Latitudinal extent and natural history characteristics of birds in Nicaragua

Study area

Nicaragua is the largest country in Central America and has been identified recently as a biodiversity hotspot (Myers et al., 2000). Nicaragua has eight habitat types that are used by terrestrial breeding birds (Gillespie, 2001): tropical dry forest, gallery forest and savanna occur on the western side of Nicaragua; upland pine–oak forest, montane or cloud forest and elfin forest occur on the tops of mountain ranges and volcanoes in central Nicaragua; pine savanna and lowland rain forest occur in eastern Nicaragua. Howell (1969) identified Nicaragua as an appropriate region to analyse the zoogeography of Central American avifauna because: (1) it is a former major geological discontinuity between northern and southern landmasses; (2) it contains major habitat types that terminate within Nicaraguan borders (i.e. upland pine–oak forest, pine savanna); and (3) it is an area that includes northern and southern limits of many species, with no taxa confined exclusively within its boundaries.

Data collection

Latitudinal extent and latitudinal midpoint were calculated for all terrestrial breeding birds in Nicaragua. Latitudinal extents were calculated as the straight-line distance between the northern and southern extremes of a species’ breeding range with migratory ranges excluded (Gaston, 1996). It is expressed in the number of degrees within which a species is known historically to breed and was determined using a variety of range maps and breeding records from North, Central and South America. Latitudinal midpoint is the central point between a species’ northern and southern extent. A complete list of sources for the Nicaraguan bird species list and for the calculation of latitudinal extent is available at www.geog.ucla.edu/birdrange/ref_nicar_birds.html.

Data on body mass, trophic guild, forest dependence and number of forest types were collected for all recorded terrestrial breeding birds in Nicaragua. Body masses were given in grams and were averages from mistnet data from Stiles & Skutch (1989) and Dunning (1992). Trophic guild classification followed Karr et al. (1990) and Stiles & Skutch (1989). Trophic guilds were classified as carnivores, frugivores, granivores, insectivores, nectarivores and omnivores. Forest dependence was based on a species’ need for forest following the classification system of Stiles (1985). Species are classified as requiring almost solid forest, patchy forest or not needing forest. Seven forest types used by terrestrial breeding birds occur in Nicaragua: lowland rain forest, montane or cloud forest, elfin forest, tropical dry forest, gallery forest, lowland pine savanna and upland pine–oak forest. All forest birds were classified as occurring in one, two, three or four or more forest types (Stotz et al., 1996; Gillespie, 2001). The majority of these data came from unpublished data from T.R. Howell, Stotz et al. (1996) and 17 months of field surveys within Nicaragua. All latitudinal and natural history data are available at www.geog.ucla.edu/birdrange ref_nicar_birds_data.html.

Data analysis

Associations among latitudinal midpoint, body mass and latitudinal extent were examined over three taxonomic scales: (1) all terrestrial breeding birds recorded in Nicaragua; (2) order and suborder (non-passerines, suboscines passerines, oscines passerines); and (3) eight large families and subfamilies with over 13 species: hawks (Accipitridae), pigeons (Columbidae), parrots (Psittacidae), hummingbirds (Trochilidae), antbirds (Formicariidae), flycatchers (Tyrannidae), tanagers (Thraupinae) and finches (Emberizinae). A Spearman's rank correlation was used to identify associations among latitudinal extent and latitudinal midpoint or body mass over the three taxonomic scales. Bonferroni corrections were employed for significance levels (Sokal & Rohlf, 1995).

Kruskal–Wallis tests and Mann–Whitney U-tests were used to test hypotheses of no differences between latitudinal extent and natural history characteristics of birds. Natural history characteristics included trophic guild, forest dependence and number of forest types. Significance levels from Mann–Whitney U-tests are reported along with significance levels with Bonferroni corrections.

According to the pattern predicted by Rapoport's rule, birds with a northern affinity or latitudinal midpoints at high latitudes should have a greater latitudinal extent than birds with a southern affinity or a latitudinal midpoint at a low latitude. Nicaragua's political boundaries are located between 15°N and 11°N. Birds that have a southern latitudinal extent or range limit between 15°N and 11°N were considered to have a northern affinity, while birds with a northern latitudinal extent or range limit between 15°N and 11°N were considered to have a southern affinity. Mann–Whitney U-tests were used to identify if there were significant differences between latitudinal extent and latitudinal midpoint for species with a northern and southern affinity. Chi-square tests were used to identify if there were significant differences among natural history characteristics (trophic guild, forest dependence, number of forest types) for species with a northern or southern affinity. Some trophic guilds (i.e. frugivores and granivores) and number of forest types (three forest types and four or more) were combined for the chi-square test to obtain an expected value greater than five.

Distributional patterns of Nicaraguan birds

There were 424 terrestrial breeding birds recorded in Nicaragua with latitudinal extents ranging from 2° to 121° (Fig. 1). Birds in Nicaragua were classified into four general categories based on latitudinal extent and affinity: birds generally restricted to Central America, birds near the edge of their range within Nicaragua, Neotropical birds and Pan-American birds. There are 84 species (20%) with a latitudinal extent of 12° or less that are generally restricted to Central America and southern Mexico. There are 143 species (34%) with a latitudinal extent between 13° and 29° that have either a Nearctic or Neotropical affinity and are at or near the edge of their latitudinal distribution in Nicaragua. There are 177 species (42%) that have a Neotropical distribution with a latitudinal extent between 30° and 68°. These species latitudinal extents are slightly greater in the southern hemisphere than in the northern hemisphere with the exception of 14 birds with a Nearctic affinity. Twenty 20 species (4%) have latitudinal extents of 70° or greater and have breeding ranges higher than 20° north and south latitude.

Natural history characteristics and latitudinal extent

Latitudinal extents, latitudinal midpoints and body mass were examined over different taxonomic levels (Table 1). There was a significant negative association between latitudinal extent and the latitudinal midpoint for all terrestrial breeding birds and for major orders and suborders. Five families or subfamilies had significant negative associations between latitudinal extent and latitudinal midpoint. There was a significant positive correlation between the latitudinal extent of terrestrial breeding birds and body mass. When orders and suborders were examined, there was a significant positive correlation between latitudinal extent and body mass of non-passerines. An analysis of the eight largest families or subfamilies in Nicaragua resulted in no significant correlation between body mass and latitudinal extent.

A Kruskal–Wallis test revealed that there were significant differences (P = < 0.0001) among natural history characteristics within trophic guilds, forest dependence classes, number of forest types and latitudinal extent. Natural history characteristics were compared with species latitudinal extent (Table 2). Carnivores and granivores had the greatest latitudinal extent of all trophic guilds while nectarivores and frugivores had the smallest latitudinal extent. There were significant differences among a majority of the trophic guilds. In particular, carnivores had significantly greater latitudinal extents than all other trophic guilds except granivores. There was no significant difference among the latitudinal extent and solid and patchy forest preference, but species that do not require forest had significantly larger ranges than patchy forest birds. Birds that occurred in only one or two habitat types in Nicaragua had a significantly smaller latitudinal extent than birds that occurred in three or more forest types.

Rapoport's rule

One hundred (24%) terrestrial breeding birds reach their range limits at the same latitude as Nicaragua's political boundaries. There were 46 birds with a northern affinity that reached the end of their southern extent in Nicaragua and 54 birds with a southern affinity that had their northern extent in Nicaragua. Birds with a northern affinity (mean latitudinal midpoint 21.8, standard deviation 7.6) had significantly (Mann–Whitney U, P = < 0.0001) higher latitudinal midpoints than birds with a southern affinity (mean latitudinal midpoint 7, standard deviation 4.2). However, species with a southern affinity (mean latitudinal extent 23.8, standard deviation 15.7) had significantly greater latitudinal extents (Mann–Whitney U, P = 0.034) than species with a northern affinity (mean latitudinal extent 17.2, standard deviation 14.4).

Natural history characteristics were examined for species with northern and southern affinities (Table 3). There were no significant differences between trophic guilds (χ² = 3.5, d.f. = 4, P = 0.485) or forest dependence (χ² = 4.0, d.f. = 2, P = 0.133) for species with a southern and northern affinity. There was a significant difference among the number of forest types used by forest birds and affinity (χ² = 13.4, d.f. = 2, P = 0.001); birds with a southern affinity were restricted to fewer forest types than birds with a northern affinity.

Latitudinal distribution of Nicaraguan birds

Birds generally restricted to Central America have latitudinal extents between 20°N and 8°N. The 8°N southern limit of these species is associated with the southern Panama isthmus, although one species extends into Colombia. The northern extent of Central American birds occurs in southern Mexico. This region contains one of the few east–west mountain ranges in the Neotropics and the region is located near the northernmost extent of cloud forest and lowland rain forest (Howell, 1969; Watson & Peterson, 1999). Although there are no endemic birds in Nicaragua, there are 24 birds restricted to Central America that do not extend into Mexico or South America. These species may be of global conservation importance (Long et al., 1996). However, only two species were included on the threatened birds of the world list and none received an urgent or high conservation priority from the list of threatened birds of the Neotropics (Stotz et al., 1996; BirdLife International, 2000). This suggests that Nicaragua may deserve a relatively low conservation priority compared to other Neotropical countries, but little quantitative abundance data exist in Nicaragua to assess the accuracy of these global conservation priority lists.

It is interesting to note that as total latitudinal extent increases, there is a great deal of fluctuation between the northern and southern affinities of species. Species with a northern affinity generally require pine or cloud forests while species with a southern affinity generally require lowland tropical forest. Species at the edge of their distributional or elevational ranges have been hypothesized to be more susceptible to local extinction as a result of habitat fragmentation (Terborgh & Winter, 1980; Christiansen & Pitter, 1997). If this is the case, the high proportion (24%) of species at their range limit within Nicaragua may suggest that many forest birds are vulnerable to local extinction. Yet, there has been evidence that species ranges contract to the edge not to the centre of their ranges (Channell & Lomolino, 2000).

There is a latitudinal asymmetry for Neotropical birds that occur in Nicaragua. The northern breeding limits occur around 20° in the northern hemisphere and at slightly higher latitudes in the southern hemisphere. A similar pattern has been noted for birds in the New World, in which species richness is pear-shaped, with greater species richness at higher latitudes in the southern hemisphere than in the northern hemisphere (Blackburn & Gaston, 1996b). Although climate could be used to explain this pattern, the true meteorological equator is located 3°N of the cartographic equator, suggesting the opposite bias (France, 1998). A more plausible explanation is that tropical forests extend to higher latitudes in the southern hemisphere. The latitudinal limits of tropical dry forest, cloud forest and lowland rain forest in the northern hemisphere are located at 27°N, 22°N and 22°N, respectively, while the limits of these forest types in the southern hemisphere are located at 22°S, 28°S and 28°S, respectively (Dinerstein et al., 1995). This most probably accounts for the slight asymmetry in Neotropical bird ranges.

Natural history characteristics and range

There is a growing body of literature that examines how abiotic and biotic factors determine avian range boundaries and the relationship among geographical range, abundance and body size or mass (Wiens, 1989; Brown, 1995; Blackburn & Gaston, 1996a). However, besides body mass, there is significantly less information on how natural histories of species are associated with range size or latitudinal extent. Natural history characteristics of body mass, trophic guild, forest preference and number of habitat types are important in explaining the latitudinal extent of birds in Nicaragua.

Body mass has been associated with geographical range size when a large number of species that occur in a region are examined, but not when a subset of these species are examined (Blackburn & Gaston, 1996a; Blackburn et al., 1998). The same appears true in Nicaragua and will occur most probably for birds in other regions. Although there is a positive correlation between terrestrial breeding birds and body mass, the pattern is less pronounced when an analysis is undertaken at the order or family level. Certain trophic guilds have significantly larger range sizes than other trophic guilds. Carnivores have the greatest latitudinal extent followed by granivores, insectivores and omnivores, while frugivores and nectarivores have the smallest latitudinal extent. There appears to be less overlap in trophic guilds within the tropical bird communities than within temperate bird communities (Brown & Maurer, 1987). Although carnivores have greater geographical ranges than other trophic groups, the order in which other trophic guilds occur may be a region specific-pattern. Forest birds have smaller ranges than non-forest birds in Nicaragua. Most species that require forest have smaller ranges than non-forest species and this may be a region-specific pattern in the tropics, although ranges of forest birds will probably decrease as tropical forest cover continues to decrease in the Neotropics (Myers et al., 2000). Species that occur in one and two forest types have a significantly smaller range than species that occur in three or more. Number of forest or habitat types may be one of the best predictors of a species’ latitudinal extent for birds in Nicaragua and for that matter in other regions.

Rapoport's rule?

Although species richness was greater for species with a southern affinity compared to species with a northern affinity, species with a southern affinity had significantly greater latitudinal extents. Furthermore, there was a significant negative relationship between latitudinal midpoint of terrestrial breeding birds and latitudinal extent over a number of taxonomic scales. Blackburn & Gaston (1996a) also noted that New World bird geographical range size was smallest at about 17°N. This study of latitudinal extent of Nicaraguan birds concurs with Gaston et al. (1998) and Harcourt (2000) in that Rapoport's rule is best referred to as Rapoport's effect, because there is too much variability in the pattern to be considered a rule.

Comparisons of natural history characteristics of birds with northern and southern affinities does not provide clear support for the argument that species’ natural histories are more specialized toward the equator (Stevens, 1989). Birds with a southern affinity comprise an increased proportion of insectivores (i.e. ant followers), a decrease in the proportion of omnivores and increases in species with solid forest preference. However, there is only a statistically significant difference between number of forest types used by birds with a northern affinity and birds with a southern affinity. A similar variable pattern using habitat and diet breadth has been shown for primates in the tropics (Eeley & Foley, 1999; Harcourt, 2000). Future research on the causes of Rapoport's effect should include natural history characteristics of trophic guild, habitat specificity and number of habitat types to identify if this pattern occurs in other regions and for other taxa.