INTRODUCTION

Beginning with the explorations of Charles Darwin, Richard Spruce, Alexander von Humboldt, Aimé Bompland, and their contemporaries, ornithologists have explored the New World tropics for more than 200 years. In that time, Neotropical ornithology has blossomed (Walters 2003, Stiles 2008), resulting in myriad publications from complete and comprehensive country field guides, regional and taxonomic monographs, extensive compilation volumes and audio publications, to hundreds of thousands of scientific papers (Vuilleumier 2003; but see Vuilleumier 2004 for a critique of the scientific development of Neotropical ornithology).

Little more than a century ago, Neotropical ornithology was predominantly focused on descriptions of new taxa and expedition reports. Natural history studies became more numerous in the mid-twentieth century, particularly in some areas (e.g., Galapagos, Barro Colorado Island, Costa Rica; Lack 1945, Skutch 1950), while in recent decades taxonomic studies made a resurgence following advances in molecular phylogenetics (Cadena et al. 2007, Parra et al. 2009; see Vuilleumier 2003). This trend continues, and many studies currently combine molecular techniques with morphological and vocal analyses, in an attempt to unravel the phylogenetic history of clades across the taxonomic hierarchy (Isler et al. 2008, Cadena and Cuervo 2010), including studies revolutionizing our understanding of relationships within the entire class Aves (Hackett et al. 2008).

During the most recent decades, development of Neotropical ornithology has been greatly aided by the publication of two international journals devoted entirely to the region's avifauna (Ornitología Neotropical and Cotinga) and further advanced by the publication of several regional journals and bulletins (Ornitología Colombiana, Boletín Sociedad Antioqueña de Ornitología, Huitzil, El Pitirre—now Journal of Caribbean Ornithology, and Ararajuba—now Revista Brasileira de Ornitologia). These fairly new publications have complemented the work of older ones (Hornero, Boletín Chileno de Ornitología), as well as that of international journals with a broader geographic focus (Journal of Field Ornithology, Wilson Journal of Ornithology, Bulletin of the British Ornithologists' Club, The Condor, among others), further improving on the advancement and dissemination of ornithological studies in the region.

Despite the accumulated wealth of published information from the Neotropics, new species are still regularly described (Carantón-Ayala and Certuche-Cubillos 2010, Whitney et al. 2010, O'Neill et al. 2011), and the systematics of all South American birds are under continued, exhaustive revision (Remsen et al. 2012). Moreover, information generated and published is not evenly distributed across the region. As noted by Winker (1998), some countries continue to accumulate a fair number of publications dedicated to their avifauna, whereas others remain poorly documented.

In 1987, P. C. James surveyed trends and deficiencies of Neotropical ornithology from a geopolitical perspective. Consulting Zoological Record, James found that 42% of papers published from 1972 to 1983 originated from only three countries (Mexico, Brazil, and Argentina), nine countries contributed 51%, and the remaining nine continental countries contributed only 9% (James did not include Caribbean countries in his assessment) (see table 2 in James 1987). A decade later, D. Winker (1998) reassessed the state of Neotropical ornithology by consulting Zoological Record from 1979 to 1995. In this study, the same three leading countries produced 41% of all publications, whereas 18 countries contributed 52%, and another 18 countries contributed 7%. Further, Winker found that no country, except for the United States, had a ratio of publications per species number higher than 1.0; in other words, no Neotropical country had at least one reference per species (see table 2 in Winker 1998).

Although these trends may have continued during the 15 years since Winker's study, the number of publications, particularly those by Latin American residents, has increased significantly (Navarro-Sigüenza et al. 2008). However, Levy (2008) reported that Caribbean ornithology is still dominated by nonresidents, with several Caribbean countries nearly or completely lacking resident ornithologists. This situation merits a new assessment of the current state of knowledge on Neotropical birds at a regional (country) level, which is provided in this paper. In addition, to offer a general assessment of the advance of different areas of knowledge across the region, we offer a summary of numbers of publications per subject during the periods 1979–1995 and 1996–2011.

METHODS

Using methods established by previous authors, we performed searches in Zoological Record (1996–2011) using two search words: birds and country name. After performing all searches, a careful visual revision, reference by reference, was performed to avoid searching pitfalls (e.g., Pluvialis dominica for Dominica; Brazil as the last name of Mark A. Brazil, a non-Neotropical ornithologist; New Mexico instead of Mexico). All filtered references for every country found in Zoological Record are available at  http://avesamericanas.lifedesks.org/node/2389.

We included all 33 Neotropical countries as currently recognized by the Organization of American States ( www.oas.org) and the United Nations ( www.un.org), as well as 9 overseas territories or departments (from the United States, France, the United Kingdom, and The Netherlands), that were not included by Winker (1998), totaling 41 countries / territories (Table 1). We incorporated 4 territories not included by Winker (1998). In addition to the journals covered in Winker's review, we also analyzed Cotinga, first published in 1994 and also covered by Zoological Record (Figure 1). Likewise, Ornitología Neotropical (first published in 1990) was analyzed separately as Zoological Record did not cover it until 2002 (Figure 2). Also, we accounted separately for papers authored by authors residing in Neotropical countries in both journals, according to authors' affiliation institutions or addresses (Table 2). We used number of publications as an indication of current knowledge, and calculated the ratio of papers per species richness for each country. To facilitate direct comparisons, we used the same species numbers as Winker (Table 1).

We performed statistical comparisons to determine whether the number of articles published in the period analyzed by Winker (1979–1995) was significantly different from that analyzed herein (1996–2011). To compare both periods, we applied a chi-square goodness-of-fit test (or one-way chi-square test) with one degree of freedom, to data gathered from the “top-12” countries. Since both datasets include different time spans (17 vs. 15 years), the test was calculated by standardizing the expected number of publications per year. To assess the statistical significance of these 12 comparisons, each comparison was evaluated by adjusting the significance level (α) after a Bonferrroni correction (α = 0.05/12).

To explore the relative advance of different areas of knowledge in ornithological research across the region, we performed statistical comparisons to determine whether the number of articles per subject, published in the period 1979–1995, was significantly different from that of the period 1996–2011. This analysis included the “top-12” Neotropical countries, as well as 10 subjects: taxonomy, systematics, biogeography, evolution, ecology, diet, breeding, migration, physiology, and conservation. Although some of these subjects are broadly overlapping (e.g., ecology, diet, breeding, migration), we preferred having a redundancy of publications among subjects, rather than missing some publications that, for example, may focus on subject “breeding” but may not be identified by topic “ecology.”

To delimit searches to Neotropical countries, as a first topic, we listed all Neotropical countries, separated by Boolean operator “OR” as a first topic (e.g., Mexico* OR Brazil* OR Venezuela*, etc.). This method allowed filtering all publications that mentioned any or more than one of the “top-12” countries. In the same search, we used Boolean operator “AND” with “bird” as a second topic and “subject” (e.g., taxonomy or systematics) as a third topic. To compare both periods, we applied a chi-square goodness-of-fit test with one degree of freedom by standardizing the expected number of publications per year. To assess the statistical significance of these 10 comparisons, each comparison was evaluated by adjusting the significance level (α) after a Bonferrroni correction (α = 0.05/10).

RESULTS AND DISCUSSION

As in previous assessments (James 1987, Winker 1998), Mexico, Argentina, and Brazil were the best-represented countries in Zoological Record (1996–2011), accounting for 54% of all references, but in a new sequence: Brazil, Argentina, Mexico (Table 1). Nine additional countries accounted for 34%, and the remaining 30 countries accounted for the remaining 12%. Most Central American countries (except for Costa Rica and Panama) remain poorly represented, whereas only Cuba, Puerto Rico, and Jamaica are fairly well-represented in the Caribbean region. Countries from the Guianan shield remain poorly studied, and Paraguay and Uruguay are significantly behind the remaining South American countries.

Although Zoological Record is not complete—i.e. it does not cover some journals, magazines, books, or regional/national publications—it provides an accurate picture of current knowledge about Neotropical birds. Preliminary surveys across databases (i.e. Zoological Record, ISI Web of Knowledge, Scopus, Latindex, SciELO) showed that title coverage of Zoological Record was, overall, more complete than other options usually associated to Neotropical or Latin American scientific literature. Only 4 out of 10 journals covered by Zoological Record were also found in ISI Web of Knowledge, Scopus, and SciELO. Journals not covered by one or all these databases include regional and non-regional journals known as important repositories of Neotropical ornithological literature (i.e. Bulletin of the British Ornithologists' Club, Journal of Caribbean Ornithology, Huitzil, Cotinga).

As previously documented by James (1987) and Winker (1998), publications concerning the avifauna of some Neotropical nations remain deficient. Fifteen of the 20 most poorly represented countries are shared between Winker's assessment and ours. Nineteen of the 41 countries did not show a significant increase, remained equal, or even decreased. It should be noted that the large number of publications concerning Dominica's avifauna documented by Winker might be better explained by the fact that Zoological Record did not discriminate Dominica, the country, and dominica, the species epithet of several taxa (Oxyura, Pluvialis, and Dendroica). In fact, the unfiltered result of our Zoological Record search was 1,271 references from Dominica (179 in Winker), but removing references to taxon names reduced it to 26.

Six countries have experienced notable increases in number of publications (Figure 3): Brazil, Argentina, Mexico, Chile, Ecuador, and Colombia. Nonetheless, the remaining 6 of the “top-12” countries experienced only small increases in publication numbers (Figure 3). The chi-square goodness-of-fit test indicated that for all “top-12” countries, the number of articles published during 1996–2011 was significantly higher than in the period 1979–1995 (all p < 0.001; α = 0.0042, adjusted after Bonferroni correction).

If the ratio of publications per species is considered, however, even well-represented countries remain under-studied; only the four better-represented countries exceed an index of 1.00 (Table 1). Even for these, however, a more careful species-by-species evaluation will undoubtedly reveal many publications for only a few well-studied species, and none for the majority of species. This trend has recently been documented for Ecuador, where about 74% of the avifauna lacked specific publications as of 2004, and the bulk of published material concerned the Galapagos' avifauna, Ecuador's least species-diverse region (Freile et al. 2006). Therefore, research needs and priorities remain on several fronts, notably poorly known countries, little-studied regions within countries, specific subjects, taxonomic groups, and species (Alves et al. 2008). A simple comparison with non-Neotropical countries from the Western Hemisphere indicates that Neotropical ornithology is still far from “maturity” (Vuilleumier 2003). Winker (1998) found that the United States (USA) alone had more than 4,100 publications from 1979 to 1995 and a 5.36 ratio of papers per species. The sum of publication totals for three of the four best-documented countries in our study (Argentina, Mexico, and Chile) barely surpasses this total. These three countries together show a 1.76 publications per species ratio, reflecting higher species richness and poorer knowledge when compared to the USA. Brazil is a notable case within the region, as the number of publications in the 1996–2011 period is vastly superior to the next countries on the list.

Totals for Cotinga differ from those of Zoological Record in several aspects. First, Ecuador and Brazil account for 37% of all publications, followed by Peru and Mexico with 20%; 8 countries account for 32%, and 30 countries account for the remaining 11%. Mexico and Argentina ranked fourth and fifth, respectively. Publication rates in Cotinga and Zoological Record were similar for Central America; aside from Costa Rica and Belize, most countries were poorly represented. Similarly, together all Caribbean countries (excepting Cuba) had only 31 references, while alone Cuba had 30 (Figure 1).

Patterns of publications in Ornitología Neotropical also differed from those of Zoological Record. Argentina and Brazil accounted for 37% of published papers, followed by 8 countries accounting for 48%, 6 countries summing 9%, and the remaining 26 countries with the remaining 6%. Again, Costa Rica and Panama accounted for 63% of all Central American references, whereas among Caribbean countries only Cuba and Puerto Rico are not on the bottom of the list. Among South American countries, the three Guianan states (Guyana, French Guiana, and Suriname), Uruguay, and Paraguay together sum fewer references than Bolivia, which ranked eighth among South American countries (24 vs. 32) (Figure 2).

Of 865 papers published in Ornitología Neotropical, a notable 74% (641) were authored by residents from Neotropical countries (this includes 59 papers in which residents are “only” junior authors; Table 2). Notably, for 11 countries, the majority (67–100%) of papers were generated by Neotropical authors. In Argentina and Brazil nearly 95% of the 322 papers published for those countries were produced by residents, primarily nationals. Similarly, Colombians and Mexicans published nearly 83% of all references on their countries' birds.

Even though not previously assessed by James or Winker, the numbers of Neotropical authors have notably increased, for most countries, over the last decade (personal observation). The exceptions are Brazil, Argentina, and Mexico, where a longer-standing tradition of ornithology in museums and academic institutions has resulted in high productivity by residents (Di Giacomo and Di Giacomo 2008, Alves et al. 2008, Navarro-Sigüenza et al. 2008; see also Naranjo 2008, Pitman et al. 2011). Some countries, such as Ecuador and Bolivia, have fairly large numbers of publications authored by residents (Table 2). It should be pointed out, however, that nearly 60% of those are authored by non-nationals, reflecting the need to continue strengthening national ornithology programs (Naranjo 2008, Pitman et al. 2011).

Comparing numbers of publications by area of knowledge between the 1979–1995 and 1996–2011 periods was informative as it allowed exploration of current and future directions of ornithology in Neotropical countries (Figure 4). Not surprisingly, numbers of publications on all subjects have increased (all p < 0.001; α = 0.005, adjusted after Bonferroni correction). Numbers of papers in taxonomy, systematics, evolution, migration, and physiology increased 160–200%; papers in ecology, diet, and breeding increased 222–253%, and papers in conservation had more than a three-fold increase, with 378%. Finally, papers in biogeography are still relatively few (in raw numbers) compared with those in all other areas, but experienced a 484% increase. It is important to notice that ecology is still the leading subject in Neotropical ornithology (Dayton 2003). Conservation is now the second-most published subject, still close to breeding, and reflecting an ever-increasing global awareness of the current biodiversity crisis.