Forest Patches in Tropical Landscapes

John Schelhas, Russell S. Greenberg, John Schelhas, Russell S. Greenberg

• Publisher: Island Press
• Year: 2013
• Print ISBN:
  9781559634267
  9781559634250
• eBook ISBN:
  9781610912693
• Topic: Biological Sciences
• Subtopic: Agriculture
• Index: Biological Sciences

While tropical forests are being cleared at an alarming rate, the clearing is rarely complete and is often not permanent. A considerable amount of tropical forest exists as remnants that have significant value both for the conservation of biological diversity and for meeting the needs of local people.

This volume brings together world-renowned scientists and conservationists to address the biological and socio-economic value of forest remnants and to examine practical efforts to conserve those remnants. An outgrowth of a year-long study by the policy program at the Smithsonian Migratory Bird Center, Forest Patches in Tropical Landscapes provides a broad overview of theory and practice, and will help foster both interdisciplinary research and more effective approaches to tropical conservation and development.

Part I

Changing Forests

Chapter 1

Preservation and Management of Biodiversity in Fragmented Landscapes in the Colombian Andes

Gustavo H. Kattan and Humberto Alvarez-López

Introduction

Land use in the Colombian Andes has greatly altered natural ecosystems. Up to 85 percent of the area in premontane and montane forests has been modified to some extent (Orejuela 1985), with the result that in most regions natural vegetation remains only in isolated patches. As a consequence, local extinctions of birds have been documented (Kattan et al., 1994), as well as population declines at the regional and national levels (Hilty 1985; Collar and Andrew 1988). Similar trends at the local level presumably occur in other animal groups (e.g., frogs and dung beetles—Kattan 1993; Escobar 1994), although data are scarce or lacking.

Despite this grim picture, the remaining fragments of natural vegetation still harbor a sizable number of species of plants and animals. While it is unquestionable that large areas need to be protected to preserve the bulk of biological diversity, forest fragments may play an important role in the preservation of a significant fraction of the original regional diversity. At the landscape level, forest fragments of different sizes and ages can be incorporated in a management scheme to maximize the persistence of species. In addition to a sheltering function, these forests would provide important environmental services, such as protection of watersheds.

In this chapter we explore the possibilities for preserving the extant biodiversity in landscape units that include highly fragmented forests in the Colombian Andes. We first describe the natural heterogeneity of Andean landscapes, the impact of land-use practices, and how fragmentation may lead to species extinction. Then we discuss how management at the landscape level may provide tools for the minimization of biodiversity losses. Although we rely mostly on examples from the Colombian Andes, the ideas presented here may be applied to similar situations in the tropical Andes and tropical mountains in general.

The Region

The three ranges of the Colombian Andes extend southwest-north and northeast from the Ecuadorean to the Venezuelan borders, in a latitudinal range from about 1 to 11 degrees north of the equator. Mountaintop elevations range from 1,800 meters at low passes to more than 5,000 meters at snow peaks, encompassing six altitudinal belts (Espinal and Montenegro 1963) from sea level to the highest Andean elevations. The lower belts are essentially continuous along the three cordilleras, but the higher ones, mostly above 3,000 meters, frequently become interrupted by topographical irregularities.

Complex patterns of interactions between altitude, temperature, and rainfall result in a correspondingly complex mosaic of plant communities. Above 1,000 meters at least 15 natural life zones (sensu Holdridge 1967; Espinal and Montenegro 1963) have been recognized. These life zones range from premontane thorn woodland in midaltitude inter-Andean valleys affected by rain shadows, to rainforests in the western and eastern foot slopes, to rain tundra adjacent to perpetual snow peaks. Edaphic and topographic factors add to this diversity in the form of an array of transitional life zones and different associations. In addition, as the three ranges of the Colombian Andes are the result of separate orogenic events (González et al. 1989), and the distributions of plants and animals were extensively affected by climatic changes during the Pleistocene (Haffer 1974), historical factors also account for biogeographical features such as complex speciation patterns (Vuilleumier 1986) and high endemism (Terborgh and Winter 1983) and biodiversity (ICBP 1992).

The generally humid mountains generate an extensive, although somewhat irregularly distributed, fluvial network, which in turn provides connectivity among altitudinal belts. Besides, these watercourses, with their associated vegetation, add to the complex interdigitation and irregularity between natural life zones (Espinal and Montenegro 1963).

Although land-use patterns might have initially created a mosaic superimposed on that of natural vegetation, the long-term effect has been one of homogenization of the landscape. While early settlers preferred fertile mountainsides, middle- and high-altitude inter-Andean valleys, and alluvial terraces (Holdridge 1967), demographic growth has forced more intensive land exploitation, and settlement of increasingly steeper slopes and higher and more rainy regions. Therefore, vast regions of the Colombian Andes are presently dominated by pastures and open areas mostly devoid of trees, and sometimes badly eroded. Exceptions to this trend are a few national parks, protected watersheds, some coffee and tree plantations, and the wettest regions of the Pacific and eastern Andean drainages.

Negative Effects of Fragmentation in Andean Ecosystems

Although very few studies have addressed the effects of habitat fragmentation in the tropical Andes, the available evidence indicates that fragmentation and current land management practices may result in the local extinction of large numbers of species. The best evidence available is for birds. The existence of historic faunal inventories dating back to 1911 and 1959 provided a rare opportunity to document avian extinctions in the region of San Antonio, a cloud forest site in the western range of the Colombian Andes (Kattan et al. 1994). The area, which was fragmented mostly during the first half of this century, at present constitutes an isolated archipelago of forest patches ranging in size from 10 to 400 hectares in a matrix of small farms and suburban houses. Forty species, or 31 percent of the original 128 forest bird species, were locally extinct in 1990.

An analysis of patterns of extinction at this site revealed two main trends (Kattan et al. 1994). First, species at the limits of their altitudinal or geographic distributions exhibited a high frequency of extinction. Of 29 species for which San Antonio was either the upper or lower limit of their altitudinal distribution, 19 are locally extinct. Second, large fruit-eating birds, such as quetzals and cotingas, were the trophic group most vulnerable to extinction. Out of 18 large frugivores originally present, 12 went locally extinct. Among 13 extinct bird species for which the study area was well within their altitudinal limits, 7 were large frugivores.

While patterns of extinction reveal groups of organisms that are particularly vulnerable, prevention of biodiversity losses requires knowledge of the processes and mechanisms responsible for such losses. The consequences of fragmentation may occur at several levels. First, there are effects on the physical environment (e.g., light incidence, temperature). Second, there are direct biological effects on the distribution and abundance of organisms, sometimes mediated through the physical effects. Third, there may be complex, indirect biological effects on the interactions among species, such as predation and parasitism (Murcia 1995).

One direct, physical consequence of fragmentation is a reduction in habitat heterogeneity, or the deterioration and disappearance of certain microhabitats. Streams, for example, are particularly vulnerable because of the desiccation process that affects isolated forest patches (Lovejoy et al., 1986; Saunders et al., 1991). In addition, in suburban areas in the Colombian Andes, streams frequently are partially or totally piped to supply water for human consumption. This may have catastrophic consequences, not only in causing the local extinction of populations of stream-dependent organisms, but in altering the fragment’s hydric dynamics.

The effect of stream disturbance is more evident on amphibians, because of the dependence of many species on water for breeding. Frogs exhibit a wide diversity of reproductive modes, which refers to a combination of egg deposition site (aquatic or terrestrial) and development mode (varying from free-swimming tadpoles to direct development). Diversity of reproductive modes contributes greatly to species richness in Neotropical anuran assemblages (Crump 1982; Kattan 1987). Thus, diversity of breeding microhabitats is an important factor in determining frog species richness.

In the Brazilian Amazon, diversity of breeding microhabitats was a very important factor determining the persistence of frogs in forest remnants (Zimmerman and Bierregaard 1986). In the San Antonio region, preliminary evaluations of frog populations in sites ranging from undisturbed streams to total piping revealed that species richness depends on disturbance level (G. Kattan and C. Murcia, unpublished data). Seven species of stream-breeding frogs were found in an undisturbed stream. In forest patches where water is partially piped, or where water runs free for a short distance (80–200 meters) and then is piped, only one to four species persisted, depending on the amount of water available. At sites where water is piped right at the stream head, no water-breeding frogs persisted.

In contrast to water-breeding frogs, most terrestrial-breeding species were present in all fragments. Frogs with terrestrial reproduction account for 50 percent or more of the species in Andean frog assemblages, in particular the genus Eleutherodactylus, which is very diverse in the tropical Andes (Lynch 1986). These frogs lay terrestrial eggs that undergo direct development (no aquatic larval stage) and are independent of water for reproduction. Thus, reproductive mode will determine the species assemblages in fragmented habitats. The existence of frogs with terrestrial reproduction means that up to 50 percent of the species may persist in small fragments in the tropical Andes, but diversity at the genera and family levels, as well as diversity in reproductive modes, will be drastically reduced.

In addition to its direct consequences, fragmentation may alter interactions among species. Studies in the temperate zone have shown that processes such as bird nest predation and brood parasitism by brown-headed cowbirds (Molothrus ater) are important factors in the decline of bird populations in small forest patches (Robinson 1992; Paton 1994). Rates of nest predation and brood parasitism are frequently elevated in small patches, when compared to large fragments or continuous forest. This frequently occurs as a consequence of an edge effect. In small patches, there is a large perimeter-to-area ratio, which makes the patch vulnerable to invasion by organisms from the matrix or to ecological changes caused by proximity to the edge.

Data on the effects of fragmentation on these ecological processes are sorely lacking in the tropics. Preliminary data suggest that parasitism by shiny cowbirds (M. bonariensis) is not a problem in cloud forest fragments (G. Kattan, unpublished data). However, a study on predation of artificial nests revealed that fragmentation may be an important factor in the extinction of some understory birds in Andean cloud forest (Arango 1991). This study, however, did not find support for an edge effect. Instead, circumstantial evidence suggested that the elevated rates of predation may result from the absence of medium-sized mammalian predators in the small fragments and a consequent abundance of small predators (Arango 1991).

Besides affecting predation rates on bird nests, changes in mammal assemblages caused by fragmentation (Malcolm 1988; Fonseca and Robinson 1990) may also have indirect consequences on other organisms. Dung beetles (Scarabaeidae) are highly sensitive to deforestation and fragmentation (Klein 1989; Escobar 1994). The species composition and activity patterns in cleared areas are drastically different from those in forests (Escobar 1994), probably reflecting parallel trends in mammals. If large and medium-sized mammals disappear from fragmented areas, dung beetles will follow a similar fate. This, in turn, may have repercussions in the enti...

Table of contents

• About Island Press
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Acknowledgments
• Introduction: - The Value of Forest Patches
• Part I - Changing Forests
• Part II - Regional Landscapes
• Part III - Human Dimensions
• Part IV - Management
• Contributors
• Index
• Island Press Board of Directors