Effects of management practices on coffee productivity and herbaceous species diversity in agroforestry systems in Costa Rica

Abstract

Agroforestry systems (AFS) have the potential to provide socioeconomic
benefits and also environmental services such as biodiversity
conservation. In agricultural ecosystems like AFS, plant productivity is
largely dependent on the type and intensity of management. Management
practices like fertilization, pruning, thinning, mulching, and certain
associations between crops and shade tree species have a strong influence
on plant productivity and therefore on crop yields; likewise management
practices have an effect on the biodiversity that occurs in AFS. This
chapter addresses the question of how to manage AFS to increase or
maintain agricultural productivity while conserving biodiversity. We
describe the effects of alternative management practices on coffee yield
and on the herbaceous species richness of an experimental coffee AFS
established in Turrialba, Costa Rica in 2000-2001. The AFS experimental
design was an incomplete factorial design with three repetitions. It
consisted of shaded coffee (Coffea arabica var caturra) plots with three
native tree species alone and in combination: Terminalia amazonia,
Erythrina poeppigiana, and Chloroleucon eurycyclum. The experimental design also included plots of non-shaded coffee monoculture. The
agroforestry plots were managed with either organic or chemical nutrient
inputs at two intensity levels. We quantified herbaceous plant diversity
and recorded coffee productivity (bean yield) across all combinations of
shade tree species and management practices. Our results indicated that:
1. Intermediate management intensity produces competitive coffee yields.
2. Organically managed plots have high herbaceous diversity and can be
as productive as chemically managed plots. 3. Shaded coffee AFS can be
almost as productive as non-shaded, monoculture coffee farming systems.
These results suggest that it is feasible to manage these AFS for
agricultural productivity while maintaining uniform soil cover and a
significant number of herbaceous species.

Full text

Pp. 115-132 In: Montagnini, F., Francesconi, W. and Rossi, E. (eds.). 2011. Agroforestry as a tool for
landscape restoration. Nova Science Publishers, New York. 201pp.
.
Chapter 9
EFFECTS OF MANAGEMENT PRACTICES ON
COFFEE PRODUCTIVITY AND HERBACEOUS
SPECIES DIVERSITY IN AGROFORESTRY
SYSTEMS IN COSTA RICA
Esteban Rossi
∗
∗∗
∗
1,2
, Florencia Montagnini
1
and
Elias de Melo Virginio Filho
3
1
Yale School of Forestry and Environmental Studies
195 Prospect St., New Haven, CT 06511, USA
2
Graduate School of Geography, Clark University
950 Main St., Worcester, MA 01610, USA
3
Tropical Agriculture Research and Higher Education Center (CATIE)
7170, Turrialba, Costa Rica
A
BSTRACT
Agroforestry systems (AFS) have the potential to provide socio-
economic benefits and also environmental services such as biodiversity
conservation. In agricultural ecosystems like AFS, plant productivity is
largely dependent on the type and intensity of management. Management
practices like fertilization, pruning, thinning, mulching, and certain
associations between crops and shade tree species have a strong influence
on plant productivity and therefore on crop yields; likewise management
practices have an effect on the biodiversity that occurs in AFS. This
chapter addresses the question of how to manage AFS to increase or
maintain agricultural productivity while conserving biodiversity. We
describe the effects of alternative management practices on coffee yield
and on the herbaceous species richness of an experimental coffee AFS
established in Turrialba, Costa Rica in 2000-2001. The AFS experimental
design was an incomplete factorial design with three repetitions. It
consisted of shaded coffee (Coffea arabica var caturra) plots with three
native tree species alone and in combination: Terminalia amazonia,
Erythrina poeppigiana, and Chloroleucon eurycyclum. The experimental
∗
Email: esrossi@clarku.edu

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 2
design also included plots of non-shaded coffee monoculture. The
agroforestry plots were managed with either organic or chemical nutrient
inputs at two intensity levels. We quantified herbaceous plant diversity
and recorded coffee productivity (bean yield) across all combinations of
shade tree species and management practices. Our results indicated that:
1. Intermediate management intensity produces competitive coffee yields.
2. Organically managed plots have high herbaceous diversity and can be
as productive as chemically managed plots. 3. Shaded coffee AFS can be
almost as productive as non-shaded, monoculture coffee farming systems.
These results suggest that it is feasible to manage these AFS for
agricultural productivity while maintaining uniform soil cover and a
significant number of herbaceous species.
I
NTRODUCTION
The conservation of biodiversity in transformed or degraded landscapes is
an active research area (Schroth et al. 2004). The effectiveness of alternative
strategies for the preservation of biodiversity is limited by our knowledge of
the mechanisms that maintain diversity and the linkages between diversity and
ecosystem processes (Tilman 2000). Numerous studies on temperate
grasslands, large-scale cross-country projects and controlled laboratory
experiments have shown that biodiversity can have strong effects on plant
productivity and on other ecosystem processes (Tilman et al. 1997; Hector et
al. 1999; Naeem et al. 2000; Loreau et al. 2001; Tilman et al. 2001). In
managed ecosystems biodiversity seems to be inversely related to management
intensity (De Fries et al. 2004). It is important to determine how biodiversity
loss in managed ecosystems will affect the flow of goods and services that
society derives from these ecosystems (Solé and Bascompte 2006).
Agroforestry systems (AFS) have the potential to be productive while
conserving a portion of the biodiversity that occurs in natural ecosystems
(Garrity 2004). Agroforestry can balance the tradeoffs between farmers’
economic needs, ecosystem services and biodiversity conservation (Steffan-
Dewenter et al. 2006). AFS provide a refuge for biodiversity, can supply other
environmental services, and are suitable for agro-ecological research (Ashton
and Montagnini 1999; Bhagwat et al. 2008; Perfecto et al. 1996).
Different management practices and shade tree species can have
contrasting effects on AFS productivity and on associated biodiversity (Nair
1993). In agroforestry experiments plant species diversity, species composition
and management practices can be manipulated or adjusted at a relatively low

Effects of Management Practices on Coffee… 3
cost (Nair 1993; Peeters et al. 2003). Such experiments can advance our
understanding of the effects of different management regimes on agricultural
productivity, biodiversity and other ecosystem processes in managed
ecosystems (Hooper et al. 2005; Perfecto et al. 2005).
Coffee, Coffea arabica L. (Rubiaceae) is a major crop and an important
income source for farmers in Central and South America. It is one of the
world’s most important agricultural commodities and provides economic
support for 20 to 25 million people (http://faostat.fao.org; Aguilar and Klocker
2000; Schroth et al. 2004). There are several advantages in growing coffee in
AFS in comparison with intensive monocultures. For example, it has been
shown that coffee AFS have higher nutrient retention than intensive systems
(Imbach et al. 1989). In addition, in coffee AFS shade improves coffee bean
quality (Muschler 2001). Coffee AFS also require less agrochemical inputs
(Young 2003) and can maintain soil fertility for longer periods of time (Siebert
2002).
Market fluctuations, price speculation, and unfair trade have recently led
many coffee farmers to develop new management practices, experiment with
organic inputs, or switch to other crops (Fernández and Muschler 1999,
Haggar et al. 2001). Rising environmental and health concerns have promoted
the interest in organic production systems and their environmental services
(Rice and Ward 1996). Organically grown coffee can be as productive as
conventional unshaded systems (Lyngbaek et al. 2001). To explore these
claims further, a research team at the Tropical Agriculture Research and
Higher Education Center (CATIE) of Costa Rica established an experiment in
2000-2001 to describe the effects of different management practices and shade
tree species on the productivity and sustainability of coffee AFS (Beer et al.
1998; Haggar et al. 2001; Haggar 2005). These AFS included different
management practices (with chemical and organic inputs) and three native
shade tree species.
The diversity of herbaceous species growing in the understory of an AFS
can serve as an indicator of the effects of different management practices on
plant diversity. Understory plants are directly influenced by management
practices including fertilization, weed control, mulching and shade. The
amount of light that reaches the understory, the availability of nutrients in the
upper layers of the soil and the physical structure of the soil also affect the
growth of understory plants.
This chapter describes the effects of management practices (chemical and
organic), at different intensity levels on understory herbaceous species
diversity and coffee productivity after six years of monitoring. We hypothesize

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 4
that it is possible to obtain competitive coffee bean yields while conserving
herbaceous species diversity.
M
ETHODS
Site Description
The study was conducted at an experimental coffee agroforestry system
located at the Tropical Agriculture Research and Higher Education Center
(CATIE) in Turrialba, Costa Rica (9°53´44´´ N; 83°40´7´´ W). The site is
located at 685 m above sea level, the rainfall averages 2600 mm/year with a
dry season of 1-2 months, and the average temperature is 21.8°C, with 88%
humidity and a solar incidence of 16.9 Mj/m² (Zuluaga Peláes 2004). The
coffee agroforestry system (AFS) is located at CATIE’S experimental and
commercial farm. The AFS is surrounded by sugar cane (Saccharum
officinarum L.) plantations, a few scattered native trees, and pasture lands. The
area had been planted with sugar cane for at least five years before the
establishment of the experiment.
The soils are Typic Endoaquults (Ultisols) derived from volcanic
alluvium. They are shallow, rocky and acidic (pH < 5.5), with high clay
content (clay > 50%). The soils had low to medium organic matter content at
the time of establishment of the experiment. The soil chemical properties were
consistent across the site. The most severe soil limitation was the poor
drainage; several ditches were built across the site to improve drainage. To
reduce acidity, calcium carbonate was applied between the coffee alleys
during the first two years.
Agroforestry System Design and Shade Tree Species
The Coffea arabica var Caturra AFS was established in 2000-2001. The
experiment had a total size of 9.2 hectares. Coffee planting density was 5000
coffee shrubs ha
-1
; with 2 m between rows and 1 m between coffee shrubs,
which is the standard planting density for coffee in Costa Rica. The
experimental treatments were defined as the combination of shade type and
management practices. Treatments were comprised of combinations of three
management interventions: organic or chemical management, input intensity
(quantity of inputs per area), and shade tree species. Not all combinations were

Effects of Management Practices on Coffee… 5
tested, resulting in an experiment with 20 different treatments, each with three
replicates (Table 1).
Table 1. Coffee AFS: The experiment has an incomplete factorial design;
the combination of the shade type and the management practices
constitutes the treatment. There were 20 treatments with 3 replicates each
Shade tree species Management type Management intensity
High intensity Chemical inputs
Medium intensity
Medium intensity
Erythrina poeppigiana
Organic inputs
Low intensity
High intensity Chemical inputs
Medium intensity
Medium intensity
Terminalia amazonia
Organic inputs
Low intensity
Chemical inputs Medium intensity Chloroleucon eurycyclum
Organic inputs Medium intensity
Chemical inputs Medium intensity T. amazonia + C. eurycyclum
Organic inputs Medium intensity
Chemical inputs Medium intensity T. amazonia + E. poeppigiana
Organic inputs Medium intensity
High intensity Chemical inputs
Medium intensity
Medium intensity
C. eurycyclum + E. poeppigiana
Organic inputs
Low intensity
High intensity Full sun (no shade) Chemical inputs
Medium intensity
Three shade tree species were used in the AFS: Chloroleucon eurycyclum
Barneby and J.W. Grimes (Fabaceae) (this species is also reported in the
literature as Abarema idiopoda); Terminalia amazonia (J.F. Gmel.) Exell
(Combretaceae) and Erythrina poeppigiana (Walp.) O.F. Cook (Fabaceae). C.
eurycyclum is a slow-growing, nitrogen-fixing timber tree species; there is
little information on its productivity in AFS (Cordero et al. 2003). T. amazonia
is a fast-growing timber species. It is widely distributed and cultivated across
Latin America and is highly adaptable to numerous soil conditions (Montero
and Kanninen 2005; Montagnini et al. 2008; Piotto et al. 2010). E.
poeppigiana is a nitrogen-fixing species, commonly used in Costa Rica in

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 6
several types of AFS (Beer et al. 1998). Shade trees were planted in alternate
rows at 6 m x 4 m with an initial shade tree density of 417 trees ha
-1
, and they
were thinned 4 years after planting. In 2008, when the present study was
conducted, the tree density had been reduced by 40%. E. poeppigiana trees
were pollarded according to conventional coffee AFS management practices to
add organic matter to the soil. C. eurycyclum and T. amazonia were pruned to
promote straight stem growth. No genetic selection or progeny tests have been
done with the selected tree species, therefore available planting materials
displayed high genetic variability (Galloway and Beer 1997).
Coffee Management Practices
Two different kinds of management practices were used in the coffee
AFS: chemical (also referred to as conventional) and organic (Lyngbaek et al.
2001). Conventional chemical management was based on agronomic
recommendations of the Costa Rica Coffee Institute, and included the use of
chemical fertilizers, fungicides and pesticides at two intensity levels: high (800
kg ha
-1
yr
-1
of chemical input) and medium (400 kg ha
-1
yr
-1
). These practices
are commonly used in unshaded coffee systems as well as in coffee AFS with
a single shade tree species (Sanchez-De León et al. 2006).
In plots with organic management chicken manure and organic matter
(coffee pulp) were used at two intensity levels: medium and low. The medium
intensity level consisted of 20 tons ha
-1
yr
-1
of coffee pulp, 7.5 tons ha
-1
yr
-1
of
chicken manure, 200 kg ha
-1
yr
-1
of a potassium-magnesium-sulfate mixture
(KMAG), and 200 kg ha
-1
yr
-1
of phosphate mineral. The low intensity level
consisted of 10 tons ha
-1
yr
-1
of coffee pulp, chicken manure and phosphate
mineral (Sanchez-De León et al. 2006). Pests were controlled by removing
infected fruits and by selective pruning. A committee of local coffee farmers
and agronomists revised and approved the chosen management procedures.
These practices represent local knowledge and aim to become agronomically
and economically feasible alternatives for small farmers (Sanchez-De León et
al. 2006).
Understory Species Richness, Composition and Abundance
To estimate the understory plant diversity a detailed survey of herbaceous
vegetation in the AFS was conducted. Preliminary field observations and local

Effects of Management Practices on Coffee… 7
sources indicated that herbaceous species richness and cover were
homogeneous within treatments. In each treatment/plot a 4 m x 4 m subplot
was located randomly. Within each subplot all plant individuals were
identified and counted to estimate understory species richness and abundance.
The herbs were identified to species level with the aid of local and regional
floras, photographic databases, and online herbarium databases. When reliable
identification to the species level was not possible, the individuals were
classified as morpho-species. The herbaceous species that displayed colonial
growth form (asexual reproduction) and the grasses that could not be identified
as individuals were recorded by measuring the size of the covered area in
meters. The area of litter, bare soil and herbaceous cover in each 16m
2
subplot
were measured using a one meter square frame.
Coffee Productivity Evaluation
The coffee yield data were obtained from harvest records. For each
treatment plot, the weight of the total coffee berry harvest was recorded every
year. Total averages per treatment were obtained from the three replicates of
each treatment. Yield data were compared on a per-area and per-plant basis to
correct for potential differences in plot sizes. These measures were used to
compare the yield data across treatments.
Data Analysis
To describe the effects of management practices on coffee bean
productivity and on herbaceous species diversity in the AFS we used the
repeated measures ANOVA procedure with a general linear model in SAS. We
sought to compare differences in coffee yield across treatments, differences in
understory plant richness across treatments, and to determine how coffee bean
yields were changing in time. We used dynamic tables in Excel to compare
yields across treatments, years, management practices, and shade types.
R
ESULTS
Herbaceous Species Richness, Abundance and Soil Cover

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 8
Our results indicated that management practices had a strong effect on
herbaceous species richness. Herbaceous species richness was consistently
lower under high intensity chemical treatments regardless of shade tree species
(Figure 1).
Figure 1. Herb diversity and coffee productivity along the management intensification
gradient (data from 2007). Grey bars correspond to coffee yield and are arranged by
management practices (from low-intensity organic to high-intensity chemical) and by
shade type. The shade tree species are: Ta-Terminalia amazonia; Ce-Chloroleucon
eurycyclum; Ep-Erythrina poeppigiana; the four letters correspond to shade tree
species combinations and FS to full sun grown coffee. The dots correspond to the
mean herb species richness per three 16m
2
sub-plots for each treatment. The dots were
displayed in a continuous line to show how the relationship between herb species
richness with management practices.
The high intensity chemically managed plot (CH) with T. amazonia
presented the lowest herb richness (2 species-mean of three 16m
2
plots) of the
AFS. The full sun treatments (FS) also had low herbaceous species richness.
The herbaceous ground cover was consistently higher in organic treatments
(OL, OM) followed by CM treatments.
The plots managed with medium intensity (Chemical Medium-CM and
Organic Medium-OM) displayed the highest herbaceous species richness with
up to 18 species (mean of three 16m
2
plots) in CM and OM with both E.
poeppigiana and C. eurycyclum (Figure 1). The CM and OM treatments with
C. eurycyclum also had consistently high species richness values. Both CM
and OM treatments also had high variability in species richness.

Effects of Management Practices on Coffee… 9
Regarding the influence of shade tree species, treatments with C.
eurycyclum had the highest herbaceous species abundance, high herbaceous
species richness and a very small area of exposed soil (Table 2). The
treatments with both C. eurycyclum and E. poeppigiana had high or
intermediate values of herb and litter soil cover and also small areas of
exposed soil.
The treatments shaded with E. poeppigiana had high litter production. The
unshaded (FS) treatments had the highest total area of bare soil; almost no
herb species grew under the high intensity, full sun management practices.
Over 90% of the herbaceous species sampled were identified to species;
the Poaceae (grasses) and the Cyperaceae (sedges) were the most abundant
families in the coffee AFS. Grasses were highly abundant in FS treatments
(Table 3) and under conditions of FS and high chemical fertilization. The
abundance of Cyperaceae and the observations in the field confirmed that
limited drainage on this site favors water tolerant species in the Cyperaceae
and Caryophillaceae families. Seedlings and saplings of coffee, E.
poeppigiana and A. idiopoda were also abundant across plots. Borreria sp
(Rubiaceae), Drymaria sp (Caryophyllaceae) and especially Hydrocotile sp
(Apiaceae) provided uniform and, in some plots, extensive soil cover.
Coffee Yields
Overall, the high intensity chemical treatments (CH) displayed
considerable variation in productivity between treatments, shade type and
years. In 2007 the highest yield was recorded in full sun (FS) chemical high
input (CH) plots with yield over 13 Mg/ha.

Table 2. Coffee AFS soil cover: herb richness (S), herb abundance and % area by cover types. Mean and +/- square
error of the mean (SEM) from three 16m
2
plots per treatment
Shade tree
species Management Mean richness
(#species)
Species richness
(SEM)
Mean abundance
(indiv/ plot)
Abund.
(SEM)
Plot herb
cover
(%)
Plot litter
cover (%)
Bare soil
area (%)
Ce Med. Chem. 15 3.46 109.67 42.74 45.7 32.0 16.3
Ce Med. Organic 13 6 56 43.09 77.5 14.2 0.0
Ce + Ep High Chem. 4.7 0.58 72.33 108.89 9.4 56.3 31.3
Ce + Ep Low Organic 17.7 1.53 71.33 57.87 68.0 20.7 5.9
Ce + Ep Med. Chem. 17.3 2.31 46.33 1.53 39.3 37.3 14.6
Ce + Ep Med. Organic 12.7 1.15 25.33 4.04 79.2 7.3 1.1
Ce + Ta Med. Chem. 15 1.73 34.17 2.57 11.6 55.2 32.1
Ce + Ta Med. Organic 13 2 29.67 15.31 77.7 17.1 1.6
Ep High Chem. 8 3.24 70.25 11.21 11.4 88.2 0.0
Ep Low Organic 9.7 5.51 22.67 17.62 60.2 35.4 0.0
Ep Med. Chem. 8.7 5.13 37 23.64 16.3 76.1 3.1
Ep Med. Organic 11 1 32.67 17.39 51.7 37.9 0.0
Ep + Ta Med. Chem. 13 2.65 49.33 7.09 14.1 65.4 13.1
Ep + Ta Med. Organic 8.3 1.15 21.33 13.65 71.9 12.5 7.9
Ta High Chem. 2.3 1.53 85.67 39.37 2.3 92.7 2.1
Ta Low Organic 13 3 53.33 20.82 85.3 10.9 3.6
Ta Med. Chem. 9.7 2.08 48.33 41.53 17.2 66.5 15.8
Ta Med. Organic 10.7 3.79 30.67 5.77 80.8 9.4 0.0
FS High Chem. 6.7 1.86 52.67 33.95 17.8 49.6 28.8
FS Med. Chem. 10.3 6.03 21 16.52 52.1 25.0 15.6

Table 3. Most common species and number of occurrences in the 4 x 4 m sampling plots
Family Species Occurrences Taxonomic summary
Caryophyllaceae Drymaria cordata 30
Number of plant families recorded in the
experiment: 27
Poaceae Paspalum conjugatum 30
Number of herb genera recorded in the
experiment: 50
Rubiaceae Coffea arabica (saplings) 30 Total Number of herb species: 58
Apiaceae Spananthe paniculata 27
Poaceae Digitatia sanguinalis 27 Most abundant herb families
Poaceae Paspalum conjugatum 27 Cyperaceae (sedges): 8 species
Apiaceae Hydrocotyle umbellata 26 Poaceae (grasses): 6 species
Rubiaceae Borreria laevis 26
Cyperaceae Cyperus tenuis 24 Most common genus: Cyperus
Euphorbiaceae Phyllantus niruri 24
Cyperaceae Cyperus luzulae 23
Cyperaceae Cyperus tenuis 23
Leg-Mimosaceae Mimosa pudica 22
Cyperaceae Dichromena ciliata 21
Leg-Fabaceae Erythrina poeppigiana (saplings) 21
Asteraceae Pseudoelephantopus spicatus 20

Esteban Rossi, Florencia Montagnini and Virginio Filho Elias de Melo 12
Figure 2. Coffee AFS mean bean yields for 2006. The bars show the yield per plots
arranged by management practices (low intensity organic to high intensity chemical)
and by shade type. The line corresponds to the 2006 mean coffee bean yield across all
treatments.
Plots with E. poeppigiana and the combinations of E. poeppigiana and
Chloroleucon eurycyclum produced the highest yields after the full sun CH
treatment (Figure 1-3). Plots shaded with the timber tree Terminalia amazonia
tended to have lower coffee yields with the exception of one OM plot.
In contrast, at intermediate management levels (CM, OM) the yields
displayed less variation between years and shade types (Figure 1-3). The
comparison between chemical medium and organic medium treatments
showed that at intermediate management levels the total yields were very
similar, regardless of shade tree species, or input type. Furthermore, in 2007,
with the exception of full sun CM treatments, medium organic treatments had
the highest total yield. The low intensity organic treatments (OL) had the
lowest coffee yields.
Because coffee presents a biannual fruiting pattern we show the results of
the last two years on record, 2006-2007 (Figures 1-2); this biannual pattern is a
physiological trait of the coffee shrub and is largely independent of the site
conditions. The year 2007 was a high yield year (Figure 1) and 2006 was a low
yield year (Figure 2). Our ANOVA showed a very strong effect of time on
productivity (Table 4), this effect was probably caused by the year-to-year
evolution of the AFS and the gradual increase in yields as the coffee shrubs
matured (Figure 3). We also observed a small positive effect of understory

Effects of Management Practices on Coffee… 13
diversity on coffee yield (Table 4). These yield data indicate that the AFS
productivity has been increasing gradually over the last 6 years and that there
is a significant effect of management on productivity as can be seen in Figures
1 and 2.
Figure 3. Historical mean yields of the coffee AFS grouped by management practices
for 2002-2007. Organic treatments (black thick lines) display smaller inter-year
oscillations than chemical treatments.
Table 4. Time, management and herb richness effects on coffee yield
Source of variation Pr > F
Time <.0001
time * management <.0001
time * herb richness 0.0020
time * herb richness * management 0.0044
time * block 0.9299
Chemically managed plots at high and medium intensity levels were
highly productive by the second harvest (2003) and displayed considerable
variation in inter-year yield. Interestingly, in organically managed plots mean
yield increased slowly (with low productivity in 2002 and 2003) but displayed

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 14
smaller inter-year variation. In the years 2004 and 2006 organic treatments
were as productive, or in some cases, more productive than chemical
treatments (Figure 3).
The productivity of the system in 2006 indicates that in low yield years,
the CM plots tend to perform poorly, especially the full sun plots (mean yield
for the full sun CH was 4.2 Mg/ha in 2006 and 13.5 Mg/ha in 2007). In
contrast, organically managed plots have higher yields than chemically
managed plots during low yield years; in 2006 medium intensity organic
treatments were significantly more productive than medium intensity chemical
(CM) plots (Figure 3). Also the year-to-year yield variability was higher for
CH than for the CM plots (Figure 3).
D
ISCUSSION
Impact of Management Intensity on Coffee Yield and on
Herbaceous Species Diversity
In general, understory herbaceous species diversity was positively
correlated with coffee yield, the only exceptions were the intensive treatments
– especially the unshaded, high intensity chemically managed treatments. This
suggests that coffee productivity and herbaceous species richness can be
maintained simultaneously in coffee AFS, especially at intermediate levels of
management intensity.
Also, it could be argued that biodiversity can have a small positive effect
on productivity, as shown by Figures 1-2. Given that some management
practices like the pollarding of E. poeppigiana tended to have negative impacts
on herbaceous growth it is interesting that a small biodiversity effect could still
be detected.
It is possible that in these AFS herbaceous diversity influences coffee
productivity in an indirect manner: for example, herbaceous cover could be
contributing to erosion control, addition of organic matter to the soil, and
enhanced nutrient cycling (Altieri 1999; Smith et al. 2008; Vandermeer 1989).
Other studies have reported that as management intensity increases,
biodiversity tends to decrease (DeFries et al. 2004). Our results indicated that a
sharp decline in herbaceous species richness was associated with chemically
intensive management practices, while medium intensity treatments (CM and
OM) produced similar coffee yields; therefore medium-intensity management

Effects of Management Practices on Coffee… 15
practices may represent a reasonable balance between coffee productivity and
biodiversity conservation for farmers in Costa Rica.
Our data indicated that as in other agricultural systems in the region the
Poaceae (grasses) were very abundant (Table 3). While some species of
Poaceae can benefit commercial crops and livestock, most grass species are
considered undesirable in coffee AFS. In the full sun AFS treatments,
controlling undesirable grasses commonly required the use of chemical
herbicides; while in shaded plots undesirable species were less frequent
(Romero, pers. com).
Whether the optimal management practices for coffee AFS should be
totally organic, chemical or a mixture of the two would depend on the
availability and use of organic and chemical inputs, the management
objectives and the preferences of the land manager. It is likely that as organic
technologies advance, the quality of organic inputs will improve. In some
tropical regions small farmers can easily produce their own high quality
organic inputs. Improved organic technology has the potential to increase the
profitability of these agroforestry systems. For example, in the Turrialba area
of Costa Rica, organic coffee farmers receive higher prices for their coffee
than conventional farmers. This price premium constitutes an incentive for
improved management practices but sometimes is not enough to cover
certification and other costs associated with organic production (Muschler
2001).
Influence of Shade Tree Species
We observed that the shade tree species C. eurycyclum apparently had a
positive effect on herb species richness and on coffee productivity. Its
relatively open canopy allows some light to reach the understory, making it a
potentially desirable tree species for AFS (Figure 4); also, preliminary
observations suggest that C. eurycyclum has root nodules and therefore could
be actively fixing nitrogen. However, the C. eurycyclum in the AFS had poor
form and shape and probably lacked quality for saw timber. Local farmers
reported that the high density of the wood made pruning difficult.
Chloroleucon eurycyclum is a promising shade tree species, however more
research and genetic selection is needed to obtain more benefits from its
association with coffee in AFS.

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 16
Figure 4. Coffee agroforestry system shaded with Chloroleucon euryclyclum.
The treatments with Erythrina poeppigiana confirmed the benefits of the
use of this well-known shade tree species in Costa Rica (Beer et al. 1998).
Although the treatments shaded with E. poeppigiana were productive, the
herbaceous species richness under E. poeppigiana was low, probably because
the abundant tree mulch suppressed the herbs in the understory. In combined
plots with E. poeppigiana and C. eurycyclum the overall herb richness was
much higher than in the treatments with E. poeppigiana shade alone (Figure
5).
The treatments shaded with Terminalia amazonia (Figure 6) presented
low herbaceous species richness and intermediate coffee yields. This species
of relatively valuable timber could probably be managed for timber and shade
in combination with other species. Overall, the combination of shade tree
species probably provides farmers with more benefits: C. eurycyclum provides
intermediate shade and organic matter to the soil, T. amazonia produces
straight, high quality timber and E. poeppigiana produces mulch and can be
pruned easily to adjust light conditions to the needs of the coffee shrubs
(Figure 7).

Effects of Management Practices on Coffee… 17
Figure 5. Coffee agroforestry system with pruned Erythrina poeppigiana.
Figure 6. Coffee agroforestry system shaded with Terminalia amazonia.

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 18
Figure 7. Coffee agroforestry system with Erythrina poeppigiana and in the back
Chloroleucon eurycyclum.
Long Term Sustainability of Coffee AFS
The full sun CH treatments had highly variable yields from year to year
and had very low herbaceous species richness. These unshaded CH treatments
can also have high erosion rates depending on the slope and soil type (Rice
1991). Often, due to the costs of chemical inputs these intensive systems are
not suitable for small farmers.
The coffee yield year to year records suggest that the AFS is maturing and
that yields are increasing with time. The organic treatments seem to be slowly
attaining productivity levels similar to those of chemical treatments. In
addition, coffee yields were less variable year to year in the organic than in the
conventional system.
This could indicate that organic treatments can have positive effects on the
nutrient cycles in these AFS. As the system matures and soil processes
continue, bean yields and biodiversity can increase and year to year variations
can decrease. Overall, these trends suggest that these coffee AFS can be more
sustainable than intensive coffee monocultures.

Effects of Management Practices on Coffee… 19
C
ONCLUSION
This study shows that coffee AFS can be managed for competitive coffee
yields while maintaining herbaceous species diversity. Our data showed that
low intensity management tended to produce poor yields, while high intensity
management has low diversity and can be expensive. Thus, an optimal
agroforestry management strategy for coffee productivity and biodiversity can
probably be attained using organic inputs at intermediate intensity and
complementing them with chemical inputs if needed. The data also
demonstrated that as the organic AFS mature they become more productive
and probably more resilient.
The comparisons of different shade tree species confirmed the benefits of
Erythrina poeppigiana in coffee AFS. Chloroleucon eurycyclum had positive
impacts on diversity and coffee productivity and could be successfully
combined with other species to provide shade in AFS.
The adaptability of the timber species Terminalia amazonia makes it a
useful tree species for coffee AFS. Terminalia amazonia benefits can probably
be maximized if it is planted in combination with other shade tree species that
can contribute with nitrogen fixation such as C. eurycyclum and E.
poeppigiana.
This management strategy, combination of timber and nitrogen fixing
species in organic coffee AFS can be easily transferred to farmers in Costa
Rica and other countries in Central America.
This is especially relevant in regions where farmers have poor access to
chemical inputs and can benefit by selling organically produced coffee. In
addition, the organic systems can promote restoration and conservation of
biodiversity.
A
CKNOWLEDGEMENTS
This study was supported by the Tropical Resources Institute of the Yale
School of Forestry and Environmental Studies. B. Elizondo, I. Romero, and C.
Finney provided help and suggestions. The staff at CATIE provided valuable
logistical support and assistance.

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 20
R
EFERENCES
Aguilar, B. and Klocker, J. (2000). The Costa Rican coffee industry. In: C. S.
Hall, P. Van Laake, G. Leclerc, and C. Leon Perez (Eds.), Quantifying
Sustainable Development: The Future of Tropical Economies (pp. 595-
627). New York: Academic Press.
Altieri, M. (1999). The ecological role of biodiversity in agroecosystems.
Agriculture, Ecosystems and Environment, 74, 19–31.
Bhagwat, A., Willis, K. J., Birks, J. B., and Whittaker. R. J. (2008).
Agroforestry: a refuge for tropical biodiversity? Trends in Ecology and
Evolution, 23, 261-267.
Beer, J., Muschler R., Somarriba E. and Kass D. (1998). Shade management in
coffee and cacao plantations - a review. Agroforestry Systems, 38, 139-
164.
Cordero, J., Mesén, F., Montero, M., Stewart, J., Dossier, D., Chanberlain.,
Pennington, T., Hands, M., Hughes, C., and Detlefsen, G. (2003).
Descripciones de especies de 90 árboles nativos de América Central. In J.
Cordero and D. H. Boshier (Eds.), Árboles de Centro América: un Manual
para Extensionistas (pp. 311-958). Oxford, UK: FRP, OFI/CATIE.
DeFries, R., Foley J. A., and Asner, G. P. (2004). Land-use choices: balancing
human needs and ecosystem function. Frontiers in Ecology and the
Environment, 2, 249-257.
Fernández, C. and Muschler R. (1999) Aspectos de la sostenibilidad de los
sistemas de cultivo de café en América Central. In B. Bertrand and B.
Rapidel (Eds.), Desafíos de la Caficultura en Centroamérica (pp 69-96).
San José, Costa Rica: Inter-American Institute for Cooperation on
Agriculture.
Garrity, D. P. (2004). Agroforestry and the achievement of the Millennium
Development Goals. Agroforestry Systems, 61, 5-17.
Galloway, G. and Beer J. (1997). Oportunidades para fomentar la silvicultura
en cafetales de América Central. Turrialba, Costa Rica: CATIE-GTZ.
Haggar, J., de Melo E., and Staver C. (2001). Sostenibilidad y sinergismo en
café agroforestal: Un estudio de interacciones entre las plagas, la
fertilidad del suelo y el estrato árbol. Turrialba, Costa Rica: CATIE,
Programa Regional MIP/AF.
Haggar, J. (2005). Investigación regional para una caficultura ecológica y
diversificada. Series Divulgativas. Turrialba, Costa Rica: Centro
Agronómico Tropical de Investigación y Enseñanza para el desarrollo y la
conservación (CATIE).

Effects of Management Practices on Coffee… 21
Hector, A., Schmid, B., Beierkuhnlein, C., Caldeira, M. C., Diemer, M.,
Dimitrakopoulos, P. G., Finn, L. A., Freitas, H., Giller, P. S., Good, J.,
Harris, R., Högberg, P., Huss-Danell, K., Joshi, J., Jumpponen, A.,
Körner, C., Leadley, P. W., Loreau, M., Minns, A., Mulder, C. P. H.,
O'Donovan, G., Otway, S. J., Pereira, J. S., Prinz, A., Read, D. J., Scherer-
Lorenzen M., Schulze, E. D., Siamantziouras, A. S. D., Spehn, E. M.,
Terry, A. C., Troumbis A. Y., Woodward F. I., Yachi, S., and Lawton J.
H. (1999). Plant Diversity and productivity experiments in European
Grasslands. Science, 286, 1123-1127.
Hooper, D. U., Chapin, F. S., Ewel, J. J., Hector, A., Inchausti, P., Lavorel, S.,
Lawton, J., H., Lodge, D. M., Loreau, M., Naeem, S., Schmid, B., Setälä,
H., Symstad, A. J., Vandermeer, J. and Wardle D. A. (2005). Effects of
biodiversity on ecosystem functioning: a consensus of current knowledge.
Ecological Monographs, 75, 3–35.
Imbach, A. C., Fassbender, H. W., Borel, R., Beer, J. and Bonnemann, A.
(1989) Modelling agroforestry systems of cacao (Theobroma cacao) with
laurel (Cordia alliodora) and cacao with poro (Erythrina poeppigiana) in
Costa Rica IV. Water balances, nutrient inputs and leaching. Agroforestry
Systems, 8, 267–287.
Loreau, M., Naeem S., Inchausti, P., Bengtsson, J., Grime, J. P., Hector, A.,
Hooper, D. U., Huston, M. A., Raffaelli, D., Schmid, B., Tilman, D., and
Wardle, D. A. (2001). Biodiversity and Ecosystem Functioning: Current
Knowledge and Future Challenges. Science, 294, 804-808.
Lyngbaek A. E., Muschler R. G., and Sinclair F. L. (2001). Productivity and
profitability of multistrata versus conventional coffee farms in Costa Rica.
Agroforestry Systems, 53, 205-213.
Montagnini, F., Cerezo, A., Lam Bent, H. C., Finney, C., and Kim, T. J.
(2008). Reforestación para control de pastos invasores y protección de
cuencas hidrográficas en el Canal de Panamá. Reforestation for control of
invasive grass and watershed protection in the Panama canal. Yvyraretá
(Argentina) 15: 33-38.
Montero, M. and Kanninen, M. (2005). Terminalia amazonia: ecología y
silvicultura. Turrialba, Costa Rica: CATIE-CIFOR.
Muschler, R. G. (2001). Shade improves coffee quality in a sub-optimal
coffee-zone of Costa Rica. Agroforestry Systems, 85, 131-139.
Naeem, S., Hann, D. R., and Schuurman, G. (2000). Producer-decomposer co-
dependency influences biodiversity effects. Nature, 403, 762-764.
Nair, R. P. K. (1993). An introduction to agroforestry. London: Kluwer
academic Publishers.

Esteban Rossi, Florencia Montagnini and Elias de Melo Virginio Filho 22
Peeters, L., Soto Pinto L., Perales, H., Montoya, G. and Ishik, M. (2003).
Coffee production, timber, and firewood in traditional and Inga-shaded
plantations in Southern México. Agriculture, Ecosystems and
Environment, 95, 481-493.
Perfecto, I., Rice, R. A., and Greenberg, R. (1996). Shade coffee: a
disappearing refuge for biodiversity. BioScience, 46, 598–608.
Perfecto, I., Vandermeer, J., Mas A. A., and Soto-Pinto, L. (2005).
Biodiversity, yield, and shade coffee certification. Ecological Economics,
54, 435–446.
Piotto, D., Craven, D., Montagnini, F., and Alice, F. (2010). Silvicultural and
economic aspects of pure and mixed native tree species plantations on
degraded pasturelands in humid Costa Rica. New Forests, 39, 369-385.
(Also published online, DOI 10.1007/s11056-009-9177-0).
Rice, R. A.. (1991). Observaciones sobre la transición en el sector cafetalero
en Centroamérica. Agroecología Neotropical, 2, 1-6.
Rice, R. A. and Ward, W. R. (1996). Coffee, conservation, and commerce in
the western hemisphere. Washington DC: Smithsonian Migratory Bird
Center and Natural Resources Defense Council.
Sánchez-de Leon, Y., De Melo, E., Soto, G., Johnson-Mayanard, J., and Javier
Lugo-Pérez, J. (2006). Earthworm populations, microbial biomass and
coffee production in different experimental agroforestry management
systems in Costa Rica. Caribbean Journal of Science, 42, 3, 397-409.
Schroth, G., da Fonseca, G., Harvey, C. A., Gascon, C., Vasconcelos, H., L.,
and Izac, A., N., (2004). Agroforestry and Biodiversity Conservation in
Tropical Landscapes. Washington, DC: Island Press.
Siebert, S. F. (2002). From shade-to sun-grown perennial crops in Sulawesi,
Indonesia: implications for biodiversity conservation and soil fertility.
Biodiversity and Conservation, 11, 1889-1902.
Smith, R. G, Gross, K. L., and Robertson, P. G. (2008). Effects of crop
diversity on agroecosystem function: crop yield response. Ecosystems, 11,
355-366.
Steffan-Dewenter, I., Kessler, M., Barkmann, J., Bos, M. M., Buchori, D.,
Erasmi, S., Faust, H., Gerold, G., Glenk, K., Gradstein, S. R., Guhardja,
E., Harteveld, M., Hertel, D., Höhn, P., Kappas, M., Köhler, S.,
Leuschner, C., Maertens, M., Marggraf, R., Migge-Kleian, S., Johanis
Mogea, J., Ramadhaniel Pitopang, R., Schaefer, M., Schwarze, S., Sporn,
S. G., Steingrebe, A., Tjitrosoedirdjo, S. S., Tjitrosoemito, S., Twele, A.,
Weber, R., Woltmann, L., Zeller, M., and Tscharntke, T. (2007). Trade-
offs between income, biodiversity, and ecosystem functioning during

Effects of Management Practices on Coffee… 23
tropical rainforest conversion and agroforestry intensification. Proc. Nat.
Acad. Sci., 104, 4973–4978.
Solé, R. and Bascompte J. (2006). Self organization in complex ecosystems.
Monographs in population Biology 42. Princeton, NJ: Princeton
University Press.
Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M. and Siemann E.
(1997). The influence of functional diversity and composition on
ecosystem processes. Science, 277, 1300-1302.
Tilman, D., Reich, P., Knops, J., Wedin, D., Mielke, T., and Lehman, C.
(2001). Diversity and productivity in a long term grassland experiment.
Science, 294, 843-845.
Tilman, D., Cassman K. G., Matson, P. A., Naylor, R., and Polasky, S. (2002).
Agricultural sustainability and intensive production practices. Nature, 418,
671-677.
Vandermeer, J.H. (1989). The ecology of intercropping. Cambridge, UK:
Cambridge University press.
Young, C. (2003). Coffee Agroforestry Systems for Conservation and
Economic Development: A Case study of the Amisconde Initiative in a
Buffer Zone Community of Costa Rica. Journal of Sustainable Forestry,
16, 39-63.