a marked transition in feeding ecology in the last two to three centuries (Mundy, 1982; Mundy & Ledger, 1976). Such a transition is mirrored in the situations of similarly endangered populations of congeners elsewhere in the world, e.g. Europe (Perco et al., 1983; Smith, 1984). The effects of the initiation and development of agricultural practices on the status and distribution of Cape vultures throughout the Cape Province of South Africa are described by Boshoff & Vernon (1980), and specifically for the southwestern Cape by Boshoff & Vernon (1979) and Robertson (1984). This paper presents the results of a study of the nature and effects of this altered food supply on the feeding ecology and aspects of reproduction of Cape vultures at a particular colony.

STUDY AREA Observations were made at a colony in the Potberg (34 ° 22' S; 20 ° 33' E) in the De Hoop Nature Reserve near Bredasdorp. About 60 vultures inhabit a ravine in the mountain and use southeast- and southwest-facing cliffs for nesting and roosting. In addition a smaller colony of about 20 vultures located at Aasvogelvlei (33 °52' S; 21 °38' E), which is 120km NE of the Potberg, was monitored for numbers of birds and breeding success (Fig. 1). The area has a temperate Mediterranean climate, receives most of its rainfall ( + 530 mm a year) from about May to September (i.e. winter), and has a warm to hot and dry summer. The year 1981 was exceptionally wet in that 64 ~o more rain fell than the average obtained for the previous 25 years (J. C. Michler, pers. comm.). The strongest winds occur in midsummer, when prevailing winds are southeasterly, whereas in winter

Ecology 0[ the Cape Vulture"

65

I

0

G

~g

0

©

©

..~

~o

o

ffl ..1 _1

© ©

o

.< o

,d

0

66

A. S. Robertson, A. F. Boshoff

northwesterly winds are frequent. In 1982, the farm Driefontein (14 km W of the Potberg) received an average of 7.2 h of sunlight per day (range 5.2 9.6h); only in April, June and July did this figure fall below 6 h per day (J. P. Pietersen, pers. comm).) With the exception of the south, the area around the mountain is now a patchwork of wheatfields and pastures. About 70 ~o of the area that was previously Coastal Rhenosterbosveld (Acocks, 1975) has been cultivated and the combined Bredasdorp-Swellendam area now contains some 600 000 sheep and 35 000 cattle (J. P. Pietersen, pers. comm.). METHODS The study was divided into three aspects, viz. an investigation of (a) the potential food available; (b) foraging behaviour; and (c) the amount of food obtained by the vultures. The first-mentioned was carried out in the farmlands surrounding the colony. For the rest, observations began in the vulture ravine in May 1981, lasted for 12 days per month, regardless of weather conditions, and ended in May 1982. Altogether, 165 days were spent at a fixed observation point, averaging about 8.5h per day. Observations were conducted with a 15-60 x telescope mounted on a tripod and 8 x 40 binoculars. Potential food available Four methods were used to determine the size of the foraging range of the Potberg colony: (a) a questionnaire, requesting information on vultures seen, was posted to the farmers in the environs of the colony; (b) personal interviews were conducted of a random sample of 45 farmers within a 40 km radius of the colony; (c) individual feeding sites were located by means of a radio-tracking study of the foraging behaviour of an adult vulture (Boshoff et al., 1984); (d) feeding times, defined as the length of time taken by an individual to leave the nest site, locate food and eat and return to the site, were documented where possible for breeding birds-the times were grouped into calendar months and analysed by multiple range tests, using a Hewlett-Packard HP-85 calculator. Information pertaining to the availability of carcasses to the vultures was obtained by personal interviews with farmers within the suspected foraging range. The farmers were questioned on, inter alia, sheep mortality rates (and temporal variations thereof), carcass management procedures and their personal attitudes to the vultures.

Ecology o[ the Cape Vulture

67

Eggs that failed to hatch after about 67 days were collected, and analysed by gas chromatography.

Food obtained, and foraging behaviour To determine the diet of the colony, the bases of the breeding and roosting cliffs were searched for regurgitated pellets and artefacts, and farmers were interviewed to obtain their personal observations. Artefacts and other indicators of food type were collected during visits to the nests. The breeding ledges were visited four times in 1981 and five times in 1982, but individual sites were visited no more than three times in any one year, in order to minimise disturbance. The Aasvogelvlei sites were visited once in each year. A number of carcasses consumed by the vultures were located by searching the farmlands close to the colony. We attempted to determine both the actual amount of food obtained by the vultures, as well as an indication of its availability. One of the methods used, the visual estimation of crop content of returning foragers, has been attempted before (Houston, 1976). The crop projects beyond the contour feathers when full and is observable in both flying and perched birds (see Houston, 1976: Fig. 3). For each observation day, estimates were made of the proportion of the colony considered to have obtained food that day, and individual breeding sites were monitored and the degree of crop distension of each occupant was scored into one of the following three categories, whenever possible: 0 = N o food obtained: 1 = Food obtained, less than about 700 g; 2 = Food obtained, more than about 700 g. At each nest visit, the nestling was weighed and its wing length measured. As nearly all 1981 and 1982 nestlings from Potberg were of known age, these figures were directly comparable to the correlations of mass and growth obtained at certain other colonies (Mundy, 1982). During the observation days, counts of birds present in the ravine were made every 30 min. The results were grouped by calendar month and compared by Tukey's multiple range tests.

RESULTS

Foraging range A breakdown of the postal survey is given in Table 1, and the results are depicted in Fig. 2. Vultures sighted by farmers in the farmlands

. .

• .:.

.



• ..':.:

,.,



Oo

.~







~

. : ... •

"



-k

o •

i

°





°

~,

0 °

(b)





t

®





,



"

®

ill

m•

/,'

~ KM

m m



. ,

m







'~ •



,

,,

..

Fig. 2. Results of the postal survey, giving the location of: (a) farmers who reported that they'never see' vultures and (b) farmers who reported that they see vultures 'once a year' or more often. The star marks the location of the Potberg colony; B--Bredasdorp, H - Heidelberg, R--Riversdale, S--Swellendam towns: the stippled line encloses the daily foraging range; the limestone ridge south of the colony is indicated by dotted lines.

69

Ecology o1 the Cape Vulture TABLE 1

Breakdown of the Postal Survey

Questionnaires mailed returned (viable) returned (incomplete) Unsolicited information supplied

Total

"/ o

1173 (A) 555 (B) 51 156

47 (of A) 4 (of A) 28 (of B)

surrounding the Potberg are presumed to be colony members. Figure 3 includes sightings of birds that are presumed to have foraged from the Aasvogelvlei colony, as the sightings were made in farmlands close to that colony. The current foraging range, an area of approximately 194 000 ha, as determined by comparison of Fig. 2(a) and (b), is shown on Fig. 2(b). The limestone ridge (see Fig. 2(b)) is uncultivated, supports a comparatively low number of stock and is not included within the foraging range in the calculations of potential food. The De Hoop Nature Reserve is also excluded because of low ungulate numbers, the behaviour of the radio-marked vulture and the observation that vultures are 'never seen' over the area (P. van der Westhuizen, pers. comm.). The overall average feeding time was 3 h 42 min (range 52 min 7 h 52min, n = 322) (Fig. 3). Analysis indicated significant differences between various months, in particular shorter feeding times for months during which lambing occurred, as compared to out-of-lambing season months (Tukey's multiple range test, 0.01 level). From the radio-tracking study, a factor of distance to feeding site per feeding time was obtained: using this, the above average corresponds to feeding sites located about 13 km from the ravine. Nature of the food source

A range of artefacts, bones and bone fragments as well as seven sheep eartags, presumed to have been brought to the ravine by the vultures, was found at the bases of the breeding and roosting cliffs, and at nests (Table 2). All 45 farmers interviewed considered that the vultures currently fed exclusively on sheep carcasses, although five were of the opinion that dead cattle could 'very likely' supplement this diet. All carcasses visited by vultures, that were located during the radio-tracking study, were of sheep (Boshoff et al., 1984).

70

A. S. Robertson, A. F. Boshoff

SIGNIFICANT

I

2

DIFFERENCES

3 '~ 4 5

I 2 6

7

8

9 10

11"

3* 6

4 6 7 8 9 11"

5 6 10

27

29

24

A

M

46 350

24

24

300

27

30

27

39

26 250. z

200 I,IJ I'-

150

100'

50

J

A

S

O

N

D

J

F

M

MONTHS

Fig. 3. The sample sizes, means and SD of times taken for feeding forays by individual vultures, averaged per calendar month. Months that are significantly different share the same number (above the Figure) (Tukey's multiple range, 0.05 level); * indicates significance at the 0.01 level.

Stock contained within the foraging range In the calculations of the theoretical amount of food available to the vultures, only sheep are considered because evidence obtained during this study indicated that only sheep carcasses were eaten. The average annual adult sheep mortality estimate obtained from the interviews was 2.2 ~o. All farmers interviewed considered sheep deaths to be temporally dependent, with a peak at the time of lambing. In addition, in the winter rainfall region an average of 16.9 ~o of lambs die between birth and weaning (Louw, 1970). All farmers considered the months of March-April to be the main lambing season, while 38 (847o) indicated that approximately 30~o of one year's lambing occurs in the September October period. Also, losses from diseases such as lamb

Ecology ol the Cape Vulture

71

TABLE 2 Material Found at Nests (Where possible, the bones have been identified: vertebra (v), metatarsal (mr) or rib (r); sizes are in mm)

1981

1982

Potberg Bone intact 3 5 x 10, 3 7 x 12, 2 5 x 35 fragment 7 5 x 15, 6 0 x 1 0 , 2 0 × 10 Horny hoof covering nil Other nil Regurgitate wool

95 x 20, 35 × 25(v), 20 × 30, 35 × 10(mr), 25 × 15(v) 70 × 30(r), 55 x 10, 55 x 17, 10 x 60 5 I - glass 20 × 30 wool

Aasvogelvlei Bone intact fragment Horny hoof covering Regurgitate

nil 75 x 10, 80 x 10

35 x 20(v) 40 x 10(r)

1 wool, skin of head (lamb)

1 nil

dysentery and enterotoxaemia are heaviest during the winter months (Monnig & Veldman, 1976). An estimated average stocking rate of 1.7 sheepha-t of cultivated farmland was obtained from the interviews. This is probably an underestimate, e.g. on the farm Mopama (35 km N of the Potberg), about 8500 sheep are pastured on about 2000 ha, an average of 4.2 sheep ha- 1 (M. G. Lourens, pers. comm.). The potential number of sheep deaths that can be expected in one year is then calculated as follows: Foraging area' 74 ..../o under cultivation: 1.7 sheep ha - 1 Mortality estimate:

194 000 ha 143 560 ha (J. P. Pietersen, pers. comm.) 244052 sheep 4264 sheep.

Of the 45 farmers questioned, 29 (62 ~0) considered it important to remove the carcass from the grazing lands as soon as possible. If the carcass was "reasonably' fresh, it would always be removed, but 16 farmers replied that carcasses in an advanced state of decomposition would either be left or covered with poison. If the factor of two-thirds is

72

A. S. Robertson, A. F. Boshoff

applied to the figure reached in the above calculations, 1790 carcasses could theoretically be available to the vultures within the foraging range in one year. This is a minimum estimate: not all carcasses are located and not all those that are removed become unavailable to the vultures (personal observation). A recent study of the daily requirements of Cape vultures (Komen, 1984) confirmed earlier estimates (Houston, 1972; Mundy, 1982) of 300-500 g of lean meat per day. Thus 60 vultures require about 1100 kg per year, or 730 sheep carcasses, assuming that one sheep carcass supplies 15kg of meat (Jarvis et al., 1974). In addition, each nestling requires about 76 kg of lean meat between hatching and leaving the nest (Komen, 1984).

Amount and frequency of food obtained The results of the visual estimation of the proportion of the colony considered to have obtained food during a particular observation day are expressed as averages of each calendar month (Fig. 4). The overall monthly average of 40~o translates into an individual feeding approximately every two and a half days. Difficulties were experienced with this method, and are considered in the Discussion. The results of individual crop estimates are presented in Table 3, where an overall average step of 1.5 crop estimates were obtained per nest per day (Table 3, column A/B). As most birds at breeding sites were not individually recognisable, only breeding sites where a fledgling was raised in the 1981 season were monitored. Of all crops recorded, 53 ~ were

212

M

J

J

A

S

0

N D J MONTHS

F

M

A

M

Fig. 4. Dailyestimations of the proportion of the colony considered to have obtained food in an observation day. The proportions are averaged per calendar month, and the mean estimate is given (:~).

,..,,

+B

©

•+?_+,.~ ~

-

-

~

,~-

~+-

~

B

¢...

,.. •

+~

:~

~ ~

~._~ ~

e",

~-

e-

~ ~.~ •

e~

,,~

~1-

.~+

~-

~ -

74

A. S. Robertson, A. F. Boshoff

classed as empty; this indicates that individuals obtain food in one of every two foraging trips. The difference between the number of times birds were recorded as having obtained food, compared to returning with an empty crop, was not significant (Z 2 = 2.66, df = 6, p > 0.25). Of the birds recorded as having obtained food, 69 ~o were estimated as having obtained more than their daily requirement (Table 3, column E/D). Body mass and wing lengths of known-age Potberg nestlings are compared to the curve of similar measurements made at other colonies

~7

i~i ///

6

20

40

60 80 100 AGE (days)

120

(a) 700 I-/B/

E 6o0 E

/

•- i - 5 0 0 uJ 4 0 0 ._1

i/"

/

• /

0•

~300

/,

/

/

/

200

,// J

100

2o

4o

6o 80 100 AGE (days)

120

(b) Fig. 5.

Growth of Potberg nestlings in (a) body mass and (b) winglength, in relation to the mean curve from the Magaliesberg (Transvaal) colonies (Mundy, 1982).

Ecology oJ the Cape Vulture

75

(Fig. 5) (Mundy, 1982). Figure 6 also shows the body mass-wing length correlation, measured at Aasvogelvlei during the study period, in relation to the mean curve of Transvaal nestling growth. In 1981, four Potberg nestlings were visited twice, and one was weighed once. In 1982, two nestlings were visited three times; the age of one of these was not known accurately. The weight of another nestling was not obtained. Thus all but -/• 8¸

IP -K.-

6

co cO ,,¢

5

3E

4/

• •

/ • 1977, 79, 8 0 ~ 1981 • 1982

/

2-

e/

* avv 1981

/

~ a v v 1982

100200300400500600 WING

LENGTH

(ram)

Fig. 6. The relationship between body mass and wing length of nestlings weighed during the study period at Potberg and Aasvogelvlei (avv), as well as Potberg nestlings weighed in 1977, 1979 and 1980 (unpublished records). The straight line is the relationship obtained for the Magaliesberg colonies (supplied by P. J. Mundy).

one of the nestlings were heavier than the mean weight at the same age that was recorded at the Magaliesberg colonies, and all but two had longer wing lengths for their age (in the two with shorter wing lengths, feathers had not yet appeared). One 1981 and one 1982 nestling had exceptionally long wings for their age: nestling periods were 139 days and 124 days (shortest recorded in this study), respectively. Three large nestlings left their nests prematurely in November 1982, most likely the result of human disturbance (Robertson, 1984). Two were

A. S. Robertson, A. F. Boshoff

76

located within a week of the event and post-mortems were carried out by a veterinary surgeon. The level of fat storage was considerable in both and was classed as score 6 and score 7 (highest possible score of 9, see Houston, 1976). The stomach of one contained several sticks, the largest measuring 180 × 10mm, an end of which had caused an ulcer in the pyloric region and most likely a small rupture in the left liver lobe (localised fibrinous peritonitis seen). No parasites were found in the intestinal tracts of either nestling. All interviewed farmers living more than 15 km from the colony noted that the proportion of carcasses they located that had been consumed by the vultures was 'particularly small'.

Foraging behaviour Proportion of time spent foraging The number of birds present in the ravine during each observation day were averaged for each calendar month (Figs 7 and 8), to indicate the proportion of each day that the colony spent foraging. During the summer months (October-March), at least 50 ~o of the birds had left the ravine by 0930 h (3 months) or 1000 h, and in winter the birds left at least half an hour (one time interval) later. Conversely, 50~o or more had returned by 1330 h or earlier (3 months) or 1400 h in the summer months, and by 1400h (5 months) or later in the winter months. At 1000h, significantly more birds were present in May 1982 than from September-February inclusive (Tukey's multiple range test, 0.05 level). The same held for the l l 0 0 h count between May 1982 and October,

60

+-~

50

~4o ~) 30O" ;¢2010-

TIME OF DAY

Fig. 7.

Foraging pattern of the whole colony for May 1981. The means ( + SD) of vultures present in the ravine at each time interval are averaged.

Ecology o/ the Cape Vulture

77

'"

JJJIl ~o~ ul'J,,,ll :

:

;

;

I

I

I

I

8 10 12 14 16

j~rlltlfl~ '

;

I

I

I

+

tl

t1982 ,

l

I

I

I

I

t

I

I

I

!

!

!

!

,

,

:

;

l !

A I

|

|

II

!i

|

!

|M=

Fig. 8. Monthly foraging patterns of the whole Potberg colony: June 1981 May 1982. The means ( _+ SD) of vultures present in the ravine at each time interval are averaged per calendar month. The axes captions are as for Fig. 7,

November, December and March. There were no significant differences at the 0.05 level between the other hourly counts. Most vultures returned to the ravine after each day's foraging, as established by the counts made at the end of each day. The mean of each month's standard deviation of vultures present at the end of each day was 2.04 individuals.

80

A. S. Robertson, A. F. Boshoff

Jarvis et al. (1974) calculated a theoretical daily foraging radius of 180-225 km from the Potberg colony. This resulted in a feeding range of about 200 000 km 2. Jarvis et al. (1974) concluded that it was possible that the colony's requirements were met within an area of 35 km radius, although that study did not consider that the Indian Ocean reduces this potential foraging area to one half. The methods used here to define the size of the current daily foraging range indicated that it was smaller than a theoretical maximum: (a) as areas of vulture 'presence' indicated by the postal survey were surrounded by areas of 'absence', the boundary was defined with relative confidence. Also, the area covered by this survey was clearly sufficient to encompass these daily foraging movements. With a maximum radius of some 40 km, this boundary is within one hour's flight of the colony. Individuals may obviously exceed the boundaries, e.g. the radio-marked vulture roosting on the Langeberg (Boshoff et al., 1984), and other sightings, documented in Robertson (1984); (b) All 'feeding sites' located during the radiotracking study were situated within the boundaries of the foraging range; (c) If the factor of 'time spent away from the ravine' per distance to 'feeding' site is valid, the average of 13 km to 'food' sources is well within a vulture's foraging capabilities. As the feeding forays were documented for breeding birds, this distance may be longer; active breeders would be expected to return to their nests and nestlings sooner (Drent & Daan, 1980). If 10 birds on average visit carcasses and each carcass supplies about 15kg (Jarvis et al., 1974), then if each bird removes about 1200g (Houston, 1976), no food remains. Adult merino sheep weigh about 50 kg and presumably more than 33 ~o of this weight is edible to vultures, e.g. 65 ~ of an impala Aepyceros melampus is edible to griffon vultures (Mundy et al., 1983), then a carcass will obviously supply more food. The figure of 15 kg is therefore a minimum estimate. In essence, the feeding range is the summation of the individual requirements of colony members; comparison of colony requirements with the expected number of carcasses available to the vultures in one year within the range tends to reflect this. Similar inverse correlations are documented between the home ranges of individual raptors and the numerical density of their prey (reviewed in Schoener, 1968). The shape of the range is probably influenced by two topographical features of the area, viz, the orientation and elongated shape of the

Ecology ~?[ the Cape Vulture

81

Potberg, and the limestone ridge to the west of the colony (Figs 1 and 2(b). The strong southeasterly winds in summer and the shape of the mountain would result in considerable wave lift over the surrounding northwestern farmlands. Also, the vultures are often seen flying close to, and in parallel with, the limestone ridge. This is consistent with the expectation of vultures exploiting orographic lift (Pennycuick, 1972), in this case ridge lift, produced by winds incident on the southern side of the ridge.

Foraging behaviour The foraging patterns of the colony as a whole are remarkably consistent over the observation period given the, albeit slight, change in colony numbers, the variation in hours of sunlight between winter and summer, and the change in breeding status of some 36 birds per year Ctying' them to sites). This is in marked contrast to rooks Corvus frugilegus, for example, which spent more than 90 ~o of the daylight feeding during summer when food was scarcest (Feare, 1972). The standard deviation of the mean of each month's daily counts provides an indication of the variability observed during individual days: as a general rule, this variability was greater during the morning than it was in the afternoon (except January). This variation likely reflects the bird's dependence on lift (thermals, or orographic lift induced by wind): in the mornings they would congregate on specific cliffs and 'wait' for suitable flying conditions. The shortest proportion of time spent foraging, in terms of food theoretically available, would be expected during lambing season (March-May), a period of above average mortality. This would result in a higher probability of a food item occurring within a given radius of the colony. Birds returned at very similar times during these months (standard deviation of the 1530 h means of vultures present of 4.2, 2.7 and 3.0). The length of feeding forays documented during these months support this indication of increased food availability. In contrast, the variability in numbers returning at the end of the day was greatest in January and February and, along with the reasonably lengthy individual feeding forays documented during these months, this probably reflects a relatively lower availability of food. Similarly, the radio-marked vulture's mean furthest foraging site and mean furthest ~feeding' site in summer were more than twice as far as those in winter (Boshoff et al., 1984).

84

A. S. Robertson, A. F. Boshoff

requirements; in general terms there is a spatial sufficiency. In terms of the temporal availability of food, the evidence indicates that both adults and nestlings obtained sufficient quantities during the nestling period. But is this food available to all age groups? Young vultures are less efficient at obtaining food than older birds (Houston, 1976; Mundy, 1982) and have low survival rates (Piper et al., 1981 ; Robertson, 1984). The lengthy post-fledging dependence period exhibited by juveniles (Robertson, 1985) may be indicative of difficulty in finding food. In this respect, comparative data are required from other colonies. We contend that an immature is more likely to be behaviourally excluded by adults from ingesting food at a small (e.g. sheep) carcass rather than at a large (e.g. hartebeest) carcass. Two factors could reinforce this disadvantage. First, the smaller carcass would presumably be eaten up faster, and by fewer birds, and secondly, adults, which outnumber juveniles by a ratio of some three to one (Robertson, 1984), generally initiate feeding at a carcass. In conclusion, whether the transition to feeding on stock carcasses has been achieved with equal success by all age groups remains to be fully investigated; this aspect is discussed further in Boshoff & Robertson (1985). ACKNOWLEDGEMENTS We thank M. Centner for advice with statistical analyses, N. Drager for performing the post-mortem inspections, A. C. de Kock of the University of Port Elizabeth for analysis of egg contaminants, G. L. Maclean, P. J. Mundy and H. E. H. Paterson for criticising manuscripts, and J. P. Pietersen of Agricultural Technical Services for supplying information. During the fieldwork, Robertson was employed by the Cape Department of Nature and Environmental Conservation, and received bursaries from the Council for Scientific and Industrial Research and the University of the Witwatersrand in 1983. REFERENCES Acocks, J. P. H. (1975). Veld types of South Africa. Mem. Bot. Surv. S. Air., No. 40. Boshoff, A. F. & Robertson, A. S. (1985). A conservation plan for the Cape vulture colony at Potberg, De Hoop Nature Reserve, southwestern Cape Province. Bontebok, 4, 25-31.

Ecology o[ the (7ape Vulture

85

Boshoff, A. F., Robertson, A. S. & Norton, P. M. (1984). A radio-tracking study of an adult Cape griffon vulture Gyps coprotheres in the south-western Cape Province. S. Afr. J. Wildl. Res., 14, 73-8. Boshoff, A. F. & Vernon, C. J. (1979). Supplementary notes on the Cape vulture in the Cape Province. Cape Town, Cape Department of Nature and Environmental Conservation. Boshoff, A. F. & Vernon, C. J. (1980). The past and present distribution and status of the Cape vulture in the Cape Province. Ostrich, 51, 230--50. Brown, k. H. & Amadon, D. (1968). Eagles, hawks and litlcons o.[ the world. Feltham, Country Life. Bryant, D. M. & Gardiner, A. (1979). Energetics of growth in house martins Delichon urbica. J. Zool., Lond., 189, 275 304. Drent, R. H. & Daan, S. (1980). The prudent parent: energetic adjustments in avian breeding. Ardea, 68, 225 52. du Plessis, S. F. (1969). The past and present geographical distribution o1 the Perissodaetyla and Artiodactyla in southern A/i'ica. MSc dissertation, University of Pretoria. Feare, C. J. (1972). The seasonal pattern of feeding in the rook (Corvus J?ugilegus) in northeast Scotland. Proc. int. Orn. Congr., 15th, 643. Houston, D. C. (1972). The ecology o[ Serengeti vultures. DPhil thesis, University of Oxford. Houston, D. C. (1974). Food searching in griffon vultures. E. AJi'. Wildl. J., 12, 63 77. Houston, D. C. (1975). Ecological isolation of African scavenging birds. A rde~, 63, 55 64. Houston, D. C. (1976). Breeding of the white-backed and Ruppell's griffon vultures, Gyps aJ?icanus and Gyps rueppellii. Ibis, 118, 14 40. Houston, D. C. (1979). The adaptations of scavengers. In Serengeti, Dynamics ot an ecosystem, ed. by A. R. E. Sinclair and M. Norton-Griffiths, 263-86, Chicago, University of Chicago Press. Jarvis, M. J. F., Siegfried, W. R. & Currie, M. H. (1974). Conservation of the Cape vulture in the Cape Province. J. S. Aji'. Wildl. Mgmt. Ass., 4, 29 34. Komen, J. (1984). The energy demands on parent Cape vultures at the Skeerpoort breeding colony. In Proceedings of the Second Symposium on Al}'ican predator)' birds, ed. by J. M. Mendelsohn and C. W. Sapsford, 215-16, Durban, Natal Bird Club. Ledger, J. A. (1979). AJrican insect liJe. Cape Town, Struik. Lockie, J. D., Ratcliffe, D. A. & Balharry, R. (1969). Breeding success and organo-chlorine residues in golden eagles in West Scotland. J. appl. Ecol., 6, 381 9. Louw, D. J. F. (1970). Prevention of lamb mortality. The Wool Grower, 24, Suppl., i-iv. Monnig, H. O. & Veldman, F. J. (1976). Handbook on stock diseases. 2nd edn. Cape Town, Tafelberg. Mundy, P. J. (1982). The comparative biology oJ southern AJrican vultures. Johannesburg, Vulture Study Group.

86

A. S, Robertson, A. F. Boshoff

Mundy, P. J., Grant, K. I., Tannock, J. & Wessels, C. L. (1982). Pesticide residues and eggshell thickness of griffon vulture eggs in southern Africa. J. Wildl. Mgmt, 46, 769-73. Mundy, P. J. & Ledger, J. A. (1976). Griffon vultures, carnivores and bones. S. Afr. J. Sei., 72, 106-10. Mundy, P. J., Morris, A. & Haxen, C. M. (1983). The proportion of an impala edible to vultures. Afr. J. Ecol., 21, 75-6. Newton, I. (1979). Population ecology of raptors. Berkhamsted, T. & A. D. Poyser. Newton, I. (1980). The role of food in limiting bird numbers. Ardea, 68, 11-30. Peakall, D. B. & Kemp, A. C. (1976). Organochlorine residues in herons and raptors in the Transvaal. Ostrich, 47, 139-41. Pennycuick, C. J. (1972). Soaring behaviour and performance of some East African birds, observed from a motor-glider. Ibis, 114, 178-218. Perco, F., Toso, Susic, G. & Apollonio, M. (1983). Initial data for a study on the status, distribution and ecology of the griffon vulture (Gyps Julvusfulvus Hablizl 1783) in the Kvarner Archipelago. Larus, 33-35, 99-134. Piper, S. E., Mundy, P. J. & Ledger, J. A. (1981). Estimates of survival in the Cape vulture Gyps coprotheres. J. Anita. Ecol., 50, 815 25. Robertson, A. S. (1983). The feeding ecology and breeding biology oJ a Cape vulture colony in the southwestern Cape Province. MSc dissertation, University of the Witwatersrand, Johannesburg. Robertson, A. S. (1984). Aspects of the population dynamics of Cape vultures in the Cape Province. Ostrich, 55, 196 206. Robertson, A. S. (1985). Observations on the post-fledging dependence period of Cape vultures. Ostrich, 56, 58-66. Schoener, T. W. (1968). Sizes of feeding territories among birds. Ecology, 49, 123-41. Schweitzer, F. R. & Scott, K. J. (1973). Early occurrence of domestic sheep in sub-Saharan Africa. Nature, Lond., 241, 547. Skead, C. J. (1980). Historical mammal incidence in the Cape Province, V. 1. The western and northern Cape. Cape Town, Cape Department of Nature and Environmental Conservation. Smith, J. K. (1984). The Aragon region of the Pyrenees--A stronghold for birds of prey in Europe. Biol. Conserv., 29, 307-20.

Biological Conservation 35 (1986) 87 96

The Distribution, Status, and Level of Exploitation of the Freshwater Turtle Dermatemys mawei in Belize, Central America Don Moll Department of Biology, Southwest Missouri State University, Springfield, Missouri, USA

ABSTRACT

Dermatemys mawei /s common to abundant in many areas of northern and central Belize and is present in two river systems of extreme southeastern Belize. It mainly exists in larger, deeper, lowland rivers and large freshwater lagoons. It has seriously declined in some areas near human population centres and in other accessible, easily collected habitats. It is mainly marketed in Belize City, Belize District, in low to moderate numbers that peak during the spring reproductive season. The main source of animals is the Belize River. Some animals from the Rio Grande River are marketed in spring in Punta Gorda, Toledo Dist. While the status of Dermatemys is relatively good in Belize, compared to Mexico, it is not currently protected by law, and is vulnerable to increasing exploitation from an increasing human population. Other commonly marketed freshwater turtle species in Belize, Staurotypus triporcatus and Pseudemys scripta venusta, are common to abundant, and widely distributed throughout the country.

INTRODUCTION The Central American river turtle D e r m a t e m y s m a w e i is a species of concern to conservationists. It is listed as 'vulnerable' in the International Union for the Conservation of Nature and Natural Resources Red Data Book, Amphibia-Reptilia, Part 1 (IUCN, 1982), is given 'highest priority' ranking for research by the Freshwater Chelonian Specialists 87 Biol. Conserv. 0006-3207/86/$03"50 © ElsevierApplied SciencePublishers Ltd, England, 1986. Printed in Great Britain

88

Don Moll

Group of the IUCN, and is given Appendix II listing by the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). It is sought for meat and eggs throughout its limited range (coastal lowlands of the western Caribbean from southern Mexico through northern Guatemala, Belize, and extreme northwestern Honduras), and is known to have declined drastically in Mexico (Alvarez del Toro et al., 1979; IUCN, 1982). Its status elsewhere remains unclear. During the summer of 1983, and the winter and spring of 1984, I conducted surveys to determine the status, distribution, and level of exploitation of this species in Belize. This paper summarises the results and conclusions of these surveys.

METHODS Searches for Dermatemys were conducted by outboard motor-powered boats, by trammel netting, and occasionally by snorkeling in suitable habitats in Belize. Habitats were selected for search based upon their likelihood of containing Dermatemys, and their accessibilty. Analysis of topographic maps, interviews with local citizens and scattered references to Dermatemys collecting at particular sites in various published and unpublished sources provided guidance in selection. Due to the abundance of prime habitat, surveys were mainly conducted in the northern half of Belize, but searches were also conducted in larger streams crossing the southern highway in south central Belize and in the Rio Grande River of southeastern Belize. Surveys were conducted along predetermined routes at slowest motor speed (i.e., 'trolling' speed) during periods of maximum Dermatemys activity (i.e., when many individuals are floating or actively foraging at the water surface). In inland lagoons or river sections, surveys were usually conducted after 2000 h due to the characteristic nocturnal behaviour of Dermatemys in these habitats. In river sections and bays, under tidal influence, surveys were conducted during a rising tide regardless of time since tidal influx stimulated foraging upon incoming floating vegetation. Time of search period and number of Dermatemys sighted along the search route during boat surveys were recorded. In nocturnal surveys observations were limited by the strength of the torch beam used for illumination of individuals (approx. 20m on either side of the boat's course). When more than one survey could be conducted, different search routes were

Exploitation of a freshwater turtle

89

followed but the possibility of recounting mobile individuals cannot be discounted. The number of specimens collected in trammel nets, set near foraging areas, was also recorded at sites where nets were used, as was the number seen collected by turtle fishermen. The Lincoln-Peterson technique of population estimation, based upon the recapture of marked individuals, was employed at the Progresso Lagoon study area to obtain an absolute size and density estimate for this population. The area of Progresso Lagoon was determined by planimetry.

D I S T R I B U T I O N OF Dermatemys mawei IN BELIZE This turtle is widely distributed in northern and central Belize, and is also present in two river systems in the extreme southeastern part of the country (Fig. 1). It has probably been present in Belize since the Pleistocene, when northern Belize and the modern coastal plain emerged above sea level (Greenfield et al., 1982), and dispersal was possible from previously emergent source areas in southern Mexico. Dermatemys in Belize is primarily found today in larger, deeper, lowland rivers, and large freshwater lagoons, most of which are located in the northern half of the country. Some are also currently present in smaller tributary streams of the larger rivers but not as abundantly as in earlier years (see account of status below). Several individuals were collected in brackish water in Corozal Bay near the mouth of the New River, and one shell from Belize was covered with barnacles (Neill & Allen, 1959), indicating regular occurrence in saltwater habitats. Despite search efforts in many of the larger streams crossing the Southern Highway between Dangriga and Punta Gorda specimens were not collected, and are believed to be extremely rare or absent in these shallow, high-gradient streams draining the Maya Mountains complex of southern Belize. Estuarine habitats were not searched in this part of Belize (Toledo District), with the exception of the Rio Grande, which did contain Dermatemys. Therefore, the possibility of undiscovered coastal populations exists. Potential Derrnatemys habitats in extreme northwestern Belize (western Orange Walk District), such as the Rio Bravo and Booth River systems, were not sampled because of inaccessibility, but unsubstantiated reports of Dermatemys in these habitats were received.

96

Don Moll

and management of this economically important species before its populations are allowed to deteriorate would allow its sustained use and survival in Belize for the foreseeable future. The relatively large, undisturbed populations of Belize also offer an opportunity to accumulate data representative of natural circumstances which may be difficult or impossible to gather in more disturbed and depleted populations (i.e., as in Mexico). Such data are vital for the formulation of useful regulations concerning commercial, and private exploitation of populations. A C K N O W L E D G E M ENTS I wish to thank Mr G. Winston Miller, Fisheries Administrator of Belize's Fisheries Unit Laboratory, for permission to conduct these surveys in Belize. The Revd Leonard E. Dieckman, SJ, of Saint John's College, Belize City, provided invaluable help in the initial stages of this project, and the Belize A u d u b o n Society, notably its secretary, Mrs Lydia Waight, provided useful information concerning the status of Dermatemys in Belize earlier in the century. Thanks are also due to Mr Alan Burn of Belize City for his assistance in surveys of the Belize River, and to Mr and Mrs Otto and Thea Maraun of Corozal Town for the generous loan of their boat and motor used in these surveys. Funding for these studies was provided by the World Wildlife Fund-US, the Southwest Missouri State University Faculty Research Fund, the SMSU Foundation, and the Fauna and Flora Preservation Society. REFERENCES Alvarez del Toro, M. (1982). Los reptiles de Chiapas, 3rd edn. Inst. Zool. del Estado. Tuxtla Gutierrez, Chiapas, Mexico. Alvarez del Toro, M., Mittermeier, R. A. & Iverson, J. B. (1979). River turtle in danger. Oryx, 15, 170-3. Greenfield, D. W., Greenfield, T. A. & Wildrick, D. M. (1982). The taxonomy and distribution of the species of Garnbusia (Pisces: Poeciliidae) in Belize, Central America. Copeia, 1982, 128-47. IUCN (1982). Red data book, Arnphibia-Reptilia, Part 1. Morges, International Union for the Conservation of Nature and Natural Resources. Neill, W. T. & Allen, R. (1959). Studies on the amphibians and reptiles of British Honduras. Publ. Res. Div. Ross Allen's Reptile Inst., 2, 1-76. Waight, Lydia (1983). The hicatee. Belize Audubon Soc. Bull., 15(3), 3-4.