Neotropical dry forest wildlife water hole use and management

Selected wildlife species diurnal use of a natural water hole (QO) and an artificial water hole (AW) were studied during 1990 dry season at Guanacaste Conservation Area, Costa Rica. In total, 9 19 individuals (six marnmal and one garne bird species) consumed water from QO, while 713 individuals (four maromal species) consumed water from AW. Estimated daily water consumption by selected wildlife species was 29.7 I at O.D and 27.3 I at AW. Estimated 2 4-h water consumed by all wildlife species or evaporated was 44.6 1 at QO and 41.1 1 at AW. This resulted from summing: a) water consumed by studied species, b) estimated 24-hour water consurned by other wildlife (QO = 14.85 1, AW = 13.65 1) and c) daily water evaporation (QO = 0.04 1, AW = 0.10 1). Ouring a 120-day dry season, AW required about 4 932 1 of water from the park administration. Management implications for neotropical dry forest water holes are discussed.

Use and rnanagernent of water sources for wildlife has been studied in serniarid and sea sonally dry environrnents on several conti nents.In Africa, water holes are critical for wildlife existence (Ayeni 1975, Hitchcock 1996, Knight 1989, Knight et al. 1988).In cen tral Mexico, white-tailed deer (Odocoileus vir ginianus) depend on water holes and pre forrned water sources in dry season (Mandujano and Gallina 1995).Wildlife use (see Krausrnan and Etchberger 1996) and rnan agernent of natural and rnan-rnade water deve1oprnents (see Kie et al. 1994, Tsukamoto and Stiver 1990, and Yoakurn et al. 1980) has been well studied in the western United States.In neotropical dry forests, McCoy et al. (1990), Vaughan and Rodriguez (1991) and Shaw (1996 unpublished rnanuscript) docurnented wild1ife water ho1e use, while Cornelius (1974) and Vaughan et al. (1982) recornrnended rnan agernent practices for water holes.The objec tives of this study were: a) to estirnate selected species diurnal use of water at a natural and an artificial water hole in the neotropical dry for est, b) to estirnate 24 h and seasonal water con surnption by all wildlife species at both water holes and e) to recornrnend water hole man agernent practices for wildlife.

MATERIALS AND METHODS
Study site: The study was carried out in Guanacaste Conservation Area (GCA), Costa Rica, located 30 km north of Liberia, the Guanacaste Province capital.GCA consists of 70 000 ha, and was created to preserve the largest area of primary dry forest (several hun dred hectares) and restore the rest to tropical dry forest, considered the most endangered tropical ecosystem (Janzen 1986).Located between Matzalan, Mexico and northwest Costa Rica, the Pacific Mesoamerican tropical dry forest receives 900-2400 mm of rainfall annually between May and December.During the January to May dry season, surface water becomes scarce.Moisture-containing vegeta tion, fmit, prey and dry season water holes (tree hollows, springs and pools ín stream and riverbeds and artificial sources) may provide water, which maintains wildlife in the region.
The two water holes monitored were located in forested areas within 800 m of the Casona (85°37"N, lO050"W) on a plateau between 220-350 m elevatíon.No other water holes were found within 500 m .The natural water hole (QD) was located in a secondary dry forest in Quebrada Duende, approximately 10 m SE of the stream intersection wíth the Indio Desnudo Nature TraiL Many tourists walk the nature trail and visit QD, hoping to observe wildlife there.QD was enlarged from about four to a 12 I capacity in 1987 using cement and rock.The artificial water hole (A W) was located about 1.5 km NE of QD in a 20-year old forest with occasional Pithecellobium sama n trees.Unknown to tourists, AW was originally installed in 1988 for use by white-faced monkeys (Cebus capucinus).It consisted of half a tractor tire with an 11 1 water capacity fed by a 500 m long 1.0 cm plastic hose from a nearby camp ing area.
Wildlife species observation: Between March 8-April 28, 1990, all mammalian species and one game bird species which visit-ed QD and A W. between 0600-1800 h were counted.For each individual observed, data on species, date and time of visit was taken.Species and numbers of small birds and rep tiles were not documented.Observations at QD were made from a large rock located approximately 10 m N of the water hole and partially hidden by large rocks and trees.The observation blind for A W was behind a stone corral fence, approximately 30 m N of the water source.One observer, using binoculars when necessary, recorded data on alternate days at each water holeo Analysis: For analysis, average number of individuals/species were calculated by di vid ing total observations for each species visiting each water hole by 18 days of observation.Estimated daily consumption (liters) by an individual of each species was calculated mul tiplying average species body weight (kg) (Robinson and Redford 1986) by the for mula proposed by Calder (1984).
r!lti�� (A+ a +E) w�apprq�lltel}';44,(d;a, ú " .QD .���l.l l; afAW.Estin)a�c1;tota1,�a,ter con� s�aruteYaPc.mtttl4auringa'12o"WlY dry sea,." , son ��6352l �I' QP an,d4 9'�21iatAW,•, ' .Preformed water: Water intake as animal tissue, succulent lea ves and fruit is called pre formed water and may meet part or all of daily water needs for some species (Calder 1984).Mandujano and Gallina (1995) found o. vir gínianus dependent on Spondias purpurea fruit for preformed water in Mexico during the late dry season when vegetation and water holes contributed liule water.However, succu lent Ieaves were uncommon during the late dry season at our study site.We analyzed dry sea son diets in dry forest for N. narica, O. vir ginianus, C. capucinus, T. tajacu (Espach and Saenz 1995, McCoy et al. 1990, Morera 1997, Moscow and Vaughan 1987, Saenz 1995).C. capucinus fed infrequently on animal tissue in the form of insects, reptiles, mammals or birds.Probably fruits of only two species of the 14 reported, Manikara chicle (N.na rica) and Solanum americana (O.virginianus), provided sufficient quantities of preformed water.Based on an apparent lack of preformed water sources for selected wildlife species in GCA, we consider water holes very important for wildlife species in GCA during late dry season.
Body size vs water usage, evaporation: Although 28% more individuals visited QD then AW (919 vs 713), only about 10% more water was consumed at QD (29.7 1 vs 27.3 1).This was because heavier species, such as T. tajacu, were proportionally more abundant at AW then QD and consumed more water then lighter species.Although C. rubra and T. tajacu visited QD equally (115 vs 116), C. rubra consumed only 25% of the water T. tajacu did because it was lighter and had a dif ferent water balance formula.Over a 120-day dry season, at QD, C. rubra would consume 228 I of water, while T. tajacu would consume 864 I of water.O. virginianus consumed as much water as six N. narica, six C. capucinus or four A. geoffroyi.T. bairdii weighs 300 kg (Robinson and Redford 1986) and requires 14.5 I of water daily, equivalent to seven O. virginianus, 44 N. na rica or 50 C. capucinus.IncIuding nocturnal species, insects, andlor birds, would more accurately calculate total water hole water consumption.Evaporation accounted for only 1-2% of water 10ss, because of the small surface area.
Water hoJe management: Natural and artificial water holes can be managed: a) to increase species and carrying capacity of wildlife using them, b) to stimulate regenera tion by attracting wildlife which defecate seeds near the water, e) to expand habitats for wildJife during dry periods, and d) to stimulate ecotourism and biological educatíon use by human users.
The manager can use plastic and/or masonry work to reduce water wastage at water holes.Shade, reducing surface area and other methods can reduce evaporaríon (see Kie et al. 1994 andTsukamoto andStiver 1990).Where applicable, domestic livestock should be excluded from wildlife water developments, because they consume at least four to five times the water deer and other large neotropical game species do and contaminate water holes with trampling, wallowing, etc (Rice 1990).If water is limited, controlling O. virginianus numbers by hunting and/or fencing could favor smaller species such as primates or coatimundis, impor tant for wildlife viewing and as seed dispersers (Brigham 1990, Ernst andTolsma 1990).
Placing artificial water holes strategically on public and private lands could increase wildlife species carrying capacity if sufficient food and shelter exists (see Kie et al. 1994, Tsukamoto and Stiver 1990, and Yoakum et al. 1980.Water hole management could also favor wildlife protection and visitor observatíon.AW and QD received comparable number of wildlife visitors throughout the 1990 dry season, indicat ing relative importance of artificial water holes.AW would require almost 5 000 I of water dur ing a normal 120-day dry season to satisfy wildlife species needs.This could be in the form of a stórage tank, well or other water source. In OCA, Vaughan et al. (1997) found 14 of 20 natural water holes monitored during the 1995 dry season dried up, sorne early in the dry season.QD has dried up at least a month early since 1990.Natural phenomena (ie El Nino?), overuse of aquifers by park officials and other factors might be responsible.However, without free or preformed water sources, wildlife might be forced to migrate or perish.Should artificial water sources replace natural water holes in sea sonally dry environments to better control the water resource?Managers should calculate water availability, local aquifer capacity, administrative and wildlife water needs on their ranches and wildlands.The present "laissez faire" approach to wildlife water management in neotropical dry forest regions is a cause for concern.