Chapter 22

PROCESSES CONSTRAINING

WOODY SPECIES SUCCESSION ON

ABANDONED PASTURES IN THE

TROPICS: On the Relevance of

Temperate Models of

Succession

Chris J. Peterson andWalter P. Carson

OVERVIEW

We review the major constraints on woody species succession in abandoned pastures in the tropics and ask whether
conceptual models developed primarily in temperate regions are useful in tropical habitats. We found that the
majority of tropical post-agricultural succession research (>60 studies) was not typically focused on testin gbroad
general hypotheses or conceptual models, particularly those developed to explain patterns of early succession in
temperate regions. Instead, the studies focused more on evaluating general constraints on woody species recruit-
ment. Amon gthe studies that do consider the major successional models, the three mechanisms of interaction
described by Connell and Slatyer (1977) are the most frequently considered. Empirical studies suggest that woody
species succession in abandoned agricultural fields is constrained primarily by the availability of woody propagules,
though studies rarely simultaneously evaluate the relative importance of other potential processes. Additional con-
straints on succession are competition with residual or resident herbaceous vegetation (e.g., graminoids) and seed
and seedlin gpredation. Most models of succession in temperate re gions were unsatisfactory because they failed to
place enough emphasis on propagule limitation and facilitation. Nonetheless, one conceptual model, the nucleation
model, appears to provide a general conceptual framework that is robust enough to encapsulate post-agricultural
succession in many tropical habitats. Under the nucleation model, a few successful early woody colonists, residual
trees, or key microsites present in the pasture early in succession facilitate the establishment and survival of woody
species in the immediate proximity of these early woody residents or on these microsites. Succession proceeds as
these patches of woody vegetation that form around recruitment foci spread and coalesce. The repeated observation
of enhanced seed input and woody species recruitment near survivin gremnant trees and on unique microsites is
consistent with this model. We urge tropical succession researchers to evaluate general models of succession and
further quantify the degree to which propagule limitation, facilitation, seed predation, and life-history trade-offs
interact to determine the rate of woody species succession into abandoned agricultural habitats. The refinement
and further development of the nucleation model in combination with quantitative models of dispersal holds much
promise.