Mastozoología Neotropical ISSN: 0327-9383 [email protected] Sociedad para el Estudio de los Mamíferos Argentina

Dellafiore, C.M.; Demaría, M.; Maceira, N.; Bucher, E. Distribution and abundance of the in San Luis Province, Argentina Mastozoología Neotropical, vol. 10, núm. 1, enero-junio, 2003, pp. 41-47 Sociedad Argentina para el Estudio de los Mamíferos Tucumán, Argentina

Available in: http://www.redalyc.org/articulo.oa?id=45710105

How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Mastozoología Neotropical / J. Neotrop. .; 10(1):41-47 ISSN 0327-9383 ©SAREM, 2003 Versión on-line ISSN 1666-0536

DISTRIBUTION AND ABUNDANCE OF THE PAMPAS DEER IN SAN LUIS PROVINCE, ARGENTINA

. C.M. Dellafiore1, M. Demaría2, N. Maceira2, and E. Bucher3

1 National Parks Administration, Alberdi 3016, 5800 Río Cuarto, Córdoba, Argentina. 2 National Institute of Agriculture and Technology, CC 17, 5730 Villa Mercedes, San Luis, Argentina. 3 National University of Cordoba, CC 198, 5000 Cordoba, Argentina.

ABSTRACT. The pampas deer was originally distributed in the open grasslands of central and southern . In the Argentine Pampas, two small populations of the endemic pampas deer (Ozotoceros bezoarticus celer) subspecies persist in the eastern and western extremes of its original distribution. We used ground and aerial surveys in the semiarid grasslands of the south of San Luis Province to assess the status of this remaining popu- lation. Population density estimates ranged from 0.43 to 0.83 deer/km2 (488-1,224 ). The female/male ratio indicated a predominance of females over males (1.23, 1.14, 1.35 and 1.54 in August and November of 1995 and February of 1996 and 1997 respectively). This study suggests that the pampas deer in Argentina is an endangered species, even consid- ering the population of pampas deer in Bahía Samborombón, in province (with nearly 200 animals).

RESUMEN. Distribución y abundancia del Venado de las Pampas en la provincia de San Luis, Argentina. El Venado de las Pampas (Ozotoceros bezoarticus) se distribuía originariamente en los pastizales abiertos del este y centro de Sudamérica. En los extremos este y oeste de la Pampa Argentina persisten dos pequeñas poblaciones de la subespecie endémicas O. bezoarticus celer. Empleamos censos aéreos y terrestres para conocer el estatus de la población en el pastizal pampeano semiárido del sur de la provincia de San Luis. La densidad poblacional estimada osciló entre 0.43 a 0.83 ciervos/Km2 (488 – 1.224 animales). La relación hembra/macho indicó un predominio de las hembras sobre los ma- chos (1.23; 1.14; 1.35 y 1.54 en agosto y noviembre de 1995 y febrero de 1996 y 1997, respectivamente). Nuestro estudio sugiere que, aun teniendo en cuenta la población presen- te en Bahía Samborombón - Buenos Aires (cercana a los 200 animales), el Venado de las Pampas en Argentina es una especie en serio peligro de extinción.

Key words: pampas deer, pampas grassland, endangered species, Ozotoceros bezoarticus celer, aerial survey, ground survey.

Palabras clave: venado de las pampas, pastizal, especie en peligro,Ozotoceros bezoarticus celer, censo aéreo, censo terrestre.

Recibido 22 junio 2002. Aceptación final 25 noviembre 2002. 42 Mastozoología Neotropical / J. Neotrop. Mammal.; 10(1):41-47 C.M. Dellafiore et al.

INTRODUCTION estimate population abundance, (b) determine the sex ratio, and c) identify the pattern of The pampas deer (Ozotoceros bezoarticus) was spatial distribution. originally distributed in the open grasslands (pampas) of South America, including central STUDY AREA and southwestern Brazil, southeastern Bolivia, Paraguay, , and northern and central This survey was conducted on an area that Argentina (Jackson and Langguth, 1987). Of extends from 33º 55' S 66º 18' W to 34º 30' S the three recognized subspecies, Ozotoceros 65º 30' W in the General Pedernera Depart- bezoarticus celer is endemic to the Pampas ment, province of San Luis, Argentina. This region of Argentina, and is the only deer spe- region is characterized by sandy soils with cies directly associated with open grasslands slightly undulating relief and sand dunes that in that country. The Pampas grasslands origi- occupy large areas. Annual rainfall averages nally covered 500,000 km2 in east-central Ar- 450 mm and is concentrated in summer (Octo- gentina. However, since 1870 most of the area ber-April). Annual mean temperature is 17ºC. was turned into cropland following a massive The absolute maximum in summer is 43º C, immigration of European settlers. Conversion whereas the absolute minimum in winter is of native grassland to agriculture still contin- -15ºC (Capitanelli and Zamorano, 1972). ues in the more arid, western edge of the re- Although this area was originally included gion. Habitat loss, unregulated hunting, com- in the Espinal region (Cabrera, 1971), Leon petition with (Jackson and Giullieti, and Anderson (1983) considered the area to be 1988), and transmission of cattle diseases the semi-arid, western extreme of the Pampas (Jungius, 1976) have caused the observed de- region. The dominant vegetation in undisturbed crease in pampas deer populations. Today the areas is open grassland dominated by cow grass species is included in the Red Data Book (Sorghastrum pellitum). Other associated (IUCN 1996), in the Red Book of Threatened grasses include Elionurus muticus, of Argentina (Diaz and Ojeda, 2000), Bothriochloa springfieldi, Chloris retusa, and is listed in CITES Appendix I. Schizachyrium plumigerum, Eragrostis lugens, At present, only two small isolated popula- Sporobolus subinclusus, Aristida spegazzini, tions of pampas deer remain in the Pampas Poa ligularis and Poa lanuginosa. A few region. The first is located in the coastal woody plants occur in the native grasslands, marshes and grasslands of the Bay of including patches of chañar (Geoffroea Samborombón in the province of Buenos Aires, decorticans), and isolated individuals of calden where a reserve of some 15,000 ha exists (Prosopis caldenia) and alpataco (Prosopis (Gimenez Dixon, 1991). The second is in our alpataco) (Anderson et al., 1970). study area in the south of San Luis province At present, the grasslands have been drasti- where the deer population is concentrated over cally altered by overgrazing, plowing, and in- 145,000 ha (Dellafiore et al., 2001). troduction of exotic grass species. Therefore, population dynamics is a function of the landscape appears to be more heteroge- population density, age structure, sex ratio, and neous, with each area reflecting its manage- recruitment rate. The evaluation of population ment history (Anderson 1973; Anderson et density and distribution is important for the al.1978) development of management programs. The San Luis population of pampas deer remained METHODS almost unknown until Jackson and Langguth (1987) estimated 300 animals remaining in a The pampas deer is distributed over 400,000 ha, sample area of 300 km2. At present the conser- but its populations are concentrated on 145,000 ha vation status of the pampas deer in San Luis (Dellafiore et al, 2001). Aerial and ground sam- needs to be evaluated because of the growingly pling efforts were made over an area ranging be- 2 degraded habitat. Our objectives were to (a) tween 1,311 to 2,848 km . THE PAMPAS DEER IN SAN LUIS PROVINCE, ARGENTINA 43

Ground and aerial surveys a total width of 80 m in 1995 and 140 m in 1997. Two observers, one on each side of the plane, made The ground survey was conducted following the the counts, recording all the deer seen inside and internal roads in the area (fire lanes), whereas par- outside the transect boundaries. allel strip transects were used for the aerial survey. In both cases, we took a GPS point to record deer Abundance observations. Data were analyzed with “use of space” Relative abundance of deer seen via air and land of the GIS CAMRIS (Ecological Consulting 1993), were determined by the method of Järvinen and which allows the creation of percent convex poly- Väisänen (1975, 1976, 1977, 1981), which is used gons, where probability levels (P) specify the quan- for low-density species. This technique is an ampli- tity of observations that lie within them (thus, the fication of the strip transect method and considers 0.90 nucleus contains 90% of all the observations both the animals seen inside and outside the sam- in the nuclei). pling band. Variance was calculated when the num- Ground sampling ber of registered deer was greater than 25, while we used the jacknife method when less than 25 deer During 1995-96, the sample area was 1,520 km2, were counted (Miller, 1974). The confidence inter- distributed over three ranches or groups of ranches. val was calculated by a t-test (D± ta Ö and R- In 1997, the sampling area was increased to 1,870 1 DF). Abundance data estimated by air and land 2 km and covered seven ranches. Deer were counted sampling were compared via the Wilcoxon test for along transect lines (Järvinen and Väisänen, 1975, comparison of two samples (Sokal and Rohlf, 1980). 1976, 1977, 1981), travelling in a pick-up truck at a constant speed of 20 km/h. Each strip transect Dispersion was 17 km long; 16 independent transects were evaluated in 1995-96, and 30 independent transects We used the data obtained from aerial and ground were surveyed in 1997. Deer counts were conducted surveys to test the hypothesis of clumped disper- by two observers, looking at opposite sides of the sion of deer in each sampling season (c2 test). The road and counting all animals seen inside and out- dispersion index “I” was used to define the vari- side the transect (by means of separate registrations) ance/mean ratio (Tellerías, 1986). of 200 m (100 m to each side). We recorded the sex of each deer observed. RESULTS

Aerial survey Ground and aerial surveys We surveyed an area of 1,311 km2 in 1995 and 2,848 km2 in 1997. In 1995, we used a Cessna 192, Abundance and in 1997 we flew a Cessna 172. The average altitude was 90 m above ground at a constant speed Estimations of density from ground sampling of 100 km/h. Deer were counted along parallel strip data varied between 0.43 and 0.55 deer/km2 transects 36 km long in 1995, and 65.5 km long in 1997. Distance between transects was 2.5 km in (488-688 animals; Table 1). In contrast, esti- 1995, and 11.6 km in 1997. Each transect covered mates of density from aerial surveys varied

Table 1 Number and density of deer estimated by land and air censuses.

TERRESTRIAL CENSUS AERIAL CENSUS

Traveled Estimated Estimated Confidence Traveled Estimated Estimated Confidence surface density number interval surface density number interval Km (deer/ km²) 95% Km (deer/ km²) 95%

August 95 206 0.55 615 +275 462 0.83 1094 +638 November 95 228 0.43 488 +295 462 0.72 943 +563 February 96 237 0.47 523 +201 ______February 97 445 0.44 688 +366 458.5 0.43 1224 +990 44 Mastozoología Neotropical / J. Neotrop. Mammal.; 10(1):41-47 C.M. Dellafiore et al.

0 5 10 15 20 25

Kilometers .9000 - .7000 .7000 - .5000 .5000 - .1027 Sampling area Observation

Fig. 1. Nucleus of density of the pampas deer for the land sampling of 1995-96. Probability levels (P) specify the quantity of observations that lie within them (thus, the 0.90 nucleus contains 90% of all the observations in the nuclei). These data were analyzed with the chapter “use of space” of the GIS CAMRIS (Ecological Consulting 1993).

0 5 10 15 20 25

Kilometers Sampling area .9000 - .7000 .7000 - .5000 .5000 - .1412 Observation

Fig. 2. Nucleus of density of the pampas deer for the land sampling of 1997. Probability levels (P) specify the quantity of observations that lie within them (thus, the 0.90 nucleus contains 90% of all the observations in the nuclei). These data were analyzed with the chapter “use of space” of the GIS CAMRIS (Ecological Consulting 1993). THE PAMPAS DEER IN SAN LUIS PROVINCE, ARGENTINA 45

0 5 10 15 20 25 Kilometers .9000 - .7000 .7000 - .5000 .5000 - .0000 Sampling area Observation

Fig. 3. Nucleus of density of the pampas deer for the aerial sampling of 1995. Probability levels (P) specify the quantity of observations that lie within them (thus, the 0.90 nucleus contains 90% of all the observations in the nuclei). These data were analyzed with the chapter “use of space” of the GIS CAMRIS (Ecological Consulting 1993).

0 5 10 15 20 25 Kilometers .9000 - .7000 .7000 - .5000 .5000 - .0000 Sampling area

Fig. 4. Nucleus of density of the pampas deer for the aerial sampling of 1997. Probability levels (P) specify the quantity of observations that lie within them (thus, the 0.90 nucleus contains 90% of all the observations in the nuclei). These data were analyzed with the chapter “use of space” of the GIS CAMRIS (Ecological Consulting 1993). 46 Mastozoología Neotropical / J. Neotrop. Mammal.; 10(1):41-47 C.M. Dellafiore et al.

between 0.43 and 0.83 deer/km2 (943-1,224 seventeen animals. The area is characterized animals; Table 1). We found no differences in by some degree of heterogeneity, which is the abundance between the two methods (Wicolxon product of several factors (e.g., different to- test, P < 0.05). pography, different degrees of grassland con- The greatest proportion of space use (per- servation, etc.). cent convex polygon), both in air and land samplings, was observed in an area of 145,000 Sex ratio ha (Figs. 1- 4). This result did not change even Sex ratios are basic considerations used in when the sampling area was doubled in 1997. planning management or in assessing the deer population dynamics (Hungerford et al., 1981). Sex ratio We observed that the female/male ratio indi- The female/male ratio was 1.23 (n: 85, un- cates a predominance of females. Moore and determined animals: 22) in August 1995, 1.14 Mueller-Schwarze (1993) found a similar rela- (n: 82, undetermined: 3) in November 1995, tionship in Uruguay, and suggested a sex-bi- 1.35 (n: 58, undetermined: 17) in February ased mortality pattern because the neonatal sex 1996, and 1.54 (n: 72, undetermined: 21) in ratio approached 1:1. Illegal hunting of pam- February of 1997. Undetermined animals were pas deer still persists in Argentina and could generally distant or young deer. support that hypothesis.

Dispersion MANAGEMENT IMPLICATIONS The individuals always presented a clumped dispersion (c2, P < 0.01). This was also ob- The total population of O. bezoarticus celer is served through the use of the space analysis still very low, and fragmented into two sub- made with GIS CAMRIS (Figs. 1-4). There populations (San Luis, 500-1,200 animals; was no seasonal pattern with respect to the Bahía Samborombon, 220 animals; Vila and size of the groups. The size of the groups varied Beade, 1997), increasing the risk of extinction between 1 and 12 in ground surveys, and be- of this subspecies. However, the San Luis tween 1 and 17 in aerial sampling. population has good potential for preservation and recovery via the development of plans DISCUSSION favoring the conservation and sustainable use of the natural grassland. Although the pampas Abundance and dispersion deer is an emblematic and charismatic species, the fact that its populations occur in the pro- The population of pampas deer in San Luis ductive region of Argentina implies a challenge province is probably between 500 and 1,200 to conservation biologists because we must animals. Dellafiore et al. (2001) observed that coordinate conservation and the sustainable the pampas deer population was concentrated development of the area. over 145,000 ha, as was also observed in this study. Even when the sampling area (aerial ACKNOWLEDGMENTS survey 1997) was doubled, the nuclei remained concentrated on a 145,000-ha area (Figs. 3 and We thank the Wildlife Management Master of Science 4). These 145,000 ha correspond to areas where Program of Cordoba University, the National Institute of Agricultural Technology (INTA) of San Luis and the natural grasslands are in better condition in San National Parks Administration (APN) for their academic, Luis. technical, and economic support. We are grateful to the Similarly to what was observed by Jackson personnel and authorities of the V Brigada Aérea of the (1985) and Vila and Beade (1997) at Bahía Argentine. Air Force who provided planes and pilots to perform the air censuses. We thank Laura Valinotti for Samborombón, the size of the pampas deer English corrections. We especially thank Dr. Guy groups varied between solitary individuals and Baldasarre for suggestions and comments on the work. THE PAMPAS DEER IN SAN LUIS PROVINCE, ARGENTINA 47

LITERATURE CITED JACKSON, J. 1985. Behavioural observations on the Argentinian pampas deer (Ozotoceros bezoarticus celer Cabrera, 1943). Zeitschrift für Säugetierkunde, ANDERSON, D.L.; J.A. DEL ÁGUILA, and A.E. 50:107-116. BERNARDÓN. 1970. Las formaciones vegetales de JACKSON, J. and A. LANGGUTH, 1987. Ecology and la Provincia de San Luis. Rev. Inv. Agropecuaria status of the pampas deer in the Argentinian pampa INTA, serie 2, Biología y Producción Vegetal, 7(3):83- and Uruguay. Biology and Managment of the Cervidae 153. (Wemmer, C.M., ed.). Smithsonian Inst. Press, Was- ANDERSON, D.L. 1973. La distribución de Sorghastrum hington DC Pp. 402-409. pellitum (Poaceae) en la Provincia de San Luis y su JACKSON, J. and J. GIULLIETI, 1988. The food of significado ecológico. Kurtziana, 12-13:37-45. pampas deer Ozotoceros bezoarticus celer in relation ANDERSON, D.L.; E.L. ORIONTE, L. ENRIQUE, and to its conservation in relict natural grassland in Ar- J.C. VERA. 1978. Una reliquia del pastizal de San gentina. Biological Conservation, 45:1-10. Luis. Ecología, 3:139-151. JÄRVINEN, O. and R.A., Väisänen. 1975. Estimating CABRERA, A.L. 1971, Fitogeografía de la Republica relative densities of breeding birds by the line transect Argentina. Boletín de la Sociedad Argentina de Bo- method. Oikos, 26:316-322. tánica, XIV (1-29) JÄRVINEN, O. and R.A. VÄISÄNEN. 1976. Finnish line CAPITANELLI, R. and M. ZAMORANO. 1972. Geogra- transect census. Ornis Fennica, 53:115-118. fía regional de la Provincia de San Luis. Boletín de JÄRVINEN, O. and R.A. VÄISÄNEN. 1977. Line transect Estudios Geográficos. (20) 74-77. Facultad de Filo- method: a standard for field work. Polish Ecological sofía y Letras. Universidad de . Studies, 3:7-17. DELLAFIORE, C.M.; M. DEMARÍA, N. MACEIRA, and JÄRVINEN, O. and R.A. VÄISÄNEN.1981. Methodology E. BUCHER. 2001. Estudio de la distribución y for censusing land bird faunas in large regions. Studies abundancia del venado de las pampas en la provincia in Avian Biology, 6:146-151. de San Luis mediante entrevistas. Revista Argentina JUNGIUS, H. 1976. Status and distribution of threatened de Producción Animal, 2(2):137-144 deer species in South America. World Wildlife DIAZ, G.B. and R. OJEDA. 2000. Libro Rojo de Mamí- Yearbook. Pp. 203-217. Morges, Switzerland: World feros Amenazados de la Argentina. SAREM. 106 pp. Wildlife Fund. ECOLOGICAL CONSULTING. 1993. CAMRIS: Siste- LEÓN, R.J.C. and D.L. ANDERSON. 1983. El límite ma de inventario de recursos asistido por computado- occidental del pastizal pampeano. Tuexenia, 3:67-83. ra. Rare Center for Tropical Conservation, U.S.A. MILLER, R.G. 1974. The jacknife - A review. Biometrika, GIMÉNEZ DIXON, M. 1991. Estimación de parámetros 61:1-15. poblacionales del venado de las pampas (Ozotoceros MOORE, D. and D. MUELLER-SCHWARZE. 1993. bezoarticus celer, Cabrera 1943 - Cervidae) en la Pampas deer (Ozotoceros bezoarticus subsp.) costa de la Bahía de Samborombón (Provincia de demographics. State University of New York College Buenos Aires) a partir de datos obtenidos mediante of Environmental Science and Forestry. Syracuse, New censos aéreos. Tesis Doctoral Universidad Nacional York, EE.UU. 11 pp de la Plata. 116 pp. SOKAL, R. and F.J. ROHLF 1980. Introducción a la HUNGERFORD, C.R., M.D. BURKE, and P.F. Bioestadística. Ed. Reverté, S.A. 362 pp. FOLLIOTT. 1981. Biology and Population dynamic TELLERÍAS, J.L.1986. Manual para el censo de los of in Southwestern United States. Pp. 109- vertebrados terrestres. Raices Editorial. 278 pp. 132. VILA, A. and M. BEADE. 1997. Situación de la pobla- IUCN, 1996. IUCN Red list of threatened animals. Gland, ción del venado de las pampas en la Bahía de Suiza. Samborombón. Boletín Técnico FVSA Nº 37, 30 pp.