Bird communities in tropical agroforestry ecosystems: an underappreciated conservation resourceby Skye M. Greenler, James J. Ebersole

Agroforestry Systems


Agronomy and Crop Science / Forestry


Bird communities in tropical agroforestry ecosystems: an underappreciated conservation resource

Skye M. Greenler • James J. Ebersole

Received: 27 July 2014 / Accepted: 31 March 2015  Springer Science+Business Media Dordrecht 2015

Abstract Tropical conservation and research focus primarily on protected areas and often ignore conservation values of surrounding agricultural lands. Understanding how species utilize agricultural land will maximize conservation efforts. We compared bird community composition in four habitats in northeastern Costa Rica: shade-grown cacao, live fences, riparian forest buffers, and preserved late-successional rainforest. Point counts over 2 months found 167 species from 36 families. Rainforest contained the most species unique to a habitat although richness per point was lower than in agricultural habitats. Half, 31, of the rainforest species did not occur in other habitats, while 106 species, mostly those preferring open areas, occurred in agroforest habitats but not rainforest.

While agricultural habitats had fairly similar species composition to each other as determined by distance in an ordination, each also contained significant numbers, 9–30, of unique species. While intact rainforest remains central to conservation of tropical birds, agricultural lands with substantial trees, e.g., live fences, riparian buffers, and plantations with shade trees, can support a high richness of birds. These avian communities are not simply subsets of the rainforest species but include substantial numbers of unique species. Conservation contributions of these lands to species richness and complexity should be considered in conservation, and trees in these habitats preserved.

Keywords Agroforest  Anthropogenic disturbance 

Avian guilds  Costa Rica  Matrix effects 



For many years deforested agricultural land was not considered to have conservation value, but recent research shows that it may hold substantial biodiversity (Donald and Evans 2006). Understanding this conservation resource can help manage these lands effectively to maximize their conservation values as complements to the central importance of preserved areas (McIntyre and Hobbs 1999; Donald and Evans 2006; Koh and Ghazoul 2010; Mendenhall et al. 2012). In addition to direct value as habitat, the agricultural matrix around a preserved forest also influences preserve health, fragment recolonization, and bird movement (Stouffer and Bierregaard 1995;

Electronic supplementary material The online version of this article (doi:10.1007/s10457-015-9805-y) contains supplementary material, which is available to authorized users.

S. M. Greenler (&)  James J.Ebersole

Department of Organismic Biology and Ecology,

Colorado College, 14 E. Cache la Poudre St.,

Colorado Springs, CO 80903, USA e-mail:; 123

Agroforest Syst

DOI 10.1007/s10457-015-9805-y

S¸ekerciog¯lu et al. 2002; Sigel et al. 2006; Laurance et al. 2012; Muhamad et al. 2013). However, only a few studies in Mexico (Estrada et al. 1997; Greenberg et al. 2000; Estrada and Coates-Estrada 2005),

Nicaragua (Harvey et al. 2006), and Indonesia (Muhamad et al. 2013) have investigated differences in avian community composition among several agricultural habitats. To improve conservation management, we need to better understand bird communities of tropical agricultural lands and how they vary among different agricultural systems.

South and Central America are key locations to study the conservation value of agricultural land because this region loses approximately 4.4 million ha of forest cover annually and is a center of current extinctions (Wilson 1988; Bradshaw et al. 2009;

Gardner et al. 2009; Eva et al. 2012). Deforestation has left Costa Rica with many discontinuous forest fragments surrounded by a complex agricultural mosaic (Bierregaard et al. 1992; Sigel et al. 2006).

Fragmentation strongly affects Neotropical birds, which are generally specialized and use narrow microhabitats (Stouffer and Bierregaard 1995; Lindell et al. 2007; Ruiz-Guerra et al. 2012).

Island biogeography theory (MacArthur and Wilson 1967) has been used to describe species richness following forest fragmentation for decades (Mendenhall et al. 2014). However, it does not adequately describe the community composition following forest fragmentation because it accounts only for patch size and isolation and assumes the matrix is incapable of sustaining native species (Laurance 2008). Assuming a barren matrix and that species cannot adapt to altered habitat limits the conservation contributions of these areas (McIntyre and Hobbs 1999; Prevedello and

Vieira 2010; Mendenhall et al. 2012).

Several types of tropical agroforest systems, including live fences, riparian forests, and shade-grown crops, have been touted to benefit avian conservation and functionally extend adjacent preserves (Van Bael et al. 2007; Seaman and Schulze 2010). Although not managed for conservation, these highly modified habitats may support some bird species (Gardner et al. 2009;

Prevedello and Vieira 2010; Laurance et al. 2012;

Kumaraswamy and Kunte 2013).

In this study we present data on bird community composition in a preserved, late successional rainforest; an organic, shade-grown cacao plantation; riparian forests; and live fence systems in northeastern Costa

Rica. We use these data to ask: (1) How do bird species assemblages differ among these agricultural habitats and the late-successional rainforest? (2) What families and guilds drive the differences in species assemblage among these habitats? Through these questions we explore the utility of the several community composition models in this system, address the conservation value of the existing agroforest ecosystem matrix, and make concrete conservation suggestions.


Study site

This study was completed in 10 km2 landscape of mixed rainforest, pasture, agroforest, and monoculture in the Caribbean lowlands of northeastern Costa Rica (102000N, 832000W) (Fig. 1). The area is approximately 35 m above sea level and is in the premontane wet forest life zone (Holdridge 1967).