Master in Biodiversity and Conservation in Tropical Regions

Biogeography and Conservation Status of Diurnal Raptors in Venezuela

Adrian Naveda-Rodríguez

A thesis submitted

In partial fulfillment of the requirements for the degree of

Master of Science

Advisor:

Keith L. Bildstein, Ph.D.

Academic year 2012 – 2013

ACKNOWLEDGMENTS

This work was possible thanks to logistical and financial support provided by the Spanish Research Council (CSIC), Hawk Mountain Sanctuary, The Peregrine Fund and Wild4Ever.

I wish to thank all those who contributed throughout the course of this work. I am absolutely indebted to my lovely wife Gabriela Lugo, my parents Maritza Rodriguez and Rafael Naveda who supported me in this process. I am very grateful to Dr. Keith Bildstein for introducing me to raptor conservation sciences and for all the opportunities he has given me since 2006. I am also grateful to Dr. Hernán Vargas and Dr. Gary Riggs for their confidence and support for this project. Marcial Quiroga-Carmona, José Gustavo León, Tony Crease, Alan Highton and Christian Olaciregui provided valuable information on raptors of Venezuela and Colombia. I sincerely appreciate the efforts of Marcial Quiroga-Carmona, Jorge Peralta and Bayron Calles for their help with statistical analyses. I appreciate the improvements in English usage made by Phillip Schwabl

I kindly appreciate Francisco Bisbal, Alexis Araujo, Miguel Lentino, Jurahimar Gamboa, Marcos Salcedo, Carlos Rengifo and Rosana Calchi for providing information on voucher specimens in ornithological collections under their care.

ABSTRACT

Aim: Natural history data is crucial to monitor the vulnerability of birds of prey. In Venezuela, this information is non-existent for almost all species of raptors. Here, I examine the biogeography and conservation status of birds of prey of the order Cathartiformes, Acciptriformes and Falconiformes in Venezuela.

Location: Venezuela, northern South America.

Method: By means of ecological niche modeling, the geographic distribution (SDM) of 65 species of diurnal raptors was described. Nine climatic and seven landscape metrics were used as environmental predictors. In a GIS environment the extent of occurrence (EOO) was estimated for each species using the minimum presence training threshold. SDM were stacked to examine species richness (S) patterns. Predictors of S were investigated using a regression model. The EOO was used to re-evaluate the conservation status of diurnal raptors in Venezuela, applying the criterion B1 of the IUCN Red List. Furthermore, a gap analysis was performed to evaluate the effectiveness of strict protected areas in the conservation of raptor diversity.

Results: The EOO ranged from 1,072 km2 to 881,918 km2. 40 species showed a continuous distribution, while a disjunct distribution was observed in 25 species. S differed among bioregions. However, six pairwise compared bioregions did not show differences. Highest S values were located in the Guiana Shield and the mountains ranges of northern Venezuela. S was explained mainly by landscape features, specifically forest canopy height, land cover and terrain slope. Five species met the B1 criterion and were re-classified.

Main conclusions: Environmental heterogeneity affects the configuration of EOO areas and the distribution of S and is therefore an important aspect to consider in conservation planning for neotropical birds of prey. The computation of EOO areas is susceptible to the applied estimation methods, and responses from environmental variables used to predict S are scale-dependent; therefore, it is necessary to standardize methods and experimental design to study the biogeography of raptor species. Priority-setting for the conservation of raptor species in Venezuela must consider rare species, even if they are not threatened. A territorial re-ordering is urgent to improve the protection of this group of birds.

INTRODUCTION

The Neotropic, an area of more than 20 million km2 comprising South and Central America, the Caribbean islands, parts of Mexico and southern Florida, is the richest biogeographic region in the world (Ojasti 2000, García-Moreno et al. 2007, Rodríguez & Rojas- Suárez 2008). In this context, Venezuela holds a special ecological significance due to its geography, which includes part of the Cordillera de Los Andes, the Guiana Shield, the Llanos, the Caribbean Sea and the Atlantic Ocean. These locations give the country a wide diversity of biomes along 27 climatic zones, giving rise to 650 natural vegetation types, 23 landforms and 37 major geological units. In consequence, Venezuela is among the 17 countries with the greatest biological diversity in the world (megadiverse countries), with 15,000 higher plant species, 351 mammal species, 1,400 bird species, 340 reptile species, 315 amphibian species and over 1,800 fish species (Mittermeier et al. 1997, MARN 2000, Aguilera et al. 2003, Rodríguez & Rojas- Suárez 2008).

Birds are the best known vertebrate group in Venezuela due to the interest of the scientific community (Lentino 2003); nevertheless, the development of the ornithological knowledge in the country is still incipient, particularly in specific cases such as with birds of prey, which include about 90 species of diurnal and nocturnal raptors (Ferguson-Lees & Christie 2001, König & Weick 2008). The diurnal raptors of Venezuela are comprised of 68 species distributed in four families of the order Cathartiformes, Accipitriformes and Falconiformes (Ascanio et al. 2012, Hilty 2003). As are other tropical raptors species, they are confronting threatening factors such as habitat loss and destruction, environmental pollution and human persecution (Bierregaard 1998, Bildstein et al. 1998). Thus, conservation action is required to prevent species loss.

In the last evaluation of extinction risk of Venezuelan wildlife, six species of diurnal raptors were included (Rodríguez & Rojas-Suarez 2008), one species was listed as critically endangered, two species as vulnerable, three species as near threatened and six species as data deficient. The conservation status assessment of Venezuelan raptors has been in some cases underestimated whereas in others overestimated. According to IUCN (2012), the evaluation of the global conservation status of any species is based on any of the following five criteria: (A)

Reduction in population size, (B) Geographic range in the form of Extent of Occurrence or Area of Occupancy, (C, D) Population size estimates and (E) Quantitative analysis showing the probability of extinction in the wild. The information required to apply any of these criteria is not available in Venezuela due to the lack of biological information on birds of prey.

The natural history of diurnal raptors in Venezuela has been poorly studied, and only a few species have been subject of detailed description and analysis of their autoecology (Alvarez- Cordero 1996, Balgooyen, 1989, Beissinger et al. 1988, Mader 1982); basic (but not less important) aspects such as species richness and geographic distribution patterns and the factors determining them, remain unstudied for most species. Consequently, little is known about how the number of species varies among the biogeographic regions of Venezuela. Available distribution maps in field guides and in Ridgely et al. (2007) lack metadata and incur omission/commission errors.

The deficiency in knowledge of the natural history and population biology of birds of prey do not allow the correct assessment of the conservation status, as well as the design and implementation of effective conservation strategies. Sound scientific evidence based on a number of biological factors, the knowledge of geographic distribution and population density (Bierregaard 1998) as well as species richness patterns are crucial in monitoring and mitigating the vulnerability of raptors species. In an attempt to fill this gap, I performed biogeographic analysis and an evaluation of conservation status by means of geographic distribution of the Cathartiformes, Accipitrifomes and Falconiformes of Venezuela. I addressed my effort by: (1) describing the geographic distribution of diurnal raptors, (2) describing patterns of species richness, (3) determining the environmental variables that shape species richness patterns, (4) assessing the current conservation status of diurnal raptors in Venezuela.

MATERIALS AND METHODS

Study area

Venezuela is in northern South America, between latitudes 00º45’ North and 15º40’ North and between meridians 59º45’ West and 73º25’ West. Its total land mass occupies 916,445 km2 and its maritime territory covers around 900,000 km2. Venezuela is bordered in the south by

Colombia and Brazil, in the west by Colombia and in the east by Guyana. The Dominican Republic, Netherlands Antilles and United States of America (Puerto Rico and Virgin Island) lie to the north and France (Martinique and Guadalupe Islands) and Trinidad and Tobago lie to the East (MARN 2000).

Delimitations of natural regions or bioregions of Venezuela vary according to the author and definition criteria. Nonetheless, there is a general consensus in recognizing at least nine spatial units with distinct environmental and geographical characteristics (PDVSA 1992, Linares 1998). In this work I included the 13 bioregions defined by Huber & Alarcón (1988). These include: Insular, Coastal, Central Coastal Range, Eastern Coastal Range, Orinoco Delta, Maracaibo Lake Basin, Llanos, Mountain Range of Mérida, Mountain Range of Perijá, Lara- Falcón Hill System, Guayana Massif, Foothills System of Guayana Massif and Amazonia.

Work scale and cartographic aspects

Working with geographic data supposes the generation of cartographic products, so it is necessary to define the applied geographical scale. All the data was identified and analyzed at macro-ecosystem scale; a cartographic scale of 1:2,000,000 was selected, considering the level of abstraction appropriate for studies of extensive areas (Velásquez et al. 2004). In this scale the minimum mapping unit is 1 km or 0.00833 degree.

The input data (species records, environmental variables) were processed in a latitude and longitude geographic coordinate system (Datum: WGS84, Coordinate units: decimal degree) and projected to Universal Transverse Mercator geographic coordinate system (Datum REGVEN UTM Zone 19N, Coordinate Units: meters).

Species data

I conducted the analyses on species of the order Cathartiformes, Accipitriformes and Falconiformes, represented in Venezuela by 68 species of 35 genera distributed in four families (Ascanio et al. 2012). A database containing the presence records of the species was built with three attribute fields: Species (taxonomic identification of the record), Latitude (geographic coordinates in decimal degree referring to the location where the species was recorded) and Longitude (idem Latitude). Presence records were obtained from voucher specimens deposited in

the Colección Ornitológica Phelps (COP), Colección de Vertebrados de la Universidad de Los Andes (CVULA), Museo de la Estación Biológica de Rancho Grande (EBRG), Museo de Biología de la Universidad del Zulia (MBLUZ), Museo de Biología de la Universidad Central de Venezuela (MBUCV), Museo de Ciencias Naturales de Caracas (MCNC) and Museo de Historia Natural La Salle (MHNLS). Additional records were obtained from eBird (Sullivan et al. 2009) and the pertinent literature. All gathered records were georeferenced, and taxonomically standardized following Remsen et al. (2013). The database containing 9,237 occurrence records was revised to reduce geographical and taxonomical bias resulting from input sources. The final edited database included 6,197 records of 65 species. Three species (Ictinia mississippiensis, Circus cyaneus and Micrastur buckleyi) were not included in the analysis because recent presence records are lacking

Species geographic distribution

Species distributions were described by means of ecological niche modeling (Peterson 2001); I used the maximum entropy method in the program MAXENT 3.3.3k (Phillips et al. 2006) to generate species distribution models (SDM). Twenty-six environmental predictors with spatial resolution of 1 x 1 km obtained from remote sensing data were used in MAXENT. These included the 19 bioclimatic variables available in WorldClim 1.4 (Hijmans et al. 2005), digital elevation model (DEM) from shuttle radar topographic mission (Jarvis et al. 2008), slope and aspect derived from DEM, topographic roughness index calculated as the surface area ratio derived from DEM (Jenness 2013), Terra MODIS MOD44B (Townshend et al. 2011), Terra MODIS MCD12Q1 (NASA 2013), and ICESat/GLAS 3D Global Vegetation (Simard et al. 2011). In order to reduce collinearity I performed a Pearson’s pairwise correlation test in SPSS 19.0 (IBM 2010) and removed one of the variables in each pair that had a pairwise correlation value higher than 0.8. The final variable set included 16 variables (Table 1). SDM were developed using MAXENT default settings and cumulative output. Model accuracy was evaluated using Area Under the Curve (AUC) of Receiver Operator Characteristic, SDM with AUC values above 0.8 were considered indicative of good accuracy.

The models generated were reclassified into models of presence/absence (binary models) using MAXENT’s minimum training presence, this threshold includes all the known localities of

the species. Presence pixels of the binary models of each species were converted to polygons in ArcGIS 9.3 (ESRI 2008), yielding one polygon per pixel. Polygons were merged into a single polygon and the area of the resulting polygon was calculated in km2; this was considered the extent of occurrence of a species (EOO).

Table 1. Environmental variables used in MAXENT.

Variable Description of variable Bio_1 Annual mean temperature Bio_3 Isothermality Bio_4 Temperature seasonality Bio_7 Temperature annual range Bio_12 Annual precipitation Bio_15 Precipitation seasonality Bio_17 Precipitation of driest quarter Bio_18 Precipitation of warmest quarter Bio_19 Precipitation of coldest quarter DEM (alt) Ground height in relation to sea level Slope (slope) Ground inclination relative to the horizontal plane Aspect (aspect) The direction in which a slope faces Surf area ratio (surf) Topographic index. Quotient of surface area/planimetric area MOD44B (forest) Percentage of tree canopy cover ICESat/GLAS (forhei) Forest canopy height MCD12Q1 (land) Categorical variable with land cover information.

The area of distribution was defined for each species, adopting the concept of disjunct distribution (a range geographically separated into two or more areas by natural elements (e.g. ecological barriers, bioregion) and continuous distribution (a range that is not separated by natural elements).

Species richness patterns

In order to obtain a species-richness grid (SRG), I stacked (summed) all the binary models generated in MAXENT using ArcGIS 9.3 in which the digital number of each pixel represents the total species number in that pixel. Differences in species richness between

biogeographic regions of Venezuela were evaluated with one-way ANOVA followed by a post hoc Dunnett’s T3 test performed in SPSS 19.0. Additionally, I used principal components analysis of the environmental variables (Table 1) from each bioregion to help explain variability in species richness among bioregions. A random stratified sampling was used to select 5% of pixels of each bioregion. A total of 53,857 pixels were used in the analysis. Selected pixels were separated with a distance of three kilometers in order to avoid the spatial autocorrelation between two contiguous pixels (Legendre 1993).

Environmental variables predicting species richness patterns

To determine the power of environmental variables to explain species richness patterns, I performed a forward stepwise regression in SPSS 19.0. One percent (10,741) pixels of the SRG and the same variables data set used in the SDM were employed in this analysis. Minimum distance between selected pixels was five kilometers. Significance values to enter and to leave the model were set at 0.05 and 0.10, respectively.

Conservation status assessment

Conservation status of diurnal raptors in Venezuela was evaluated using the geographic range in the form of extent of occurrence criteria (B1) (IUCN 2012). Furthermore, I performed a GAP analysis (Scott et al. 1993) to evaluate the effectiveness of protected areas in habitat protection, I used the digital cartography of national parks, natural monuments and wildlife refuges (Rodríguez et al. 2005), defined as strict protected areas (SPA) (Rodríguez & Rojas- Suárez 1998). Using ArcGIS 9.3, I intersected the species data set with SPA to estimate the amount of EOO and protected species richness.

RESULTS

Species geographic distribution

The 65 SDM were accurate, mean AUC values averaged 0.91 with a range from 0.82 to 0.98. EOO had values from 1,072 km2 to 881,918 km2. Andean Condor (Vultur gryphus) and Semicollared Hawk (Accipiter collaris) had the lowest EOO: 1,072 km2 and 8,570 km2, respectively; Turkey Vulture (Cathartes aura) and Roadside Hawk (Rupornis magnirostris) had

the greatest EOO (>850,000 km2). On the other hand, 40 species showed a continuous distribution while a disjunct distribution was observed in 25 species. Values of AUC, EOO and distribution of each species are shown in Table 2. Thirty-seven (24%) species hold a wide distribution within the country and were present in more than ten bioregions while seven (5%) species had distributions restricted to one or two bioregions (Table 3).

Table 2. Number of species occurrence records (#OC), accuracy of species distribution models (AUC), extent of occurrence (EOO), EOO protected distribution area type, and red list status of 65 species of diurnal birds of prey in Venezuela.

Species #OC AUC EOO, km2 EOO Protected, km2 (%) Distribution Status Cathartes aura 434 0.89 881,918 147,060 (16.7) Continuous LC Cathartes burrovianus 82 0.92 704,727 106,181 (15.1) Continuous LC Cathartes melambrotus 94 0.95 319,559 81,539 (25.5) Continuous LC Coragyps atratus 578 0.91 762,454 123,916 (16.3) Continuous LC Sarcoramphus papa 116 0.88 701,752 111,301 (15.9) Continuous LC Vultur gryphus 6 0.98 1,072 923 (86.1) Disjunct EN B1ab(i,iii,iv) Pandion haliaetus 160 0.92 651,629 86,351 (13.3) Continuous LC Elanus leucurus 134 0.94 699,518 106,133 (15.2) Continuous LC Gampsonyx swainsonii 148 0.9 633,698 68,103 (10.7) Disjunct LC Chondrohierax uncinatus 77 0.87 846,492 137,623 (16.3) Continuous LC Leptodon cayanensis 63 0.86 755,334 127,604 (16.9) Continuous LC Elanoides forficatus 126 0.93 713,430 136,745 (19.2) Continuous LC Morphnus guianensis 21 0.96 174,453 30,095 (17.3) Disjunct LC Harpia harpyja 25 0.91 363,756 81,822 (22.5) Disjunct LC Spizaetus tyrannus 72 0.91 759,368 129,788 (17.1) Continuous LC Spizaetus melanoleucus 31 0.94 197,305 40,618 (20.6) Disjunct LC Spizaetus ornatus 50 0.89 590,373 80,812 (13.7) Continuous LC Spizaetus isidori 14 0.89 36,810 13,031 (35.4) Disjunct LC Busarellus nigricollis 95 0.93 538,344 58,515 (10.9) Disjunct LC Rostrhamus sociabilis 86 0.93 741,753 113,435 (15.3) Continuous LC Helicolestes hamatus 12 0.96 155,464 13,322 (8.6) Disjunct LC Harpagus bidentatus 51 0.9 453,222 111,888 (24.7) Disjunct LC Harpagus diodon 7 0.87 275,408 62,056 (22.5) Disjunct LC

Table 2. (continued)

Species #OC AUC EOO, km2 EOO Protected, km2 (%) Distribution Status Ictinia plumbea 178 0.89 742,537 124,494 (16.8) Continuous LC Circus buffoni 17 0.94 189,369 14,310 (7.6) Continuous LC Accipiter poliogaster 8 0.94 80,126 10,003 (12.5) Disjunct LC Accipiter superciliosus 28 0.95 216,236 48,470 (22.4) Disjunct LC Accipiter collaris 6 0.97 8,570 4,985 (58.2) Disjunct VU B1ab(i,iii) Accipiter striatus 63 0.96 191,918 61,855 (32.2) Disjunct LC Accipiter bicolor 38 0.84 833,695 139,802 (16.8) Continuous LC Geranospiza caerulescens 89 0.91 626,678 86,873 (13.9) Disjunct LC Buteogallus schistaceus 7 0.98 36,889 2,404 (6.5) Disjunct LC Buteogallus anthracinus 72 0.89 404,510 31,602 (7.8) Continuous LC Buteogallus aequinoctialis 24 0.98 28,110 2,963 (10.5) Continuous LC Buteogallus meridionalis 152 0.93 664,598 84,997 (12.8) Continuous LC Buteogallus urubitinga 125 0.88 660,580 88,595 (13.4) Continuous LC Buteogallus solitarius 12 0.9 71,421 21,459 (30.0) Disjunct LC Rupornis magnirostris 468 0.86 858,709 139,952 (16.3) Continuous LC Parabuteo unicinctus 44 0.94 422,719 23,116 (5.5) Continuous LC Parabuteo leucorrhous 18 0.9 30,507 9,422 (30.9) Disjunct LC Geranoaetus albicaudatus 124 0.88 678,269 94,772 (14.0) Continuous LC Geranoaetus melanoleucus 23 0.95 15,973 5,719 (35.8) Continuous NT Pseudastur albicollis 59 0.92 540,400 96,358 (17.8) Continuous LC Leucopternis melanops 17 0.94 184,563 51,602 (28.0) Continuous LC Buteo nitidus 132 0.93 444,478 57,247 (12.9) Continuous LC Buteo platypterus 142 0.95 594,096 132,318 (22.3) Disjunct LC Buteo albigula 7 0.93 8,660 3,603 (41.6) Disjunct VU B1ab(i,iii) Buteo brachyurus 80 0.93 514,370 81,024 (15.8) Continuous LC Buteo swainsoni 13 0.96 17,274 4,929 (28.5) Continuous VU B1ab(i,iii) Buteo albonotatus 58 0.88 503,194 49,424 (9.8) Continuous LC Herpetotheres cachinnans 125 0.9 644,049 101,281 (15.7) Continuous LC Micrastur ruficollis 67 0.91 463,773 107,091 (23.1) Disjunct LC Micrastur gilvicollis 17 0.93 295,296 44,718 (15.1) Continuous LC Micrastur mirandollei 12 0.89 174,796 22,667 (13.0) Disjunct LC Micrastur semitorquatus 41 0.88 732,868 125,036 (17.1) Continuous LC

Table 2. (continued)

Species #OC AUC EOO, km2 EOO Protected, km2 (%) Distribution Status Caracara cheriway 248 0.9 704,729 89,490 (12.7) Continuous LC Ibycter americanus 76 0.91 601,845 113,923 (18.9) Disjunct LC Daptrius ater 64 0.93 448,520 71,529 (15.9) Disjunct LC Milvago chimachima 355 0.89 694,253 111,880 (16.1) Continuous LC Falco sparverius 350 0.9 641,724 83,770 (13.1) Disjunct LC Falco columbarius 54 0.95 433,466 57,368 (13.2) Continuous LC Falco rufigularis 144 0.9 817,109 135,869 (16.6) Continuous LC Falco deiroleucus 12 0.83 261,784 80,822 (30.9) Continuous LC Falco femoralis 105 0.9 672,116 100,580 (15.0) Continuous LC Falco peregrinus 41 0.93 697,037 118,744 (17.0) Continuous LC

Species richness patterns

According to the one-way ANOVA there are significant differences (F12, 53,844 = 1561.1, p = 0) in mean species richness between the bioregions; nonetheless, the post hoc Dunnett’s T3 test did not exhibit significant differences in six pairwise compared bioregions (Table 3). The principal component analysis of the environmental variables used in this study (Table 1) did not exhibit differences among the regions: Coastal Ranges and Foothills System of Guayana Massif (Figure 1A); Insular and Coastal (Figure 1B); and Orinoco Delta and Foothills System of Guayana Massif (1C).

Maracaibo Lake System has species richness and elevation range from 1 to 44 species and 0 to 799 m.a.s.l., respectively; Lara-Falcon Hill System has an elevation range from 0 to 1782 m.a.s.l. and a species richness range from 1 to 41 species. Considering the number of species below 799 m.a.s.l in Lara-Falcon Hill System, a range of 7 to 41 species was observed. Thus, the total number of species in Lara-Falcon Hill System is located in the same elevation range as Maracaibo Lake System. Similarly, this is observed in Orinoco Delta and Coastal Range, with 44 species distributed from 0 to 685 m.a.s.l. in Orinoco Delta and 45 species found below 685 m.a.s.l in the Coastal Range. However, although this analysis compared species number, it is necessary evaluate species similarity.

Figure 1. Principal component analysis of the environmental variables to evaluate heterogeneity among bioregions of Venezuela.

A. Blue: Coastal Range. Red: Foothills System B. Blue: Coastal. Red: Insular. of Guayana Massif

C. Blue: Orinoco Delta. Red: Foothills System D. Blue: Coastal Range. Red: Orinoco Delta. of Guayana Massif.

E. Blue: Lara-Falcon Hill System. Red: Maracaibo

Lake Basin.

Lower values (<10) of species richness are located in the Coastal, Insular, southwest of Amazonia and south of Guayana Massif; the greatest species richness (>40) of diurnal raptors in Venezuela is distributed along the mountain ranges, northeast of Amazonia and north and east of Guyana Shield (Table 4, Figure 2).

Figure 2. Distribution of Species Richness of Diurnal Raptors in Venezuela.

Table 3. Non-significant comparisons of the post hoc Dunnett’s T3 test of Bioregions. Significance value was set at 0.05. Bioregions: 1. Central Coastal Range; 2. Eastern Coastal Range; 3. Mountain range of Mérida; 4. Orinoco Delta; 5. Maracaibo Lake Basin; 6. Insular; 7. Coastal; 11. Mountain Range of Perijá; 12. Lara-Falcon Hill System; 13. Foothills System of Guayana Massif.

(I) Bioregion_ID (J) Bioregion_ID (I-J) Differences p

1 2 .2246 1.000 1 4 -.2153 1.000 1 13 .2650 1.000 2 4 -.4399 1.000 2 13 .0404 1.000 3 11 -.6528 1.000 4 13 .4803 .787 5 12 -.4998 .571 6 7 1.9253 .911

Table 4. Distribution of 65 species of diurnal raptors in Venezuela (based on species distribution models). Bioregions: 1. Central Coastal Range; 2. Eastern Coastal Range; 3. Mountain range of Mérida; 4. Orinoco Delta; 5. Maracaibo Lake Basin; 6. Insular; 7. Coastal; 8. Llanos; 9. Guayana Massif; 10. Amazonia; 11. Mountain Range of Perijá; 12. Lara-Falcon Hill System; 13. Foothills System of Guayana Massif.

Total Species 1 2 3 4 5 6 7 8 9 10 11 12 13 Bioregion/Sp. Cathartes aura x x x x x x x x x x x x x 13 Cathartes burrovianus x x x x x x x x x x x x 12

Cathartes melambrotus x x x x 4

Coragyps atratus x x x x x x x x x x x x x 13 Sarcoramphus papa x x x x x x x x x x x 11

Vultur gryphus x x 2

Pandion haliaetus x x x x x x x x x x x x x 13

Table 4. (continued)

Total Species 1 2 3 4 5 6 7 8 9 10 11 12 13 Bioregion/Sp. Elanus leucurus x x x x x x x x x x x 11

Gampsonyx swainsonii x x x x x x x x x x x x x 13 Chondrohierax uncinatus x x x x x x x x x x x x 12

Leptodon cayanensis x x x x x x x x x x x x 12

Elanoides forficatus x x x x x x x x x x x 11

Morphnus guianensis x x x x x x x x 8

Harpia harpyja x x x x x 5

Spizaetus tyrannus x x x x x x x x x x x 11

Spizaetus melanoleucus x x x x x x x 7

Spizaetus ornatus x x x x x x x x x x x 11

Spizaetus isidori x x x 3

Busarellus nigricollis x x x x x x x x x x x 11

Rostrhamus sociabilis x x x x x x x x x x x x x 13 Helicolestes hamatus x x x x x x 6

Harpagus bidentatus x x x x x x x x x x 10

Harpagus diodon x x x x 4

Ictinia plumbea x x x x x x x x x x x x 12

Circus buffoni x x x x x x x x x x x x 12

Accipiter poliogaster x x x x 4

Accipiter superciliosus x x x x x x 6

Accipiter collaris x x 2

Accipiter striatus x x x x x x x x x 9

Accipiter bicolor x x x x x x x x x x 10

Geranospiza caerulescens x x x x x x x x x x x x 12

Buteogallus schistaceus x x x 3

Buteogallus anthracinus x x x x x x x x x 9

Buteogallus aequinoctialis x x 2

Buteogallus meridionalis x x x x x x x x x x x x 12

Buteogallus urubitinga x x x x x x x x x x x x 12

Buteogallus solitarius x x x x x x 6

Rupornis magnirostris x x x x x x x x x x x x 12

Table 4. (continued)

Total Species 1 2 3 4 5 6 7 8 9 10 11 12 13 Bioregion/Sp. Parabuteo unicinctus x x x x x x x x x x x x 12

Parabuteo leucorrhous x x 2

Geranoaetus albicaudatus x x x x x x x x x x x x x 13 Geranoaetus melanoleucus x 1

Pseudastur albicollis x x x x x x x x x x x x 12

Leucopternis melanops x x x 3

Buteo nitidus x x x x x x x x x x x x 12

Buteo platypterus x x x x x x x x x 9

Buteo albigula x 1

Buteo brachyurus x x x x x x x x x x x x 12

Buteo swainsoni x 1

Buteo albonotatus x x x x x x x x x x x x 12

Herpetotheres cachinnans x x x x x x x x x x x x x 13 Micrastur ruficollis x x x x x x x 7

Micrastur gilvicollis x x x 3

Micrastur mirandollei x x x 3

Micrastur semitorquatus x x x x x x x x x x x x 12

Caracara cheriway x x x x x x x x x x x x x 13 Ibycter americanus x x x x x x x x x 9

Daptrius ater x x x x x x x x x 9

Milvago chimachima x x x x x x x x x x x x x 13 Falco sparverius x x x x x x x x x x x x x 13 Falco columbarius x x x x x x x x x x x 11

Falco rufigularis x x x x x x x x x x x x 12

Falco deiroleucus x x x 3

Falco femoralis x x x x x x x x x x x x 12

Falco peregrinus x x x x x x x x x x x x 12

Total Species/Bioregion 47 43 53 44 44 15 31 42 55 53 50 41 56

Environmental variables and species richness

The stepwise regression model incorporating 13 of 16 variables explained 45% of the variation in species richness (Table 5). Forest canopy height (forhei), land cover (land), annual mean temperature (bio_1) and slope emerged as the most important predictors of species richness. Characteristics of pixels with low (<10) and high (>40) species richness are shown in Table 6.

Table 5. Forward stepwise regression model of the environmental determinants of species richness. Significance values to enter and to leave the model were set at 0.05 and 0.1, respectively.

2 R = 0.45, F 13, 10613 = 689.302, p = 0.000

Variable Beta Standard Error t p

Constant 42.583 4.303 9.896 0.000 bio_19 -0.007 0.000 -25.823 0.000 bio_17 -0.041 0.002 -27.119 0.000 bio_15 -0.243 0.011 -23.118 0.000 forhei 0.143 0.010 14.879 0.000 bio_4 0.010 0.001 16.824 0.000 alt 0.009 0.001 9.555 0.000 bio_7 -0.168 0.008 -21.337 0.000 forest -0.028 0.003 -9.117 0.000 bio_18 -0.008 0.001 -10.198 0.000 bio_1 0.098 0.016 6.260 0.000 bio_12 0.001 0.000 5.884 0.000 land 0.104 0.025 4.180 0.000 slope 0.087 0.028 3.161 0.002

Table 6. Characteristics of areas (pixels, n) with low (<10) and high (>40) species richness (No. species). Mean (min-max) values of the four most important predictors of species richness are presented.

n No. species Forhei No. Land cover types Bio_1 Slope

1900 >40 21 (0-41) 12* 23.9 (2-28) 3.8 (0-27) 340 <10 25 (0-35) 13** 26.1 (18-28) 1 (0-15) *Open Schrublands and Deciduous Broadleaf Forest retain the greatest mean value of species richness

(ANOVA F 11, 1,888 = 8.6, p = 0). ** Mixed Forest, Evergreen Broad Leaf Forest and Grasslands hold the

greatest mean values of species richness (ANOVA F 12, 327 = 5.43, p = 0).

Conservation status assessment

Five species meet the B1 red list criterion (Table 2). Andean Condor meets the criteria to be classified as Endangered B1ab(i,iii,iv), Semicollared Hawk, White-throated Hawk (Buteo albigula) and Swainson’s Hawk (Buteo swainsoni) meet the criteria of Vulnerable B1ab(i,iii); and the Black-chested Buzzard-eagle (Geranoaetus melanoleucus) is classified as Near Threatened because it meets the B1 criteria and one of two required conditions. The remaining 60 species are classified as Least Concern.

The gap analysis revealed that the representation of species’ EOO in SPA varied from 5.5% (Harris’s Hawk, Parabuteo unicinctus) to 86.1% (Vultur gryphus) with a mean value of 19.5%. (Table 2). The size and numbers of SPA varied among bioregions, altogether the SPA cover 149,910 km2 corresponding to 16.3% of Venezuela; SPA cover a disproportionate percentage of bioregion areas (Table 7). The Foothills System of the Guyana Massif (the richest bioregion) is covered by SPA in only 0.9% of its extent without SPA in the northeast where the species hotspot of the region and the country is located (Figure 2); nonetheless, the Guyana Massif with 55 species is covered by SPA in 52.9% of its extent. The mountain ranges protection area averaged 22.9%. The remaining bioregion accounts for 1.6% to 26.5% surface protected.

Table 7. Coverage of strict protected area (SPA) in Bioregions of Venezuela

Bioregion No. Species Area, km2 SPA, km2 (%) Central Coastal Range 47 36,194 4,827 (13.3) Eastern Coastal Range 43 14,070 1,734 (12.3) Mountain Range of Merida 53 40,626 11,009 (27.1) Orinoco Delta 44 44,623 3,182 (7.1) Maracaibo Lake Basin 44 39,164 2,297 (5.9) Insular 15 810 213 (26.2) Coastal 31 5,061 966 (19.1) Llanos 42 250,600 12,831 (5.1) Guyana Massif 55 165,475 87,491 (52.9) Amazonia 53 100,833 20,635 (20.5) Mountain Range of Perijá 50 6,321 2,447 (38.7) Lara -Falcon Hill System 41 44,836 731 (1.6) Foothills of Guayana Massif 56 171,469 1,546 (0.9)

DISCUSSION

Species geographic distribution

Although there are 68 species of diurnal raptors in Venezuela, this study provides information on 65 species with confirmed occurrence records in the country, which represent the 95% and 70% diurnal raptors of Venezuela and South America (Ascanio et al. 2012, Remsen et al. 2013), respectively. The presence of Ictinia mississippiensis and Micrastur buckleyi is considered hypothetical as there are no confirmed records of these species in the country. The occurrence of Circus cyaneus is known from only one museum record (COP 65,704) from 1903 from the Mountain Range of Mérida (Hilty 2003); it is considered a vagrant species. Harpagus diodon has been considered a vagrant species (Ascanio et al. 2012) with only two voucher specimens (COP 41106, EBRG 59) from 1950; I included this species in the analysis because a recent confirmed record is available (Espinosa & Rodríguez 2012).

The geographic distribution of raptors in Venezuela has been poorly documented. Ornithological field guides (Hilty 2003, Restall et al. 2006) and electronic databases (Ridgley et al. 2007) constitute the main sources of information on this subject; nonetheless, as I stated 18

before, the available maps from these sources lack metadata and/or they are not georeferenced which make them useless for management and conservation purposes. This situation has been resolved – at the national scale – with the results of this work; 65 species distribution maps indicating the extension of occurrence and cartographic details have been produced. Furthermore, distribution area types and EOO were described and estimated for each species, respectively; this information is not available in field guides cited above.

There are no references on previous estimates of EOO of raptors species for Venezuela. Red Lists of Threatened Species are references of EOO given that EOO is used as evaluation criteria but in spite of this, the last edition of the Red List of Venezuelan Fauna (Rodríguez & Rojas-Suarez 2008) did not report estimates of EOO. The most complete study on the biology of a raptor species in Venezuela, the Harpy Eagle (Alvarez-Cordero 1996), includes an analysis of the geographic and ecological range of this species (Chapter 2) but it does not provide estimates of EOO. Other references on the geographic distribution of diurnal raptors in Venezuela are available in the form of distribution extensions (Barrowclough et al. 1997, Calchi & Viloria 1991, Rios et al. 2010, Hilty 1999, Kirwan & Sharpe 1999, Pérez-Emán et al. 2003, Sharpe et al. 2001, Zimmer & Hilty 1997, Naveda-Rodríguez & Bisbal 2008) and only document distribution points.

EOO estimates of raptor species in northern South America are available for only four species (Vultur gryphus, Buteogallus solitarius, Spizaetus isidori and Accipiter collaris) in Colombia (Renjifo et al. 2005). EOO estimates ranged between 0.3% and 26.7% with respect to EOO estimated for these species in this study. Although the methods employed included all the known localities of any species, I attribute this variation to two reasons: The methods used by Renjifo et al. (2005) are a modification of the minimum convex polygon method of the IUCN (2012). The method I used in this study (ecological niche modeling) did not consider the minimum convex polygon. Moreover, the heterogeneous environmental conditions (this includes human impact) of each geopolitical unit influence the niche variables and dispersal dynamics which shape the species geographic range boundaries.

Species richness patterns and environmental predictors

Macroecological studies determining bird species richness patterns and environmental factors predicting species richness have been developed at different spatial scales, with most studies performed at scales of lower resolution (>0.5° or >50 km minimum mapping unit, e.g. Cueto & Lopez de Casenave 1999, Rahbek & Graves 2001, Diniz-Filho et al. 2002, Meynard et al. 2004, Bellocq & Gómez-Insausti 2005, Ruggiero & Hawkins 2008, Ramirez & Telleria 2003, Rios-Muñoz & Navarro-Singüenza 2012) and very few fine-scale (<0.04165° or <5 km minimum mapping unit, e.g. Böhning-Gaese 1997, Lee et al. 2004, Koh et al. 2006, this study) studies. This is worth highlighting because richness patterns and environmental variable responses vary according to the scale, affecting conclusions on the drivers of patterns and processes.

The distribution of species richness of diurnal raptors in Venezuela is similar to other vertebrates within the country with highest richness located in the mountain ranges and the Guiana Shield and lower richness in the Coast and Insular regions (Aguilera et al. 2003). The species richness I found with the modeling method is consistent with species numbers reported by Thiollay (1996), Alvarez et al. (1996) and Jensen et al. (2005) in the central Mountain Range of Mérida, Imataca and southern Llanos, respectively, indicating a good performance of the models. As a functional group, diurnal and nocturnal (Strigiformes) raptors share an analogous pattern of distribution in species richness in Venezuela (Naveda-Rodríguez & Torres in press), suggesting a possible response to the same environmental variables driving species richness which could be interpreted as a convergence in patterns and processes among these two distinct taxonomic groups (Carnicer & Díaz-Delgado 2008).

Although landscape heterogeneity did not have effects on carnivore bird species richness in North America at a large scale (Carnicer & Díaz-Delgado 2008), Rahbek & Graves (2001) found it to be an important predictor of bird species richness in South America. The differences in raptors species number observed among bioregions in Venezuela can be attributed to the environmental heterogeneity, which has been described as a determinant of neotropical raptor species diversity at both coarse and fine-scale resolution (Jullien & Thiollay 1996, Anderson 2001, Diniz-Filho et al. 2002).

Huber & Alarcón (1988) defined the bioregions of Venezuela based on floristic, edaphic and climatic traits, recognizing four regions with similar characteristics: the Central Coastal Range and Eastern Coastal range are considered subregions of a major region denominated Coastal Range, the same is observed with Mountain Range of Mérida and Mountain Range of Perijá which are subregions of the Venezuelan Andes; these subregions are separated by the Unare depression and Táchira depression in the Coastal Range and the Venezuelan Andes, respectively (Meier 2011, Schargel 2011), therefore, the species richness is not expected to be different among subregions.

The similarities of environmental conditions among the regions: Coastal Ranges and Foothills System of Guayana Massif (Figure 1A); Insular and Coastal (Figure 1B); and Orinoco Delta and Foothills System of Guayana Massif (Figure 1C), would explain the similarity in mean number of species among these bioregions. Moreover, Coastal Ranges and Orinoco delta (Figure 1D) as well as Maracaibo Lake Basin and Lara Falcon Hill System (Figure 1E) show partial similarity in respect with the environmental variables values, the dissimilarities between this pair-compared bioregions are defined by the mountain areas of Sierra de San Luis in Lara-Falcon Hill Systems and Coastal Range.

As mentioned before, the highest values of species richness are distributed along mountainous regions of the country, this tendency was expected since the patterns of mountains harboring large number of species has been documented (Lomolino 2001, Rahbek and Graves 2001, Ruggiero and Hawkins 2008, Davies et al. 2008); this pattern is associated to climatic and topographic feature gradients (see Ruggiero & Hawkins 2008, and references therein).

With more than 100 hypotheses proposed to explain species richness patterns (Palmer 1994), species richness of neotropical birds of prey have been contrasted with the ambient- energy hypothesis, the productivity hypothesis, the mid-domain hypothesis and the environmental heterogeneity hypothesis (Bellocq & Gómez-Insausti 2005, Diniz-Filho et al. 2002, Meynard et al. 2004, Rangel & Diniz-Filho2004). In Venezuela, diurnal raptors species richness was linked to three landscape features (forest canopy height, land cover type and slope) and one climatic variable (mean annual temperature). Forest canopy height and land cover types were the best environmental predictors of raptor species richness. This is consistent with the

hypothesis of environmental heterogeneity, which asserts that the greater the heterogeneity the greater the species richness (Houston 1979) and has been found to be an important predictor of raptor species richness at coarse spatial scales (Bellocq & Gómez-Insausti 2005, Diniz-Filho et al. 2002, Meynard et al. 2004). At true fine-scale resolution, habitat heterogeneity is an important predictor of bird species richness (Böhning-Gaese 1997, Anderson 2001, Lee et al. 2004, Koh et al. 2006). The references cited used elevation as a measure of habitat or landscape heterogeneity.

Although elevation seems to have no effect on the overall raptor species richness in Venezuela, Thiollay (1996) and Meynard et al. (2004) found a negative effect of this variable on the species richness of birds of prey in northern and southern South America, respectively; I attribute this inconsistency to differences in the methodological approach, the geographical extent of the analysis, and the scale of analysis.

Other variables are used as measures of landscape heterogeneity (vegetation structure, Ott 2007, topographic roughness, Ruggiero & Hawkins 2008 and references therein), but they have been used alone, and depending on the scale of analysis, they did not necessarily describe spatial heterogeneity. Rather than using one variable, this study used seven variables considered landscape features, one corresponding to land cover type, two related to vegetation structure and four to ground description, which is more informative on landscape heterogeneity; from this number, one variable of each group was the most important predictor of species richness. Altogether, the three landscape variables selected in the regression model constitute 73% of the total variability explained (45%). This supports my previous assumption on landscape heterogeneity and diurnal raptors richness in Venezuela.

In Honduras, Anderson (2001) found a positive correlation between diurnal raptor species richness and habitat heterogeneity when studying this relationship at very fine-scale resolution (1:50,000). Anderson measured habitat heterogeneity using Shannon index and identified five habitats types describing the core forest and grass cover as the most important in explaining species richness. In comparison, the land cover variable I used here included 17 cover types. The open scrublands and deciduous broadleaf forest were detected to be the richest classes.

Mean annual temperature (bio_1, MAT) had influences on raptor species richness in southern South America. In Argentina, species richness increases with temperature (Bellocq & Gómez-Insausti 2005), while in Chile this relationship was negative (Meynard et al. 2004). In Venezuela, bio_1 is not correlated more strongly with species richness, explaining 9.8% of the variability of the dependent variable. Rahbek & Graves (2001) identified MAT in only one of 10 regression models developed to explain bird species richness in South America, but it was not an important variable. As with elevation, I found that the inconsistences in variable responses are an artifact of experimental design.

Conservation status assessment

Previous conservation assessments of diurnal raptors in Venezuela lacked expert knowledge. The Red List of Venezuelan Fauna (Rodríguez & Rojas-Suarez 1999, 2008) and the IUCN Red List of Threatened Species (IUCN 2013) did not provide detailed information on arguments used when selecting the criteria for the evaluation of raptor species. This lack of sound scientific evidence leads the evaluation process to a wrong estimation of extinction risk, rendering it useless. The knowledge of species’ natural history is crucial for the development of an accurate conservation assessment, as well as effective conservation actions (Bierregaard 1998). Certainly, little is known about the biology and ecology for almost all raptor species in Venezuela, particularly information regarding geographic distribution and population density.

The geographic distribution, in the form of extent of occurrence (EOO, criterion B1) and area of occupancy (AOO, criterion B2), is one of the five criteria used to evaluate conservation status (IUCN 2012). Based on my preliminary assessment, I recommend to the Red List authorities to change the status of Andean Condor, Semicollared Hawk, White-throated Hawk, Swainson’s Hawk and Black-chested Buzzard-eagle Harpy Eagle (Harpia harpyja), Crested Eagle (Morphnus guianensis), Solitary Eagle (Buteogallus solitarius) and Black-and-chestnut Eagle (Spizaetus isidori) in Venezuela from the current status in Rodríguez & Rojas-Suarez (2008) and IUCN (2013) to that proposed in this study (Table 8).

The changes proposed are based on the criterion B1; nevertheless, the conservation status assessment needs to consider the other IUCN criteria. I am evaluating the use of other MAXENT thresholds to determine the AOO which is defined as “the smallest area essential at any stage to

the survival of the taxon”. This has been considered as highly suitable habitat that would be estimated with the 10 percentile training presence logistic threshold or maximum training sensitivity plus specificity threshold in MAXENT (Renjifo et al. 2002, Nazeri et al. 2012, Trisurat et al. 2013).

Table 8. Current and proposed Red List Status of diurnal birds of prey in Venezuela. CR=Critically Endangered,, EN=Endangered, VU=Vulnerable, NT=Near Threatened, LC=Least Concern, DD=Data Deficient, NE=Not Evaluated.

Red List Status and Criteria Used Species IUCN 2013 Rodríguez & Rojas Suarez 2008 This study Andean Condor CR EN NT Vultur gryphus D1 B1ab(i,iii,iv) Semicollared Hawk VU NT NT Accipiter collaris B1ab(i,iii) White-throated Hawk Vulnerable LC DD Buteo albigula B1ab(i,iii) Swainson’s Hawk VU LC NE Buteo swainsoni B1ab(i,iii) Black-chested Buzzard-eagle LC NE LC Geranoaetus melanoleucus Harpy Eagle NT VU C2a(ii); D1 LC Harpia harpyja Crested Eagle NT VU C2a(ii); D1 LC Morphnus guianensis Solitary Eagle LC NT LC Buteogallus solitarius Black-and-chestnut Eagle VU NT LC Spizaetus isidoris C2a(i)

From the gap analysis I conclude that the establishment of strict protected area (SPA) was not a planned activity in Venezuela; although it is an act covered in the Forest Law and Wildlife Protection Law (República de Venezuela 1970, República Bolivariana de Venezuela 2013), the main reason for the creation of National Parks and Natural Monuments used to be the protection of soils and waters. The richest region located south of the Orinoco River (Foothills of Guayana

Massif) is the least protected – in fact it is protected in only 0.9% of its extent. The Rio Grande area (in Imataca) holds around 42 species of diurnal raptors (Alvarez et al. 1996) and more than 40 Harpy Eagle nests (Alvarez-Cordero, 1996), but this natural heritage is not protected; instead of a SPA, a forest reserve with serious problems of illegal logging and mining covers the area.

Fortunately, the montane regions of northern Venezuela have a different situation; these mountains represent 10% of the national territory and they are protected in 20% of their extent. In spite of this, areas with high values of species richness (>41) are best represented – by chance – in the SPA system. Mountains of northern Venezuela, specifically the Coastal Range of Mérida and Central and Eastern Coastal Ranges retain the greatest number of protected areas in contrast with other bioregions of the country, but these large numbers of SPA are not well connected. A priority for bird conservation at landscape scale is the maintenance of connectivity between protected areas (Tobias et al. 2013).

Rodríguez et al. (2004) suggested that northern Venezuela should receive priority conservation attention because a number of threatened bird species inhabit the region; moreover, this analysis is skewed to species listed as threatened, excluding an important number of unthreatened rare species. Although species rareness is often controversial due to relationships of abundance and distribution, Van Auken (1997) considers a species to be rare if it has a limited distribution. This is the case of Rufous Crab-Hawk (Buteogallus schistaceus), Black-Faced Hawk (Leucopternis melanops), Lined Forest-Falcon (Micrastur gilvicollis), Slaty-backed Forest-Falcon (Micrastur mirandollei) and Orange-breasted Falcon (Falco deiroleucus), with distribution restricted to the south of Venezuela.

Forest-Falcon (Micrastur mirandollei) and Orange-breasted Falcon (Falco deiroleucus), with distribution restricted to the south of Venezuela.

On the other hand, areas with 31-40 species are barely covered with SPA. The distribution of areas with that number of species matches the distribution of tropical dry forest. This ecosystem is mainly distributed in the Llanos and is considered the most threatened ecosystem in Venezuela (Fajardo et al. 2005). Only 5% of the Llanos (Venezuela’s largest bioregion) is protected, and its biggest SPA provides protection for less than 50% of diurnal raptor species in Venezuela.

In theory, strictly protected areas are assuming an effective role in the protection of species’ EOO, with 20% of species’ EOO protected on average. Seven species have low percentage (<10%) of EOO protected. This could be critical for Buteogallus schistaceus, whose 36,889 km2 EOO is only 6% protected. Conservation goals proposed for biodiversity protection range from 10% to 70% of species habitats (Pearce et al. 2008) or 30% of the species distribution (Douglass et al. 2011). For bird species, Pearce et al. (2008) proposed a conservation goal of 40% and 20% of restricted and wide distribution, respectively. 57% of the species with restricted distribution have less than 40% of EOO protected. Andean Condor and Semicollared Hawk, which have a restricted distribution and are classified as Endangered and Vulnerable, respectively, have more than 50% of their EOO under protection. A large number of species (35) with wide distribution in the country have more than 80% of their distribution unprotected.

Final thoughts and suggestions

Seventy percent of neotropical diurnal raptor species are found in Venezuela. As in other areas of South America, raptor diversity is influenced by environmental heterogeneity; however, this has been described primarily at coarse-scale resolution. Because there is a strong influence of spatial scales in variable responses, we need to standardize methods that allow for higher resolution results, including a precise definition of variables measuring environmental heterogeneity. For instance, cartographic scales for the analysis of macroecological patterns and processes could be predefined, so that biogeographers would recognize the minimum mapping scale required for studies of varying spatial extent. The method proposed by IUCN (2012) to

define EOO must be revised because it incurs commission/omission errors. With the available technology and GIS data, ecological niche modeling would be the best current option.

Raptor conservation in Venezuela must be thoroughly planned. Criteria used to create a natural protected area must go beyond water or soil conservation. In the same context, territorial re-ordering is urgent to improve the protection of birds of prey. Priorities for species conservation must be guided by multicriteria, not only threatened species or endemism; species rareness and commonness as well as a complementarity analysis will provide better results when proposing priority geographical areas for conservation. Raptor conservationists should remember early (1934) advice from Rosalie Edge – “The time to protect a species is while it is still common”.

LITERATURE CITED

Aguilera, M., Azocar, A. & González, E. (2003) Biodiversidad en Venezuela. Fundación Polar, Ministerio de Ciencia y Tecnología. Caracas.

Alvarez, E., Ellis, E., Smith, D. & Larue, C. (1996) Diurnal raptors in the fragmented rain forest of the Sierra Imataca, Venezuela. Raptors in human landscapes (ed. D. Bird, D. Varland and J. Negro), pp. 263-273. Academic Press Inc., San Diego.

Alvarez-Cordero, E. 1996. Biology and conservation of the harpy eagle in Venezuela and Panama. Ph.D. Thesis. University of Florida, Gainesville.

Anderson, D. (2001) Landscape heterogeneity and diurnal raptor diversity in Honduras: the role of indigenous shifting cultivation. Biotropica, 33, 511-519.

Ascanio, D., Miranda, J., León, J.G., Marantz, C., Crease, T., Kvarnbäck, J. & Rodríguez, G. (2012) Species lists of birds for South American countries and territories: Venezuela. Version 14/June/2012. http://www.museum.lsu.edu/~Remsen/SACCCountryLists.html

Balgooyen, T. G. (1989) Natural history of the American kestrel in Venezuela. Journal of Raptor Research, 23, 85-93.

Barrowclough, G., Lentino, M. & Sweet, P (1997) New records of bird from Auyantepui, Estado Bolivar, Venezuela. Bulletin of the British Ornithologist’s Club, 117, 194-198.

Beissinger, S., Thomas, B. & Strahl, S. (1988) Vocalization, food habits, and nesting biology of the Slender-billed Kite with comparisons of the Snail Kite. The Wilson Bulletin, 100, 604- 616.

Bellocq, M., & Gómez-Insausti, R. (2005) Raptorial birds and environmental gradients in the southern Neotropics: A test of species-richness hypotheses. Austral Ecology, 30, 892– 898.

Bildstein, K., Scheslsky, W., Zalles, J. & Ellis. (1998). Conservation status of tropical raptors. Journal of Raptor Research, 32, 3-18,

Bierregaard, R. (1998) Conservation status of birds of prey in the South American tropics. Journal of Raptor Research, 32, 19-27.

Böning-Gaese, K. (1997) Determinants of avian species richness at different spatial scales. Journal of Biogeography, 24, 49-60.

Calchi, R. & Viloria, A. L. (1991) Occurrence of the Andean condor in the Perija Mountains of Venezuela. The Wilson Bulletin, 103, 720-722.

Carnicer, J. & Díaz-Delgado, R. (2008) Geographic differences between functional groups in patterns of bird species richness in North America. Acta Oecologica, 33, 253-264.

Cueto, V. & Lopez de Casenave, J. (1999) Determinants of bird species richness: role of climate and vegetation structure at a regional scale. Journal of Biogeography, 26, 487-492.

Davies, R., C. Orme, D. Storch, V. Olson, G. Thomas, S. Ross, T. Ding, P. Rasmussen, P. Bennett, I. Owens, T. Blackburn & K. Gaston. 2008. Topography, energy and the global distribution of bird species richness. Proceedings of The Royal Society B, 274, 1189– 1197.

Diniz-Filho, J., De Sant'ana, C., De Souza, M. & Rangel., T. (2002) Null models and spatial patterns of species richness in South American birds of prey. Ecology Letters, 5, 47-55.

Espinosa, A., & Rodríguez. G. A. (2012) Rapaces de Venezuela. Intenso Offset, Caracas.

ESRI. 2008. ArcGIS, version 9.3. Environmental Systems Research Institute, Redlands, C.A.

Fajardo, L., González, V., Nassar, J.M., Lacabana, P., Portillo, C., Carrasquel, F. & Rodríguez, J.P. (2005) Tropical dry forest of Venezuela: characterization and current conservation status. Biotropica, 37, 531-546.

Ferguson-Lees, J. & Christie, D.A. (2001) Raptors of the World. Christopher Helm, London.

García-Moreno, J., Clay, R. P. & Ríos-Muñoz, C.A. (2007) The importance of birds for conservation in the Neotropical region. Journal of Ornithology, 148, S321-S326.

Hijmans, R., Cameron, S., Parra, J., Jones, P. & Jarvis, A (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965–1978.

Hilty, S. L. (1999) Three bird species new to Venezuela and notes on the behaviour and new distribution of other poorly known species. Bulletin of the British Ornithologist’s Club, 119, 220-235.

Hilty, S.L. (2003) Birds of Venezuela. Princeton University Press, New Jersey.

Houston, M. (1979) A general hypothesis of species diversity. The American Naturalist, 113, 81- 101.

Huber, O. & Alarcón, C. (1988) Mapa de vegetación de Venezuela. Ministerio del Ambiente y de los Recursos Naturales Renovables, The Nature Conservancy, Fundación Bioma, Caracas.

IBM (2010) IBM SPSS Statistics for Windows, Version 19.0. IBM Corporation, Armonk, NY.

IUCN (2012) IUCN red list categories and criteria, version 3.1. Second edition. IUCN, Gland, Switzerland and Cambridge, UK..

IUCN (2013) IUCN red list of threatened species, version 2013.1. . Downloaded on 04 September 2013.

Jarvis, A., Reuter, H.I., Nelson, A. & Guevara, E. (2008) Hole-filled seamless SRTM data, version 4. International Centre for Tropical Agriculture (CIAT).

Jenness, J. 2013. DEM Surface Tools. Jenness Enterprises, Flagstaff, AZ. Available at: http://www.jennessent.com/arcgis/surface_area.htm.

Jensen, W., Gregory, M., Baldassare, G., Vilella, F. & Bildstein, K. L. (2005) Raptor abundance and distribution in the llanos wetlands of Venezuela. Journal of Raptor Research, 39, 417-428.

Jullien, M., & Thiollay, J. M. (1996) Effects of rain forest disturbance and fragmentation: comparative changes of the raptor community along natural and human-made gradients in French Guiana. Journal of Biogeopgraphy, 23, 7-25.

Kirwan, G. & Sharpe, C. (1999) Range extensions and notes on the status of little-known species from Venezuela. Bulletin of the British Ornithologist’s Club, 119, 38-47.

Koh, C., Lee, P. & Lin, R. (2006) Bird species richness patterns of northern Taiwan: primary productivity, human population density, and habitat heterogeneity. Diversity and Distributions, 12, 546–554.

König, K. & Weick, F. (2008) Owls. A guide to the Owls of the World. Second Edition. Christopher Helm, London.

Lee, P., Ding, T., Hsu, F. & Geng, S. (2004) Breeding bird species richness in Taiwan: distribution on gradients of elevation, primary productivity and urbanization. Journal of Biogeography, 31, 307–314.

Legendre, P. (1993) Spatial autocorrelation: trouble or new paradigm?. Ecology, 74, 1659-1673.

Lentino, M. (2003) Aves. Biodiversidad en Venezuela (ed. M. Aguilera, A. Azocar & E. González), pp. 610-648. Fundación Polar, Ministerio de Ciencia y Tecnología, Caracas.

Linares, O. (1998). Mamíferos de Venezuela. Sociedad Conservacionista Audubon de Venezuela, Caracas.

Lomolino, M. (2001) Elevation gradients of species-density: historical and prospective views. Global Ecology & Biogeography, 10, 3-13

Mader, W. (1982) Ecology and breeding habits of the Savanna Hawk in the Llanos of Venezuela. The Condor, 84, 261-271.

MARN (2000) Primer informe de país para la convención sobre diversidad biológica. Ministerio del Ambiente y de los Recursos Naturales, Caracas.

Meier, W. (2011) Los bosques nublados de la cordillera de la costa en Venezuela. Biollania, 10, 106-121.

Meynard C., Samaniego H. & Marquet, P. (2004) Biogeografía de aves rapaces de Chile. Aves rapaces de Chile (ed. A. Muñoz, J. Rau & J. Yánez), pp. 129-143. Ediciones CEA, Valdivia.

Mittermeier, R. A., Robles, P. & Mittermeier, C.G. (1997) Megadiversidad: los países biológicamente más ricos del mundo. CEMEX, Ciudad de México.

NASA Land Processes Distributed Active Archive Center (LP DAAC) (2013) MODIS MCD12QQ. USGS/Earth Resources Observation and Science Center, Sioux Falls, South Dakota.

Naveda-Rodríguez, A. & Bisbal, F. (2008). Avifauna of Dinira National Park, Venezuela. Check List, 4, 373-381.

Naveda-Rodríguez-A. & Torres, D.A. (En prensa). Los Búhos en Venezuela. Los Búhos Neotropicales: Diversidad y Conservación (ed. P. Enríquez), pp-578-593. El Colegio de la Frontera Sur, San Cristóbal de las Casas.

Nazeri M., Jusoff, K., Madani, N., Mahmud, A., Bahman, A., & Kumar, K. (2012) Predictive Modeling and Mapping of Malayan Sun Bear (Helarctos malayanus) distribution uing maximum entropy. PLoS ONE, 7, e48104.

Ojasti, J. (2000) Manejo de Fauna Neotropical. Simab Series No. 5. Smithsonian Institution, Washington.

Ott, T. (2007) Landscape heterogeneity as a determinant of range size utilization by African elephants (Loxodonta africana) in mesic savannas. MSc Thesis. University of Pretoria, Pretoria.

Palmer, M. (1994) Variation in species richness: towards unification of hypotheses. Folia Geobotanica et Phytotaxonomica, 29, 511-530.

PDVSA (1992) Imagen de Venezuela, una visión espacial. Petróleos de Venezuela, Caracas.

Pérez-Emán, J. Sharpe, C., Lentino, M., Prum, R. & Carreño, I. (2003) New records of bird from the summit of Cerro Guaiquinima, Estado Bolivar, Venezuela. Bulletin of the British Ornithologist’s Club, 123, 78-90.

Peterson, T. (2001) Predicting species’ geographic distributions based on ecological niche modeling. The Condor, 103, 599-605.

Phillips, S., Anderson, R. & Schapire, R. (2006) Maximum entropy modeling of species geographic distributions. Ecological Modeling, 190, 231–259.

Rahbek, C., & Graves, G. (2001) Multiscale assessment of patterns of avian species richness. Proceedings of the National Academy of Sciences, 98, 4534-4539.

Ramirez, A. & Telleria, J. L. (2003). Efectos geográficos y ambientales sobre la distribución de las aves forestales ibéricas. Graellsia, 59, 219-231.

Rangel, T. & Diniz-Filho, J. (2004) Worldwide patterns in species richness of Falconiformes: analytical null models, geometric constraints, and the mid-domain effect. Brazilian Journal of Biology, 64, 299-308.

Remsen, J. V., Jr., Cadena, C. D., Jaramillo, A. Nores, M., Pacheco, J. F., Pérez-Emán, J., Robbins, M. B., Stiles, F. G., Stotz, D. F. & Zimmer, K. J. (2013) A classification of the bird species of South America. American Ornithologists' Union. http://www.museum.lsu.edu/~Remsen/SACCBaseline.html.

Renjifo, L. M., Franco-Maya, A. M., Amaya-Espinel, J. D., Kattan, G. H. & López-Lanús, B. (2002) Libro rojo de aves de Colombia. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt y Ministerio del Medio Ambiente. Bogotá.

República Bolivariana de Venezuela. 2013. Ley de Bosques Gaceta Oficial N° 40,222. Caracas 6 de agosto de 2013.

República de Venezuela. 1970. Ley de Protección a la Fauna Silvestre. Gaceta Oficial N° 29,289. Caracas 11 de agosto de 1970.

Restall, R., Rodner, C. & Lentino, M. (2006) Birds of Northern South America, and identification guide. Volume 1: species accounts. Yale University Press, New Haven, USA and London, UK.

Ridgely, R. S., Allnutt, T. F., Brooks, T., McNicol, D. K., Mehlman, D. W., Young, B. E. & Zook, J. R. (2007) Digital Distribution Maps of the Birds of the Western Hemisphere, version 3.0. Nature Serve, Arlington, VA.

Ríos, G. A., Serrano, J., Canelon, M. J., Briceño, E. & Vargas, J. J. (2010) Primer registro del Aguila Crestada Morphnus guianensis en los llanos occidentales de Venezuela. Cotinga, 32, 160-161.

Rios-Muñoz, C. & Navarro-Singüenza, A. (2012). Patterns of species richness and biogeographic regionalization of the avifaunas of the seasonally dry tropical forest in Mesoamerica. Studies on Neotropical Fauna and Environment, 47, 171–182.

Rodríguez, J. P. & Rojas-Suárez, F. (1998) Las áreas protegidas estrictas y la conservación de la fauna venezolana amenazada. Acta Científica Venezolana, 49, 173-178.

Rodríguez, J. P. & Rojas-Suárez, F. (1999) Libro Rojo de la Fauna Venezolana, Segunda Edición. Provita, Fundación Polar, Caracas.

Rodríguez, J. P. & Rojas-Suárez, F. (2008) Libro Rojo de la Fauna Venezolana, Tercera Edición. Provita, Shell de Venezuela, S.A., Caracas.

Rodríguez, J. P., Rojas-Suárez, F. & Sharpe, C. (2004) Setting priorities for the conservation of Venezuela’s threatened birds. Oryx, 38, 373-382.

Rodríguez, J. P., Lazo, R., Solórzano, L.A. & Rojas-Suárez, F. (2005). Cartografía Digital Básica de las Áreas Naturales Protegidas de Venezuela: Parques Nacionales, Monumentos Naturales, Refugios de Fauna, Reservas de Fauna y Reservas de Biósfera. Centro Internacional de Ecología Tropical, Instituto Venezolano de Investigaciones Científicas, Conservación Internacional Venezuela, UNESCO y Oficina Nacional de Diversidad Biológica del Ministerio del Ambiente y de los Recursos Naturales, Caracas.

Ruggiero, A. & Hawkins, B. (2008. Why do mountains support so many species of birds?. Ecography, 31, 306-315.

Schargel, R. (2011) Una reseña de la geografía física de Venezuela, con enfasis en los suelos. Biollania, 10, 11-26.

Scott, J., Davis, F., Csuti, B., Noss, R., Butterﬁeld, B., Groves, C., Anderson, H., Caicco, S., D’Ericha, F., Edwards, T., Ullman J. & Wright, R. (1993) Gap analysis: a geographic approach to the protection of biological diversity. Wildlife Monograph, 23, 1-41.

Sharpe, C., Ascanio-Echeverría, D. & Rodríguez, G. (2001) Further range extensions and noteworthy records for Venezuelan birds. Bulletin of the British Ornithologist’s Club, 121, 50-62.

Simard, M., Pinto, N., Fisher, J., Baccini, A. (2011) Mapping forest canopy height globally with spaceborne lidar. Journal of Geophysical Research, 116, 1-12.

Sullivan, B.L., Wood, C. L., Iliff, M. J., Bonney, R. E., Fink, D. & Kelling, S. (2009) eBird: a citizen-based bird observation network in the biological sciences. Biological Conservation, 142, 2282-2292.

Thiollay, J. (1996) Distributional patterns of raptors along altitudinal gradients in the northern Andes and effects of forest fragmentation. Journal of Tropical Ecology, 12, 535-560.

Tobias, J., Şekercioğlu, Ç. & Vargas, H. (2013) Bird conservation in tropical ecosystems. Key Topics in Conservation Biology 2 (ed D. W. Macdonald and K. J. Willis), pp. 258-276. John Wiley & Sons, Oxford.

Townshend, J.R.G., Carroll, M., Dimiceli, C., Sohlberg, R., Hansen, M. & DeFries, R. (2011) Vegetation Continuous Fields MOD44B, 2010 Percent Tree Cover, Collection 5, University of Maryland, College Park, Maryland,

Trisurat, Y., Chimchomea, V., Pattanaviboola, A., Jinamoya, S., Thongareea, S., Kanchanasakhaa, B., Simcharoena, S., Sribuaroda, K., Mahannopa, N. & Poonswad P. (2013) An assessment of the distribution and conservation status of hornbill species in Thailand. Oryx, 47, 441–450.

Van Auken, O. (1997) Species rareness and commonness along spatial and temporal gradients. The Southwestern Naturalist, 42, 369-374.

Velásquez, J., Ochoa, G., Oballos, J., Manrique, J. & Santiago, J. (2004) Metodologia para la delineacion cartografica de suelos. Revista Forestal Latinoamericana, 36, 15-34.

Zimmer, K. & Hilty, S. L. (1997) Avifauna of a locality in the upper Orinoco drainage of Amazonas, Venezuela. Ornithological Monographs, 48, 865-885.