Mimicking Nature: A Review of Successional Agroforestry Systems as an Analogue to Natural Regeneration of Secondary Forest Stands

Abstract

Over the past 30 years, successional agroforestry systems (SAFS) have been increasingly promoted in Latin America as an approach for recovering soils and improving agro-ecosystems in degraded landscapes. Successional agroforestry systems (SAFS) are complex, multi-strata systems composed of species assemblages that resemble native forest structures. The concept of SAFS integrates indigenous knowledge of intercropping multi-purpose subsistence species, modern agroforestry techniques, and applications of assisted natural regeneration to emphasize biodiversity, adaptive management, and the use of ecological succession to establish a productive system. Much like the management of assisted regeneration of forest stands, mimicking natural ecosystems in agroecosystems requires the knowledge of species survival, growth, functional traits, and niche resource requirements in order to appropriately select multi-functional species and to develop spatial arrangements for stratified stand structures. In recent years, conceptual theories have been proposed that support parallels drawn between natural succession models of forest stand development to management of SAFS. This chapter summarizes background theory to ground the reader in key principles of ecological regeneration and silvicultural management; provides examples that have tested biomimicry hypotheses in agroforestry systems in the tropics; and introduces three case studies from current SAFS in Brazil, Nicaragua, and Belize to examine their potential to promote agro-biodiversity, regenerate severely disturbed agricultural landscapes, diversify harvest yields, and reduce ecological and economic risks associated with conventional agricultural systems.

Full text

179© Springer International Publishing AG 2017
F. Montagnini (ed.), Integrating Landscapes: Agroforestry for Biodiversity
Conservation and Food Sovereignty, Advances in Agroforestry 12,
https://doi.org/10.1007/978-3-319-69371-2_8
Chapter 8
Mimicking Nature: AReview ofSuccessional
Agroforestry Systems asanAnalogue
toNatural Regeneration ofSecondary Forest
Stands
KatherineJ.Young
1 Introduction
Agroecological systems can be designed to mimic nature (Alvim and Nair 1986;
Ewel 1999; Somarriba etal. 2001; McNeely 2004; Malézieux 2011). One agroeco-
logical strategy that seeks to facilitate tropical forest recovery by mimicking pat-
terns and processes of natural succession of native forests is successional agroforestry
(Götsch 1992; Schulz et al. 1994; Peneireiro 1999; Ashton and Montagnini 2000;
Dufty etal. 2000; Vaz 2000; Vieira etal. 2009; Schulz 2011). Successional agrofor-
estry systems (SAFS) are composed of stratied multifunctional species assem-
blages that collectively appear to have a similar structure to native forests.
Management of SAFS emphasizes building species assemblages that contain func-
tional characteristics of key natural successional stages to diversify functional
groups, promote tree-growth and crop productivity, and offset reforestation costs
(Götsch 1992; Peneireiro 1999; Vaz 2000; Vieira etal. 2009; Schulz 2011).
The concept of SAFS integrates indigenous knowledge of intercropping multi-
purpose subsistence species (Nair 1991; Schulz etal. 1994; Senanayake and Jack
1998; Senanayake and Beehler 2000), modern agroforestry techniques (Alvim and
Nair 1986; Ashton and Montagnini 2000; McNeely and Schroth 2006), and assisted
natural regeneration (Parrotta etal. 1997; Shono etal. 2007) to emphasize biodiver-
sity, and the use of ecological succession to establish a productive forest system
(Senanayake and Jack 1998; Dufty et al. 2000; Senanayake and Beehler 2000;
Schulz 2011; Dickinson 2014). Much like the adaptive management of assisted
regeneration of forest stands, mimicking natural ecosystems in agroecosystems
requires the knowledge of species-specic survival, growth, functional traits, and
niche resource requirements in order to appropriately select multi-functional species
K. J. Young (*)
School of Forestry and Environmental Studies, Yale University, New Haven, CT, USA
e-mail: katherine.young@yale.edu

180
and to develop ideal spatial arrangements for stratied stand structures (Ewel 1999;
Somarriba etal. 2001; Kraft etal. 2008).
Conceptual theories have drawn parallels between natural succession models of
forest stand development and management of successional agroforestry systems
(Perera and Rajapakse 1991; Senanayake and Jack 1998; Ashton and Ducey 2000;
Dickinson 2014). However, to date, studies on SAFS are limited to conceptual
papers and descriptive case studies without replication, most often featured in the
grey literature (see Götsch 1992; Peneireiro 1999; Vaz 2000; Vieira etal. 2009).
In this chapter, information is rst summarized from publications that have
advanced our understanding of natural succession. The chapter reviews theoretical
frameworks that have contributed to the development of the biomimicry concept,
and introduces a silvicultural approach to agroforestry systems highlighting ana-
logue forest hypotheses. Case studies from current agroforestry systems in Latin
America are provided, demonstrating the theory of successional agroforestry sys-
tems as anthropogenic analogues to secondary forest stand development, and its
potential to regenerate severely disturbed agricultural landscapes to bountiful, bio-
diverse agroforestry systems.
2 Conceptual Background
2.1 Theory ofNatural Succession
Succession refers to the changes observed in an ecological community following a
disturbance. Successional processes of natural regeneration are driven by the type,
intensity, timing, and duration of a disturbance to a site; pre- and post-disturbance
of propagules; climate; and species competition for resources such as sunlight,
water, and nutrients (Gleason 1939; Egler 1954; Connell and Slatyer 1977; Halpern
1989; Camp and Oliver 2004). Within mixed species stands, each species responds
to microenvironmental changes in resource availability and stress through time,
while competing with neighboring plant species for access to limited resources
(Gleason 1939; Egler 1954; Connell and Slatyer 1977; Halpern 1989). In his botani-
cal eld studies, Gleason observed:
1. Every species of plant has reproductive powers in excess of its need;
2. Every species of plant has some method of migration;
3. The environment in any particular station is variable; and
4. The development of a vegetational unit depends on one or the other of two condi-
tions, the appearance of new ground or the disappearance of an existing associa-
tion (1939).
Over time, associations between groups of plants in natural succession are tem-
porary and uctuating—dependent upon their origin, structure, and the particular
disturbance in the environment and surrounding vegetation (Gleason 1939).
K. J. Young

181
Conventional interpretation of “old-eld plant succession” has followed the idea of
relay oristics– involving a succession of incoming and outgoing plant associations,
each group establishing itself and out-competing predecessor plant groups. This
model suggests each plant groups modies or “prepares” the site for a subsequent
group. Alternatively, in the initial oristic composition (IFC) hypothesis, species
representing all guilds colonize soon after an initial disturbance but reach domi-
nance at different times according to their growth rates and longevities. Egler (1954)
found evidence for both models in old elds, however, relay oristics was less fre-
quent: “several instances are now known where the incoming relay is responsible
for less than 5% of the eventual forest stand. Conversely, 95% was determined years
before, at the time of abandonment, and developed completely independently of the
grasslands and shrublands that have temporally preceded them.” IFC is largely
dependent upon the availability of seeds lying dormant in the soil and/or from seed
trees adjacent to the site, and to the necessary conditions to break the seed coats for
germination. In a successional agroforestry system, these seeds and/or seedlings are
purposely introduced by the manager, rather than being naturally present in the site
following a disturbance.
In a seminal 1977 paper, Connell and Slatyer proposed alternative models for
natural succession: the tolerance model, the inhibition model, and the facilitation
model. Mechanisms that determine changes during succession (such as interactions
with herbivores, predators, and pathogens) and the relationship between succession
and community dynamics were explored. Papers published before 1977 all agreed
that certain species will usually appear rst because they have evolved “colonizing”
characteristics (such as the ability to produce large numbers of propagules). The
models differed in how new species appeared later in the sequence. In the facilita-
tion model, species modify the environment making it more suitable for other spe-
cies to invade and grow. In the tolerance model, new arrival of species from
modications in the environment neither increased nor decreased the rates of recruit-
ment and growth of later colonizing species. Examples include shade tolerance, and
resilience to other environmental factors such as moisture, nutrients, allelopathy, or
grazing. In the inhibition model, species do not grow to maturity when previously
established species are present; they only appear later because they live longer and
gradually accumulate as they replace earlier species. Early species are killed by
local disturbances or by natural enemies (herbivores, parasites, pathogens, etc.).
The facilitation model can be seen in primary succession (newly exposed sites),
such as weedy species with quickly spreading roots or umbel-like owers that dis-
perse wind-blown propagules rapidly and far, quickly covering the ground. This
maintains moisture and adds organic matter, which is critical for nutrient cycling
and nitrogen xation. This is the dominant model in agricultural fallows following
slash-and-burn (swidden) agriculture, but is unlikely to be found in forest stands.
The tolerance model is supported by evidence that late successional species are
often able to establish without any preparation of the site by earlier species. This is
true in timber plantations, or in the installation of native tree species for reforesting
disturbed forests in tropical regions. This is also true for simple agroforestry sys-
tems, where trees that are highly tolerant to poor soils and high UV (such as
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

182
Mangifera indica, Citrus spp., Inga spp., and Eugenia uniora) establish well in
former agricultural elds without prior preparation/modication to the environmen-
tal conditions. However, species that are less tolerant to extremes (Theobroma
cacao, Persea spp., Cinnamomum zelanicum, or other genus in the Malvaceae or
Lauraceae families) may require some modication of the environment (shade, leaf
litter accumulation, water inltration and soil quality) to establish strong root sys-
tems and healthy shoots, foliage, and fruits.
In management of SAFS following agricultural fallows, selective removal or
thinning of ora can produce new plant assemblages (IFC) by modifying the envi-
ronment, and can also open up niches for plants with complementary structures to
thrive. Management approaches of SAFS are consistent with the individualistic
concept of plant associations, whereby plants are selected for individual phenologi-
cal traits (timing of owering, shade tolerance, drought resistance, etc.) and struc-
tural traits (height, growth rate, etc.) and are interplanted with other multifunctional
species with complementary traits. Rather than monocropped plantations, succes-
sional agroforestry techniques seek to take advantage of individual plant traits and
adaptively manage the variation in the environment.
2.2 Theory ofBiomimicry
The idea of simulating the diversity of the forest for agricultural purposes—or bio-
mimicry—has been debated since the 1960s (Geertz 1963; Harris 1971; Nations
and Nigh 1978; Beckerman 1983; Vickers 1983; Eden 1987; Ewel 1999; Malézieux
2011). Citing examples from indigenous forest cultivation practices in Indonesia
and Japan (Geertz 1963), Venezuela and New Guinea (Harris 1971), and Mexico
(Nations and Nigh 1978), researchers suggested that using traditional methods of
shifting cultivation (clearing and burning forest land to cultivate a short period of
mixed cropping, followed by an extended period of forest fallow) as a strategy to
intercrop subsistence products in gaps, develops complex, polycultural community
structures that protect the soil from leaching and erosion, and directly utilize nutri-
ents stored in living matter–– much like a tropical forest (Geertz 1963). Subsequent
studies challenged this hypothesis, citing evidence from indigenous communities in
tropical South America (Beckerman 1983; Vickers 1983) to argue that while many
crop species may be present in swidden agriculture, the distribution and composi-
tion is often patchy, ‘zoned’, and dominated by the crop of primary subsistence
(such as manioc), and therefore is not comparable to the native forest (Beckerman
1983; Vickers 1983; Eden 1987). Recent research has attempted to bridge the argu-
ments to demonstrate how productive agricultural systems can better mimic native
ecosystems by developing a working framework of high species diversity, complex
structure, low management intensity with minimal inputs, and diversied output of
yields (Perera and Rajapakse 1991; Ewel 1999; van Noordwijk and Ong. 1999;
McNeely 2004; Malézieux 2011).
K. J. Young

183
Agroforestry systems have borrowed from biomimicry theory since the 1970s
and have been promoted as a land use strategy for addressing food security and sus-
tainable agricultural production (Nair 1993; Somarriba and Beer 2011), biodiversity
conservation (Nair 1993; McNeely and Schroth 2006; Bhagwat etal. 2008), and
restoring connectivity to fragmented landscapes (Nair 1993; Laurance 2004;
McNeely 2004; Schroth 2004; Montagnini etal. 2011). Integrating indigenous prac-
tices of intercropping non-timber, multi-purpose tree species, and managing long
fallow periods (e.g. traditional homegardens and swidden agriculture) into modern
techniques that t within linear conventional agricultural systems (e.g. alleycrop-
ping, windbreaks, and living fences), agroforestry techniques range from simple sys-
tems that intercrop one tree species with one agricultural crop species to systems that
interplant tree crops of high value under existing or introduced canopy cover (e.g.
cabruca cacao or shade-grown coffee systems). However, many conventional agro-
forestry systems do not necessarily mimic native forest structure and biodiversity.
Trees are the primary focus in plantation agroforestry (e.g. cacao, Theobroma cacao;
coconut, Cocos nucifera; rubber, Hevea brasiliensis; and oil palm, Elaeis guineen-
sis), however the contributions of other forest components (such as herbs, shrubs,
lianas, epiphytes, etc.) are often ignored (Ewel 1999). It should be noted that non-
tree biodiversity has important roles to play in enhancing sustainability and struc-
tural stability of landscapes (Dufty etal. 2000), and also provides many high market
value products such as culinary and medicinal herbs, berries, vanilla, and bers.
3 Mimicry Hypotheses inSAFS
Several studies have shown the potential for agroforestry systems to better mimic
natural ecosystems, whose components are the results of natural selection and are
therefore sustainable models on which to base the design of new systems of land use.
3.1 Three Case Studies Testing theBiomimicry Hypothesis
inAgroforestry Systems
In 1986, Alvim and Nair examined intercropping of cacao with rubber, clove, black
pepper, and coconuts on various farms in the cacao-producing region of Bahia,
Brazil. They found that overall, crop combinations were preferred most with clove
and least with coconuts, which reects the synergistic growing requirements of
cacao and clove. The authors point to further examples of highly productive inter-
cropping that includes: coffee, passion fruit, papaya, clove, peach palm, beans,
patchouli, cardamom, and vanilla—which all have high market value. Yields were
measured in pure stands (conventional, un-shaded cacao production; improved
management; experiment station sites) and mixed stands of rubber and cacao. The
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

184
paper suggests that benets to crop combinations include: more efcient use of
labor and equipment over a relevant time and space, reduction of weeding, more
efcient use of soil fertility resources, decrease in erosion hazards and increase in
land’s total productivity. Socio-economic benets included increasing overall eco-
nomic yield and protecting against threats of pest/diseases. The authors also identify
management considerations, such as choosing appropriate intercrops, identifying
their agronomic requirements, developing adaptive management practices for crops
with varied needs over time and space, and ensuring marketability of the various
products (Alvim and Nair 1986). Overall, the authors found that complex crop com-
binations are more demanding in labor, capital, and technical/managerial skills.
Ewel (1999) articulated the ‘mimicry hypothesis’ when he argued that “agricul-
tural systems should benet from imitating the structure and function of natural
ecosystems, since components of the latter result from natural selection towards
sustainability and the ability to adjust to perturbations” (as cited by van Noordwijk
and Ong 1999). Van Noordwijk and Ong extended that hypothesis on structure and
function to include a second mimicry hypothesis: “agroecosystems shall benet
from resembling the diversity of natural ecosystems” (van Noordwijk and Ong
1999). Both hypotheses refer to the potential benets of combining different plant
life forms in the place of monoculture:
(i) Larger total production with staggered harvests throughout the year
(ii) Higher efciency (particularly labor) over time, as less effort is needed to
achieve the same results where one works ‘along with’ rather than ‘against’
natural processes;
(iii) Reduction of downstream effects (e.g. groundwater ow patterns can be
grossly affected by both additional and reduced water use) if the total resource
use (water, nutrients, biodiversity) simulates that of the natural system of
which it replaces;
(iv) Improved maintenance of environmental service functions in the landscape.
Ewel (1999) examined plant communities designed to imitate the structure and
function of secondary succession in humid tropical lowlands over 5 years.
Comparing natural succession vegetation following slash-and-burn of existing veg-
etation; a monoculture maize eld; an “imitation system” (“investigators’ attempted
to build [a system that contains] the same mix of life forms that appeared in the
successional vegetation [i.e. dominated by herbaceous vegetation, shrubs, or trees],
but which consisted of species alien to the site [Zea mays, Manihot esculenta, and
Cordia alliodora, respectively], selected and planted by the scientists (naturally
occurring colonists being weeded out”); and species-enriched treatments (“in which
vast quantities of seeds were regularly added to natural successional vegetation in
an effort to assess the functional responses of further enrichment of diversity”); and
measured responses to pests, productivity, and soil fertility. The ndings indicated
that the imitation system performed on par with the natural succession plot in
respects to pests and soil fertility. However, the imitation system was not as produc-
tive as the successional vegetation, although much more productive than the mono-
culture maize at times.
K. J. Young

185
A subsequent study (Somarriba etal. 2001) advanced Ewel’s ndings to com-
pare six production systems: (i) taungya under inga (Inga edulis), acacia (Acacia
mangium), or in association with crops cycles consisting of three cycles of maize (Z.
mays), one cycle of ginger (Zingiber ofcinale), followed by permanent installation
of a perennial fruit shrub, araza (Eugenia stipitata); (ii) new cacao plantings
(Theobroma cacao) under the shade of valuable tropical timber trees such as laurel
(Cordia alliodora), inga (Inga edulis), tropical almond (Terminalia ivorensis), and
rosy trumpet tree (Tabebuia rosea), with laurel and cacao planted simultaneously
with maize (Z. mays), cassava (M. esculenta) and pigeon pea (Cajanus cajan) as
temporary shade for the cacao; (iii) old cacao plantations, (iv) cacao-laurel-plantain
(Musa AAB) systems, (v) line plantings of laurel, A. mangium and Tectona grandis
compared with ve trials of laurel, T. ivorensis and Eucalyptus deglupta; and (vi)
pure plantations of just laurel (C. alliodora). The authors evaluated mortality,
growth, site index and site variables and found that the laurel growth was highest in
the intercropped cacao-laurel-plantain system, followed by taungya, new cacao
plantations, old cacao plantations, line plantings, and pure plantations, respectively.
This study suggests that intercropping valuable timber in agroforestry systems ben-
ets growth when planted in association with other species.
Also in 1999, van Noordwijk and Ong tested the two “mimicry” hypotheses ((i)
Agricultural systems benet from imitating the structure and function of natural
ecosystems and relative components of natural succession; and (ii) Agroecosystems
benet from resembling the diversity of natural ecosystems) in the savannah zone of
sub-Saharan Africa, an area where agriculture and native vegetation are difcult to
maintain due to temperature extremes, limited water source, and other disturbances
such as wind, grazing pressures, and human disturbance. The authors sought to
understand how diversity of species might be used to design or improve agrofor-
estry systems for sub-Saharan Africa. To answer this question, they examined the
structure and function of sequential fallow-crop rotation agroecosystems in the
savannah zone, comprised of gum arabic (Acacia senegal), desert date (Balanites
aegyptica), shea butter nut (Vitellaria paradoxa) and néré (Parkia biglobosa), and
studied species competition and diversity, and structure and function of the African
savannah ecosystem (dominated by thorny shrubs and trees in the Acacia genus
such as Acacia senegalensis; long-lived canopy trees such as the baobab, Adansonia
digitata) and elephant grass (Pennisetum purpureum). Van Noordwijk and Ong con-
cluded that agricultural systems do benet from imitating the structure and function
of natural ecosystems, but the relative benet from resembling the diversity of natu-
ral ecosystems is dependent upon the perspective and ability of farmers to derive
value from a complex system with many components (van Noordwijk and Ong
1999). These ndings echo the warnings of other studies that acknowledge the chal-
lenges of designing ecologically intensive agroecosystems due to the necessity of
in-depth knowledge of biological processes (Alvim and Nair 1986; Malézieux
2011), and the knowledge-intensive management required for each tree species to
be selected based on the site conditions and an understanding of survival, growth,
and other dynamics (Alvim and Nair 1986; Perera and Rajapakse 1991).
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

186
In an opinion paper published in Trends in Ecology and Evolution, Bhagwat
etal. (2008) examined the role of agroforestry systems in maintaining biodiversity,
and argued that agroforestry systems can re-connect fragmented landscapes such as
nature reserves to human-dominated landscapes, reduce resource-use pressure from
these forests, and provide habitats outside formally protected land. The authors
introduced the concept of conservation in human-dominated landscapes by chal-
lenging conceptions of “untouchable wilderness” through their analysis of land-
scape management practices known from ancient, pre-Columbian civilizations (i.e.
the Maya in Mesoamerica, and native populations in central Amazonia) and hunter-
gatherer societies in Cameroon and Papua New Guinea. They reviewed recent
research in tropical agroforestry vis a vis a literature search (185 references, using
keywords ‘agroforestry’ and ‘biodiversity’), and found that many studies sampled
multiple taxa in forest and agroforestry landscapes: 36 articles made a direct com-
parison between agroforestry systems and neighboring forest reserves, and 12 types
of agroforestry systems and 9 taxa were represented in 14 tropical countries. For
each agroforestry study, the authors calculated the species richness as a percentage
of the species found in the neighboring forest reserve, and examined the similarity
in species composition between the two system types. This work demonstrated: (i)
many agroforestry systems are important for the protection of species and habitats
outside protected areas; (ii) agroforestry systems maintain heterogeneity at the habi-
tat and landscape levels; (iii) trees in agroforestry landscapes reduce pressure on
protected forests (Bhagwat etal. 2008). Agroforest matrices provide a rich habitat
for native species to survive outside the connes of a nature reserve.
3.2 Theory of(Agro)Forestry Analogues
Are successional agroforestry systems analogous to natural regeneration of native
forests? An Analogue Forestry system is dened as “an environmentally sustainable
land management approach that assists farmers in developing multi-species plots of
both native and exotic crop species that, over time, mature to approximate the struc-
ture, ecological functions and environmental integrity of a natural forest”
(Senanayake and Beehler 2000). Analogue Forestry uses organic principles to pro-
duce a suite of high-value foods, spices, herbs, and medicinal plants, as well as
fuelwood and timber stratied amongst canopy trees, vines, understory shrubs and
herbs without compromising biodiversity (Senanayake and Beehler 2000). The
choice and placement of each species is determined by its contribution of specic
ecological and economic functions. Creating multi-strata agro-forests improves
wildlife habitat, thus becoming ideal buffer zones with multiple functions (ecologi-
cal, economic, etc.). Benets include increased ecosystem services from a mature
forest system; economic opportunities from the production of timber and non-tim-
ber forest products; social benets stemming from exchange of information, best
practices, and strategies among practitioners (Senanayake and Beehler 2000;
Dickinson 2014). Each of these principles is consistent within SAFS.
K. J. Young

187
Can silvicultural management of forest stands be applied to successional agro-
forestry models? Because the dominance of long-lived, woody trees and shrubs in
agroforestry systems adds structural and functional complexity over a long period
of time, successional agroforestry is best understood from a silvicultural perspec-
tive, in which multistoried stand structures are managed for complexity (Ashton and
Montagnini 2000). Successional processes of natural regeneration are driven by the
type, intensity, timing and duration of a disturbance to a site; pre- and post-
disturbance of propagules; weather conditions; and individual species competition
for limited resources such as sunlight, water, and nutrient availability (Gleason
1939; Egler 1954; Connell 1978; Tilman 1985). In successional agroforestry, these
disturbances are largely anthropogenic management interventions (such as burning,
planting, pruning, thinning, coppicing and pollarding), and therefore require an
understanding of forest stand dynamics to manage them appropriately.
Management of forest stand development has been described as occurring in four
phases: (i) stand initiation, (ii) stem exclusion, (iii) understory re-initiation, and (iv)
oscillating steady state or “old growth” (Oliver and Larson 1990; Kozlowski 2002;
Camp and Oliver 2004) (Table8.1). In recent years, conceptual theories have been
proposed that draw parallels between management of secondary forest stands
(Oliver and Larson 1990; Ewel and Bigelow 1996; Kozlowski 2002; Camp and
Oliver 2004) and management of successional agroforestry systems (Perera and
Rajapakse 1991; Senanayake and Jack 1998; Ashton and Ducey 2000; Ashton and
Montagnini 2000; Kelty 2000; Jones 2001).
Following the conceptual model proposed by Ashton and Ducey (2000), the nurse
phase—combining annual crops and perennial agroforestry species—is analogous to
“stand initiation”; the training phase—establishing timber, fuelwood, fruits, spices,
and medicines in multistoried stands of trees—is analogous to “stem exclusion”; and
the shade tree/crop phase—preferentially managing valuable overstory timber spe-
cies and understory perennial crops such as cacao or coffee—is analogous to the
“understory initiation” of stand development (Ashton and Ducey 2000; Kelty 2000)
(Table8.1). As gaps in the understory and canopy are lled with multi-functional
agroforestry species, and management efforts shift over time from planting (stand
initiation), to thinning/pruning (stem exclusion), to installation of understory crops
(understory re-initiation), towards harvesting and maintaining the system (oscillating
stable state/old growth), silvicultural management of SAFS may resemble silvicul-
tural management of multi-species forest stands (Perera and Rajapakse 1991;
Senanayake and Jack 1998; Ashton and Ducey 2000; Kelty 2000; Dickinson 2014).
Table 8.1 Parallels between
successional stages of forest
stand dynamics (Oliver and
Larson 1990; Camp and
Oliver 2004) and analogue
agroforestry (Senanayake and
Jack 1998; Ashton and Ducey
2000; Kelty 2000)
Forest stand dynamics Analogue (agro-)forestry
Stand initiation Nurse phase
Stem exclusion Training phase
Understory re-initiation Shade-tree crop phase
Old growth Management for harvest/
structural diversication
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

188
Mimicking natural ecosystems in agroecosystems similarly requires develop-
ment of multi-strata, multi-functional species stand structures based on age, size,
spatial arrangements, functional traits, and species-specic niche requirements of
plants (Kraft etal. 2008) as well as a clear denition of desired stand structures
based on developmental stages. These systems combine the horizontal arrangement
of complementary species with vertically overlapping species, mainly determined
by light needs and the available rooting zone (Schulz 2011). Through the mixture of
trees, shrubs and herbs or crops, a broader spectrum of functionally different species
can lead to a more efcient use of the available nutrients, solar energy and water,
giving more stability to the system and reducing the need for external inputs (Schulz
etal. 1994; Schulz 2011).
3.3 The Missing Link?: Functional Diversity Theory
What traits and environmental variables are most important in determining funda-
mental niches of agroforestry species? To answer this, we can look at agro-
biodiversity in terms of functional traits that are present in the dominant species in
a successional age class. The functional biodiversity of the system emphasizes
dynamics at various levels and can reveal implications for its functioning as a whole
(Callo-Concha 2009). Within mixed species stands, each species responds individu-
ally to niche micro-environmental changes in resource availability and stress
through time, while competing with neighboring plant species for access to limited
resources (Gleason 1939; Ashton and Larson 1996). Changes in species composi-
tion and diversity will affect the functioning of ecosystems most strongly when
species differ in their effects on ecosystem processes or in their response to environ-
mental change (Chapin etal. 1997; Huang etal. 2002; Lavorel etal. 1998; Lavorel
and Garnier 2002; McGill etal. 2006; Lebrija-Trejos etal. 2010). Ecosystem func-
tioning is the end result of multiple stages of lters that, by assembling individuals
with appropriate responses (functional traits), result in communities with varying
trait composition (Huang etal. 2002; Lavorel etal. 1998; Lavorel and Garnier 2002;
McGill etal. 2006). As species richness increases, productivity and biomass of the
system also increase (Chisholm etal. 2013).
Functional biodiversity is operationalized by identifying functional groups, indi-
cating productive, ecological and operational functions in the ecosystem (Lavorel
etal. 1998; Tilman 2001; Callo-Concha 2009). Plant functional groups are classi-
ed by functional traits that reveal their strategies in varying environments and in
the face of changing resources (Tilman etal. 1997; Lavorel et al. 1998; Tilman
2001; Cornelissen etal. 2003; Lebrija-Trejos etal. 2010; Lohbeck etal. 2012). Plant
functional groups are cohorts of species that share similar functioning at the stand
level, and demonstrate similar responses to environmental factors and/or similar
effects on ecosystem processes (Tilman etal. 1997; Lavorel et al. 1998; Tilman
2001; Cornelissen etal. 2003; Lebrija-Trejos etal. 2010; Lohbeck etal. 2012).
K. J. Young

189
3.4 Successional Agroforestry Systems (SAFS)
Incorporating a range of agroecology and agroforestry techniques as a transition
phase early in forest restoration could be used to overcome socioeconomic and eco-
logical obstacles to restoring these lands (Vieira etal. 2009). The management prac-
tices used in SAFS parallel those used in many forest restoration efforts. A range of
agroforestry systems could be used as a transitional phase in restoration that simul-
taneously helps provide for human livelihoods, reduces the initial costs of restora-
tion, and extends the time period of management of restoration. What does this look
like, conceptually?
In a typical agricultural production systemfollowing disturbance using slash-
and- burn methods of clearing land prior to cultivation (known as swidden agricul-
ture), short-lived perennials (such as cassava) are planted after slash-and-burn. Fires
intentionally ignited for agricultural management are a lethal or sub-lethal distur-
bance to the site for short-term agricultural production purposes (Fig. 8.1).
Immediate consequences include the loss of plant and animal diversity from removal
of habitat and, may also result in soil erosion and nutrient leaching, ultimately pos-
ing an ecological, nutritional, and economic threat to farming communities.
In contrast, SAFS seek to imitate the structure of natural ecosystems and take
advantage of ecological interactions among species to mimic patterns of natural
regeneration to provide long-term food security. Management of SAFS focuses on
the selection, combination, and management of crops, and managing structural and
functional diversity to promote complementary ethnobotanical plant associations,
mitigating negative relationships such as allelopathy, and harnessing positive rela-
tionships such as cycling of carbon and nitrogen (Schulz et al. 1994). As such,
Typical swidden
agricultural
production in the
tropics
Before:
Severe
anthropogenic
disturbance
(slash and burn)
After:
Short-lived
perennials
(such as cassava)
are planted after
slash and burn
Fig. 8.1 Management transition of swidden agriculture
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

190
SAFS have been identied as an adaptive socio-ecological management approach
that has the potential to improve tree-growth, stabilize crop yield productivity,
regenerate degraded landscapes, and offset reforestation costs by providing seed
sources (Schulz 2011; Vieira etal. 2009; Kelty 2000; Vaz 2000; Peneireiro 1999;
Götsch 1992) (Fig.8.2).
Silvicultural management of forest stands may be applied to SAFS models by
following analogous phases of natural regeneration in adjacent native secondary
forest stands to favor high-value ethnobotanical species and cash tree-crops, maxi-
mizing nutrient cycling and vertical growing space (Fig.8.3). In successional agro-
forestry management, enriching the growing site with leguminous annuals and
perennials mimics stand initiation by planting species tolerant to high UV radiation
and poor soil quality (A). Nitrogen xing woody trees, shrubs, and leguminous
herbaceous ground cover can be integrated into the understory to provide ecosystem
services (preventing erosion, moderating topsoil temperatures and soil moisture,
and supporting nitrogen cycling for surrounding woody species—critical in tropical
soils such as oxisols and alsols), while also providing a food source (e.g. beans) for
humans or livestock, and/or a source of seasonal income. Leguminous ground cov-
ers benet the stand initiation of young agroforestry species by providing nutrients
and moderating extreme temperatures and weather events. The shade from the
newly established trees and shrubs also create an ideal “nursery” for the establish-
ment of valued trees—such as fruit, nut, or timber species (B). As the system
matures, available growing space is lled, and sunlight, water, and soil nutrients
may become limited by heavy competition—much like in the stem exclusion phase
of forest stand dynamics. This is the stage in which adaptive management and tree
training (such as pruning, coppicing, pollarding, and thinning the stand) becomes
critical to support the growing space and nutrient needs of high-value species (C).
In the analogous “understory re-initiation” phase of SAFS, native secondary or
selectively planted leguminous overstory trees (such as Inga spp. or Erythrina spp.)
may serve as a shade-tree phase, promoting the establishment of high-market value
understory crops (such as cacao, Theobroma cacao; coffee, Coffea arabica; yerba
Regeneration of highly
disturbed agricultural systems
Before:
Site characterized by loss of
biodiversity and complex
resource cycling
After:
Sustainable
agro-ecosystem characterized
by high agro-biodiversity,
complex resource cycling, and
extended harvest periods
Fig. 8.2 Potential for SAFS to regenerate degraded agricultural systems
K. J. Young

A.
Stand initiation/
Nurse phase I
Anthropogenic
disturbance opens
growing space for
high-UV tolerant
species.
Short-lived perennials
tolerant of high UV and
poor soilsare planted
after disturbance.
B.
Stand initiation/
Nurse phase II
Intensive enrichment
plantings of
herbaceous, nitrogen-
fixing ground covers,
short-lived perennials
(e.g. bananas, cassava)
and long-lived
perennials (e.g. fruit or
nut trees)
C.
Stem exclusion/
Training phase
Intensive pruning,
coppicing, and
pollarding(up to 95%
of leaf canopy), and/or
selectivethinning.
Trimmings are
“chopped” into smaller
pieces and “dropped”
under high-value
species to promote
decomposition,
moisture retention, and
biomass production of
regrowth to support
light, moisture, and
nutrient requirements
of high value speciesat
each vertical strata of
the system.
Fig. 8.3 Management transitions (A-E) of SAFS following successional analogues

192
D.
Understory
Re-initiation/
Shade-tree crop
Phase
Natural regeneration of
seedlings are permitted
in understory until
saplings surpass
understory strata.
Integrative and
adaptive management
(intensive pruning,
selective thinning) is
repeated [C]to
maximize benefits of
biomass accumulation
and light, moisture, and
nutrient requirements
between
complementary plant
associations.
E.
Target goal:
Diversified,
agroforestry
Systems
Multi-strata perennial
polycultures maximize
available growing
space, nutrient
resources, and diversify
multiple harvestsper
year.
Fig. 8.3 (continued)
K. J. Young

193
mate, Ilex paraguariensis; or cardamom, Elettaria cardamomum) (D). Many con-
ventional agroforestry systems stop there, but this is where SAFS stands apart.
Niche micro-environments in mature agroforestry systems can be advantageously
exploited to ll the spaces with additional species, diversifying the system with both
ecological and economical outputs from other valuable woody and herbaceous spe-
cies to spread the potential yields of food, fuels, bers, medicines, precious timbers,
and livestock fodder across the seasonal harvest periods and over time (E). As such,
short term gains from annual crops and fruit harvests and long-term investments
into precious hardwoods (such as mahogany, Swietenia macrophylla; teak, Tectona
grandis; and rosewood, Dalbergia spp.) are achieved.
Many farmers focus on one or two harvest seasons per year with yields from just
one or two crops, but they could spread out the returns in their yearly income to
prot from a diversied harvest regime. While the management of a complex agro-
ecosystem requires complex planning and initial investment, raising initial opportu-
nity costs (particularly for smallholder farmers), single-species cropping systems
are a risky business: they are highly susceptible to threats such as pests, diseases,
and seasonal irregularities associated with climate change. SAFS offer an inte-
grated, adaptive management approach to reduce ecological damage from intensive
farm management, while diversifying yields and reducing the risks of harvest losses.
All these translate into potential gains over time in farm income.
Management of SAFS can be designed to better mimic the structure and function
of natural secondary successional development in native forests by selecting pro-
ductive species that share the same functional traits within a plant functional group.
Careful observation of both native forests adjacent to the agroforestry system and of
the specic species chosen for the SAFS has the potential to increase agro-
biodiversity, conserve native ethnobotanical species, add ecological complexity to
strata structure and provide ample functional traits to maintain important ecosystem
services such as nutrient cycling, soil quality, and water retention, as well as build-
ing biomass and habitat for animal species. For proper design and management of
successional agroforestry systems it is vital to understand individual species require-
ments and how to pair them in successional plant associations to improve the func-
tioning of agroforestry systems.
4 Case Studies ofSuccessional Agroforestry Systems
What does a theoretical SAFS that mimics the structure and function of secondary
forest regeneration look like at the farm level? Below, three experimental case stud-
ies are presented that highlight the theory of successional agroforestry principles in
practice, and examine the limitations of further adoption of these systems.
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

194
4.1 Bahia, Brazil: Fazenda Olhos d’Agua
Successional agroforestry systems (SAFS) have been promoted by agronomist
Ernst Götsch in Bahia, Brazil over the last 30years as an approach for regenerating
degraded soils and improving productive agro-ecosystems in degraded agricultural
landscapes. In 1994, B.Schulz and B.Becker collaborated with Götsch to highlight
the management of Fazenda Olhos d’Agua—Götsch’s 500ha farm in the Gandu
District of southeastern Bahia. The farm was converted in 1985 from degraded pas-
ture and secondary forest—with remnant fragments of primary forest—to 100ha of
cultivated agroforest over a 10-year time period (at time of 1994 publication).
Initially, the farm was planted with cocoa as the main cash crop; however, due to the
crash in the cocoa economy from the witch’s broom fungus outbreak in the late
1980s, bananas became the primary source of income (Schulz et al. 1994).
Management decisions were guided by the following objectives:
1. Protection of the soil from direct inuence of weather (sun, wind, rain), resulting
in reduced erosion and increased microbial activity.
2. Decrease of water loss by reduced run-off, improved drainage, higher water
retention capacity, and decreased evaporation.
3. Humus accumulation.
4. Maximization of nutrient cycling.
5. Management of light and space for the cultivated species.
6. Management of positive and negative growth inhibition effects.
Plant selection followed two primary strategies: (i) Native species without com-
mercial value [were] deliberately left in or incorporated into the system, and (ii) A
portion of the native species [were] substituted by eco-physiologically similar cul-
tivated ones. Management of the agroforest system was further guided by Götsch’s
two “working hypotheses”: rst, there is a “growth reducing effect of maturing
plants at the end of their life cycle on their neighboring plants, and (conversely)
there is a growth stimulating effect by young plants on the vegetative development
of adjacent plants”. This likely refers to Connell and Slatyer’s (1977) “inhibition”
model (growth reducing effect) and the “facilitation” model (growth stimulating
effect), which may be correlated with nutrient cycling associated with particular
plant combinations. The latter assumption is echoed by Götsch’s second working
hypothesis: “There is a positive relationship between carbon and nitrogen cycles
within the cropping system and its productivity.” According to Schulz, Becker, and
Götsch, removing plants just before they reach maturity may accelerate succession
by “shortcutting the homeostatic phases of the different succession stages”, remov-
ing any antagonistic growth reduction. In Götsch’s system, 6ha of land are man-
aged at a time. Overstory species are intensively pruned (up to 95%) each year, and
the trimmings are cut into smaller pieces and evenly distributed on the ground with
larger branches placed on contour to prevent erosion. Intensive pruning regimes
stimulate high biomass production (increasing foliar growth and fruit production),
intensies photosynthesis, increases moisture retention in soil and moderates
K. J. Young

195
ambient temperatures. Additionally, mulching made from “chopping and dropping”
selected plants (such as banana stems, and maturing plants at the end of their life
cycles) provides organic material from which microbiological activity can rapidly
mineralize nutrients, building a rich topsoil supporting in situ nutrient cycling—
much like a natural forest under assisted regeneration.
What methodology is needed to quantify the validity of these claims? Schulz,
Becker, and Götsch attempted to estimate the annual amount of i) natural litter fall
and plant material cut during cocoa (Theobroma cacao) cultivation in Götsch’s
SAFS, and ii) compared yields to conventional cocoa plantations of neighboring
farms (Table 8.2). Results indicated that the ‘forest garden’ cacao management
approach provides cocoa yields without the need for external inputs “at a level
which, in the surrounding cocoa plantations, can only be attained by the use of con-
siderable amounts of fertilizer and pesticides” (Schulz etal. 1994).
Whereas assisted natural regeneration relies upon seed trees and release methods
of natural regeneration as the primary sources of seed dispersal and seedling establish-
ment, SAFS like Götsch’s use existing seed trees as nurse trees, and interplant col-
lected seeds or seedlings from ethnobotanical and cash crop plants to occupy the gaps
in the understory and ll in canopy levels over time. Natural regeneration of saplings
in the understory beneath the cacao is left to grow until it has either i) surpassed the
cacao, or ii) is occupying horizontal or vertical space. The natural regeneration is then
incorporated into the system, either as a long-term investment for timber, or integrated
into the intensive pruning regime. As such, SAFS rely on both active observation and
keen knowledge of the particular site, as well as future trajectories based on physio-
logical characteristics of the timber and non-timber forest crops.
In 2011, Schulz, J. compared two climatic regions in the northeast of Brazil,
where smallholders have applied successional agroforestry methods. Schulz found
that at the start, the SAFS had been very difcult to manage, as very limited yields
of the plot coincided with a high workload. The rst years of development of the
successional system could only be supported by a newly introduced honey produc-
Table 8.2 Annual litter and harvest yields and associated input quantities from a successional
agroforestry system, a conventional cacao production system, and as noted in literature
recommendations
Fazenda Olhos d’Agua:
SAF Cacao System
Conventional Cacao
Plantation
Literature
recommendations
Dry matter/mulch
(t/ha/year)
8–16 1.5–5 5–20a
Cocoa yield (kg/
ha)
110–370 225b–
Inputs:
Fertilizers 0 130kg/ha (N)b–
Pesticides 0 Fungicidesb–
aBeer (1988)
bCEPLAC (Commissao Executiva do Plano da Lavoura Cacaueira),the Brazilian Cocoa Research
Center
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

196
tion initiative, with the help of a local NGO (Schulz 2011). The workload of the
system decreased proportionally with the development of the perennial plants and
after 10years the system required only low maintenance efforts, leaving sufcient
time for the processing of food. After 10years, the regeneration of the site had been
achieved, enabling the provision of a wide variety of food for the subsistence of four
persons, as well as sufcient supply of construction timber and rewood (Schulz
2011). Results suggest that degraded agricultural lands can be regenerated with
SAFS.At the same time, crop diversication and increases in agricultural produc-
tion lead to the regeneration of the subsistence basis of smallholder farmers.
4.2 Toledo District, Belize- Maya Mountain Research Farm
(MMRF)
Maya Mountain Research Farm (MMRF) is a registered NGO and research/training
center in southern Belize that promotes the imitation of primary forest structure and
services to create complex agroforestry systems as a means of ensuring food sover-
eignty and (agro)forest regeneration. MMRF is located approximately 3.2km west
from the predominantly K’ek’chi Mayan village of San Pedro Columbia in the
District of Toledo in the foothills of the Maya Mountains, which range from 244 to
400 m.a.s.l. where limestone hills meet the plains. It is situated between a large
unbroken tract of forest reserve (Columbia Forest Reserve) to the north and the
Columbia branch of the Rio Grande to the south. The population of the Columbia
Forest Reserve’s fringe is composed almost entirely of K’ek’chi and Mopan Maya
Indians. Sixteen villages are considered as being within daily communication dis-
tance, approximately 8km from the management area. All are subsistence and/or
cacao-producing farming communities, practicing some livestock grazing and rear-
ing pigs, chickens, and some ducks. The local population in the Toledo District is
home to the Garifuna, mixed communities with creole, Hispanic, East Indian,
American, European, and further inland to the K’ek’chi and Mopan Maya.
MMRF was founded in 1988 by Christopher Nesbitt (Director of MMRF) rst as
a family homestead and later developed into an applied agroforestry experimenta-
tion farm. MMRF now manages approximately 28ha of land, which functions as a
genetic seed bank for indigenous species and as a buffer zone to the adjacent
Columbia River Forest Reserve with his wife, Celini Logan. The only entrance is by
foot trail or canoe. Across the river from MMRF is K’ek’chi Maya land in a reserva-
tion system, under the “alcalde” (town mayor) system introduced from Guatemala.
In addition to agroforestry development, MMRF managed the Belizean Maya
Ethnobotanical Research Project for the University of Florida, hosts interns, stu-
dents, groups and volunteers, works with local NGO/CGO’s, and, from 1997 to
2004, Christopher managed the Toledo Cacao Growers Association on behalf of the
Green & Black Chocolate Company. MMRF currently provides training in
renewable energy and holds an annual Permaculture Design Course for local,
regional, and international students.
K. J. Young

197
In the 1940s to late 1950s, the land was a “mahogany works” or area just used to
harvest mahogany, being run by a family of creole/East Indian descent. Prior to
MMRF, the site was cleared and planted as a citrus plantation, and used for grazing
cattle until the soil was so leached and compacted that the intensive management
supported very little production. Other disturbances included anthropogenic res
(swidden agriculture) and frequent hurricanes (in 2001, the area was devastated by
Hurricane Iris).
MMRF’s restoration strategy is to emulate the form and function of the primary
rainforest (Fig.8.4). Since 1988, management at MMRF has focused on “induced
patchiness”, with concentrations of species in certain areas to facilitate easier har-
vesting, especially of seasonal crops, and better rates of pollination. MMRF is
divided into different management and restoration strategies dening four different
zones: (i) wamil (secondary forest), (ii) multi-level (“stacked”) polycultures/com-
plex agroforestry systems; (iii) sheep pasture; and (iv) alley cropping. Much of the
food they eat themselves, and most of the food they raise is also used to feed their
animals (pigs, chickens, ducks). By utilizing perennial crops, they maximize the
“calorie production to energy expended” ratio (Heichel 1976). With a total of 28ha,
12 ha are under adaptive management (primarily agroforestry). Of those 12 ha,
approximately 11ha are over 20years old and consist of over 500 mixed species
used for food, ber, biofuels, rewood, precious timber, medicinal crops, market-
able crops, and ornamentals. Approximately 1.4ha has been cleared for sheep pas-
ture. The remaining 12ha are native secondary forest stands. Future plans include
4ha of future sheep pasture for Barbados black belly sheep, comprised of humidi-
Fig. 8.4 Diverse polyculture with native and exotic fruit and timber trees (Photo by Christopher
Nesbitt)
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

198
cola (Brachiaria humidicola) and brizanta (Urochloa brizantha) grass for forage.
Of the 28ha, MMRF actively manages approximately 10ha of land at a time.
Cacao (Theobroma cacao) production at MMRF is the main cash crop, providing
high quality heirloom Mayan cacao varieties for local chocolate production. The
seed parents of these rare varieties were collected by Nesbitt in the upper Bladen
watershed with helicopter support from the British Army in 1999 as a way to address
the devastation of the witch’s broom fungus (Taphrina betulina) on hybrid cacao
varieties introduced by USAID to southern Belizean K’ek’chi and Mopan Mayan
growers in 1986 to foster sales to Hershey Foods. Cacao is an integral part of the
agroforestry system at MMRF, and is placed beneath the shade of many species,
including inga (Inga edulis). Amidst the cacao and inga, caimito (Chrysophylum
caimito and C. mexicanus), banana (Musa spp.), jippy jappa palm (Carludovica
palmata), coconut (Cocos nucifera), biriba (Rollinia deliciosa), avocado (Persea
americana), coffee (Coffea arabica), cardamom (Elettaria cardamomum), bukut
(Cassia grandis), and Pride of Barbados (Ceasalpinia spp.) are planted.
In MMRF’s SAFS, diverse polycultures are created with fruit and timber trees.
Both native timber species such as mahogany (Swietenia macrophylla), cedar
(Cedrela odorata), guanacaste (Enterolobium cyclocarpum), mayower (Epigaea
spp.), and “samwood” or laurel (Cordia alliodora), and exotics such as teak (Tectona
grandis) are utilized to maximize utility and value. Between timber trees, fruit trees
such as mango (Mangifera spp.), avocado (Persea americana), mamey sapote
(Pouteria sapota), Rollinia spp., soursop (Annona muricata), noni (Morinda citrifo-
lia), breadfruit (Artocarpus altilis), breadnut (Brosimum alicastrum), golden plum
(Spondias dulcis), lime (Citrus aurantifolia), etc. are interplanted and, amongst
those, understory species such as coffee (Coffea arabica), cacao (T. cacao), jippy
jappa palm (Cardulovica palmata), coconut (Cocus nucifera) and other species ll
micro-environment niches to maximize utility of sunlight, shade, and leaf litter
decomposition. At the ground level, ginger (Zingiber ofcinale), pineapple (Ananas
comosus), coconut (C. nucifera), turmeric (Curcuma longa), and leguminous plants
like wild peanuts (Arachis pintoi) and Desmodium spp. are planted, and other her-
baceous perennials such as banana and papaya (Carica papaya) serve as pioneer
species, and are placed in the system where conditions are favorable for them, giv-
ing a quick return and providing biomass to the farm when harvested. MMRF dem-
onstrates optimum SAFS management, whereby farmers carefully observe species’
niche growing requirements, growth heights, widths, and rates, to take advantage of
spatial niches and temporal windows of access to sunlight (or shade), rainfall, air-
ow (to prevent fungal disease), and leaf litter decomposition below and amongst
species. In this way, MMRF maximizes use of vertical and horizontal space and
potential for both short- and long-term returns.
In 2013, MMRF pioneered a former “wamil”, or second growth, and converted it
into “food forest” using Erythrina spp., golden plum (Spondias dulcis), banana, polly
redhead (Hamelia erecta), sugarcane (Saccharum barberi), coconut (C. nucifera),
peach palm (Bactris gasipaes), mango (Mangifera spp.), laurel (C. alliodora), and
interplanted with “pioneer-like species” such as cocoyam or taro (Xanthosoma sagit-
tifolium), cassava (M. esculenta), banana (Musa spp.) and plantain (Musa x
K. J. Young

199
paradisiaca). The back is a section of existing agroforestry, formerly citrus, with
cacao, coffee, turmeric, vanilla and ginger, amongst others, underneath.
To achieve their goals, MMRF starts disturbances to create niche environments to
grow their agroforestry species of choice (Fig.8.5). Using a method called “chop and
drop”, existing vegetation is cut by hand (machete) and “dropped” in place to allow
for decomposition and prevent loss of biomass and nutrients from leaving the site.
The newly cleared site is established with leguminous woody shrubs such as pigeon
pea (Cajanua cajan). To mitigate erosion and slow water ow, pineapple (A. como-
sus), lemongrass (Cymbopogon citratus), and vetiver (Chrysopogon zizanioides) are
planted on contour with banana, plantain, and Anjali (Artocarpus hirsutus).
Lemongrass is planted in between rows of pineapple to stagger (or “stack”) comple-
mentary structural proles, and is particularly useful for holding the soil with its deep
taproots, while also providing pest control with high aromatic oil content. Arachis
pintoi is planted in alleys below this. When the rainy season begins, MMRF adds
Caesalpinia spp. Each tree is mulched with lots of biomass (mostly dry leaves, but
also some banana stems, a small amount of biochar and compost) in a “V” shape—a
method known as “Mascarenhas Mulching method” after Kevin Mascarenhas, whom
developed this specic application at MMRF in 2009. This shape slows the move-
ment of water from uphill, dropping nutrients in the form of soil, manure, leaves, dirt,
clay, owers, stems and fruit, in the areas beneath the newly establishing trees where
Fig. 8.5 MMRF restoration strategies and management
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

200
it is needed most. This mulching approach fosters healthy mycorrhizal fungal com-
munities at the intersection of soil and decaying biomass.
Currently, monitoring is done daily (harvesting, site walks, installations), sea-
sonally, and yearly. However, data are limited to qualitative evidence of growth/health
of indicator species, an increase in soil litter, biomass, soil moisture retention and/
or drainage, and forest cover and gap dynamics. Nevertheless, empirical evidence
demonstrates that from 1989–2016, MMRF has achieved an increase in biomass
(critical for carbon sequestration), improved soil fertility and soil moisture reten-
tion, the creation of wildlife habitat in a previously severely disturbed agricultural
landscape, and has mitigated ooding. One critical indication for improved ecosys-
tem functioning was discovered in 2014, when MMRF observed their vanilla plants
had been naturally pollinated and were producing large quantities of valuable vanilla
beans. This suggests the restoration strategies have successfully regenerated the
habitat necessary for the vulnerable bee species (Melipona spp.) associated with
naturally pollinating vanilla.
MMRF demonstrates that agro-reforestation using successional agroforestry
principles can restore ecosystem services, improve soil quality, mitigate soil ero-
sion, and provide food sovereignty. However, limitations in quantitative data hinder
a clear understanding and evaluation of the effects of the management strategies
over time. To improve assessment of the efcacy of management strategies, a con-
sistent monitoring and evaluation protocol should be established. Additionally, mar-
ket development may be addressed by working with pre-existing producer
cooperatives for other non-timber forest products to increase economy of scale at
the landscape level. These cooperatives could also put resources towards coopera-
tively managed tool “libraries”, transportation, and processing into value-added
products (such as dried fruit, etc.). Other potentials for incentivizing the adoption of
successional agroforestry systems for biodiversity conservation and food sover-
eignty include payment for ecosystems services.
4.3 Isla de Ometepe, Nicaragua- Project Bona Fide
Project Bona Fide is a multi-strata agroforestry research and demonstration farm
that seeks to promote food sovereignty and an example of integrated agro- ecological
design systems that mimic native forest structure and function. It is located in the
southwest corner of Nicaragua, on a twin-peaked volcanic island in the tenth largest
fresh water lake in the world (Lago Nicaragua). The island is home to approxi-
mately 40,000 inhabitants. Project Bona Fide seeks to be an experience-based learn-
ing center that aims to inspire local partners and people from all over the world to
undertake their education through hands-on learning experiences with permaculture
design, agroforestry, and community collaboration. Additionally, Project Bona Fide
runs an organic agroforestry nursery with over 250 species of rare and/or endan-
gered tropical fruits, nuts, hardwoods, and other multifunctional tree species. This
nursery serves to provide the farm and community valuable seed and seedling stock,
K. J. Young

201
and is shared with local farmers from the island at their annual seed and plant
exchanges during the rainy season. Saplings from the nursery are also sold to refor-
estation and landscaping projects throughout Nicaragua to earn additional income
to support staff members.
In 2001, Michael Judd puchased a 10.5ha property on the south-eastern side of
Volcán Maderas on the twin-peaked volcanic island Isla de Ometepe. The site had
been abandoned by local farmers, who had cleared the land of timber, and practiced
swidden agriculture (primarily for maize and sorghum) until the land could no lon-
ger support agriculture. It was then heavily grazed with cattle until compacted and
eroded, and no longer provided adequate nutrition for the cows. Judd cover-cropped
approximately 7 ha with cow pea (Vigna unguiculata L. Walp.) and divided the
property into ve different management zones. In 2002, Christopher Shanks joined
Judd to further develop agro-restoration strategies into more than a dozen different
experimental management zones, to assay species suitability in the creation of niche
micro-environments through initial windbreaks and dense plantings (Table8.3).
Project Bona Fide’s management approach for a newly opened site is to create
“triangulated guild systems” to promote nutrient cycling (particularly nitrogen)
through the introduction of leguminous groundcovers, and multifunctional shrubs
and trees; build rapid biomass through selection of short-lived perennials (such as
bananas or pigeon pea) and fast-growing woody species (ideal for rewood), that
can be used to “chop and drop” branches and leaves for rapid in situ mulching
around valuable fruit, nut, or hardwood trees planted on contour. The nitrogen- xing
shrubs or trees (such as Gliricidia sepium) provide wind protection and shade, x
nitrogen, and provide rewood and stakes through coppice and pollarding. The
short-lived perennials also provide shade and windbreaks, and bananas have the
added benet of maintaining moisture in soil, providing moisture-rich organic mate-
rial for mulch, and provide edible fruits. Once the desired trees begin to establish
(approximately the same length of time it takes the bananas to lose their produc-
tion), the bananas are chopped and dropped around the trees and removed. The gaps
in the system are lled with other non-timber forest product species (fruits, nuts,
oils, forage, etc.) and/or native or non-native timber species. Below, Fig.8.6 illus-
trates this “guild” planting technique after one year of planting and seven years later.
Current practices of monitoring and evaluation follow adaptive management
most appropriate to farmers. Daily farm walks, seasonal observations of weather
events (such as heavy rains or extreme droughts) allow the farmers to tailor their
management approach for the following season. Water ow has been directed to
areas of species with higher moisture requirements and, with contour planting, has
helped to keep soils from eroding. After 15years of experimentation with both native
and non-native timber and non-timber forest species, Project Bona Fide provides an
example of how interplanting species of value can provide an extremely biodiverse
agroforestry system that in time resembles a multi-strata forest. Future goals include
developing systems to support the harvest and export of their products.
Like with other SAFS, management is focused on the daily operations of the
farm. Therefore, data available on the performance of this system are limited to
qualitative evidence of growth/health of indicator species, an increase in soil litter,
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

202
Table 8.3 Project bona de experimental zones and species
Experimental
zones (year
began) Principal species Supporting species
Mixed agroforest
(2002)
Malabar chestnut (Pachira aquatic), pejibaye
(Bactris gasipaes), coffee (Coffea arabica, C.
robusta), banana (Musa paradisiaca),
plantain (Musa x paradisiaca), mango
(Mangifera indica), tamarind (Tamarindus
indica), cashew (Anacardium occidentale),
ackee (Blighia sapida), guanabana (Annona
muricata), custard apple (Annona reticulata),
cinnamon (Cinnamomum zelanicum), nispero
(Manilkara sapota), jackfruit (Artocarpus
heterophyllum), canistel (Pouteria
campechiana), star apple (Chrysophyllum
cainito)
Mango with palm
overstory (2002)
Mango (Mangifera indica, var. Ford, Hayden,
and Baptista), palma real (Sabal mexicana),
coconut (Cocos nucifera), pejibaye (Bactris
gasipaes)
Nitrogen-xing trees:
Madero negro (Gliricida
sepium),
Other: madroño
(Calycophyllum
candidissimum), coffee
(Coffea arabica and C.
robusta), and cacao
(Theobroma cacao)
Citrus trials
(2002)
Many Citrus spp. hybrids (grafted) and
seedling trials (3–9 varieties of each): orange,
grapefruit, mandarins, limes, lemons,
kumquat, pomelo, calamondin
Nitrogen-xing trees:
Leucaena spp., Cassia
siamea, Gliricidia sepium,
Delonix regia
Other: Moringa oleifera
Research triangle
(2003)
Grafted citrus, Wampi (Clausena lansium),
Pink wampi (Clausena excavata), Atemoya
(Annona annona x atemoya), Bael fruit
(Aegle marmelos), khirni (Manilkara
hexandra), Natal plum (Carissa grandiora),
Pomegranate (Punica granatum), curry tree
(Murraya koenigii), Bignai fruit (Antidesma
bunius), grafted nispero (Manilkara huberi),
grafted canistel/ross sapote (Pouteria
campechiana), white sapote (Casimiroa
edulis), grafted jackfruit (Artocarpus
heterophyllus)
Nitrogen-xing trees:
Leucaena (Leucaena
leucocephala),
pheasantwood (Cassia
siamea).
Other: Moringa (Moringa
oleifera)
Bamboo alley
windbreak (2005)
Dendrocalamus spp., Bambusa stenostachya,
Gigantachloa atter, Guadua spp., Bambusa
textilis, Bambusa oldhamii, Dendrocalamus
giganteus, Bambusa dolichoclada, and one
unknown Bambusa spp.
Other: Canistel (Pouteria
campechiana), mamey
(Pouteria sapota), ackee
(Blighia sapida), nispero
(Manilkara huberi), and
caimito (Chrysophyllum
caimito)
(continued)
K. J. Young

203
Table 8.3 (continued)
Experimental
zones (year
began) Principal species Supporting species
Guanabana and
avocado (2006)
Guanabana (Annona muricata), avocado
(Persea americana), coconut (Cocos
nucifera)
Nitrogen xing species:
Malinche (Delonix regia),
Erythrina spp., Leucaena
spp., Bauhinia spp
Pig forage system
(2005)
Avocado (P. americana), coconut (C.
nucifera), jackfruit (A. heterophyllus), star
fruit (Averrhoa carambola), canistel (P.
campechiana), guanabana (A. muricata),
jocote (Spondias purpurea), Mayan breadnut
(Brosimum alicastrum), citrus, guajilote
(Paramentiera edulis), Malabar chestnut
(Pachira aquatica).
Erythrina spp., Gliricidia
sepium, Leucaena spp.
Breadfruit and
palmyra (2006)
Breadfruit (Artocarpus altilis) and Palmyra
palm (Borassus abellifer)
Leucaena spp., banana
(Musa spp.), Hibiscus
tiliaceus
Jackfruit and
pejibaye (2007)
Jackfruit (A. heterophyllum), pejibaye
(Bactris gasipaes) Brazilian retree
(Schizlobium parahybum)
Cacao (Theobroma cacao),
False mangosteen (Garcinia
xathochumus), Cassia
siamea, Erythrina spp.,
Albizia guachapele
Resilient salad
perennials (2007)
Chaya (Cnidosculus chayamansa), moringa
(Moringa oleifera), katuk (Sauropus
androgynous), Haitian basket vine
(Trichostigma octandrum), sunset hibiscus
(Abelmoschus esculenta), Brazilian spinach
(Alternanthera sisoo), Okinawan spinach
(Gynura crepioides), cranberry hibiscus
(Hibiscus acetosella)
Mixed agroforest
(2007)
Biriba (Rollinia deliciosa), Custard apple
(Annona reticulata), Amla fruit (Phyllanthus
emblica), Nispero (Manilkara zapota), Rose
apple (Syzygium jambos)
Brazilian retree
(Schizlobium parahybum),
Mountain sugar apple
(Annona spp.)., Jinocuavo
(Bursera simaruba), Yellow
bamboo (Bambusa vulgaris)
Homestead nut
system (2008)
Candlenut (Aleurites molucana), jackfruit
(Artocarpus heterophyllus), Mayan bread nut
(Brosimum alicastrum), pejibaye (Bactris
gasipaes), Malabar chestnut (Pachira
aquatica), rubber tree (Hevea brasiliensis)
Schizlobium parahybum,
coffee (Coffea arabica and
C. robusta)
Sapotaceae
conservation area
(2009)
Pan de vida fruit (Pouteria hypoglauca) Gliricida sepium, Delonix
regia, Leucaena
leucocephala
(continued)
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

204
Table 8.3 (continued)
Experimental
zones (year
began) Principal species Supporting species
Annual alley
croppping (2010)
Leucaena spp., Caesalpinia velutina,
Gliricidia sepium, Acacia albidia, Sabal
mexicana, Delonix regia, Yellow trumpetbush
(Tecoma stans), Madroño (Calycophyllum
candidissimum), vetiver (Chrysopogon
zizanioides), Moringa (Moringa oleifera)
Maize (Zea mays), Sorghum
(Sorghum bicolor), Pigeon
pea (Cajanus cajan), velvet
bean (Mucuna pruriens),
coconut (C. nucifera)
Heritage Jocote
Conservation
(2010)
Spondias purpurea (both cultivated and wild) Native groundcovers
Coffee,
cinnamon, and
coconuts (2010)
Coconut nucifera, Coffea arabica,
Cinnamomum zeylanicum
Turmeric (Curcuma longa),
palma real (Sabal
mexicana), leucaena
(Leucaena leucocephala),
and Brazilian retree
(Schizlobium parahybum)
Oil polyculture
(2010)
African oil palm (Elaeis guineensis), South
American oil palm (Elaeis oleifera), and
Jatropha (Jatropha curcas)
Malinche (Delonix regia),
Gliricidia sepium, pigeon
pea (Cajanus cajan)
Tree cereals trials
(2011)
Malabar chestnut (Pachira aquatica), Maya
nut or ojoche (Brosimum alicastrum), Ackee
(Blighia sapida)
Gliricidia sepium and
Delonix regia
Pulasaan and Pili
nut (2011)
Pili nut (Canarium spp.), Pulasaan
(Nephelium mutabile)
Acacia mangium, Leucaena
spp., cassava (Manihot
esculenta), taro (Colocasia
esculenta)
Surinam cherry
slope (2011)
Surinam cherry (Eugenia uniora) Papaya (Carica papaya),
Citrus spp.
Psidium
collection (2011)
Strawberry guava- yellow and red (Psidium
cattleianum) Cas guava (Psidium
friedrichsthalianum), common guava
(Psidium guajava), Brazilian guava (Psidium
guineense)
Pheasantwood (Cassia
siamea), rose apple
(Syzygium jambos)
Plantain orchard
(2011)
Plantain (Musa x paradisiaca), red sapote,
nispero, canistel (P. campechiana), Bambusa
spp., caimito, and Maya nut (P. aquatica)
Nitrogen xing species:
Malinche (Delonix regia),
Erythrina spp., Leucaena
spp., and Bauhinia spp.
Mesoamerican
tree crop
conservation
(2011)
Maya nut (P. aquatica), canistel, jocote,
caimito, sugar apple (Annona squamosa)
Neem (Azadiracta indica),
elephant grass, nitrogen
xing trees
Rambutan trial
orchard (2012)
Rambutan (Nephelium lappaceum) Leucanea leucocephala,
pheasantwood (Cassia
siamea), Moringa oleifera
(continued)
K. J. Young

205
biomass, soil moisture retention and/or drainage, and forest cover and gap dynam-
ics. Nevertheless, empirical evidence demonstrates that from 2001–2016, Project
Bona Fide has achieved an increase in biomass (critical for carbon sequestration),
improved soil fertility and soil moisture retention, mitigation of erosion and
ooding, wildlife corridor connectivity, the creation of wildlife habitat in a previ-
ously severely disturbed agricultural landscape, and food sovereignty.
Table 8.3 (continued)
Experimental
zones (year
began) Principal species Supporting species
Jackfruit and
nitrogen-xer
windbreak (2012)
Jackfruit (Artocarpus heterophylum), Cassia
siamea, Delonix regia
Taro and Moringa
Agro- silvopasture
(2012)
Native pasture grasses, elephant grass
(Pennisetum purpureum), canistel (P.
campechiana), mamey (P. sapota), red
sapote, jackfruit (A. heterophyllus), nispero
(Manilkara sapota), Mayan breadnut
(Brosimum alicastrum), Malabar chestnut (P.
aquatica)
Delonix regia, Leucaena
spp., Moringa oleifera
Native bamboo
conservation
(2012)
Guadua amplexifolia
Greywater forest
garden (2014)
Taiwan grass (Pennisetum purpureum), Citrus
(Citrus spp.), Soursop (Annona muricata),
vetiver grass (Chrysopogon zizanioides),
banana (Musa paradisiaca), Surinam cherry
(Eugenia uniora)
Acacia mangium and
Flemingia spp.
Fig. 8.6 Triangulated guild technique at initial planting (left) and 7years later (right)
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

206
5 Conclusion
Successional agroforestry has great potential to (i) promote biodiversity in agricul-
tural systems, (ii) reduce risks associated with mono-cropped commodities, (iii)
regenerate degraded agricultural landscapes, and (iv) reconnect fragmented land-
scapes to native forest stands of vulnerable ecosystems. The concept of successional
agroforestry integrates indigenous knowledge of intercropping multi-purpose sub-
sistence species, modern agroforestry and horticultural techniques such as alley
cropping and intensive pruning, and applications of assisted natural regeneration to
emphasize maturity, biodiversity, and the use of ecological succession to establish a
productive forest system. Various succession models and silviculture practices give
us insights into managing SAFS.
Much like the adaptive management of assisted regeneration of forest stands,
mimicking natural ecosystems in agroecosystems requires the knowledge of
species- specic survival, growth, functional traits, and niche requirements in order
to appropriately select multi-functional species and to develop ideal spatial arrange-
ments for stratied stand structures. For its successful adoption, agroforestry man-
agers must have a working knowledge of natural successional development and
tropical forest stand dynamics to manage their system appropriately. Experimental
case studies from SAFS indicate a dearth of quantitative data, critical for under-
standing long-term effects of management approaches. These systems are highly
complex, and knowledge and labor intensive—particularly during the rst ve to
ten years of establishment—until management focuses on thinning, pruning, cop-
picing, and pollarding for improved harvest. Without quantitative evidence of the
socio-economic and ecological benets for adopting a more complex agroforestry
system at the farm level, adoption by local farmers will continue to be limited. Data
on soil erosion, nutrient leaching, soil quality, crop yields, and farmer income before
and after agroforestry development as compared to other local/regional data would
be an important rst step along these lines.
Ultimately, widespread adoption is undoubtedly hindered by poor local and
regional infrastructure (such as processing and transportation), and by a lack of
local, regional, and global market incentives. Markets are currently unable to sup-
port the high biodiversity of farm yields that do not reach economy of scale, thus
attention is needed to harness cooperative approaches to address these economic
and infrastructure limitations. Farmer-to-farmer eld school models, utilized in
more conventional agricultural development, may prove promising for addressing
the knowledge gap to attract new adopters of SAFS at the local and regional lev-
els—thereby achieving economy of scale at the landscape level. Nevertheless,
despite these current limitations, SAFS show great promise as an innovative
approach to increase agro-biodiversity, regenerate severely disturbed agricultural
landscapes, diversify harvest yields, and reduce ecological and economic risks asso-
ciated with conventional agricultural systems.
K. J. Young

207
References
Alvim R, Nair PK (1986) Combination of cacao with other plantation crops- an agroforestry sys-
tem in Southeast Bahia, Brazil. Agrofor Syst 4:3–15
Ashton MS, Ducey MJ (2000) Ch 10: Agroforestry systems as successional analogs to native for-
ests. In: Ashton MS, Montagnini F (eds) The silvicultural basis for agroforestry systems. CRC
Press, Boca Raton, pp207–228
Ashton MS, Larson BC (1996) Germination and seedling growth of Quercus (section
Erythrobalanus) across openings in a mixed-deciduous forest of southern New England,
USA.For Ecol Manag 80:81–94
Ashton MS, Montagnini F (eds) (2000) The silvicultural basis for agroforestry systems. CRC
Press, Boca Raton
Beckerman S (1983) Does the swidden ape the jungle? Hum Ecol 11:1–12
Beer J(1988) Litter production and nutrient cycling in coffee (Coffea arabica) or cocoa (Theobroma
cacao) plantations with shade trees. Agrofor Syst 7:103–114
Bhagwat SA, Willis KJ, Birks HJB, Whittaker RJ (2008) Agroforestry: a refuge for tropical biodi-
versity? Trends Ecol Evol 23:261–267
Callo-Concha D (2009) An approach to environmental services assessment: functional biodiver-
sity in tropical agroforestry systems: the case of Tomé-Açú, Northern Brazil. PhD Diss, Bonn
University. Bonn, Germany
Camp AE, Oliver CD (2004) Ch. 17: Forest dynamics. In: Jeffrey B (ed) The encyclopedia of
forestry. Elsevier Sciences, Oxford, pp1053–1062
Chapin FS, Walker BH, Hobbs RJ, Hooper DU, Lawton JH, Sala OE, Tilman D (1997) Biotic
control over the functioning of ecosystems. Science 277:500–504
Chisholm RA, Muller-Landau HC, Abdul Rahman K, Bebber DP, Bin Y, Bohlman SA et al (2013)
Scale-dependent relationships between tree species richness and ecosystem function in forests.
J Ecol 101:1214–1224
Connell JH (1978) Diversity in tropical rain forests and coral reefs. Science 199:1302–1310
Connell JH, Slatyer RO (1977) Mechanisms of succession in natural communities and their role in
community stability and organization. Am Nat 111:1119–1144
Cornelissen JHC, Lavorel S, Garnier E, Diaz S, Buchmann N, Gurvich DE etal (2003) A hand-
book of protocols for standardized and easy measurement of plant functional traits worldwide.
Aust JBot 51:335–380
Dickinson AK (2014) Analog forestry: creating productive landscapes. In: Towards productive
landscapes, European tropical forest research network news 56, pp103–109
Dufty AC, Senanayake FR, Jack JB, Melvani K (2000) Analogue forestry-a total ecosystem
management approach that maximises biodiversity within plantation agriculture. Planter
76:671–696
Eden MJ (1987) Traditional shifting cultivation and the tropical forest system. Trends Ecol Evol
2:340–343
Egler FE (1954) Vegetation science concepts: I.Initial oristic composition—a factor in old-eld
vegetation development. Vegetatio 4:412–417
Ewel JJ (1999) Natural systems as models for the design of sustainable systems of land use.
Agrofor Syst 45:1–21
Ewel JJ, Bigelow SW (1996) Plant life-forms and tropical ecosystem functioning. In: Biodiversity
and ecosystem processes in tropical forests. Springer, Berlin/Heidelberg, pp 101–126
Geertz C (1963) Agricultural involution: the process of ecological change in Indonesia, vol 11.
University of California Press, Berkeley/Los Angeles, pp1–176
Gleason HA (1939) The individualistic concept of the plant association. Am Midl Nat 21:92–110
Götsch E (1992) Natural succession of species in agroforestry and in soil recovery. http://www.
agrooresta.net/artigos/agroforestry_1992_gotsch.pdf
Halpern CB (1989) Early successional patterns of forest species: interactions of life history traits
and disturbance. Ecology 70:704–720
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…

208
Harris DR (1971) The ecology of swidden agriculture in the upper Orinoc rainforest,Venezuela.
Geogr Rev 61:475–495
Heichel GH (1976) Agricultural production and energy resources: current farming practices
depend on large expenditures of fossil fuels. How efciently is this energy used, and will we be
able to improve the return on investment in the future? Am Sci 64:64–72
Huang W, Luukkanen O, Johanson S, Kaarakka V, Räisänen S, Vihemäki H (2002) Agroforestry
for biodiversity conservation of nature reserves: functional group identication and analysis.
Agrofor Syst 55:65–72
Jones CM (2001) Analog forestry as a conservation and development approach: lessons learned
from the international analogue forestry network. Course paper: international agriculture and
development 200N, University of California-Davis. Davis, California, pp16
Kelty MJ (2000) Ch 9: species interactions, stand structure, and productivity in agroforestry sys-
tems. In: Montagnini F, Ashton MS (eds) The silvicultural basis for agroforestry systems. CRC
Press, Boca Raton, pp183–205
Kozlowski TT (2002) Physiological ecology of natural regeneration of harvested and disturbed
forest stands: implications for forest management. For Ecol Manag 158:95–221
Kraft N, Valencia R, Ackerly DD (2008) Functional traits and niche-based tree community assem-
bly in an Amazonian forest. Science 322:580–582
Laurance SG (2004) Landscape connectivity and biological corridors. In: Agroforestry and bio-
diversity conservation in tropical landscapes, vol 1. Island Press, Washington, DC, pp50–63
Lavorel S, Garnier E (2002) Predicting changes in community composition and ecosystem func-
tioning from plant traits: revisiting the holy grail. Funct Ecol 16:545–556
Lavorel S, Touzard B, Lebreton JD, Clément B (1998) Identifying functional groups for response
to disturbance in an abandoned pasture. Acta Oecol 19:227–240
Lebrija-Trejos E, Pérez-García EA, Meave JA, Bongers F, Poorter L (2010) Functional traits and
environmental ltering drive community assembly in a species-rich tropical system. Ecology
91:386–398
Lohbeck M, Poorter L, Paz H, Pla L, van Breugel M, Martínez-Ramos M, Bongers F (2012)
Functional diversity changes during tropical forest succession. Perspect Plant Ecol Evol Syst
14:89–96
Malézieux E (2011) Designing cropping systems from nature. Agron Sustain Dev 32:15–29
McGill BJ, Enquist BJ, Weiher E, Westoby M (2006) Rebuilding community ecology from func-
tional traits. Trends Ecol Evol 21:178–185
McNeely JA (2004) Nature vs. nurture: managing relationships between forests, agroforestry and
wild biodiversity. Agrofor Syst 61:155–165
McNeely JA, Schroth G (2006) Agroforestry and biodiversity conservation–traditional practices,
present dynamics, and lessons for the future. Biodivers Conserv 15:549–554
Montagnini F, Francesconi W, Rossi E (eds) (2011) Agroforestry as a tool for landscape restora-
tion. Nova Science Publishers, NewYork
Nair PKR (1991) State-of-the-art of agroforestry systems. For Ecol Manage 45:5–29
Nair PKR (1993) An introduction to agroforestry. Kluwer Academic Publishers, Dordrecht, p499
Nations JD, Nigh RB (1980) The evolutionary potential of Lacandon Maya sustained-yield tropi-
cal forest agriculture. JAnthropol Res 36:1–30
Oliver CD, Larson BC (1990) Forest stand dynamics. McGraw-Hill Inc, NewYork, p467
Parrotta JA, Turnbull JW, Jones N (1997) Catalyzing native forest regeneration on degraded tropi-
cal lands. For Ecol Manag 99:1–7
Peneireiro FM (1999) Sistemas agroorestais dirigidos pela sucessão natural: um estudo de caso.
Doctoral dissertation, Universidade de São Paulo, San Pablo, Brazil
Perera AH, Rajapakse RN (1991) A baseline study of Kandyan forest gardens of Sri Lanka: struc-
ture, composition and utilization. For Ecol Manag 45:269–280
Schroth G (ed) (2004) Agroforestry and biodiversity conservation in tropical landscapes. Island
Press, Washington, p523
K. J. Young

209
Schulz J (2011) Ch 1: imitating natural ecosystems through successional agroforestry for the
regeneration of degraded lands—a case study of smallholder agriculture in northeastern Brazil.
In: Montagnini F, Francesconi W, Rossi E (eds) Agroforestry as a tool for landscape restora-
tion. Nova Science, NewYork, pp3–17
Schulz B, Becker B, Götsch E (1994) Indigenous knowledge in a ‘modern’ sustainable agrofor-
estry system-a case study from eastern Brazil. Agrofor Syst 25:59–69
Senanayake FR, Beehler BM (2000) Forest gardens-sustaining rural communities around the
world through holistic agro-forestry. Sustain Dev Int 1:95–98
Senanayake FR, Jack J (1998) Analogue forestry: an introduction. Monash publications in geogra-
phy and environmental science 49. Monash University, Melbourne, pp 1–145
Shono K, Cadaweng EA, Durst PB (2007) Application of assisted natural regeneration to restore
degraded tropical forestlands. Restor Ecol 15:620–626
Somarriba E, Beer J(2011) Productivity of Theobroma cacao agroforestry systems with timber or
legume service shade trees. Agrofor Syst 81(2):109–121
Somarriba E, Valdivieso R, Vásquez W, Galloway G (2001) Survival, growth, timber produc-
tivity and site index of Cordia alliodora in forestry and agroforestry systems. Agrofor Syst
51:111–118
Tilman D (1985) The resource-ratio hypothesis of plant succession. Am Nat 125:827–852
Tilman D (2001) Functional diversity. Enc Biodivers 3:109–120
Tilman D, Knops J, Wedin D, Reich P, Ritchie M, Siemann E (1997) The inuence of functional
diversity and composition on ecosystem processes. Science 277:1300–1302
van Noordwijk M, Ong CK (1999) Can the ecosystem mimic hypotheses be applied to farms in
African savannahs? Agrofor Syst 45:131–158
Vaz P (2000) Regenerative analog forestry in Brazil. Iliea Newsletter 2000:14–16
Vickers WT (1983) Tropical forest mimicry in swiddens: a reassessment of Geertz’s model with
Amazonian data. Hum Ecol 11:35–45
Vieira DL, Holl KD, Peneireiro FM (2009) Agro-successional restoration as a strategy to facilitate
tropical forest recovery. Restor Ecol 17:451–459
8 Mimicking Nature: AReview ofSuccessional Agroforestr y Systems…