Predictive distribution maps of rodent reservoir species of zoonoses in Southern America

Ximena Porcasi¹, Gladys E. Calderón², Mario Lamfri¹, Marcelo Scavuzzo¹, Marta S. Sabattini³, and Jaime J. Polop⁴

¹ Instituto Gulich - Comisión Nacional de Actividades Espaciales, Centro Espacial Teófilo Tabanera, Córdoba, Argentina. ² Instituto Nacional de Enfermedades Virales humanas "Dr. Julio I. Maiztegui", ANLIS "Dr. Carlos G. Malbrán", Pergamino, Bs. As., Argentina. ³ Universidad Nacional de Córdoba, Córdoba, Argentina. ⁴ Departamento de Ciencias Naturales, Universidad Nacional de Río Cuarto, Agencia Postal N°3, 5800 Río Cuarto, Córdoba, Argentina. <jpolop@exa.unrc.edu.ar>

ABSTRACT: We model potential distribution for three species of rodents known to be reservoirs of zoonotic diseases: Calomys musculinus, Oligoryzomys flavescens and O. longicaudatus. These models provide general distribution hypotheses obtained using environmental data from record localities. Satellite remote sensing is then used to extrapolate climatic and ecological features of potentially suitable habitats for these rodents. In the three species mapped, we found high overall correspondence between predicted (based on environmental data) and specimen based distributions. The maps proposed here provide several advantages over dot and shaded outline maps. First, the predictive maps incorporate geographically explicit predictions of potential distribution into the test. Second, the validity of the predictive map can be appreciated when localities of previous records of the studied species, not used as training sites or used as control sites, are overlaid on the map. In this approach, environmental factors, criteria and analytical techniques are explicit and can be easily verified. Hence, we can temporally fit data in more precise distribution maps.

Key words. Environmental factors. Geographic distribution. Rodent reservoirs.

INTRODUCTION

   The distributions of many taxa are poorly known in most of the Neotropics in genera land Argentina in particular (e.g., Redford and Eisenberg, 1992). For example, most regions in Argentina have only received cursory and unevenly distributed surveys. However, the geographic distributions of species have a growing number of applications in conservation biology, evolutionary studies, viral disease host and invasive-species management (Mc Arthur,1972; Ricklefs and Schluter, 1993; Kerr and Ortrovsky, 2003).
   Museum records constitute the primary documentation in the process of outlining the boundaries of a species. Distribution maps constructed based on these data represent the probable geographic space inhabited by individuals of the same species. Historically, two types of maps based on museum records have been produced: dot maps and shaded outline maps. Dot maps (documented localities plotted on a map) depict a species' range of distribution in a very conservative way, leaving the reader to draw conclusions regarding the true distribution. By contrast, shaded outline maps attempt to extrapolate a species range among and beyond known localities. They are highly dependent on subjective knowledge of the group under consideration, on the study region (Jamas and McCullock, 2002), and on the assumption that a species' spatial distribution is linked to specific habitat requirements. These associations constitute approximations of a species' realized ecologic niche, defined as the conjunction of ecological conditions within which it can maintain populations without immigration (Grinnell, 1917a, b; Mac Arthur, 1972;Peterson et al., 1999; Peterson et al., 2002a, b).
   Recent advances in remote sensing of climatic and ecological features via satellite have begun to be used to identify particular environments that are suitable for different species. Techniques that model species' requirements using environmental characteristics of localities of known occurrence represent a great advance in the production of range maps(Walker and Cocks, 1991; Carpenter et al.,1993; Skov, 2000; Peterson, 2001). Plant species composition and vegetation structure are important determinants of habitat quality. Furthermore, the spatial arrangement of different habitat types within heterogeneous landscapes has been shown to be a strong influence on the size and persistence of animal populations. Factors such as terrain slope and elevation, temperature, humidity, soil texture, and vegetation, exert a profound influence over the distribution of many species (Wilcove et al.,1986). Geographic information systems (GIS) use these environmental data from locations of a species' known occurrence to produce a model of its requirements in those environmental dimensions, and then projects them onto geographic space in order to create a map of the species potential distribution based on the modeled environmental parameters (Hay et al.,1996; Engelthaler et al., 1999; Pikula et al.,2002). Predictive distribution maps suggest where the species is likely to be present (Ricklefs and Schluter, 1993). Models used in ecology to predict species abundance have been based on linear relationships among environmental variables (Manel et al. 1999), and assumed to be normally distributed, raising statistical and theoretical concerns (Austin and Meyers, 1996; Lek et al., 1996). New modeling paradigms (e.g., logistic regressions) account for these shortcomings by accommodating binomial errors and have been successfully used (Osborne and Tigar, 1992; Green et al.,1994; Austin and Meyers, 1996). Anderson et al. (2003) suggest that use of the Genetic Algorithm for Rule selection (GARP), which works with stochastic elements and yields multiple solutions, can be successfully employed in cases where only presence data are available.
   Modeling approaches provide a new global framework in epidemiologic studies by accessing new variables which could improve epidemiological modeling for emerging infectious diseases (Linthicum et al., 1987; Croner et al.,1996; Beck et al., 1997; Cheek et al., 1998;Dale et al., 1998; Engelthaler et al., 1999; Becket al., 2000; Boone et al., 2000; Glass et al.,2000; Anderson et al., 2002; Chaput et al.,2002; Pikula et al., 2002). One of the most basic pieces of information for designing and directing a prevention program for any disease is a more accurate knowledge of the geographic distribution of the reservoir species that defines the potential endemic area of the disease(Mills, 1999; Singleton et al., 1999).
   Two rodent-borne diseases have been considered major public health problems in Argentina (Sabattini and Maiztegui, 1970; Sabattini et al., 1977; Maiztegui et al., 1986; Mills et al., 1994; Calderón et al., 1999; Mills, 1999; Enria et al., 2000). Argentine Hemorrhagic Fever (AHF) is a human disease produced by Junín virus. The virus is maintained in nature in the sigmodontine rodent Calomys musculinus and is transmitted to humans predominantly through aerosolized particles of contaminated rodent excreta. The disease occurs in a small but expanding area on the central Argentine pampa, representing only a small portion of the total distributional range of the reservoir species.
   Calomys musculinus has been captured in disturbed fields of crops and stable habitats (De Villafañe et al., 1977; Kravetz, 1978; Kravetz and Polop, 1983; Mills et al., 1991; Polop and Sabattini, 1993). The mice exhibit distinct habitat associations within their range, and detailed studies of physical and environmental factors influencing their distribution have been carried out (Crespo et al., 1970; De Villafañe, 1970; Contreras and Rossi, 1980; Kravetz and de Villafañe, 1981; Kravetz and Polop, 1983; Busch et al., 1984; Zuleta et al., 1988; Bilenca, 1993; Polop and Sabattini, 1993).
   The other rodent-borne disease is Hantavirus Pulmonary Syndrome (HPS). In Argentina, five genotypes are associated with human disease: Orán (ORN) and Bermejo (BMJ) viruses in northwestern Salta province, Lechiguanas (LEC) and HU 39694 viruses in central Buenos Aires and Santa Fe provinces, and Andes (AND) virus in southwestern Río Negro province (Enría and Levis, 2004). Hantaviruses are maintained in nature in their species-specific rodent hosts and transmitted to humans through aerosolized particles from contaminated excreta. The natural reservoirs are rodents of the genus Oligoryzomys: O. longicaudatus is the host for ORN and AND; O. chacoensis for BMJ, and O. flavescens for LEC and HU 39694 (Levis et al., 1997, 1998). These rodent species have been captured in disturbed (cultivated fields and edge areas), as well as grasslands and forests areas (Massoia and Fornes, 1966; Massoia, 1973; Contreras and Berry, 1983; Pearson, 1983; Mills et al., 1991; Redford and Eisenberg, 1992; Polop and Sabattini, 1993; Monjeau et al., 1998; Pardiñas et al., 2000; Pardiñas et al., 2003; Piudo et al., 2005).
   Oryzomys longicaudatus and O. flavescens are found predominantly in stable disturbed habitats (pastures, forests, roadsides, and railroad right-of-way) (Kravetz and Polop, 1983; Mills et al., 1991; Andrades-Miranda et al., 2001; Delfraro et al., 2003).
   The regional distribution of C. musculinus, O. longicaudatus and O. flavescens is poorly known (Thomas, 1927; Hershkovitz, 1962; Massoia and Fornes, 1966; Olrog and Lucero, 1981; Redford and Eisenberg, 1992; Massoia and Pardiñas, 1994; Monjeau et al., 1994, 1997, 1998; Saba et al., 1995; Pardiñas, 1999; Pardiñas et al., 2000; Pardiñas et al., 2003; Pardiñas et al., 2004). Because these species are known from localities with differing ecological conditions, extrapolation is necessary to estimate the species' distributions. The objective of this study was to develop a predictive ecological distribution map of C. musculinus, O. flavescens and O. longicaudatus using digital databases of potential environmental determinants.

MATERIALS AND METHODS

   Appendix 1 shows the localities of field trapping used to define each species' presence sites as "training sites", these localities were obtained from both authors' field data and bibliography. In order to evaluate the potential distributional range of C. musculinus, O. flavescens and O. longicaudatus we used geographic computer databases on spatial distribution of environmental factors including the following data:
    1) Precipitation: A latitude-longitude rasterized grid of 30-year-monthly mean rainfall (1930-1960) with a spatial resolution of 30 min (Leemans and Cramer, 1991; Cramer and Leemans, 2001). IIASA database, Luxemburg, Austria. Available from / /www.daac.ornl.gov
    2) Vegetation Index: A 1982-1992 time series of Normalized Difference Vegetation Index (NDVI) from meteorological satellite of National Oceanic and Atmosphere Administration/Advanced Very High Resolution Radiometer (NOAA/AVHRR) with a pixel of 8 x 8 km. NDVI=(Ch2-Ch1)/(Ch2+Ch1), Ch is the channel of AVHRR sensor. This index represents greenness but does not distinguish between primary, agricultural, or disturbed habitats
    3) The Land Surface Temperature (LST) of the former temporal series LST= Ch4+3.33(Ch4-Ch5) (Price, 1984).
    4) Digital Elevation Model (DEM): 1 km x 1 km of spatial resolution data from AVHRR sensor, provided by USGS, 1998. Eros Data Center. Available from http://edcdaac.usgs.gov/gtopo30/ gtopo30.asp
    Environmental data were geo-referenced to a latitude- longitude coordinate system and were resampled using nearest neighbor algorithm to adjust the pixels size to 1 km x 1 km. ENVI 3.5 (System Research) was used in all geographic analyses. The resulting raw data have 737 x 1037 pixels and include South America between 13° and 56° S and 33° and 82° W. The analysis includes data derived from monthly averaged satellite images. Temporal series covered the period 1982-1992. Eastman and Fulk (1993) suggest the use of standardized principal component analysis (PCA), because it decomposes the time-series into a sequence of spatial and temporal components that may often be interpreted as corresponding to particular environmental features or events. Typically the first component indicates the characteristic value of the variable, whereas subsequent components represent change elements of decreasing magnitude. In PCA, spatial variation is registered by the first component, and the temporal variation of this vegetation index is represented by the second and third components. Due to the different units of measures for the variables under consideration herein, a standardization of each new variable (i.e. the components obtained from PCA) and DEM product was done before integrating them into an image composed of seven bands, and analyses carried out afterwards. We used a PCA of a NDVI time series of South America obtained by Gorla (2002). Here, the first two components of each environmental variable accounted for more than 99% of the total variance of each series. Thus, both first and second components of each environmental variable and DEM were used in classification analysis. We elaborated a multi-dimensional map in which each geographic area was characterized by this set of seven variables (first and second component of NDVI, precipitation and LST, and DEM) and where "training sites" or "classes" represent the environmental conditions for the presence or absence of C. musculinus, O. flavescens, and O. longicaudatus.
    Minimum size of training sites (number of pixels) required for classification analysis increase directly with the number of variables included. Thus, presence sites (Appendix 1, 2, 3) were composed of the pixel with the exact geographical coordinates of field sampling sites and its closest eight neighbor pixels (we assume similar values in those pixels due to the fact that they are environmental variables).
    Field absent sites were not homogeneously distributed over a wide area. To characterize all geographic regions involved in the "absence" class (which is the largest area in all species at the considered scale), we decided to add to that absence localities, random points over South America (excluding the area where presence of each species was confirmed). As the field absence points represent a low proportion of total absence training sites, and random points need to be confirmed as true absence, we could name it "pseudo absence class" as is suggested by Anderson et al. (2003).
    Mean and dispersion of environmental variables characterized presence and absence sites. The multivariate distribution function of presence and absence sites should be different enough to enable us to unambiguously identify between the classes (presence and absence) that are analyzed to characterize the habitat of each species. To verify this point, Divergence Transformed (Chuvieco, 1996) and Jeffries-Matusita (Matusita, 1966) tests were performed between presence-absence class pairs.
    After the suitability test of each site was defined, we selected the number of variables and performed Maximum Likelihood algorithm for classification. The method considered the likelihood distribution function of each class defined and also the covariance among different variables intervening in the analysis. The probability of each pixel belonging to a certain class is calculated, and then the pixel is assigned to those with major likelihood. It was also possible to choose the threshold for the inclusion of a pixel in each class based on its probability function. Values of threshold closer to 1 indicated strict inclusion of pixels whose values were closer to the means of training sites; while when threshold values tended to 0, the inclusion of pixels took into account major dispersion of the values.
    The accuracy of the thematic map obtained (presence- absence map) was evaluated with other presence control points (Appendix 1, 2, 3) that were taken at random among all localities available and that were not used to characterize training sites. A confusion matrix that shows discrepancies between "real data" with those derived from classification was made; throughout the matrix, commission errors (false positives), omission errors (false negatives), as well as an overall accuracy (the percentage of correctly classified over all map) could be calculated (Congalton, 1991; Chuvieco, 1996). Due to the asymmetry in the input data (because of "pseudo-absence class"), commission error includes true commission (overprediction) and apparent commission errors (correctly predicted areas not verifiable as such, primarily because of the lack of adequate sampling). The interpretation of these measures needs to be done carefully (Anderson et al., 2003).
    Due to the wide distribution of the rodent species (particularly C. musculinus and O. flavescens, which cover diverse biogeographic regions) classifications were performed with a strict inclusion threshold (between 0.9 and 0.99) to limit the resulting presence area to very similar "training site" values. Cabrera's phytogeographic regions (Cabrera, 1976) were used in this work.

RESULTS

   The PCA analysis performed in NDVI, LST and rainfall time series are summarized in Table 1; 99.70% of the variance in rainfall, 99.72 % in NDVI, and 99.97 % in LST was explained by first principal component.

    The Jeffries-Matusita test among all pairs of training sites is shown in Table 2. Values greater than 1.9 indicate good differentiation among classes. A good definition was observed between O. longicaudatus and C. musculinus (1.99), and O. longicaudatus and O. flavescens (1.99), but only a small one between C. musculinus and O. flavescens (1.52). Calomys musculinus and O. longicaudatus presence training sites had the highest value. These classes could be used together if, for example, the possibility of competitive exclusion between these species is considered.

   We generated potential distribution maps using the method of Maximum Likelihood for C. musculinus (Fig. 1); O. flavescens (Fig. 2) and O. longicaudatus (Fig. 3). In Argentina, a widespread distribution of C. musculinus was observed (Fig. 1). The estimated area of presence covered from northern Formosa province (25º 30' S) and north-central Salta and Jujuy provinces, to 45º S latitude south of Deseado river in Santa Cruz province. In north and northwest Argentina, C. musculinus is predicted in high and contiguous patches. In combination, the predicted distribution encompassed several phytogeographic units, included in two major Domains: the Chacoan Domain and the Andean-Patagonic Domain. The predicted C. musculinus distribution also includes Uruguay, western Bolivia and Peru, and several sites in Paraguay.

Fig. 1. Estimated geographic distributions for Calomys musculinus in Southern America. Areas predicted by model appear in gray.

Fig. 2. Estimated geographic distributions for Oligoryzomys flavescens in Southern America. Areas predicted by model appear in gray.

Fig. 3. Estimated geographic distributions for Oligoryzomys longicaudatus in Southern America. Areas predicted by model appear in gray.

   In C. musculinus, the confusion matrix with 11 control points was accurate overall (score 90.13). Only one control point was unclassified and 7% of pseudo-absence classes were classified as presence (likely over-prediction). Threshold for pixel inclusion in presence class was 0.95.
    O. flavescens had a higher number of presence sites to characterize the habitat but was also the widest distributed. It covered from 22°4' S in the north to 40° 30' in the south. Oligoryzomys flavescens ' distribution extended to the north of Santa Cruz province on the Atlantic coast. It included Chaco, El Espinal and Pampean phytogeographic provinces; El Monte in the Chacoan Domain, and only a small band of Patagonic province in the Andean-Patagonic Domain. The estimated area also included Uruguay, and southern Paraguay and Brazil.
    A strict limit of similarity among the pixels analyzed was used (Threshold=0.99), resulting in a 95% overall classification accuracy. Most errors observed were commission errors (3.5% of pseudo-absence class classified as presence).
    In O. longicaudatus the map predicted a more restricted distribution than in the former species. The distribution occurred in the east Andean zone of the Patagonic forest, from San Juan (30°) to the center of Chubut (44°) provinces, and it extended to the Atlantic Ocean in Río Negro and Chubut provinces. The map did not predict the presence of O. longicaudatus in the north of Argentina (Salta and Jujuy). The O. longicaudatus distributions include a portion of the Patagonic province in the Andean-Patagonic Domain, and a portion of El Monte region in de Chacoan Domain.
    In O. longicaudatus, few points of control were taken because of the low number of presence sites. In Chile, the potential distribution registered was between 33° and 44° S, and was extended in discontinued form to 22° at North. The overall accuracy was 95.9 % with Argentinean control points correctly classified but one assigned to pseudo absence class in Chile but only 2% incorrectly in pseudo absence class. The threshold used for the classification was 0.95.

DISCUSSION

   This work was based on a combination of field data of species occurrence and environmental data collected by satellite remote sensing. The model is based on the species' fundamental niche with factors controlling distributions (Grinnell, 1917a, b; MacArthur, 1972). In the Grinnell (1917b) and MacArthur (1972) multidimensional ecological/environmental space concept, the ecological niche of a species can be defined as the intersection of ecological conditions within which it can maintain populations without immigration. This is a habitat or distribution concept. The approach we used herein for mapping rodent distributions cannot identify the variables that best predict areas of apparent suitability. However, the distribution maps do show large areas of suitable habitats for the species with the variables used. Previous works (Crespo, 1966; Crespo et al., 1970; Bonaventura and Kravetz, 1989; Paterson et al., 1990; Kelt, 1996; Monjeau et al., 1997; Castellarini and Polop, 2002; Castellarini et al., 2002; Pardiñas et al., 2003) suggested that rainfall, temperature, and vegetation are important ecological variables for the distribution of Argentinean rodents. Accurate knowledge of these variables in relation to the distribution and abundance of the rodents is imperative for a better mapping of the species in the future. A better understanding of these ecological factors might enhance our ability to predict how the distribution of rodent species might be affected in the light of the current environmental changes in the region.
   Several factors require these distribution maps to be interpreted with caution. First, the original resolution of climatic and vegetation data is quite coarse. Even though we carefully chose the pixels adjacent to presence data (excluding pixels with water or great changes in the digital numbers values compared with its neighbor) to increase the Training site size, this procedure could produce some biases on the resulting map. On a fine scale at sub-regional or local level the situation will be different. The density of the required information is high and remote sensing based on mapping approaches has to include additional knowledge from field investigations, ecological information and a new generation of high-resolution sensors.
   We give the overall performance of classifications even knowing they do not provide an effective means for identifying the best model (Anderson et al., 2003). We prefer to point out the correct classification of control points and small commission errors, suggesting little over-prediction of presence data. It could give an idea of the accuracy of the classification. Furthermore, if populations in other parts of the species range occupy environmental conditions outside those typified by the present sample, the exclusion of records from those distributional areas may have led to an underestimation of the species' niche in the examined environmental dimensions (Anderson, 2003). Moreover, only moderate numbers of places for species occurrence were used as training sites. Given these limitations, we compared the predictive maps with data from literature for species occurrence that were not used as training sites in our study. Despite these caveats and possible sources of error, the predictive maps yielded a realistic prediction of the species' potential distribution. In the C. musculinus, O. flavescens and O. longicaudatus maps, the high overall accuracy (90.1, 95.0 and 97.4 respectively) indicated high predictive values.
   Olrog and Lucero (1981) indicated the Colorado river was the southern distributional edge of C. musculinus, and Redford and Eisenberg (1992) cited the southern limit of the species as near the Negro river in Río Negro province. However, Thomas (1927) and Hershkovitz (1962) had already recorded the species in Patagonia. Saba et al. (1995) registered C. musculinus in Puerto Madryn sandbank, and Massoia and Pardiñas (1994) extended the distribution to the southwest of Santa Cruz Province. Moreover, Pardiñas et al. (2003) registered 14 occurrence sites of Calomys sp. from raptor pellet material in Patagonia not used as training sites in our study. These authors listed the specimens as Calomys sp., but suggested referring them to C. musculinus in accordance with Puerto Madryn specimens collected previously (Salazar-Bravo et al., 2001). Our C. musculinus distribution map (Fig. 1) predicted these occurrences, and its presence in sites such as Punta Delgada in Península Valdez (42°46´13"S; 63°38´25"W) where it was recently registered by Pardiñas et al. (2003). Our predictive map extended the C. musculinus distribution edge further than the Negro river, inhabiting extremely heterogeneous and complex geographical areas.
   In northern Argentina Thomas (1913) and Cabrera (1961) registered C. musculinus in Jujuy, Salta and Catamarca, and Olrog and Lucero (1981) considered the species present in Salta and Jujuy, and in all Formosa province. But, Mares et al. (1981, 1989), and Ojeda and Mares (1989) did not register the species in Salta and Jujuy Provinces, and Pardiñas and Teta (in press) did not register C. musculinus presence in Formosa province. Our study predicted its presence only in some parts of Formosa, Salta and Jujuy Provinces. On the other hand, our map predicted C. musculinus in Tucumán province, not used as a training site in our study, where the species has been captured by Barquez et al. (1991). Moreover, the estimated area of our map also includes Uruguay, and western Bolivia and Peru, where C. musculinus has not been recorded. Uruguay has similar environmental condition as the pampean region in Argentina. Thus, the absence of this species in Uruguay may be related to historical and ecological factors that are not taken into account in our approach. The predicted presence of C. musculinus in Paraguay is corroborated by William et al. (1997), who captured eight specimens in Toledo, Loma Plata and Monte Palma localities. Moreover, the Bolivia presence is corroborated by C. musculinus specimens captured in Tarija (Salazar-Bravo et al., 2001). Its presence needs to be confirmed in Peru.
   Our results predict the presence of O. flavescens from -23° to -47° latitude, to the east of the 67° W meridian, extending the southern edge as a thin border on the Atlantic Ocean coast up to the north of Santa Cruz province. The distribution of O. flavescens encompasses large variation in climate, soil, and vegetation variability. Oligoryzomys flavescens ' predictive distribution map agreed with occurrence sites not used as training sites in our study (Thomas, 1927; Hershkovitz, 1962; Saba et al., 1995; Monjeau et al., 1997; Pardiñas et al., 2003). The distribution map also predicts O. flavescens in the Chacoan region, including Formosa province and east of Salta province, but Pardiñas and Teta (in press), and Mares et al. (1989) did not register the species in these provinces.
   The estimated distribution of O. flavescens also included Uruguay and southern Brazil, where the presence of this rodent was mentioned by Andrades-Miranda et al. (2001), Becker and Vieira (2002), and Marinho et al. (2002), and the South of Paraguay, where its presence needs to be confirmed. Delfraro et al. (2003) showed that O. flavescens captures in the southern areas of Uruguay could be the host for a pathogenic hantavirus called Central Plata.
   Even though the potential distributions of O. flavescens and C. musculinus greatly overlapped, the areas of high support for the respective predicted distributions was not the same (Figs. 1 and 2). The distribution of these species does not fit the distribution of phytogeographic units. This is likely due to the great plasticity in the distributional behavior of these species, as they appear to be greatly influenced by anthropogenic modifications (Pardiñas, 1999; Pardiñas et al., 2000; Pardiñas et al., 2004). The high training site numbers that characterized these species habitats could diminish the classification accuracy because of the high training site variability.
   Previous studies (Olrog and Lucero, 1981; Redford and Eisenberg, 1992) have recognized O. longicaudatus along an extensive geographical range in the west of Argentina from Salta in the north to Santa Cruz and Tierra del Fuego in the south. Pearson (1983), Carleton and Musser (1989), and Monjeau et al. (1998) considered O. longicaudatus as ranging in the west, from southern Mendoza (32° S) to northern Santa Cruz (48° S) provinces, associated with Nothofagus forests as well as the ecotone between these forests and the Patagonian steppe. Recently, populations from southern Argentina and Chile have been considered as O. longicaudatus (Galliari et al., 1996). All O. longicaudatus sites cited by Pardiñas et al. (2003) and Monjeau et al. (1997) have been predicted by our distribution map. Out of 28 occurrence sites listed by Palma et al. (2005), only the north Pacific coast (La Silla, Fray Jorge, Quebrada del Tigre and San Antonio) are not included in our map (Fig. 3). Tiranti (1988) also reported O. longicaudatus from owl pellets in Curacó and Lihuel Calel departments (La Pampa province) near the north edge of our distribution map. On the other hand, Monjeau et al. (1997) did not register O. longicaudatus specimens in nine study sites in Patagonia steppe: our map predicted the presence in two of these sites: 15 km at Mencué northeast (Río Negro province) (40° 21´ S; 69° 31´ W), and 100 km at Dolavon west (Chubut) (43° 17´ S; 67° 04´ W).
   The Orán Hantavirus strain was isolated from putative O. longicaudatus in northern Argentina. However, a number of authors suggested that this Oligoryzomys belongs to a distinct species (Capllonch et al., 1997; Díaz, 1997; Levis et al., 1998; Massoia, 1998; Calderón et al., 1999; Enria et al., 2000; Ortiz and Pardiñas, 2001). Phylogenetic studies supported by analysis of restriction sites of mitochondrial DNA (Gonzalez Ittig et al., 2002) and partial sequences of cytochrome b (Myers et al., 1995) concluded that specimens from northern Argentina showed differences from those in Patagonia, a conclusion also reached by Palma et al. (2005). Espinosa and Reig (1991) distinguished between karyotyped specimens of Tucumán Province from O. longicaudatus of Chile and south of Argentina. Specimens from Bolivia ascribed as O. longicaudatus were then identified as O. andinus, O. destructor or as undetermined species of the " flavescens complex" (Carleton and Musser, 1989).
   The absence of potential habitats beyond 48°S registered in our map agrees with Gallardo and Palma (1990) and Steppan (1995), who suggested separation between O. longicaudatus (38° S to 48° S) and O. magellanicus (48° to Tierra del Fuego). However, Palma et al. (2005) refers O. magellanicus to synonymy with O. longicaudatus, thus enlarging the range of this species at least as far as latitude 51°S in both Argentina and Chile. These authors concluded that the phylogenetic relationship of O. longicaudatus in Chile and nearby areas in Argentina confirms the existence of a single species, suggesting strong gene flow among populations that would account for the genetic uniformity of population along their range of distribution.
   Our predictive maps showed large areas of apparent suitability in which none of the rodent species was recorded, either due to them not actually occurring there or in which no information is yet available on these areas. This is a feature of many prediction maps in which species do not always occupy habitats that are suitable for them or are not always found even when they do occur (Rogers et al., 1996). Moreover, the use of different trapping methods by different collectors and time of collection may determine the absence or presence of a species in a given area (Gebre-Michael et al., 2004).
   However, the maps proposed here afford several advantages. Clearly, objective modeling of a species' geographic distribution is an important improvement over subjective, broadstroke shaded outline maps. First, the predictive maps incorporate geographically explicit predictions of potential distribution into the test. In addition, although fine-resolution base environmental data are always desirable, the development of modeling techniques that provide general distributional hypotheses using the relatively coarse environmental data currently available is important (Anderson and Martinez- Meyer, 2004). Second, the validity of the predictive map can be appreciated when localities of previous records of the studied species, not used as training sites or used as control sites, overlay in the map. In this approach, environmental factors, criteria and analytical techniques are explicit and can be easily verified. Hence, we can temporally fit data in more precise distribution maps.
   Future research should continue to evaluate these approaches with other C. musculinus, O. flavescens and O. longicaudatus occurrence records, as well as with other species and their necessary environmental related data requirements.

APPENDIX 1

   Presence and control sites of Oligoryzomys flavescens expressed as Latitude and Longitude in centesimal degrees.

APPENDIX 2

   Presence and control sites of Calomys musculinus, expressed as Latitude and Longitude in centesimal degrees.

APPENDIX 3

   Presence and control sites of Oligoryzomys longicaudatus expressed as Latitude and Longitude in centesimal degrees.

ACKNOWLEDGMENTS

   We want to thank Drs. Noemí Gardenal, Ulyses Pardiñas, Enrique Martinez-Meyer, Luis A. Ruedas and one anonymous reviewer for comments and suggestions to earlier versions of this work. The Instituto Nacional de Enfermedades Virales Humanas "Dr. Julio I. Maiztegui" (INEVH) provided field data for rodent and virus distributions and the CONAE (Comisión Nacional de Actividades Espaciales) and National University of Río Cuarto provided support for this study. The National University of Córdoba provided partial funding for this work through the Extension Activities scholarship. Remote sensor data used by the authors in this study include data produced through funding from the Earth Observing System Pathfinder Program of NASA's Mission to Planet Earth in cooperation with National Oceanic and Atmospheric Administration. The remote sensor data were provided by the Earth Observing System Data and Information System (EOSDIS), Distributed Active Archive Center at Goddard Space Flight Center which archives, manages, and distributes this data set.

LITERATURE CITED

ANDERSON RP, D LEW, and AT PETERSON. 2003. Evaluating predictive models of species' distributions: criteria for selecting optimal models. Ecological Modelling 162:211-232.         [ Links ]

ANDERSON RP. 2003. Real vs. artefactual absences in species distributions: tests for Oryzomys albigularis (Rodentia: Muridae) in Venezuela. Journal of Biogeography 30:591-605.         [ Links ]

ANDERSON RP, AT PETERSON, and M GÓMEZVALVERDE. 2002. Using niche-based GIS modeling to test geographic predictions of competitive exclusion and competitive release in South American pocket mice. Oikos 98:3-16.         [ Links ]

ANDERSON RP and R MARTÍNEZ-MEYER. 2004. Modeling species' geographic distributions for preliminary conservation assessments: an implementation with the spiny pocket mice (Heteromys) of Ecuador. Biological Conservation 116:167-179.         [ Links ]

ANDRADES-MIRANDA J, LFB OLIVEIRA, CAV LIMA-ROSA, AP NUNES, NIT ZANCHIN, and MS MATTEVI. 2001. Chromosome studies of seven species of Oligoryzomys (Rodentia: Sigmodontinae) from Brazil. Journal of Mammalogy 82:1080-1091.         [ Links ]

AUSTIN MP and JA MEYERS. 1996. Current approaches to modelling the environmental niche of eucalyptus: Implications for management of forest biodiversity. Forest Ecology Management 85:95-106.         [ Links ]

BARQUEZ RM, MA MARES, and RA OJEDA. 1991. Mammals of Tucumán. Norman, Oklahoma, U.S.A. 282 pp.         [ Links ]

BECK LR, MH RODRÍGUEZ, SW DISTER, AD RODRÍGUEZ, RK WASHINO, DR ROBERTS, and MA SPANNER. 1997. Assessment of a remote sensing based model for predicting malaria transmission risk in villages of Chiapas, Mexico. American Journal of Tropical Medicine and Hygiene 56:99-106.         [ Links ]

BECK LR, BM LOBITZ, and BL WOOD. 2000. Remote sensing and human health: New sensors and new opportunities. Emerging Infectious Diseases 6:217-227.         [ Links ]

BECKER R and EM VIEIRA. 2002. Vertical stratification of small mammals in the Atlantic rain forest of Southeastern Brazil. XVII Jornadas Argentinas de Mastozoología, Mar del Plata, Argentina.         [ Links ]

BILENCA DN. 1993. Caracterización de los nichos ecológicos y organización de las comunidades de roedores cricétidos de la Región Pampeana. Tesis Doctoral Inédita, Universidad de Buenos Aires, Buenos Aires, Argentina.         [ Links ]

BONAVENTURA SM and FO KRAVETZ. 1989. Relación roedor vegetación: importancia de la cobertura verde para Akodon azarae. Physis 112:1-5.         [ Links ]

BOONE JD, KC MCGWIRE, EW OTTESON, RS DE BACA, EA KUHN, P VILLARD, PD BRUSSARD, and SC ST. JEOR. 2000. Remote Sensing and Geographic Information Systems: Charting Sin Nombre Virus Infections in Deer Mice. Emerging Infectious Diseases 6:248-257.         [ Links ]

BUSCH M, FO KRAVETZ, RE PERCICH, and G ZULETA. 1984. Propuestas para un control ecológico de la FHA a través del manejo del hábitat. Medicina 44:34-40.         [ Links ]

CABRERA A. 1961. Catálogo de los Mamíferos de América del Sur. I y II. Revista del Museo Argentino de Ciencias Naturales "Bernardino Rivadavia", Zoología 4(1-2):1-732.         [ Links ]

CABRERA AL. 1976. Regiones fitogeográficas Argentinas. Enciclopedia Argentina de Agricultura y Jardinería, ACME, Buenos Aires 2.         [ Links ]

CALDERÓN G, N PINI, J BOLPE, S LEVIS, J MILLS, E SEGURA, N GUTHMANN, G CANTONI, J BECKER, A FONOLLAT, C RIPOLL, M BORTMAN, R BENEDETTI, M SABATTINI, and D ENRIA. 1999. Hantavirus reservoir hosts associated with peridomestic habitats in Argentina. Emerging Infectious Diseases 5:792-797.         [ Links ]

CAPLLONCH P, AG AUTINO, M DÍAZ, RM BARQUEZ, and M GOYTIA. 1997. Los mamíferos del Parque Biológico Sierra de San Javier, Tucumán, Argentina: Observaciones sobre su sistemática y distribución. Mastozoología Neotropical 4:49-71.         [ Links ]

CARLETON MD and GG MUSSER. 1989. Systematic studies of oryzomyne rodents (Muridae, Sigmodontinae): A synopsis of Microryzomys. Bulletin of the American Museum of Natural History 191:1- 83.         [ Links ]

CARPENTER G, AN GILLISON, and J WINTER. 1993. DOMAIN: a flexible modelling procedure for mapping potential distributions of plans and animals. Biodiversity and Conservation 2:667-680.         [ Links ]

CASTELLARINI F and JJ POLOP. 2002. Effects of extra food on population fluctuation patterns of the muroid rodent Calomys venustus. Austral Ecology 27:273- 283.         [ Links ]

CASTELLARINI F, MC PROVENSAL, and JJ POLOP. 2002. Effect of climatic variables on the population fluctuation of muroid Calomys venustus in central Argentina. Acta Oecologica 23:385-391.         [ Links ]

CONGALTON R 1991. A Review of Assessing the Accuracy of Classifications of Remotely Sensed Data. Remote Sensing of Environment 37:35-46.         [ Links ]

CONTRERAS JR and MI ROSI. 1980. Una nueva subespecie del ratón colilargo para la provincia de Mendoza: Oligoryzomys flavescens occidentalis (Mammalia, Rodentia, Cricetidae). Historia Natural 1:157-160.         [ Links ]

CONTRERAS JJ and LM BERRY. 1983. Notas acerca de los roedores del género Oligoryzomys de la Provincia del Chaco, República Argentina (Rodentia- Cricetidae). Historia Natural 3:145-148.         [ Links ]

CRAMER WP and R LEEMANS. 2001. Global 30-Year Mean Monthly Climatology, 1930-1960, Version. 2.1 (Cramer and Leemans). Data set. Available online http://www.daac.ornl.gov from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, U.S.A.         [ Links ]

CRESPO JA. 1966. Ecología de una comunidad de roedores silvestres en el Partido de Rojas, Provincia de Buenos Aires. Revista Museo Argentino de Ciencias e Instituto Nacional de Investigación en Ciencias Naturales. Ecología 1:79-134.         [ Links ]

CRESPO JA, MS SABATTINI, MJ PIANTANIDA, and G DE VILLAFAÑE. 1970. Estudios ecológicos sobre roedores silvestres. Observaciones sobre densidad, reproducción y estructura de comunidades de roedores silvestres en el sur de Córdoba. Ministerio de Bienestar Social, Secretaría de Estado de Salud Pública 45 pp.         [ Links ]

CRONER CM, J SPERLING, and FR BROOME. 1996. Geographic information systems (GIS): New perspectives in understanding human health and environmental relationships. Statistics in Medicine 15:1961-1977.         [ Links ]

CHAPUT EK, JI MEEK, and R HEIMER. 2002. Spatial analysis of Human Granulocytic Ehrlichiosis near Lyme, Connecticut. Emerging Infectious Diseases 8:113-123.         [ Links ]

CHEEK J, R BRYAN, and G GLASS. 1998. Geographic distribution of high risk, HPS areas in the U.S. Southwest. The Fourth International Conference on HFRS and Hantaviruses, March 5-7, 1998. Atlanta, Georgia, USA (Abstract).         [ Links ]

CHUVIECO E. 1996. Fundamentos de teledetección espacial. II. RIALP Eds. Alcalá, Madrid. Spain 2.         [ Links ]

DALE PER, SA RITCHIE, BM TERRITO, CD MORRIS, A MUHAR, and BH KAY. 1998. An overview of remote sensing and GIS for surveillance of mosquito vector habitats and risk assessment. Journal of Vector Ecology 23:54-61.         [ Links ]

DELFRARO A, C MARIO, L TOMÉ, F ACHAVAL, S LEVIS, G CALDERÓN, D ENRÍA, M LOZANO, J RUSSI, and J RABIZA. 2003. Yellow pygmy rice rat (Oligoryzomys flavescens) and Hantavirus Pulmonary Syndrome in Uruguay. Emerging Infectious Diseases 9:846-852.         [ Links ]

DE VILLAFAÑE G. 1970. Ecología de roedores silvestres (Cricetidae) en el sur de la provincia de Córdoba, completada con observaciones sobre reproducción y desarrollo obtenidas por cría experimental. Tesis Doctoral Inédita. Universidad Nacional de Córdoba, Córdoba, Argentina.         [ Links ]

DE VILLAFAÑE G, FO KRAVETZ, O DONADIO, RO PERCICH, L KNECHER, M TORRES, and N FERNÁNDEZ. 1977. Dinámica de las comunidades de roedores de agroecosistemas pampásicos. Medicina 37:128-140.         [ Links ]

DÍAZ MM. 1997. Mamíferos de la Provincia de Jujuy: sistemática, distribución y ecología. Tesis Doctoral Inédita. Facultad de Ciencias Naturales e Instituto Miguel Lillo, Universidad Nacional de Tucumán, Tucumán, Argentina.         [ Links ]

EASTMAN JR and M FULK. 1993. Long sequence time series evaluation using standardised principal components. Photogrammetric Engineering and Remote Sensing 59:991-996         [ Links ]

ENGELTHALER DE, DG MOSLEY, JE CHEEK, CE LEVY, KK KOMATSU, P ETTESTAD, T DAVIS, DT TANDA, L MILLER, JW FRAMPTON, R PORTER, and RT BRYAN. 1999. Climatic and environmental patterns associated with Hantavirus Pulmonary Syndrome, four corners region, United States. Emerging Infectious Diseases 5:263-269.         [ Links ]

ENRIA DAM and SC LEVIS. 2004. Zoonosis virales emergentes: las infecciones por Hantavirus. Revue Scientifique et technique Office International des Epizooties 23: 595-611.         [ Links ]

ENRIA D, M BOWEN, J MILLS, WJ SHIEB, D BAUSCH, and CJ PETERS. 2000. Arenavirus Infections. Pp.1191-1212, in: Tropical Infectious Diseases: Principles, Pathogens, and Practice (R Guerrant, DH Walker, PF Weller and WB Saunders, eds.) 2: Chapter 111.         [ Links ]

ESPINOSA MB and OA REIG. 1991. Cytogenetics and karyosystematics of South American oryzomyine rodents (Cricetidae, Sigmodontinae) III. Banding karyotypes of Argentinian Oligoryzomys. Zeitschrift für Säugetierkunde 65:1-19.         [ Links ]

GALLARDO MH and E PALMA. 1990. Systematics of Oryzomys longicaudatus (Rodentia: Muridae) in Chile. Journal of Mammalogy 71:333-342         [ Links ]

GALLIARI CA, UFJ PARDIÑAS, and FJ GOIN. 1996. Lista comentada de los mamíferos argentinos. Mastozoología Neotropical 3:39-61.         [ Links ]

GEBRE-MICHAEL T, JB MALONE, M BALKEW, A ALI, N BERHE, A HAILU, and AA HERZI. 2004. Mapping the potential distribution of Phlebotomus martini and P. orientalis (Diptera: Psychodidae), vectors of kala-azar in East Africa by use of geographic information systems. Acta Tropica 90:73- 86.         [ Links ]

GLASS EG, JE CHEEK, JA PATZ, TM SHIELDS, TJ DOYLE, DA THOROUGHMAN, DK HUNT, RE ENSCORE, KL GAGE, C IRLAND, CJ PETERS, and R BRYAN. 2000. Using remotely sensed data to identify areas at risk for Hantavirus Pulmonary Sindrome. Emerging Infectious Diseases 6:238-247.         [ Links ]

GONZÁLEZ ITTIG RE, GR THEILER, and CN GARDENAL. 2002. A contribution to the subgeneric systematics of Oligoryzomys (Rodentia, Muridae) from Argentina by means of PCR-RFLP patterns of mitochondrial DNA. Biochem. Systematic Ecology 30:23-33.         [ Links ]

GORLA DE. 2002. Variables ambientales registradas por sensores remotos como indicadores de la distribución geográfica de Triatoma infestans (Heteroptera: Reduviidae). Ecología Austral 12:117-127.         [ Links ]

GREEN RE, PE OSBORNE, and EJ SEARS. 1994. The distribution of passerines birds in hedgerows during the breeding season in relation to characteristics of hedgerows and adjacent farmlands. Journal of Applied Ecology 31:677-692.         [ Links ]

GRINELL J. 1917a. Field tests of Theories Concerning Distributional Control. American Naturalist 51:115- 128.         [ Links ]

GRINNELL J. 1917b. The niche-relationships of the California trasher. Contribution from the Museum of Vertebrate Zoology of the University of California 34:427-433.         [ Links ]

HAY S, C TUCKER, D ROGERS, and M PACKER. 1996. Remotely sensed surrogates of meteorological data for the study of the distribution and abundance of arthropod vectors of disease. Annals of Tropical Medicine and Parasitology 90:1-19.         [ Links ]

HERSHKOVITZ P. 1962. Evolution of Neotropical cricetine rodents (Muridae) with special reference to the phyllotine group. Fieldiana: Zoology 46:1-524.         [ Links ]

JAMAS FC and CE MCCULLOCH. 2002. Predicting species presence and abundance. Pp. 161-165, in: Predicting species occurrences: issues of accuracy and scale (JM Scott, PJ Heglund, ML Morrison, JB Haufler, MG Raphael, WA Wall, and FB Samson, eds.). Island Press, Washington, DC.         [ Links ]

KELT DA. 1996. Ecology of small mammals across a strong environmental gradient in southern South America. Journal of Mammalogy 77:205-219.         [ Links ]

KERR JT and M OSTROVSKY. 2003. From space to species: ecological applications for remote sensing. Trends in Ecology and Evolution 18:299-305.         [ Links ]

KRAVETZ FO. 1978. Ecología de las comunidades de roedores involucrados en la Fiebre Hemorrágica Argentina. Tesis Doctoral Inédita, Universidad Nacional de Buenos Aires, Buenos Aires, Argentina.         [ Links ]

KRAVETZ FO and G DE VILLAFAÑE. 1981. Poblaciones de roedores en cultivo de maíz durante las etapas de madurez y rastrojo. Historia Natural 1:213-232.         [ Links ]

KRAVETZ FO and JJ POLOP. 1983. Comunidades de roedores en agroecosistemas del Departamento de Río Cuarto, Córdoba. Ecosur 10:1-18.         [ Links ]

LEEMANS R and WP CRAMER. 1991. The IIASA database for mean monthly values of temperature, precipitation and cloudiness of a global terrestrial grid. Research Report RR-91-18 November 1991, International Institute of Applied Systems Analyses, Laxenburg. 61pp         [ Links ]

LEK S, A BELAUD, P BARAN, I DIMOPOULOS, J LAUGA, and S AULAGNIER.1996. Application of neural networks to modelling nonlinear relationships in ecology. Ecological Modelling 1634:1-13.         [ Links ]

LEVIS S, JE ROWE, S MORZUNOV, D ENRÍA, and SC ST JEOR. 1997. New Hantaviruses causing hantavirus pulmonary syndrome in central Argentina. The Lancet 349:998-999.         [ Links ]

LEVIS S, SP MORZUNOV, JE ROWE, D ENRIA, N PINI, G CALDERÓN, MS SABATTINI, and SC ST JEOR. 1998. Genetic diversity and epidemiology of hantaviruses in Argentina. Journal of Infectious Diseases 177:529-538.         [ Links ]

LINTHICUM KJ, CL BAILEY, FG DAVIES, and CJ TUCKER. 1987. Detection of Rift Valley fever viral activity in Kenya by satellite remote sensing imagery. Science 235:656-1659.         [ Links ]

MAC ARTHUR RH. 1972. Geographical ecology: patterns in the distribution of species. Harper and Row, New York.         [ Links ]

MAIZTEGUI J, MR FEUILLADE, and A BRIGGILER. 1986. Progressive extension of the endemic area and changing incidence of Argentine Hemorrhagic Fever. Medical Microbiology and Immunology 175:149-152.         [ Links ]

MANEL S, JM DIAS, and S JORMEROD. 1999. Comparing discriminant analysis, neural networks and logistic regression for predicting species distributions: a case study with a Himalayan river bird. Ecological Modelling 120:337-347.         [ Links ]

MARES MA, RA OJEDA, and RM BARQUEZ. 1989. Guide to the mammals of Salta Province, Argentina. University of Oklahoma Press. 303 pp.         [ Links ]

MARES MA, RA OJEDA, and MP KOSCO. 1981. Observations on the distribution and ecology of the mammals of Salta Province, Argentina. Annals of the Carnegie Museum 50:151-206.         [ Links ]

MARINHO JR, AS CUNHA, GL GONÇALVES, and TRO FREITAS. 2002. Diversidad comparada de la fauna de roedores en un gradiente altitudinal de la Mata Atlántica a lo largo de la Rota do Sol, en el estado de Río Grande do Sul. P. 56, in: Libro de Resúmenes, XVII Jornadas Argentinas de Mastozoología, Mar del Plata, Argentina.         [ Links ]

MASSOIA E. 1973. Presencia y rasgos bioecológicos de Oryzomys longicaudatus pampanus, nueva subespecie en la provincia de Buenos Aires (Mammalia, Rodentia, Cricetidae). Revista de Investigación Agropecuaria, INTA, Serie I, Biología y Producción Animal 10(1):43-49.         [ Links ]

MASSOIA E. 1998 Roedores vinculados con las virosis humanas en la República Argentina. 2do. Congreso Argentino de Zoonosis y 1er Congreso Argentino y Latinoamericano de Enfermedades Emergentes. Buenos Aires. Temas de Zoonosis y Enfermedades Emergentes 243-246.         [ Links ]

MASSOIA E and A FORNES. 1966. Nuevos datos sobre la distribución geográfica y ecología del género Calomys (Waterhouse) (Rodentia-Cricetidae). IDIA, Instituto de Investigaciones Agropecuarias 227:55-57.         [ Links ]

MASSOIA E and UFJ PARDIÑAS. 1994. La depredación de mamíferos por Bubo virginianus y Tyto alba en Cerro Casa de Piedra, Lago Burmeister, Parque Nacional Perito Moreno, provincia de Santa Cruz. Boletín Científico, Asociación para la Protección de la Naturaleza 26:6-12.         [ Links ]

MATUSITA KA. 1966. A distance and related statistics in multivariate analysis. Pp. 178-200, in: Multivariate Analysis. Krish-naiah, P.R. Academic Press, New York.         [ Links ]

MILLS JN. 1999. The role of rodents in emerging human disease: examples from the Hantaviruses and Arenaviruses. Pp. 134-160, in: Ecologically-based Rodent Management. (G Singleton, L Hinds, H Leirs and Z Zhang, eds) Australian Centre for International Agricultural Research.         [ Links ]

MILLS JN, BA ELLIS, JE CHILDS, KT MCKEE, JI MAIZTEGUI, CJ PETERS, TG KSIAZEK, and PB AHRLING. 1994. Prevalence of infection with Junin virus in rodent population of endemic area of AHF. American Journal of Tropical Medicine and Hygiene 51:554-562.         [ Links ]

MILLS JN, BA ELLIS, KT MCKEE, JI MAIZTEGUI, and JE CHILDS. 1991. Habitat associations and relative densities of rodent populations in cultivated areas of central Argentina. Journal of Mammalogy 72:471-479.         [ Links ]

MONJEAU JA, EC BIRNEY, L GHERMANDI, RS SIKES, L MARGUTTI, and CJ PHILLIPS. 1998. Plants, small mammals, and the hierarchical landscape classifications of Patagonia. Landscape Ecology 13:285-306.         [ Links ]

MONJEAU JA, N BONINO, and S SABA. 1994. Annotated checklist of the living land mammals in Patagonia, Argentina. Mastozoología Neotropical 1:143-156.         [ Links ]

MONJEAU JA, RS SIKES, EC BIRNEY, N GUTHMANN, and CJ PHILLIPS. 1997. Small mammal community composition within the major landscape divisions of Patagonia, southern Argentina. Mastozoología Neotropical 4:113-127.         [ Links ]

MYERS P, B LUNDRIGAN, and PK TUCKER. 1995. Molecular phylogenetics of Oryzomyine rodents: The genus Oligoryzomys. Molecular Phylogenetic Evolution 4:373-382.         [ Links ]

OJEDA RA and MA MARES. 1989. A biogeographic analysis of the mammals of Salta Province, Argentina: patterns of community assemblage in the Neotropics. Special Publications, The Museum, Texas Tech University 27:1-66.         [ Links ]

OLROG CC and MM LUCERO. 1981. Guía de los mamíferos argentinos. Ministerio de Cultura y Educación. Fundación M. Lillo, Argentina.         [ Links ]

ORTIZ PE and UFJ PARDIÑAS. 2001. Sigmodontinos (Mammalia: Rodentia) del Pleistoceno tardío del valle de Tafí (Tucumán, Argentina): Taxonomía, tafonomía y reconstrucción paleoambiental. Ameghiniana 38:3-26.         [ Links ]

OSBORNE PE and BJ TIGAR. 1992. Interpreting bird atlas data using logistic models: an example from Lesotho, Southern Africa. Journal of Applied Ecology 29:55-62.         [ Links ]

PALMA RE, E RIVERA-MILLA, J SALAZAR-BRAVO, F TORRES-PEREZ, UFJ PARDIÑAS, PA MARQUET, AE SPOTORNO, AP MEYNARD, and TL YATES. 2005. Phylogeography of Oligoryzomys longicaudatus (Rodentia: Sigmodontinae) in temperate South America. Journal of Mammalogy 86:191-200.         [ Links ]

PARDIÑAS UFJ. 1999. Los roedores muroideos del Pleistoceno tardío-Holoceno en la región pampeana (secto este) y Patagonia (República Argentina): aspectos taxonómicos, importancia bioestratigráfica y significación paleoambiental. Tesis Doctoral inédita, Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata. La Plata, Argentina.         [ Links ]

PARDIÑAS UFJ, AM ABBA, and ML MERINO. 2004. Micromamíferos (Didelphimorphia y Rodentia) del sudoeste de la provincia de Buenos Aires (Argentina): taxonomía y distribución. Mastozoología Neotropical 11:211-232.         [ Links ]

PARDIÑAS UFJ, G MOREIRA, C GARCÍA-ESPONDA, and LJM DE SANTIS. 2000. Deterioro ambiental y micromamíferos durante el Holoceno en el nordeste de la estepa patagónica (Argentina). Revista Chilena de Historia Natural 72:541-556         [ Links ]

PARDIÑAS UFJ, P TETA, S CIRIGNOLI, and DH PODESTÁ. 2003. Micromamíferos (Didelphimorphia y Rodentia) de norpatagonia extra andina, Argentina: taxonomía alfa y biogeografía. Mastozoología Neotropical 10:69-113.         [ Links ]

PARDIÑAS UFJ and P TETA. (In press). Roedores sigmodontinos de la región oriental de Formosa, Argentina: aspectos taxonómicos y distribución geográfica. In: Inventario de la Biodiversidad de la Reserva Ecológica El Bagual, Formosa, Argentina (AG Di Giacomo and S Krapovickas, eds.). Temas de Naturaleza y Consevación 4, Aves Argentinas / AOP, Buenos Aires.         [ Links ]

PATERSON BD, PL MESERVE, and BK LANG. 1990. Quantitative habitat associations of small mammals along an elevation transect in temperate rainforest of Chile. Journal of Mammalogy 71:620-633.         [ Links ]

PEARSON OP. 1983. Characteristics of mammalian faunas from forests in Patagonia, southern Argentina. Journal of Mammalogy 64:476-492.         [ Links ]

PETERSON AT. 2001. Predicting species' geographic distributions based on ecological niche modeling. Condor 103:599-605.         [ Links ]

PETERSON AT, MA ORTEGA-HUERTA, J BARTLEY, V SANCHEZ-CORDERO, J SOBERON, RH BUDDEMEIER, and DRB STOCKWELL. 2002a. Future projections for Mexican faunas under global climate change scenarios. Nature 416:626-629.         [ Links ]

PETERSON AT, V SANCHEZ-CORDERO, CB BEARD and JM RAMSEY. 2002b. Ecologic Niche Modeling and Potential Reservoirs for Chagas Disease, Mexico. Emerging Infectious Diseases 8:662-667.         [ Links ]

PETERSON AT, J SOBERON and V SANCHEZ-CORDERO. 1999. Conservatism of Ecological Niches in Evolutionary Time. Science 285:1265-1267.         [ Links ]

PIKULA J, F TREML, M BEKLOVA, Z HOLESOVSKA and, J PIKULOVA. 2002. Geographic Information Systems in Epidemiology - Ecology of common vole and distribution of natural foci of Tularemia. Acta Veterinaria Brno 71:379-387.         [ Links ]

PIUDO L, M MONTEVERDE, S GONZÁLEZ CAPRIA, P PADULA, and P CARMANCHAHI. 2005. Distribution and abundance of sigmodontine rodents in relation to hantavirus in Neuquén, Argentina. Journal of Vector Ecology 30:119- 125.         [ Links ]

POLOP JJ and MS SABATTINI. 1993. Rodent abundance and distribution in habitats of agrocenosis in Argentina. Studies of Neotropical Fauna and Environments 28:39-46.         [ Links ]

PRICE JC. 1984. Land surface temperature measurements from the split window channels of the NOAA 7 advanced very high resolution radiometer. Journal of Geophysics Research 89:7231-7237.         [ Links ]

REDFORD KH and JF EISENBERG. 1992. Mammals of the Neotropics: The Southern Cone. University of Chicago Press, Chicago.         [ Links ]

RICKLEFS RE and D SCHLUTER. 1993. Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago, IL.         [ Links ]

ROGERS DJM, SI HAY, and MJ PACKER. 1996. Predicting the distribution of tsetse flies in West Africa using temporal Fourier processed meteorological satellite data. Annals of Tropical Medicine and Parasitology 90:225-241.         [ Links ]

SABA S, D PEREZ, E CEJUELA, V QUIROGA, and A TOYOS. 1995. La piosfera ovina en el extremo austral del desierto del monte. Naturalia Patagónica, Ciencias Biológicas 3:153-174.         [ Links ]

SABATTINI MS, LE GONZALEZ DE RÍOS, G DÍAZ, and VR VEGA. 1977. Infección Natural y Experimental de Roedores con Virus Junin. Medicina (Buenos Aires) 37:149-59.         [ Links ]

SABATTINI MS and JI MAIZTEGUI. 1970. Fiebre Hemorrágica Argentina. Medicina (Buenos Aires) 30:111-123.         [ Links ]

SALAZAR- BRAVO J, JW DRAGOO, DS TINNIN, and TL YATES. 2001. Phylogeny and evolution of the Neotropical genus Calomys: inferences from mitochondrial DNA sequence data. Molecular Phylogenetics and Evolution 20:173-184.         [ Links ]

SINGLETON GR, H LEIRS, LA HINDS, and Z ZHANG. 1999. Ecologically-based management of rodent pests - re-evaluating our approach to an old problem. Pp. 17-30, en: Ecologically-based Rodent Management (G Singleton, L Hinds, H Leirs and Z Zhang, eds.). Australian Centre for International Agricultural Research.         [ Links ]

SKOV F. 2000. Potential plant distribution mapping based on climatic similarity. Taxon 49:503-515.         [ Links ]

STEPPAN SJ. 1995. Revision of the tribe Phyllotini (Rodentia: Sigmodontinae), with a phylogenetic hypothesis for the Sigmodontinae. Fieldiana, Zoology 80:1-112.         [ Links ]

THOMAS O. 1913. On small Mammals collected in Jujuy by Señor E. Budin. The Annals and Magazine of Natural History (London) (20) 9:136-143.         [ Links ]

THOMAS O. 1927. On further Patagonian mammals from Neuquén and the Río Colorado collected by Señor E. Budin. The Annals and Magazine of Natural History (London) (20) 17:199-205.         [ Links ]

TIRANTI SI. 1988. Análisis de regurgitados de Tyto alba de la provincia de La Pampa. Boletin Científico, Asociación para la Protección de la Naturaleza 11:8- 12.         [ Links ]

WALKER PA and KD COCKS. 1991. HABITAT: a procedure for modelling a disjoint environmental envelope for a plant or animal species. Global Ecology and Biogeography Letters 1:108-118.         [ Links ]

WILCOVE DS, CJ MCLLEAN, and AP DOBSON. 1986. Habitat fragmentation in the temperate zone. In: Conservation Biology, the Science of Scarcity and Diversity (ME Soulé, edit.). Sunderland: Sinauer Associates 237-256.         [ Links ]

WILLIAM RJ, RT BRYAN, JN MILLS, RE PALMA, I VERA, F VELASQUEZ, E BAEZ, WE SCHMIDT, RE FIGUEROA, CJ PETERS, SR ZAKI, AS KHAN, and TG KSIAZEK. 1997. An outbreak of Hantavirus Pulmonary Sindrome in western Paraguay. American Journal of Tropical Medicina and Hygiene 57:274- 282.         [ Links ]

ZULETA GA, FO KRAVETZ, M BUSCH, and RO PERCICH. 1988. Dinámica poblacional del ratón del pastizal pampeano (Akodon azarae), en ecosistemas agrarios de Argentina. Revista Chilena de Historia Natural 61:231-244         [ Links ]

Recibido 5 octubre 2004.
Aceptación final 1º setiembre 2005.