From the Ground Up: Herbaceous Community Diversity and ...

From the Ground Up: Herbaceous Community Diversity and Management in Coffee Agroforestry Systems

by

Sarah Archibald

A thesis submitted in conformity with the requirements for the degree of Master of Science

Department of Geography University of Toronto

© Copyright by Sarah Archibald 2019

ii

From the Ground Up: Herbaceous Community Diversity and Management in Coffee Agroforestry Systems

Sarah Archibald

Master of Science

Geography Department University of Toronto

2019

Abstract

The herbaceous community (HC) is an understudied yet critical aspect of tropical

agroecosystems. I measured the diversity and perceptions of the HC within organic coffee

systems in the Central Valley of Costa Rica. The HC was taxonomically and functionally

diverse; comprised of 39 species from 20 taxonomic groups. Farms below the regional mean size

and those with canopy openness of 20-30% had higher HC functional diversity. Farmers

perceived tall species with low SLA and LNC, but high height and LDMC to be undesirable, due

to slow decomposition rates and management limitations. Farmers’ cognitive map complexity

was positively related to HC functional richness, and negatively related to functional evenness

and functional dissimilarity. All farmers placed higher emphasis on soil health and organic

matter than coffee yield, which may be indicative of their role as land stewards. Workshops are

needed to disseminate HC management information to optimize labour and ecosystem

functioning.

iii

Acknowledgments There are so many people and plants who have made my masters an incredibly meaningful

experience. First of all, I am very grateful to my thesis supervisor, Dr. Marney Isaac, for her

ongoing mentorship, encouragement and thoughtful insight into my research project. Thanks to

my committee members, Dr. Ryan Isakson and Dr. Scott MacIvor for providing their

knowledgeable feedback to strengthen my work. It was a privilege to work with the Isaac

Agroecology Lab team whose ongoing teachings and encouragement motivated me every step

the way. Special thanks to Kira Borden, Serra Buchanan, Astrid Galvez Ciani, Adam Dickinson,

Stephanie Gagliardi, Karabi Pathak, Stuart Livingstone, Adam Martin and Emma Windfield.

This work would not have been possible without the support from partners at CATIE and

APOYA. Thank you to Dr. Clémentine Allinne, Dr. Jacques Avelino and Lucie Durand for

sharing your local knowledge and connections. Gracias à Jorge Arturo Ramirez Naranjo and

Patricia Leandro for the generous use of lab space at CATIE. Gracia también à Luis Romero,

Stewar Nuñez y Victor Hugo Méndez Sanabria por su ayuda en el campo y su amistad. I am so

grateful to the Naturalba and APOYA network and each producer who shared their time and

knowledge with me. It is my hope that this research captures these farmers’ practices of caring

for the earth and their skill of producing some of the world’s best coffee. I am very grateful to all

of the plants, soil and microorganisms that gave up their home in Costa Rica to join me to the lab

on this research journey.

I am grateful to the Natural Sciences and Engineering Research Council, Ontario Graduate

Scholarship, the Department of Geography for funding this work. Thanks to the University of

Toronto’s Department of Environmental Science and Geography Department for all of the

logistical support and friendly guidance.

Finally, I’d like to thank my friends, family and community for their endless interest, support and

love. Thanks to all of you for making this experience possible, it has helped me become a better

researcher and person.

iv

Table of Contents Acknowledgements …………………………………………………………………………….. iii

Table of Contents...……………………………………………………………………………….iv

List of Tables……………………………………………………………………………………..vi

List of Figures………………………………………………………………………………….....ix

List of Appendices……………………………………………………………………………..…xi

Chapter 1 - Introduction .............................................................................................................. 1

1.1 Background ..................................................................................................................... 1

1.2 Research questions and objectives ................................................................................... 2

1.3 Research significance ...................................................................................................... 3

Chapter 2 - Literature Review ..................................................................................................... 5

2.1 Agriculture and biodiversity ............................................................................................ 5

2.2 Transitions in coffee production ...................................................................................... 5

2.2.1 Rise in coffee monoculture .................................................................................. 5

2.2.2 Herbicide use in coffee monoculture .................................................................... 6

2.2.3 The transition towards organic coffee production ................................................. 7

2.3 The emergence of the herbaceous community: ecosystem services and disservices .......... 7

2.4 Herbaceous community functional diversity .................................................................... 9

2.5 Farmer perception and management of herbaceous communities ................................... 11

2.6 Mental models ............................................................................................................... 13

2.7 Gaps in literature ........................................................................................................... 14

Chapter 3 - Description of Sites and Methodology .................................................................... 15

3.1 Description of sites ........................................................................................................ 15

3.1.1 Turrialba region of Costa Rica ........................................................................... 15

3.1.2 CATIE research plots ......................................................................................... 15

3.1.3 Organic farms in the Turrialba region ................................................................ 19

3.1.4 Land size and management practices in the region ............................................. 19

3.2 Study design .................................................................................................................. 20

3.2.1 Aboveground herbaceous community identification and sampling ..................... 20

3.2.2 Leaf functional trait analysis .............................................................................. 23

3.2.3 Coffee plant measurements ................................................................................ 23

3.2.4 Shade tree and canopy measurements ................................................................ 24

v

3.2.5 Soil sampling and analysis ................................................................................. 24

3.3 Farmer perspectives of the herbaceous community ........................................................ 25

3.3.1 Participant selection ........................................................................................... 25

3.3.2 Interview format ................................................................................................ 26

3.3.3 Participant information ...................................................................................... 26

3.3.4 Interview processing: cognitive mapping and valuation of services .................... 28

3.4 Statistical analysis ......................................................................................................... 29

Chapter 4 - Results .................................................................................................................... 31

4.1 Taxonomic and functional composition of the herbaceous community ........................... 31

4.2 Herbaceous community functional diversity, soil conditions and coffee health

correlates ....................................................................................................................... 31

4.3 Functional diversity across farm sizes, weeding and canopy openness ........................... 35

4.4 Farmer perception of the herbaceous community ........................................................... 39

4.5 Social and ecological predictors of herbaceous community composition and farm

health ............................................................................................................................ 44

Chapter 5 - Discussion…………………………………………………………………………...48

5.1 Unrecorded diversity in organic agroforestry systems .................................................... 48

5.2 Farm management and attributes strongly predict herbaceous community diversity...…..50

5.3 The role of farmer perception in driving diversity……………………………………..…52

5.4 The potential for agroecosystems service provisioning………...…………………...……54

Chapter 6 - Conclusion ............................................................................................................. 57

6.1 Conclusions ................................................................................................................... 57

6.2 Areas of future research ................................................................................................. 59

References ................................................................................................................................ 60

Appendices ............................................................................................................................... 75

vi

List of Tables

Table 3.1 Descriptions of farm sites involved in study including the years the farm had been

organic in organic coffee production, the altitude and size of the farm. The size of coffee farm

indicates the land in organic coffee production. The variety of coffee and shade tree species

varied between farm and are included.……...………...…………...………………………….18

Table 3.2 Ecological variables measured in this study including leaf functional traits, whole-plant

traits and environmental data.…………………………………………….…………………...22

Table 4.1 Most frequent herbaceous species found in plots are listed here in order of frequency.

Taxonomic family and scientific name are given (Laurito et al. 2016), as well as place of origin.

Farmer perception of species as well as notes from farmer interviews are presented. ….…...32

Table 4.2 Bivariate relationships among functional diversity indices, herbaceous community, soil

and coffee health metrics (where n=45 for all values except for FEve and FDiv* where n=42).

Indices and metrics that have been log-transformed are marked with an asterisk (*). The upper

section of this matrix displays the slopes and associated 95% confidence intervals for each

relationship, based on standardized major axis regression analysis. The lower section of the

matrix displays model r2 and one-tailed p-values (in brackets) for each bivariate model.

Significant relationships (p ≤ 0.05) are bolded……………………………………………….33

Table 4.3 One-way analysis of variance of herbaceous community multi-trait and community

weighted mean (CWM) single-trait indices across farm size. Regional mean farm size

determined from data on all organic farms in the Central Valley region (1.57 ha). Farms this

study are thus describes as above regional mean and below regional mean. Mean and standard

errors (where n=45 for all values except for FEve and FDiv* where n=42) are presented. Indices

that have been log-transformed are marked with an asterisk (*). F and p-values are provided.

A Tukey post-hoc test was used and is denoted by superscripts a and b, whereby letters indicate

significant difference between levels (where p ≤ 0.05). Significant p values are

bolded…………………………………………………………………………………………36

vii

Table 4.4 One-way analysis of variance of herbaceous community multi-trait and community

weighted mean (CWM) single-trait indices across weeding intensities. Mean and standard errors

(where n=45 for all values except for FEve and FDiv* where n=42) are presented. Indices that

have been log-transformed are marked with an asterisk (*). F and p-values are provided. A

Tukey post-hoc test was used and is denoted by superscripts a and b whereby letters indicate

significant difference between levels (where p ≤ 0.05). Significant p values are

bolded.…………………………………………………………………………………...……37

Table 4.5 One-way analysis of variance of herbaceous community multi-trait and community

weighted mean (CWM) single-trait indices across levels of canopy openness. Mean and standard

errors (where n=45 for all values except for FEve and FDiv* where n=42) are presented. Indices

that have been log-transformed are marked with an asterisk (*). F and p-values are provided. A

Tukey post-hoc test was used and is denoted by superscripts a and b whereby letters indicate

significant difference between levels (where p ≤ 0.05). Significant p values are bolded...…..38

Table 4.6 Guiding motivations that farmers indicated for switching to organic agriculture ranked

from most frequent response to least frequent. Some farmers had multiple motivations for

transitioning from conventional to organic agriculture.…………………………...………….40

Table 4.7 Number of connections, variables, connection-to-variable ratio, and map density

derived from cognitive maps. Value of ecosystem services determined through interview analysis

is also provided.…………………………………………………………………...………….41

Table 4.8 Highest domain and centrality variable derived from cognitive maps. Highest domain

variables are concepts with most in and out linkages to other concepts in the cognitive maps.

Highest centrality variables are concepts with the highest number of direct and indirect links to

other variables. ……………………………………………………………………..…………43

Table 4.9 Stepwise and multiple regression model analysis to determine the farmer attributes that

best predict indices of herbaceous community functional diversity. Indices that have been log-

transformed are marked with an asterisk (*). Parameter estimates and p-values are shown for

viii

parameters retained in most parsimonious AIC-selected model. Parameters in bold are significant

(p<0.05) in a multiple regression analysis. AIC values for full model and most parsimonious

model are presented and ΔAIC values representing the difference between the two. Full model

was of the form: functional diversity response ~ farmer attributes [years organic (yorg) +value

of ecosystem services (ves) + cognitive map connection-to-variable ratio (cv) + cognitive map

density (density)] ……………………………………………………………………….……46

ix

List of Figures

Figure 3.1 A map of the Central Valley region of Costa Rica from Satellite view with yellow

dots representing research sites (Google Map Pro 2019). Research was conducted on 9

independent organic farm sites (F2, F4, F5, F8, F10, F11, F13, F14, F15) in the region and 6 sites

at the Tropical Agriculture Research and Higher Education Centre (F1, F3, F6, F7, F9, F12) for a

total of 15 farm sites. …………………………………………………….………………..….16

Figure 3.2 Map of CATIE sites with plots involved in study highlighted. The six green boxes

represent areas of study at the CATIE farm (F1, F3, F6, F7, F9, F12). Areas of study had

different management practices and shade-tree species.……………………………………….17

Figure 3.3 Example of 1m x1m sampling quadrat. Plant cover and exposed soil was recorded,

and each species was identified. All herbaceous community biomass within the quadrat was cut

and taken to the lab for further analysis. Soil sampling was conducted in the middle of the

quadrat (where meter sticks cross). ………………………………………………………...……21

Figure 3.4 Farmer demonstrates his knowledge of the herbaceous community within his organic

coffee agroforestry system during interview. This farmer specifically discussed his values of

ecosystem services of pollinating herbaceous species such as the flowers he holds in hand, and

the beneficial ground coverage and soil erosion protection of Commelina diffusa, which he calls

“canutillo” or “oreja de ratón” (mouse ear), on his farm. ……………………………………….27

Figure 4.1 A sample cognitive map demonstrating a farmer’s perception of ecosystem services

(in green text), disservices (red text), and key factors that affect decision making and control of

weeds (blue text) solid lines represent positive relationships whereas dashed lines represent

negative relationships………………………………………………………….……………42

Figure 4.2 Principal component analysis of the community weighted mean values for four leaf-

level traits (specific leaf area, leaf dry-matter content, leaf nitrogen concentration, leaf carbon

concentration) and one whole-plant trait (plant height) of 39 herbaceous species found in 45

organic plots across the Turrialba region. Colours represent farmer perspective on herbaceous

x

species as being good, neutral or bad within their organic coffee agroforestry farms. Ellipses

correspond to 95% confidence ellipses for community weighted mean values for herbaceous

species sampled in this study.

……………………………………………………………………………….………45

xi

List of Appendices

Appendix A. All herbaceous species found in organic coffee agroforestry plots (n=45) in this

study, listed in order of frequency. Range and mean functional traits are given……………..75

Appendix B. Interview questions used in all farmer interviews, approved by University of

Toronto Human Research Ethics Program. Italicized questions were prompts used to support

further elaboration on questions from farmers……………………………………….…… ..78

1

Chapter 1 - Introduction

1.1 Background Industrial agriculture is responsible for detrimental environmental degradation such as

decreasing biodiversity (Jones et al. 2018), fragmenting habitats (Newbold et al. 2015), and

contributing to high levels of air and water pollution (Tilman et al. 2017). Increasing concerns

about industrial agricultural have catalyzed the development of sustainable agricultural systems

that promote conservation and biodiversity (Sandhu et al. 2010). Sustainable agriculture systems,

including organic agriculture, foster an increase in ecosystem functioning and the provisioning of

ecosystem services (Naeem et al. 1994; IFOAM 2009; Isbell et al. 2017). Organic agricultural

practices can provide greater food security while fostering nutrient cycling, promoting soil health

and reducing deforestation (Altieri 2009; Foley et al. 2011). Global organic production is

growing at a rate of approximately 7% per year (Willer & Lernoud 2017). As of 2015, there were

2.4 million organic farmers worldwide who were managing 50.9 million hectares of land across

179 countries (Willer & Lernoud 2017).

Coffee has become a leading crop in organic agriculture and now represents over 25% of all

permanent organic cropland (Willer & Lernoud 2017). Coffee is a crop of great importance from

a conservation perspective as it is produced on 11 million hectares of land (Bertrand et al. 2016)

located within tropical biodiversity hotspots (Myers et al. 2000). Since its introduction to the

Americas in the early 19th century, coffee has been a force in shaping policies, economies,

landscapes and livelihoods (Sick 1999; Perfecto et al. 2014). Recently, many smallholder

farmers across Latin America have shifted to agroecological coffee production, including organic

production (Toledo & Moguel 2012), which mimics coffee’s natural growing conditions by

fostering shade-tree diversity, herbaceous communities and closed nutrient cycles (Tully & Ryals

2017). Research has found that the inclusion of shade-trees in coffee systems improves

ecosystem function (Perfecto et al. 1996; Perfecto et al. 2014) and ecosystem services such as

biodiversity conservation (Méndez et al. 2010; Tscharntke et al. 2011; De Beenhouwer et al.

2013), erosion control (Meylan et al. 2013) and carbon sequestration (Wardle et al. 2012).

2

Organic coffee production protocol ensures the elimination of the synthetic herbicides that cause

water and soil contamination (Nestel 1995; Relyea 2005). Without the use of synthetic

herbicides, the herbaceous communities which naturally occur in traditional coffee systems re-

emerge. Herbaceous communities include cover crops, weeds, flowers and other ground covers.

A few studies have found positive impacts of the emergence of the herbaceous community in

coffee systems such as increasing biodiversity (Soto-Pinto et al. 2002) and reducing soil erosion

(Meylan et al. 2013). Other studies indicate negative impacts of herbaceous communities

including suppressing coffee yield (Ronchi & Silva 2006) and creating increased labour demands

for farm workers (Labrada 1997). Overall, however, the role of herbaceous communities in

coffee agroforestry systems is understudied (Ronchi & Silva 2006). New approaches to

understanding the herbaceous community, such as through a functional ecology lens, and the

ecosystem services it provides within coffee agroforestry systems will provide significant

benefits to farmers, scientists and policymakers alike.

Functional ecology has been an essential approach to advancing the understanding of natural

plant community dynamics yet has only recently been applied to agroecosystem research

(Garnier & Navas 2012; Martin & Isaac 2015). The measure of ecologically relevant plant traits,

for example leaf traits, through a functional ecology lens can provide insight into how plants

respond to environmental variation and how plants, in turn, affect ecosystem functions (Lavorel

& Garnier 2002; Garnier et al. 2004; Martin & Isaac 2015). A functional traits approach can

provide evidence to answer questions around plant community assembly and ecosystem

functioning in a way that can be more meaningful than species richness or composition (Cadotte

et al. 2011; McGill et al. 2006). Researchers have used functional traits to provide important

insights into coffee leaf expression among a range of management scenarios as well as response

to and influence on ecosystem functions (DaMatta 2004; Matos et al. 2009; Gagliardi et al. 2015;

Martin et al. 2017; Isaac et al. 2017) but the herbaceous community remains understudied from a

functional traits approach.

1.2 Research questions and objectives

My research questions are: (1) How taxonomically and functionally diverse is the herbaceous

community in organic coffee agroforestry systems? (2) What factors best explain the level of

3

diversity in organic coffee agroforestry systems’ herbaceous community? (3) What are farmer

perceptions of the herbaceous community in their coffee systems and how do these perceptions

affect management practices?

To answer these questions, I conducted a field study in the Turrialba Region of Costa Rica to

record the taxonomic and functional diversity of the herbaceous community; to measure soil

conditions including soil carbon, nutrients and moisture; and to determine the impact of

herbaceous community diversity on coffee health indicators by measuring coffee yield and

coffee leaf rust incidence. I paired this field study with semi-structured interviews with organic

coffee farmers (n=11), all members of the local organic cooperative the Organic and Sustainable

Producers Association (APOYA) and performed subsequent cognitive mapping analysis.

On these organic farms in the Central Valley of Costa Rica, I proposed the following objectives:

1. Measure the taxonomic and functional diversity of the herbaceous community within

organic coffee agroforestry systems.

2. Determine how farm management practices (weeding intensity, shade tree management)

and farm characteristics (farm size) affect organic coffee agroforestry systems’

herbaceous communities.

3. Present farmer perceptions of the herbaceous community through a cognitive mapping

approach.

1.3 Research significance

My research aims to provide some of the first insights into herbaceous community functional

diversity in tropical agroecosystems to the growing body of community ecology literature (Soto-

Pinto et al. 2000; DaMatta 2004; Beer et al. 1998; Haggar et al. 2011; Gagliardi et al. 2015).

There is currently a gap in knowledge about the role that herbaceous communities in coffee

agroforestry systems play in ecosystem service provisioning such as enhancing soil carbon.

Functional traits, such as leaf physiological, morphological and chemical characteristics (McGill

et al. 2006; Westoby & Wright 2006; Violle et al. 2007) of the herbaceous community will

contribute to an understanding of its ecological function. By measuring functional traits, soil

conditions and coffee health indicators, this project explores how the herbaceous community

contributes to ecosystem services.

4

From a social-agroecological perspective, weeds are the number one obstacle for farmers to shift

from conventional to organic coffee production due to increased labour and potential yield

decline (Lyngbæk et al. 2001; Ronchi & Silva 2006; personal correspondence, 5 July 2018).

While coffee has been cultivated without chemicals since its arrival in Latin America in the 18th

century (Perfecto et al. 2014), currently, most support for weed management comes from

chemical companies, such as Bayer CropScience, which promote herbicide use (Bellamy 2011).

My research aims to provide farmers with an agroecological alternative by presenting the

ecological and cultural elements of farmers’ current management practices and their methods of

evaluation in a way that allows for collective knowledge to guide interests of the local

community (Stirling et al. 2017). To better understand the role that organic farmers play in the

management of the herbaceous community and the provisioning of ecosystem services, this

study employs a cognitive mapping approach to investigate farmer perspectives (Isaac et al.

2009).

Finally, my research aims to inform future iterations of payment for ecosystem services.

Specifically, with payments for ecosystem services as a growing mechanism for conservation in

Costa Rica, insight into the herbaceous community’s role in ecosystem service provisioning is

essential. For example, the role of the herbaceous community in increasing soil carbon may be

important information for Costa Rica to reach its goal to become the first carbon neutral country

by 2021 (Defrenet et al. 2016). Including the herbaceous community in payments for ecosystem

service programs may offer more comprehensive supports for smallholder farmers who currently

do not benefit from these payments (Lansing 2017). My research will be disseminated in the

Central Valley of Costa Rica through partnerships with the Organic and Sustainable Producers

Association of Turrialba (APOYA), Nationally Appropriate Mitigation Action (NAMA) and the

Tropical Agricultural Research and Higher Education Center (CATIE). Finally, since Costa Rica

is known as a leader in environmental and conservation policy, such as its payment for

ecosystem service program (Keenan 2017), it is my hope that this research will inspire other

countries to consider the ecosystem services provided by their own herbaceous communities.

5

Chapter 2 - Literature Review

2.1 Agriculture and biodiversity

Agriculture is currently the largest contributor to biodiversity loss (Dudley & Alexander 2017)

and is responsible for over 30% of greenhouse gas emissions worldwide (Balmford et al. 2009;

Tillman et al. 2017). As the world’s single largest land use, agriculture must change to meet the

needs of future generations (Dudley & Alexander 2017). Farmers, particularly small-holder

farmers in the global South, foster a variety of crops and herbaceous species within in their fields

and thereby sustain and conserve agricultural biodiversity (Isakson 2014). Practices such as

conservation agriculture, agroforestry and organic agriculture foster biodiversity (Dudley &

Alexander 2017). Organic coffee agroforestry systems, for example, promote ground covers and

tree intercropping to support soil health, ecosystem services and local livelihoods (De

Beenhouwer 2013; Atangana et al. 2014).

2.2 Transitions in coffee production

Coffee is one of the world’s most important agricultural commodities with an estimated annual

retail value of 70 Billion USD (Talhinhas et al. 2017). Coffee also has significant ecological and

social influence with more than 11 million hectares of land in coffee production (Bertrand et al.

2016) and more than 100 million people who depend on coffee as their main source of income

(ICO 2016; Talhinhas et al. 2017). Coffee originates from the montane rainforest of Ethiopia,

where it naturally grew as an understory crop beneath shade-trees and among shrubs and

herbaceous communities (Toledo & Moguel 2012). When the crop was first introduced to the

Americas, it was planted within forest ecosystems to mimic coffee’s natural habitat of a diverse

shade tree and herbaceous species (Vandermeer & Perfecto 2015). Since its original introduction,

coffee production has dramatically shaped the landscapes of countries across the Americas (Sick

1999; Perfecto et al. 2014).

2.2.1 Rise in coffee monoculture

In the 1980s, countries across Latin America, including Costa Rica, experienced economic

stagnation and financial crisis (Conroy et al. 1996). In an attempt to boost the economy, USAID

and local governments encouraged farmers to increase coffee production for export. The increase

6

in coffee production was achieved through agricultural intensification including increased

chemical inputs and deforestation (Conroy et al. 1996; Perfecto et al. 2014). This monoculture

system ensured that coffee plants would not have to compete with shade-trees and herbaceous

communities for light, water or nutrients. While coffee monoculture yielded three to four times

the harvest compared to traditional agroforestry systems, it disrupted natural ecological processes

and lowered global coffee prices which resulted in economic vulnerability for many smallholder

farmers (Haggar et al. 2011).

2.2.2 Herbicide use in coffee monoculture

Coffee intensification promotes the use of synthetic chemical herbicides to control herbaceous

communities. Herbicides are applied to more hectares of land than any other category of

chemical pesticide (National Research Council 1989; Bellamy 2011). The two most dominant

herbicides used in coffee systems are paraquat and glyphosate (Bellamy 2011). Paraquat is a

water-soluble herbicide which has been banned in the United States due to its high toxicity and

long persistence in the soil (US EPA 1997; Bellamy 2011; Green 2014). Glyphosate is a broad-

spectrum systemic herbicide that is used world-wide. Recently, glyphosate was found to be a

probable carcinogen due to its long half-life in water and soil (Myers et al. 2016). As with many

herbicides used in coffee production, both glyphosate and paraquat are non-selective and can

cause damage to the coffee plant itself (Njoroge 1994).

The application rates of herbicides are not economically, ecologically, nor socially sustainable.

During the Expert Consultation on Weed Ecology and Management Conference, Gerowitt (1997)

presented findings that farmers were spraying up to 20-50% more herbicides than needed.

Moreover, an estimated 85–90% of herbicides never reach target species and instead move into

the air, soil and water (Moses et al. 1993; Bellamy 2011). Many farmers are aware of the

dangers of herbicides. A recent study in the Turrialba region of Costa Rica found that 80% of

conventional and organic farmers interviewed believed that the use of herbicides decreased the

fertility or changed the structure of their soil (Cerdán et al. 2012). With agrochemical availability

predicted to decline as a result of reduced access to fossil fuels and phosphorus in the coming

years, herbicides will become even more costly (Cordell et al. 2009; Woods et al. 2010; Garnier

& Navas 2012).

7

2.2.3 The transition towards organic coffee production

Due to the damaging economic, ecological and social impacts of monoculture coffee production,

many smallholder farmers worldwide have transitioned back to diversified agroforestry coffee

systems including certified organic coffee production (Toledo & Moguel 2012). The organic

movement has been a prominent agroecological alternative for coffee producers over the past

three decades with 8.9% of the world’s harvested coffee now being produced organically (Willer

& Lernoud 2017). The area under organic coffee production has increased four-fold since 2004

to almost one million hectares which represents 25% of the organic permanent cropland globally

(Willer & Lernoud 2017).

Organic systems prohibit the use of synthetic chemicals and rely solely on biological nutrient

sources (Ayalew 2014). Therefore, organic producers integrate nutrients through practices such

as planting and pruning shade-trees, mechanically or manually cutting their herbaceous

community and integrating it as green manure (Bellamy 2011). Organic practices provide many

ecological benefits but often require more labour and can result in lower coffee yields (Haggar et

al. 2011; Cerda et al. 2017a). To compensate for increased labour and potential yield decline,

organic coffee farmers receive a price premium (Mendéz et al. 2010). While these premiums

may not be enough incentive for farmers to transition to organic farming yet, there are emerging

opportunities. The rapid growth of the global organic market (Willer & Lernoud 2017), local

network support (personal correspondence, July 5 2018), and the potential for payment for

ecosystem service programs (Pagiola 2008), all provide farmers with opportunities to benefit

from organic production. However, with weed management named as the number one barrier for

conventional farmers to transition to organic agriculture in the Central Valley of Costa Rica

(personal correspondence, July 5 2018), there is an urgent need to understand the role the

herbaceous community plays in coffee agroforestry systems.

2.3 The emergence of the herbaceous community: ecosystem services and disservices

The elimination of herbicides in organic agriculture has led to the emergence of herbaceous

communities in coffee agroforestry systems. Grasses and broad-leaf weeds are the most

8

prominent herbaceous species that emerge (Njoroge 1994), though tree seedlings and woody

perennials also appear within coffee systems globally.

In coffee agroforestry systems, the herbaceous community provides provisioning services (MEA

2005) including coffee yield, forage for animals and medicine (Soto-Pinto et al. 2002).

Herbaceous communities also provide supporting services including the enhancement of bird and

small mammal diversity (Gordon et al. 2007) and cultural services including spiritual or aesthetic

elements to farms (Toledo & Moguel 2012). Herbaceous communities provide many regulating

services including promoting pollination (Gordon et al. 2007) and improving soil fertility (Sarno

et al. 2004). Research from vineyards indicates that after being cut, the herbaceous community

itself becomes a layer of organic matter which supports soil moisture (Morlat & Jacquet 2003)

and nutrient cycling (Garcia et al. 2018) and can be a viable alternative to the application of

chemical fertilizers (Hartwig & Ammon 2002). Herbaceous covers also play an important role in

soil protection by physically covering topsoil and strengthening the belowground soil which

protects soil from water and wind erosion (Novara et al. 2011). Within coffee-agroforestry

systems, the roots of herbaceous species have been found to support belowground soil biota and

macroporosity (Sarno et al. 2004; Meylan et al. 2013; Martins et al. 2015). Across

agroecosystems, the herbaceous community has been found to support carbon sequestration by

increasing soil organic matter (Kaye & Quemada 2017). There are many herbaceous species that

do not have known services or disservices and therefore, are considered neutral by farmers (Filho

et al. 2013).

The herbaceous community can also create ecosystem disservices, which are ecosystem

generated functions or attributes that have negative impacts on human wellbeing (Shackleton et

al. 2016). Herbaceous community ecosystem disservices including competition for limited

resources and reduction of coffee yields, are accordingly referred to as weeds (Ronchi & Silva

2006). Coffee plants respond to water and nutrient competition by pushing their roots further

into the soil, which transfers energy from fruit production to below-ground growth thereby

reducing yields (Njoroge 1994). Moreover, herbaceous communities may decrease the amount of

plant available nitrogen by immobilizing nitrogen, particularly in dry soil (Celette et al. 2009),

though this depends greatly on the leaf nitrogen concentration of the plant substrate (Yadvinder-

Singh et al. 2005). Nutrient and water competition can stress coffee plants and predispose the

9

crop to severe pathogen attacks (Zambolim et al. 1997). Moreover, herbaceous communities can

interfere with management practices including fertilization and harvesting (Njoroge 1994).

Controlling the herbaceous community can be the most labour-intensive aspect of coffee

production (Lyngbæk et al. 2001), particularly for conventional farmers who employ both

chemical and mechanical practices to weed (Bellamy 2011). In the Central Valley of Costa Rica,

hiring labour for weeding is common and can be a significant expense (Bellamy 2011). For

example, hiring labour to control the herbaceous community may cost farmers approximately

$30 USD per day, which is the equivalent price farmers receive for approximately 15lbs of green

organic coffee (personal correspondence, June 2018). Appropriate education and incentive

programs could encourage shifts management practices to decrease labour demands while

reducing ecosystems disservices and promoting ecosystem services of herbaceous communities

(Power 2010).

Costa Rica is leading the movement to address and foster ecosystem services through a payment

for ecosystem services (PES) program (Pagiola 2008). The PES program is organized through

the country’s National Fund for Forest Financing (FONAFIFO), a semi-autonomous government

agency with independent legal status (Pagiola 2008). The FONAFIFO organization supports

farmers and other land stewards who provide ecosystem services such as water filtration,

greenhouse gas mitigation, biodiversity conservation, the creation of natural beauty (FONAFIFO

2018). Costa Rica’s payment for ecosystem service program is seen as a conservation success

story (Pagiola 2008). However, there are still challenges with the program including the payment

for services provided only by trees and the skewed benefits towards large landowners (Lansing

2017). The role of the herbaceous community has been overlooked in ecosystem function and

service provision in Costa Rica, though is starting to be explored in other countries (Woodbury et

al. 2017). Further exploration of the role of the herbaceous community could support the

development of a more equitable and comprehensive payment for ecosystem service program.

2.4 Herbaceous community functional diversity

Many studies have supported the positive relationship between herbaceous diversity and

ecosystem function (Smith et al. 2009), but the drivers behind this relationship are not clearly

understood (Cadotte 2017). Taxonomic diversity measures into herbaceous communities

10

provide useful information about species richness, native versus non-native species and impact

of management practices (Kazakou et al. 2016). Assessing leaf functional traits and trait trade-

offs in managed systems can inform a more mechanistic understanding of plant responses to

environmental conditions and the influence of plants on ecosystem function, than taxonomic

species diversity alone (Cadotte et al. 2011; Granier & Navas 2012; Martin & Isaac 2015). Three

commonly measured leaf traits - specific leaf area (SLA), leaf dry-matter content (LDMC), and

leaf nitrogen concentration (LNC) are considered functional markers as they can assess the

impacts of community changes on ecosystem properties (Garnier et al. 2004). These traits

provide an understanding of plant strategies for resource use and ecosystem functioning (Wright

et al. 2004). Plants with high LNC and high SLA indicate fast-growing, short-lived plants with

faster decomposition rates. Plants with low LNC and low SLA are often longer-living plants with

lower decomposition rates (Wright et al. 2004; Garnier & Navas 2012). Leaf carbon

concentration (LCC) provides important insight into community carbon accumulation (Derroire

et al. 2018) and plant height provides intel into competitive capacity (Garnier & Navas 2012).

Collectively, these traits inform both functional diversity indices and community weighted mean

approaches to better understand herbaceous community diversity.

Functional diversity metrics – functional richness, functional evenness, functional divergence

(Mason et al. 2005) and functional dissimilarity (Botta-Dukát 2005) – are central to describing

the functional diversity of a plant community. Functional richness (FRic) is related to the number

of species present in a plot and indicates how much niche or trait space is filled (Mason et al.

2005). Functional evenness (FEve) indicates the distribution of mean values of species traits

within occupied niche space (Mason et al. 2005; Schleuter et al. 2010). Functional divergence

(FDiv) indicates the specialization of the functional traits, for instance, high functional

divergence signals that there is a high amount of niche differentiation and low resource

competition (Mason et al. 2003; Mason et al. 2005). The quadratic entropy of Rao (1982), also

referred to FDQ (Schleuter et al. 2010), incorporates the relative abundance of species and

measures pairwise functional differences between species to quantify the functional similarity of

individuals in the trait space (Shimatani 2001; Botta-Dukát 2005). A high FDQ value signifies

that individuals are less similar and therefore do not fill the same functional role (Karadimou

2016). Interestingly, FRic and FDQ may be negatively related as pairwise differences between

species may decline as more species are introduced (Botta-Dukát 2005).

11

Another useful tool for understanding herbaceous community diversity is through community

weighted single-trait indices. Community weighted means (CMW) are plot-level single-trait

values weighted by the relative abundance of the species present. Given that the mass-ratio

hypothesis suggests that the most abundant species are most important in driving ecosystem

functioning (Grime 1998; Díaz et al. 2007), CMWs are a useful measure in complex landscapes

(Butterfield & Suding 2013). Community weighted means are frequently used as an indicator of

functional composition in order to understand trait variation of plant communities (Díaz et al.

2007).

2.5 Farmer perception and management of herbaceous

communities Farmer knowledge and management decisions influence biodiversity conservation and

ecosystem functions within coffee agroforestry systems (Cerdán et al. 2012; Valencia et al.

2015). Local farming knowledge is often developed within the community (Raedeke & Rikoon

1997), with producer networks as important spaces for the transfer of knowledge and adoption of

management practices (Isaac 2012; Cadger et al. 2016; Isaac & Matous 2017). The importance

of producer networks supports the need to substantively include the perspectives of farmers in

any and all agriculture related policy and practice (Halbrendt et al. 2014; Stirling et al. 2017).

Current farmer knowledge intersects with the management of herbaceous communities in three

ways: management of shade trees, mechanical and biological control of the herbaceous

community, and the farm labour/farm engagement nexus.

Shade trees can affect the herbaceous community by reducing the amount of light that filters

through the canopy and by forming a litter layer through a leaf fall and pruning residues (Beer et

al. 1998; Staver et al. 2001). Nestel and Altieri (1992) found that the biomass of herbaceous

community in coffee monoculture was two times the amount compared to a diverse agroforestry

system due to light reduction and shade-tree pruning litter. Shade trees can also affect the types

herbaceous species present, with shaded plots fostering more broad-leaf species, often

considered good herbaceous species, whereas full sun plots foster the growth of more grasses,

which are often considered bad weeds (Nestel & Aliteri 1992). A local guide encourages farmers

to use shade-trees and prune them at least once a year (Montagnini et al. 2015), but no specific

12

level of canopy openness is suggested. Since herbicides are prohibited within organic systems,

mechanical and biological controls are permitted in organic production.

The mechanical management of the herbaceous community includes the use of a machete, weed-

wacker, and/or shovel (Bellamy 2011). A Costa Rican coffee weed management guide suggests

that farmers should use a machete and weed-wacker to cut herbaceous species down to

encourage decomposition and nutrient cycling of herbaceous biomass. The guide also suggests

that farmers use a shovel, hoe or their hands to remove herbaceous species with rhizomes during

the dry season to reduce their spreading (Filho et al. 2013). Other mechanical management

options include burning herbaceous communities, letting animals graze, and planting cover crops

(Filho et al. 2013). Research suggests that herbaceous species competition only occurs during the

early stages of growth for young coffee plants (Ruthenberg 1971; Terry 1984) and throughout

the months of crop flowering and fructification for adult coffee plants (Ronchi & Silva 2006).

As herbaceous community management can constitute over 50% of farm labour time (Labrada

1997), techniques that reduce labour while providing ecosystem services are essential to support

coffee production and farmer livelihoods. It is of importance to note that organic coffee systems

may be able to tolerate a higher level of herbaceous community biomass compared to

conventional systems due to fertility management within organic systems (Ryan et al. 2009;

Rossi et al. 2011). Moreover, conventional farmers often spend significantly more time on

herbaceous species control than organic producers, as they utilize both chemical and mechanical

management (Lyngbæk et al 2001; Bellamy 2011). Recent studies have determined that plant-

based indices to diagnose the success of farm management practices is highly related to a

farmer’s level of engagement with crops (Isaac et al. 2018), leading to decisions on fertilization,

pruning, species selection (Isaacs et al. 2016; Dickinson 2017) and soil health (Valencia et al.

2015). These plant-based indices are at least partly derived from physical engagement with

plants in the field (Isaac et al. 2018). Interestingly, research into hired labour has observed that

there can be a negative effect on agricultural biodiversity (Isakson 2011), perhaps due to a

greater variety of agricultural tasks (Van Dusen & Taylor 2005) or a reduction in regular

engagement with plants on the land (Isaac et al. 2018).

13

2.6 Mental models

Integrative frameworks which bridge biophysical and social domains are needed to understand

human-altered ecosystems, particularly agroecosystems (Collins et al. 2010). The integrated

social-ecological framework for agroecosystem services proposed by Lescourret et al. (2015)

highlights the interconnectedness between social systems, agroecosystem management,

ecosystem structure and ecosystem services. Building off of socio-ecological theory and

research into coffee agroforestry systems, farmers are seen as important agroecosystem managers

(Bandeira et al. 2002; Cerdán et al. 2012; Lescourret et al. 2015). Understanding how farmers

generate and apply knowledge to management practices has significant impacts for biodiversity

conservation of agroforests (Valencia et al. 2015).

Research on knowledge, skills and attitude (Greiner 2015) gives a more complete understanding

of intrinsic motivations for decision making (Ingram et al. 2013) and management practices

(Hoffman et al. 2014). Furthermore, it is essential to consider farmers’ perspectives when

introducing or supporting agricultural development programs (Pretty 1995; Halbrendt et al.

2014), including payment for ecosystem programs (Lansing 2017). Mental models, such as

cognitive mapping, are a widely accepted approach to understanding individual and group

decision-making processes (Jones et al. 2011; Gray et al. 2014; Halbrendt et al. 2014) and have

recently provided important insights into sustainable agricultural (Isaac et al. 2009; van Winsen

et al. 2013) and food systems management (Stier et al. 2017).

Cognitive mapping is a valuable tool to visually represent how local knowledge and human

actions affect ecosystems (Özesmi & Özesmi 2004; Christen et al. 2015). Cognitive mapping, an

approach first coined by Tolman (1948), supports the creation of participatory management plans

in ecological systems and helps to represent individual’s understanding of the systems around

them (Özesmi & Özesmi 2004; Gray et al. 2014). The cognitive mapping approach has been

successfully used to include farmer perception of management practices (Isaac et al. 2009; van

Winsen 2013) and provides important insights for planning needed to increase sustainability

(Dodouras & James 2007). Since cognitive maps allow for the modelling of relationships

between variables that are not known with certainty and are ever-evolving, they are useful in an

agricultural context. Cognitive mapping approaches allow for the inclusion of complex ideas yet

facilitate the ease of obtaining information which is useful in adapting to farmers’ busy schedules

14

(Özesmi & Özesmi 2004). One disadvantage of cognitive mapping is that the approach requires

the researcher to construct the map without a standardized methodology (Eden 2004; van Winsen

et al. 2013). Therefore, the quality of the interviewer as listener and interpreter (Eden 2004) and

investment in the time-consuming process of constructing cognitive maps is essential (Isaac et al.

2009).

2.7 Gaps in literature

While there is substantial research into ecological benefits of coffee agroforestry systems (Soto-

Pinto et al. 2002; Cerdán et al. 2012; Gagliardi et al. 2015; Cerda et al. 2017a), the role of the

herbaceous community is understudied (Ronchi & Silva 2006; Rossi et al. 2011). The majority of

research on herbaceous species in coffee systems in Costa Rica has been conducted by chemical

companies, which does not meet the needs of smallholder organic farmers (Bellamy 2011). Few

studies have looked at the taxonomy of herbaceous communities within coffee agroforestry

systems and none have taken a trait-based approach. Therefore, understanding the herbaceous

community from a functional ecology lens will provide important insight into its role in

ecosystem functioning and service provisioning.

Recent research indicates that leaf functional traits knowledge provides mutually beneficial

insights for farmers and scientists alike (Martin & Isaac 2015; Dickinson 2017; Isaac et al.

2018). However, these studies have focused on crop leaf traits rather than the traits of the

herbaceous community. This study will contribute information to the important field of

functional diversity and farmer knowledge. Finally, recent research on payment for ecosystem

programs have indicated that there is an urgent need to incorporate smallholder farmers into this

payment scheme (Lansing 2017). This research aims to provide insight to inform payment for

ecosystem programs to better support for smallholder farmers providing ecosystem services

through the management of their herbaceous communities.

15

Chapter 3 - Description of Sites and Methodology

3.1 Description of sites 3.1.1 Turrialba region of Costa Rica Research was conducted at sites throughout the Turrialba region, located within the Central

Valley of Costa Rica. Turrialba is a prominent coffee-growing region with an annual temperature

of 22.2◦C and small variations across months. The region has a mean annual rainfall of 2800mm

(Cerda et al. 2017a), though rainfall patterns have become increasingly unpredictable (Cerdán et

al. 2012; Isaac et al. 2018) due to climate change (IPCC 2013). Coffee is grown from altitudes of

600 to 1400m, with farms at higher elevations having slightly more rain and cooler temperatures

than farms at lower elevations (Cerda et al. 2017a). Soils in Turrialba region are generally acidic

and have moderate fertility (CIA 2016; Cerda et al. 2017a) with risk of nutrient depletion due to

monoculture (Isaac et al. 2018). Recently, coffee leaf rust (CLR), a fungal pathogen has

destroyed 12-25% of coffee yield annually by hindering vegetative development and causing

death of branches (Avelino et al. 2015; Allinne et al. 2016). The CLR crisis, known locally as

“La Roya” affected all farm sites involved in this study. The 45 sites in this experiment took

place within six organic plots at CATIE’s experimental farm (Figure 3.1) and on nine organic

farms within from the Asociación de Productores Orgánicos y Agrosostenibles (APOYA)

network all located in Turrialba region (Figure 3.2).

3.1.2 CATIE research plots

The CATIE experimental farm is located at 685m above sea level. Soils at the CATIE site are

Typic Endoaquults (Ultisols) derived from volcanic alluvium. Soils are acidic (pH<5.5) and have

clay content greater than 50% (Rossi et al. 2011). Until 2000, sugar cane (Saccharum

officinarum) was the main crop grown on the site (Mora & Beer 2013). Both organic and

conventional coffee production has been taking place at the CATIE site for 18 years. For this

study, six organic treatments with different shade tree combinations were chosen (Table 3.1).

These plots contained the Caturra variety of coffee which is very susceptible to CLR.

16

Figure 3.1 A map of the Central Valley region of Costa Rica from Satellite view with yellow

dots representing research sites (Google Map Pro 2018). Research was conducted on 9

independent organic farm sites (F2, F4, F5, F8, F10, F11, F13, F14, F15) in the region and 6 sites

at the Tropical Agriculture Research and Higher Education Centre (F1, F3, F6, F7, F9, F12) for a

total of 15 farm sites.

17

Figure 3.2 Map of CATIE sites with plots involved in study highlighted. The six green boxes

represent areas of study at the CATIE farm (F1, F3, F6, F7, F9, F12). Areas of study had

different management practices and shade-tree species.

F 1 3

9 12

7

6

18

Table 3.1 Descriptions of farm sites involved in study including the years the farm had been organic in organic coffee production, the altitude and size of the farm. The size of coffee farm indicates the land in organic coffee production. The variety of coffee and shade tree species varied between farm and are included.

Farm Years Organic

Altitude (m)

Coffee Farm

Size (ha)

Variety of Coffee

Shade Tree Species Scientific name (Common name to farmers)

F1 18 685 3 Caturra Erythrina poeppigiana (Poró); Chloroleucon eurycyclum (Chloroleucon)

F2 14 700 1 Caturra Citrus reticulata (Mandarina) Psidium guajava (Guayaba); Musaceae (Banano)

F3 18 685 3 Caturra Terminalia amazonia (Terminalia); Chloroleucon eurycyclum (Chloroleucon)

F4 18 750 0.5 Caturra, Típica Erythrina poeppigiana (Poró); Psidium guajava (Guayaba); Theobroma cacao

(Cacao)

F5 18 700 2 Catuai Erythrina poeppigiana (Poró); Citrus

reticulata (Mandarina); Musaceae (Banano)

F6 18 685 3 Caturra Erythrina poeppigiana (Poró)

F7 18 685 3 Caturra Erythrina poeppigiana (Poró); Terminalia amazonia (Terminalia)

F8 22 800 2 Caturra Erythrina poeppigiana (Poró); Musaceae; (Banano) Carica papaya (Papaya)

F9 18 685 3 Caturra Terminalia amazonia (Terminalia)

F10 18 720 1 Obota Laurus nobilis (Laurel), Erythrina

poeppigiana (Poró), Bactris gasipaes (Palma)

F11 6 700 2.5 Caturra,

Catimore, Costa Rica 95

Laurus nobilis (Laurel), Bactris gasipaes (Palma); Psidium guajava (Guayaba); Carica papaya (Papaya); Musaceae

(Banano) F12 18 685 2 Caturra Erythrina poeppigiana (Poró)

F13 15 775 1 Catimore Laurus nobilis (Laurel); Musaceae (Banano); Citrus reticulata (Mandarina)

F14 20 750 0.75 Esperanza, Centroamericana

Erythrina poeppigiana (Poró); Musaceae(Banano)

F15 15 1000 1 Esperanza, Milenio,

Centroamericana

Erythrina poeppigiana (Poró); Musaceae (Banano)

19

3.1.3 Organic farms in the Turrialba region

This study also included nine organic farms the Turrialba region. Sites were chosen based on the

criteria that farms i) are part of the APOYA network, ii) are owned by smallholder farmers, iii)

implement organic practices, and iv) have an herbaceous community present. All farms in this

study integrated shade-tree intercropping. The shade tree species and coffee varieties differed

between farms and were documented (Table 3.1). Due to the coffee leaf rust crises, most

producers replaced the susceptible Caturra variety of coffee with more resistant varieties

including Esparanza, Costa Rica 95, Milenio, Centroamericano, and Obata between 2015-2018.

3.1.4 Land size and management practices in the region

In this study, the area of organic coffee production on each farm ranged from 0.5 to 3 ha, which

falls within the global definition of “smallholder” farms (World Bank 2003; Conway 2011;

Graeub et al. 2016; Lowder et al. 2016). This size range is representative of organic farms in the

Turrialba region (eco-LOGICA 2017), however is much smaller than nearby countries including

Nicaragua and El Salvador (Méndez et al. 2010). Using data from organic farms in the region

(eco-LOGICA 2017), I determined the regional mean organic coffee farm size as 1.57 ha. In

analysis, farms were divided into those above the regional mean and those below the regional

mean.

The farms in this study all followed guidelines provided by the eco-LOGICA® certification

(Naturalba 2018), however farmers had different approaches to their management practices.

Particularly, the frequency of weeding and canopy openness varied between farms. Farmers

controlled their herbaceous community by chopping it with a machete, weed-wacker or a shovel

two times to six times per year. Farms that weeded two times per year or less were classified as

having low weeding intensity (Soto-Pinto et al. 2002). Farms that were weeded between three

and five per year were classified having medium weeding intensity, which is the recommended

weeding schedule by the local weed management guide (Filho et al. 2013). Farms that weeded

more than six times per year were considered as having high weeding intensity. Farms also had

different levels of canopy management from varied shade tree planting (as seen in Table 3.1) and

pruning practices. Canopy openness was measured using hemispherical canopy image analysis; n

= 3 per plot. Overall, canopy openness ranged 10.6% to 40.47%. Canopy openness of less than

20

20% was considered low, canopy openness of 20-30% was classified as medium. Farms with

over 30% canopy openness were classified has high canopy openness.

3.2 Study design

At each site, three sampling quadrats of 1m x 1m were randomly selected (Nkoa et al. 2015)

within which all sampling of the herbaceous community and soil was conducted (Figure 3.3).

This resulted in a total of 45 quadrats in the study design sampling (Table 3.2).

3.2.1 Aboveground herbaceous community identification and sampling

Photos of herbaceous species were taken in the field and identified using a local identification

guide (Laurito et al. 2016), CATIE resources and on-site expertise. Herbaceous species cover

was determined by the percentage of physical space each species covered in the plot, determined

by visual inspection using a 1m x 1m grid system (Carmona et al. 2015). The vegetative height

was based off of the tallest individual from each species to account for competitive capacity. The

vegetative height of each species was determined by measuring the distance between the upper

boundary of the main photosynthetic tissues of the tallest herbaceous species and the soil (Pérez-

Harguindeguy et al. 2013).

To determine the herbaceous community biomass, water content and functional traits, all

herbaceous plants within the quadrat were clipped with scissors and transported in coolers

(Butterfield & Suding 2013) to the CATIE lab. Upon arrival at the lab, the wet mass was

recorded for each herbaceous species. In cases where plant material could not be weighed

immediately, samples were wrapped in moist paper towel and placed in sealed plastic bags. To

reduce transpiration water loss, I breathed into the plastic bags to increase CO2 concentration and

air humidity before placing bags in a dark refrigerator (as suggested by Pérez-Harguindeguy et

al. 2013). After wet weight was recorded, herbaceous species biomass was dried at 65°C for 72

hours and weighed to determine dry mass.

21

Figure 3.3 Example of 1m x1m sampling quadrat. Plant cover and exposed soil was recorded,

and each species was identified. All herbaceous community biomass within the quadrat was cut

and taken to the lab for further analysis. Soil sampling was conducted in the middle of the

quadrat (where meter sticks cross).

22

Table 3.2 Ecological variables measured in this study including leaf functional traits, whole-plant

traits and environmental data. Name Abbreviation Ecological Function Unit

1. Leaf Traits

Specific Leaf Area

(herbaceous community) SLA

Related to growth capacity and photosynthesis

activity.

Mm2

mg-1

Leaf dry-matter content

(herbaceous community) LDMC Related to resource acquisition mg g-1

Leaf nitrogen

concentration

(herbaceous community)

LNC Related to resource acquisition mg g-1

Leaf carbon concentration

(herbaceous community) LCC Related to rates of carbon accumulation mg g-1

2. Whole-plant Traits

Vegetative height

(herbaceous community) H

Competitive capacity and species response to

environmental conditions

cm

Basal diameter

(coffee only) BD Used to determine age of plant mm

Biomass (herbaceous

community only) Biomass

Competitive capacity and species response to

environmental conditions g

Yield (coffee only) Yield Productivity of coffee plant g plant-1

3. Environmental Data

Soil moisture sm Indicates soil water availability %

Total soil carbon SoilC Indicates soil carbon storage mg g-1

Total soil nitrogen SoilN Pool of potentially mineralizable N mg g-1

Soil phosphorous SoilP Mediates the availability of P to plants mg g-1

Soil ammonia/nitrates SoilAN

Pool of readily available mineral nitrogen to

plants

mg g-1

Distance to shade trees Distance

Potential for nitrogen-fixation and organic

material addition, and related to canopy

coverage

m

Canopy Openness CO Indicator of light that reaches herbaceous

community %

23

3.2.2 Leaf functional trait analysis

The leaf functional trait sampling of all herbaceous species was performed following protocols in

Pérez-Harguindeguy et al. 2013. To determine specific leaf area (SLA) and leaf dry-matter

content (LDMC), I selected five fully expanded young leaves from the upper 20% of height from

different plants of the same species within the quadrat to ensure consistency across sites.

Whenever possible, leaves with pathogenic or pest attack symptoms were avoided (Pérez-

Harguindeguy et al. 2013). The five fresh leaves from each species per plot were blotted with

dry paper towel to remove surface water, flattened and photographed alongside a ruler. These

images were later analyzed using ImageJ software to determine average leaf area (mm2) per each

species per quadrat (Abramoff et al. 2004). These leaves were then dried at 65°C for 72 hours.

The dry leaves’ mass was measured and recorded in mg. Specific leaf area was determined using

by leaf area (mm2) /oven dry mass of leaf (mg). Using the same leaves in SLA analysis, LDMC

was determined as the mass of dry leaves (mg)/wet mass of leaves (g). Overall, 1105 herbaceous

leaves from the 45 plots were analyzed.

After measuring SLA and LDMC, dry leaves were placed in labelled envelopes and transferred

to the Isaac Agroecology Lab for chemical analysis, following Canadian import permit

regulations. Each replicate was ground separately with a mechanical grinder and dried at 65°C

for 12 hours before chemical analysis (Pérez-Harguindeguy et al. 2013). The material was then

analyzed with a LECO CN628 analyzer to determine leaf carbon concentration (mg g-1) and leaf

nitrogen concentration (mg g-1) (LECO Corporation, Minnesota, USA). Throughout the analyses

aspartic acid was tested to ensure accuracy.

3.2.3 Coffee plant measurements

Variety, plant height, yield, age and coffee leaf rust incidence were measured for all coffee plants

within a 1m radius of the experiment quadrat. Coffee variety was determined through discussions

with farmers and confirmation using the World Coffee Research Coffee Variety Guide (2016).

Coffee plant height was measured from the highest photosynthetic leaf and the base of the coffee

plant (Cornelissen et al. 2003). The age of coffee plants was measured using a digital caliper

approximately 10cm above ground or stump-pruned growth. Based on allometric data, the

equation:

24

coffee age= (10.36- diameter in mm)/4.64 (1)

This equation was used to determine plant age (Audebert 2011). The age of coffee plants

determined through this equation matched farmer knowledge.

Coffee yield was determined by the equation:

coffee yield = (8.58 + 3.88*(Number of Productive Stems per Plant) + 1.95×(Number of Fruiting Nodes) + 0.03*(Number of Fruiting Nodes per Plant) − 0.18 Number of Dead Branches)2,

(2)

developed by Cerda et al. (2017b). The equation provides grams of fresh coffee per plant, which

is helpful in determining the impact of herbaceous communities on the yields of surrounding

coffee plants. Coffee leaf rust was determined by choosing three random branches, one from

upper, middle, and lower heights of coffee plant (Soto-Pinto et al. 2002). Number of leaves on

branch with rust were measured and incidence was determined through equation:

coffee leaf rust incidence = number of leaves with infection / (number of leaves with infection + number of leaves without infection) *100.

(3)

3.2.4 Shade tree and canopy measurements

As shown in Table 3.1 a variety of shade trees were present on farm sites. The distance to shade

trees within a 10m diameter of the centre of the quadrat was measured and significant

characteristics (i.e. pruning, disease) were noted. To measure the light levels reaching the

herbaceous community, digital fisheye photographs were taken at a height of 60cm at the centre

of each quadrat. Photographs were taken with a Nikon Coolpix 950 digital camera was used with

a Nikon Fisheye Converter FC-E8 0.21x lens. Gap Light Analyzer (GLA) software (Frazer et al.

1999) was used to determine total light transmission and canopy openness (%).

3.2.5 Soil sampling and analysis

Once the biomass of the herbaceous community was cleared for analysis, soil samples were

taken. Using a soil corer (111cm3), samples were taken from the centre of each quadrat at a

depth of 10cm. Soil bulk density was determined by drying soil at 105˚C for 72 hours and

25

dividing the dry soil weight (g), sieved to 2mm, by soil volume (cm3). Soil moisture was

determined by using the equation:

soil moisture = ((wet soil mass (g) – dry soil mass) (g) / dry soil mass (g)* 100%). (4)

For soil phosphorous analysis, 4g of soil was airdried. For soil carbon and soil nitrogen analysis,

10g of soil was dried. Soil samples were brought to the Isaac Agroecology Lab following

Canadian import permit regulations.

In the lab, soil available nitrates and phosphorous were determined with a flow injection analyzer

(Lachat QuikChem, Colorado USA). For nitrate analysis, a soil subsample of 2 g was placed in

Erlenmeyer flasks and 20 mL of potassium chloride (KCl) was added. This solution was shaken

for 30 minutes and then filtered through #1 Whatman filter paper into glass vials. Subsamples

from each vial were analyzed with a flow injection analyzer (Lachat QuikChem, Colorado USA)

to determine ammonium (mg g-1) and nitrates (mg g-1) colourmetrically. For soil available

phosphorus analysis, air dried and sieved soil was placed in in Erlenmeyer flasks and 20 mL of

Brays 1 was added, shaken for 5 minutes and the mixture was filtered through #1 Whatman filter

paper into glass vials. Subsamples from each vial were analyzed with a flow injection analyzer

(Lachat QuikChem, Colorado USA) to determine soil phosphorus (mg g-1) colourmetrically.

Total soil carbon and total soil nitrogen were measured by weighing 100mg of dried soil and

running through a CHN628 analyzer (LECO Corporation, Minnesota, USA).

3.3 Farmer perspectives of the herbaceous community 3.3.1 Participant selection The APOYA network of organic coffee farmers was contacted to determine the initial list of

project participants. Using a snowball technique, all connections were made with farmers

through consensual introductions. All farmers followed the criteria outlined in 3.1.4. Ethics

approval from the University of Toronto Social Sciences, Humanities, and Education Research

Ethics Board for research involving human participants was obtained. Participant selection and

interview process followed protocols outlined in the approved Ethics application including

ensuring informed consent and confidentiality of project participants.

26

This research project was assisted by a significant trust developed between me, an international

research student, and organic coffee farmers who have long been implementing agroecological

practices. Building trust was facilitated by the Isaac Agroecology Lab’s 10-year relationship with

CATIE and local partners including the APOYA network. Moreover, my positionality as a

farmer in Canada and a Spanish-speaker provided the skills to support farmers with their on-farm

activities, allowing for relationships to build prior to interviews.

3.3.2 Interview format

Interviews were conducted at locations and times that were convenient for participant (Bryman

2012). Interviews took place in participants’ houses, offices, in car rides to farms, or on farm

(Figure 3.4). These semi-structured interviews lasted between 20 and 100 minutes depending on

farmers’ availability and elaboration on interview questions. All questions (Appendix B) were

asked to farmers, though many participants answered multiple questions in one response. The

words “monte” (greenery/cover crop) was used in place of “hierba” (“weed”) at the start of the

interview to avoid influencing interviewees towards a negative association with the word “weed”

and to discuss the herbaceous community in general. Throughout this paper the term “herbaceous

community” includes all good, neutral and bad herbaceous species. All interviews were

conducted in Spanish, recorded and saved in a password protected encrypted folder. Interviews

were translated directly into English and anonymized.

3.3.3 Participant information

Information on participant demographics, history of the land, farm characteristics, management

practices and participant perspectives on herbaceous communities was collected. All participants

were asked questions about their management practices including their transition to organic

agriculture, herbaceous community management strategy and overall perspective of the

herbaceous community within their farm. Farmers were encouraged to discuss their perspective

on the ecosystem services and disservices provided by herbaceous communities. In cases when

participants’ response to one question answered later questions, these questions were not asked

27

Figure 3.4 A farmer demonstrates his knowledge of the herbaceous community within his

organic coffee agroforestry system during an interview. This farmer specifically discussed his

values of ecosystem services of pollinating herbaceous species such as the flowers he holds in

hand, and the beneficial ground coverage and soil erosion protection of Commelina diffusa,

which he calls “canutillo” or “oreja de ratón” (mouse ear), on his farm.

28

again. Interview format ensured that all participants had responded to all questions. Interviews

ended in a discussion about current educational resources available to farmers to support their

herbaceous community management and what resources would be useful in the future.

3.3.4 Interview processing: cognitive mapping and valuation of services To best understand farmer decision-making practices within their complex agroecosystem, a

cognitive mapping approach using Decision Explorer software was employed (Banxia Software

Ltd. 2014). Key concepts on farmer values, perspectives on role of ecosystem (dis)services of

the herbaceous community and management practices were identified from interview transcripts

and coded by giving common labels to reoccurring themes (Özemi & Özemi 2004; Bryman

2012). Through an iterative process of re-listening to the interviews, these coded labels reflected

farmer-identified concepts as much as possible, resulting in a total 45 concepts. Based on

interviews, I determined start and end points as the basis of each cognitive map (Isaac et al.

2009). The starting point for each map was “conversion to organic agriculture” and the ending

point was “healthy coffee plants” which relates to economic viability including coffee yields and

quality.

Cognitive maps were analyzed for connection-to-variable ratio, density and domain and

centrality variables. The connection-to-variable ratio is the number of links compared to the

amount of farmer listed variables in each map. This ratio helps to determine the complexity of

participant thinking (Dodouras & James 2007) about the interconnectedness of their farm (Isaac

et al. 2009). The density of cognitive map is determined by the number of connections that

farmers see between concepts compared to the total possible number of connections (Hage &

Harary 1983). The measure of cognitive map density is a useful tool to determine the

comparable complexity of farmer’s cognitive map (van Winsen et al. 2013). If the density of a

map is high, this signifies that farmers will see many relationships between the variables and will

have more options for implementing change (Özesmi & Özesmi 2004; Isaac et al. 2009). To

determine density, I used the equation:

density= connections/number of variables*(number of variables-1) (Hage & Harary 1983)

(5)

29

I determined domain and centrality variables for each of the cognitive maps using analysis within

Decision Explorer software. The highest domain variables are the concepts with the most in-and-

out linkages to variables, whereas the highest centrality variables are those with the highest

number of direct and indirect links to other variables. These results help to determine prominent

variables and trends across interviews (van Winsen et al. 2013).

Finally, values of ecosystem services were quantified by determining the time farmers spoke

about each ecosystem service and disservice in relation to the total time of the interview

(Goodwin & Hertiage 1990; Bryman 2012) and is therefore presented in a percentage. Since

farmers spent a significant amount of time discussing the farm history and management

practices, value of ecosystem services never accounted for more than 10% of the interview.

3.4 Statistical analysis

Statistical analysis was performed in RStudio statistical analysis software version 3.3.3. All trait

and environmental data were checked for normality using fitting distributions approach

(Delignette-Muller & Dutang 2015). Where data were not normally distributed (i.e. FRic, FDiv,

LDMC, LNC, soilC, soilN, soiP, herbaceous community biomass) log-transformed values were

used in analysis.

The FD package (Laliberté et al. 2015) was employed to determine FRic, FEve, FDiv, FDQ and

community weighted means for the herbaceous community traits per plot. The Functional

Diversity package utilized data from the relative abundance of each species and trait-data per

species per quadrat to determine functional diversity indices and community weighted mean

values. These outputs were used in standardized major axis bivariate analysis, one-way analysis

of variance and stepwise-regression analysis.

Principal component analysis (PCA) was employed using the “vegan” r package (Oksanen et al.

2016) to determine the relationship between farmer perception of the herbaceous species and the

species’ functional traits. Four leaf traits (SLA, LDMC, LNC and LCC) and one whole-plant

trait (height) of the herbaceous species were used. Based on these analyses, PCA axis 1 and 2 for

each species were calculated.

30

To determine the relationship between herbaceous community functional diversity metrics and

biomass, soil conditions and coffee health correlates, I employed standardized major axis

bivariate analysis. To understand the effect of management approaches on the herbaceous

community trait indices, I employed one-way analysis of variance (ANOVA) and Tukey post-

hoc test to determine the significant differences within the herbaceous community’s traits across

farm size, weeding intensity and canopy openness. To determine if and how farmer perceptions

and attributes may predict functional diversity of the herbaceous community, Akaike’s

Information Criteria (AIC) was employed. The full model was of the form:

Functional Diversity response ~ farmer attributes [years organic +value of ecosystem services +

cognitive map connection-to-variable ratio + cognitive map density]

Using the full model, AIC analysis provided most parsimonious model fit to each response

variable. Significance of predictor variables in each AIC selected model was then assessed using

multiple regression.

31

Chapter 4 – Results

4.1 Taxonomic and functional composition of the herbaceous community

In total 39 herbaceous species were present across the 45 plots, with a mean of 4.82 (± 0.23)

species per plot. Of the species present, 36% were native to Central America and 55% native to

the Americas. The remaining 45% of species were native to Asia, Africa, Australia and Europe.

The herbaceous species in the plots represented a total 20 taxonomic families. Nearly 55% of

herbaceous species found in this study were considered beneficial plants (“buena hierba” or

“buena cobertura”), 21% were considered bad weeds (“mala hierba” or “hierbas competidoras”)

and 24% of species were perceived to neither have a beneficial role nor cause harm to their

coffee plants and were considered neutral (“hierba regular”) by farmers in this study. All

herbaceous communities present were naturally occurring and had not been planted by farmers

(personal correspondence, May 2018).

Herbaceous community biomass ranged from 71.4-3524.1g per plot with a mean of 88.5 (± 15.7)

percent cover per plot. The five most common species (Table 4.1) were found in over 33% of the

sampled plots. All 39 species and their functional traits are presented in Appendix A. The mean

herbaceous community species leaf dry-matter content (LDMC) ranged from 95-500mg g-1 and

the mean specific leaf area (SLA) ranged from 8.18 to 59.30mm2 g-1. The mean leaf nitrogen

concentration (LNC) ranged from 22.32-83.10 mg g-1 and the mean leaf carbon concentration

(LCC) ranged from 316.35-526.20mg g-1. The height of the tallest species in each plot ranged

from 3cm to 138cm. Herbaceous community functional richness ranged from 0.003 to 4.32. The

range of functional evenness was 0.01 to 0.98, and functional divergence was from 0.37 to 1.

Across all plots, FDQ values ranged from 0.07 to 3.9.

4.2 Herbaceous community functional diversity, soil conditions and coffee health correlates

Standardized major axis regression analysis revealed many relationships between herbaceous

community functional diversity, soil conditions and coffee health (Table 4.2). Total herbaceous

community biomass and functional richness were significantly positively correlated (r2=0.163;

p=0.003). Herbaceous community functional richness and soil moisture were significantly

32

Table 4.1 Most frequent herbaceous species found in plots are listed here in order of frequency. Taxonomic family and scientific name are

given (Laurito et al. 2016), as well as place of origin. Farmer perception of species as well as notes from farmer interviews are presented.

Scientific Name Taxonomic

Family Place of origin

Frequency

(n=45)

Farmer

Perception of

Species

Farmer perception from interviews

Commelina diffusa Commelinaceae Asia 36 Good

Soft plant that is easy to work with.

Good ground cover. Contains

beneficial soil nutrients. Edible.

Brachiaria

platyphylla Poacea North America 26 Bad

Challenging grass to work with. Grows

quickly and spreads easily.

Hydrocotyle

mexicana Araliaceae North America 20 Good

Soft plant that is easy to work with.

Good ground cover. Contains

beneficial soil nutrients. Medicinal.

Pseudelephantopus

spicatus Asteraceae Central America 18 Bad

Grows quickly and can be difficult to

work in. However, can attract

pollinators to coffee plants.

Cyperus tenuis Cyperaceae Central America 15 Bad Very competitive. Grows and spreads

quickly.

33

Table 4.2 Bivariate relationships among functional diversity indices, herbaceous community, soil and coffee health metrics (where n=45 for all values except for FEve and FDiv* where n=42). Indices and metrics that have been log-transformed are marked with an asterisk (*). The upper section of this matrix displays the slopes and associated 95% confidence intervals for each relationship, based on standardized major axis regression analysis. The lower section of the matrix displays model r2 and one-tailed p-values (in brackets) for each bivariate model. Significant relationships (p ≤ 0.05) are bolded.

Functional diversity indices H.C. metrics Soil metrics Coffee health metrics

FRic* FEve FDiv* FDQ Biomass* SoilC* SoilN* Soil Moisture SoilP* Yield CLR

FRic* - -2.64

(-3.61, -1.93)

6.15 (4.50, 8.39)

-1.59 (-2.05, -1.23)

1.51 (1.14, 1.99)

-5.01 (-6.74, -3.72)

-5.97 (-7.98 -4.47)

-0.030 (-0.04, -0.02)

1.20 (0.90, 1.60)

-6.78 (-9.17, -5.01)

35.54 (26.26, 48.09)

FEve 0.018 (0.201) -

-2.33

(-3.14, -1.72)

4.06 (3.07, 5.36)

-0.57 ( -0.75, -0.44)

-1.87 (-2.57, -1.37)

-2.25 (-3.08 -1.64)

0.011 (0.01, 0.02)

-0.46 (-0.63, -0.34)

-17.40 (-23.65, -12.80)

-94.2 (-128.93, -68.86)

FDiv* 0.021 (0.182)

0.083 (0.032) -

9.43 (6.90, 12.91)

0.25 (0.18, 0.34)

-0.81 (-1.09, -0.60)

-0.97 (-1.30, -0.72)

-0.005 (-0.006, -0.003)

0.20 (0.15, 0.27)

40.47 (29.66, 55.21)

-219.18 (-299.08, -160.63)

FDQ 0.299

(<0.001) 0.213

(0.001) 0.005

(0.324) -

-2.30

(-3.10, -1.86)

7.95 (5.91, 10.70)

9.49 (7.10, 12.69)

0.05 (0.036, 0.063)

-1.90 (-2.52, -1.43)

0.23 (0.18, 0.31)

-0.04 (-0.06, -0.03)

Biomass* 0.163 (0.003)

0.264 (<0.001)

0.004 (0.352)

0.292 (<0.001) -

-3.32 (-4.46, -2.47)

-3.96 (-5.29, -2.96)

-0.020 (-0.03, -0.015)

0.79 (0.59, 1.07)

-10.22 (-13.83, -7.56)

53.60 (40.01, 71.82)

SoilC* 0.035 (0.109)

<0.001 (0.500)

0.089 (0.027)

0.044 (0.085)

0.056 (0.059) -

0.84 (0.76, 0.92)

166.99 (127.43, 218.82)

-4.19 (-5.62, -3.12)

33.93 (25.43, 45.28)

-177.91 (-236.65, -133.74)

SoilN* 0.086 (0.025)

<0.001 (0.497)

0.108 (0.017)

0.081 (0.029)

0.085 (0.026)

0.888 (<0.001) -

199.27 (152.81, 259.86)

-5.00 (-6.69, -3.73)

40.49 (30.19, 54.30)

-2.12.30 (-281.15, -160.32)

Soil Moisture 0.116 (0.011)

0.077 (0.037)

0.028 (0.144)

0.187 (0.001)

0.184 (0.002)

0.208 (<0.001)

0.237 (<0.001) -

-0.25 (-0.03, -0.19)

0.20 (0.15, 0.27)

-1.07 (-1.41, -0.80)

SoilP* 0.105 (0.015)

0.043 (0.094)

0.001 (0.409)

0.138 (0.006)

0.037 (0.103)

0.060 (0.052)

0.067 (0.042)

0.209 (<0.001) -

-8.11 (-10.90, -6.03)

42.50 (32.10, 56.28)

Yield 0.002 (0.390)

0.048 (0.082)

0.026 (0.160)

0.082 (0.029)

0.005 (0.317)

0.096 (0.020)

0.063 (0.049)

0.199 (0.001)

0.047 (0.077) -

-0.20 (-0.25, -0.14)

CLR 0.003 (0.364)

0.005 (0.327)

0.023 (0.168)

0.012 (0.010)

0.069 (0.041)

0.116 (0.011)

0.144 (0.005)

0.135 (0.006)

0.145 (0.005)

0.172 (<0.001) -

34

negatively correlated (r2=0.116; p=0.011), as were functional richness and soil nitrogen (r2=

.086; p=0.025). Herbaceous community functional richness and soil phosphorous were

significantly positively correlated (r2=0.105; p=0.015).

The herbaceous community biomass and functional evenness were significantly negatively

correlated (r2=0.264; p<0.001). There was a significantly positive relationship between

functional evenness and soil moisture (r2=0.077; p=0.037). Total herbaceous community

functional evenness and functional divergence had a significantly negative correlation (r2=0.083;

p=0.032). Total herbaceous community functional divergence and soil nitrogen were

significantly negatively correlated (r2=0.108; p=0.017). Functional divergence and soil carbon

were also significantly negatively correlated (r2=0.089; p=0.027). The herbaceous community

FDQ and functional richness was significantly negatively correlated (r2=0.299; p<0.001).

However, FDQ was significantly positively correlated with functional evenness (r2=0.213;

p=0.001). Herbaceous community FDQ and biomass were negatively correlated (r2=0.292;

p<0.001). Herbaceous community FDQ was significantly positively correlated with both soil

nitrogen (r2=0.081; p=0.029) and soil moisture (r2=0.187; p<0.001) but was significantly

negatively correlated with soil phosphorus (r2=0.138; p=0.006). The FDQ of the herbaceous

community was significantly positively correlated with coffee yield (r2=0.082; p=0.029) and

significantly negatively correlated with coffee leaf rust (r2=0.012; p=0.010).

The herbaceous community biomass and soil moisture were negatively correlated (r2=0.184;

p=0.002). Total herbaceous community biomass and soil nitrogen were also negatively

correlated (r2=0.085; p=0.026). Total herbaceous community biomass and coffee leaf rust were

significantly positively correlated (r2=0.069; p=0.041). As expected, soil carbon and soil

nitrogen were significantly positively correlated (r2=0.888; p<0.001). Soil carbon was positively

correlated with soil moisture (r2=0.208; p<0.001) and coffee yield (r2=0.096; p=0.020), whereas

soil carbon and coffee leaf rust were significantly negatively correlated (r2=0.116; p=0.011).

Soil nitrogen and soil moisture were significantly positively correlated (r2=0.237; p<0.001),

whereas soil nitrogen was significantly negatively correlated with soil phosphorous (r2=0.067;

p=0.042). Soil nitrogen was positively correlated with yield (r2=0.063; p=0.049) and negatively

correlated with coffee leaf rust (r2=0.144; p=0.005).

35

Soil moisture and soil phosphorous were significantly negatively correlated (r2=0.209; p<0.001).

Soil moisture was significantly positively correlated with coffee yield (r2=0.199; p=0.001); and

was significantly negatively correlated with coffee leaf rust (r2=0.135; p=0.006). Soil

phosphorous and coffee leaf rust were significantly positively correlated (r2=0.145; p=0.005).

As expected, coffee yield and coffee leaf rust were negatively correlated (r2=0.172; p<0.001).

4.3 Functional diversity across farm sizes, weeding and canopy openness

Multi and single-trait indices of the herbaceous community varied across farm size (Table 4.3).

The herbaceous community on farms less than the regional mean had significantly higher FDQ

values (2.23 ±0.90) than farms greater than the regional mean (1.16 ±0.97). Farms less the

regional mean size also had higher values of CWMLNC (log-value 1.58 ±0.42 mg g-1) as

compared to farms greater than the regional mean farm size (log-value 1.51 ±0.13 mg g-1). There

were no significant differences in herbaceous community functional richness, functional

evenness, functional divergence, CWMLDMC, CWMSLA or CWMLCC across farm size.

Analysis of variance showed significant (p<0.05) effect of weeding intensity for three multi-and

single-traits functional traits of the herbaceous community (Table 4.4). Herbaceous community

functional richness was lowest (-1.18 ± 0.83) on farms with high weeding intensity, compared to

farms with low (-0.27 ± 0.54) and moderate (0.18 ± 0.39) weeding intensities. Farms with

moderate weeding intensity had the lowest FDQ (1.20 ± 0.99) compared to low (2.24 ± 0.78) and

high (2.56 ± 0.76) weeding intensities. The herbaceous community CWMLDMC was lowest on

farms with low weeding intensity (log-value 2.17 ± 0.10 mg g-1) as compared to moderate

weeding intensity (log-value 2.31 ± 0.14 mg g-1).

The level of canopy openness had an effect on three multi-and-single-traits of the herbaceous

community (Table 4.5). The herbaceous community under a high level of canopy openness

exhibited significantly lower functional evenness (0.37 ± 0.30) than herbaceous community

under a low level of canopy openness (0.69 ± 0.19). A high level of canopy openness also had

an herbaceous community with low FDQ (1.03 ± 0.99) compared to plots with medium canopy

openness (2.02 ± 1.01). As well, high canopy openness had an herbaceous community with

36

Table 4.3 One-way analysis of variance of herbaceous community multi-trait and community weighted mean (CWM) single-trait

indices across farm size. Regional mean farm size determined from data on all organic farms in the Central Valley region (1.57 ha).

Farms this study are thus describes as above regional mean and below regional mean. Mean and standard errors (where n=45 for all

values except for FEve and FDiv* where n=42) are presented. Indices that have been log-transformed are marked with an asterisk

(*). F and p-values are provided. A Tukey post-hoc test was used and is denoted by superscripts a and b, whereby letters indicate

significant difference between levels (where p ≤ 0.05). Significant p values are bolded.

Trait Indices Farm Size

Below regional mean Above regional mean F-value p-value

Multi Trait

Indices

FRic* -0.27 ± 0.81a 0.07 ± 0.49a 3.028 0.089

FEve 0.55 ± 0.22a 0.41 ± 0.27a 3.32 0.076

LogFDiv* -0.10 ± 0.079a -0.13 ± 0.13a 0.739 0.398

FDQ 2.23 ± 0.90a 1.16± 0.97b 14.266 <0.001

Single Trait

Indices

CMWLDMC* 2.28 ± 0.13a 2.29 ± 0.14a 0.064 0.801

CMWSLA 30.61 ± 5.03a 27.27 ± 10.31a 1.713 0.198

CWMLNC* 1.58 ± 0.04a 1.51 ± 0.13b 4.893 0.033

CWMLCC 421.19 ± 11.01a 400.44 ± 45.29a 3.973 0.053

37

Table 4.4 One-way analysis of variance of herbaceous community multi-trait and community weighted mean (CWM) single-trait indices

across weeding intensities. Mean and standard errors (where n=45 for all values except for FEve and FDiv* where n=42) are presented.

Indices that have been log-transformed are marked with an asterisk (*). F and p-values are provided. A Tukey post-hoc test was used and

is denoted by superscripts a and b whereby letters indicate significant difference between levels (where p ≤ 0.05). Significant p values are

bolded.

Trait Indices Weeding Intensity

Low Medium High F-value p-value

Multi-trait

indices

FRic* -0.27 ± 0.54a 0.18 ± 0.39a -1.18 ± 0.83b 20.07 <0.001

FEve 0.53 ± 0.21a 0.45 ± 0.29 a 0.48 ± 0.10 a 0.33 0.724

FDiv* -0.13 ± 0.86a -0.12 ± 0.12 a -0.08 ± 0.04a 0.37 0.694

FDQ 2.24 ± 0.78a 1.20 ± 0.99b 2.56 ± 0.76a 7.91 0.001

Single-trait

indices

CWMLDMC* 2.17 ± 0.10a 2.31 ± 0.14b 2.27 ± 0.11ab 4.42 0.018

CWMSLA 29.65 ± 12.47a 28.02 ± 7.00 a 30.92 ± 5.91a 0.38 0.687

CWMLNC* 1.55 ± 0.04a 1.54 ± 0.12a 1.58 ± 0.05a 0.26 0.770

CWMLCC 417.63 ± 11.74a 407.29 ± 40.65a 421.66 ± 6.86a 0.63 0.535

38

Table 4.5 One-way analysis of variance of herbaceous community multi-trait and community weighted mean (CWM) single-trait indices

across levels of canopy openness. Mean and standard errors (where n=45 for all values except for FEve and FDiv* where n=42) are

presented. Indices that have been log-transformed are marked with an asterisk (*). F and p-values are provided. A Tukey post-hoc test

was used and is denoted by superscripts a and b whereby letters indicate significant difference between levels (where p ≤ 0.05).

Significant p values are bolded.

Trait Indices Canopy Openness

Low Medium High F-value p-value

Multi-trait

indices

FRic* -0.15 ± 0.69 a -0.25 ± 0.74 a 0.17 ± 0.50 a 1.849 0.170

FEve 0.69 ± 0.19a 0.47 ± 0.20ab 0.37 ± 0.30b 4.391 0.019

FDiv* -0.12 ± 0.061 a -0.11 ± 0.10 a -0.11 ± 0.14 a 0.0306 0.970

FDQ 1.53 ± 0.95ab 2.02 ± 1.01a 1.03 ± 0.99b 4.361 0.019

Single-trait

indices

CWMLDMC* 2.22 ± 0.20 a 2.30 ± 0.13 a 2.31 ± 0.09 a 1.0253 0.368

CWMSLA 31.04 ± 12.32 a 29.43± 4.63 a 26.38 ± 8.88 a 1.0789 0.349

CWMLNC* 1.59 ± 0.062ab 1.57 ± 0.11a 1.49 ± 0.08b 4.0911 0.029

CWMLCC 420.11 ± 15.35a 412.15 ± 37.37 a 404.73 ± 39.72 a 0.579 0.057

39

significantly lower CWMLNC (1.49 ± 0.08mg g-1) than medium canopy openness (1.57 ± 0.11 mg

g-1). The level of canopy openness did not have a significant effect on functional richness,

functional divergence or on the CWMLDMC, CWMSLA or CWMLCC of the herbaceous community.

4.4 Farmer perception of the herbaceous community

I interviewed 11 farmers and farm-workers who represent over 50% of organic coffee farmers in

the region. Of the organic farmers interviewed 64% were farm owners and 36% were farm

workers who did not own their land. Only 9% of farmers interviewed were female, however 27%

of land was owned by females. Whereas 91% of farmers interviewed were male, however 73%

of land was owned by males. The mean age of farmers interviewed was 57.8 ± 5.4 years. All

participants had been farmers for their whole lives and had farmed both conventionally and

organically. All farmers interviewed had participated in workshops on organic coffee production

lead by regional institutions (iCafe, CATIE, and APOYA). The farms’ time since conversion to

organic coffee production ranged from 6 to 22 years with a mean of 16.9 ± 3.6 years. During

informational interviews, farmers responded to the question “what motivated you to transition to

organic coffee production?” Overall, there were eight motivations named (Table 4.6). The most

common motivation to switch to organic production, stated by 45% of participants, was that it

was “better for the earth and/or soil”, followed by for “better health.”

The cognitive maps derived from interviews with farmers had a mean of 17 ± 2.4 variables and

26.8 17 ± 4.5 connections between them (Table 4.7; see Figure 4.1 for example). The highest

domain value was organic matter followed by healthy coffee plants. The highest centrality

variable was healthy coffee plants, followed by soil nutrients, soil moisture and organic matter,

soil moisture and erosion control (Table 4.8). The mean density of the maps was 0.77 ± 0.05,

which suggests that level of complexity in herbaceous community management was similar

between participants and that there are many options to implement change.

During farm tours and interviews, as well as in a local guidebook (Filho et al. 2013), farmers

referred to herbaceous species as bad when they were competitive “hierbas competidoras,” or

when they were difficult to manage, “malas hierbas,” such as species within the Poaceae (grass)

family. Herbaceous communities were considered good when they were seen as a good

40

Table 4.6 Guiding motivations that farmers indicated for switching to organic agriculture ranked

from most frequent response to least frequent. Some farmers had multiple motivations for

transitioning from conventional to organic agriculture.

Value Number of

responses

Better for earth and/or soil 5

Better health 3

To better provide for my family 2

To join a movement/good culture with organic

agriculture 2

Out of conviction/religious reasons 2

Higher prices /better quality coffee 2

Educational purposes 2

To diversify farm 1

41

Table 4.7 Number of connections, variables, connection-to-variable ratio, and map density derived from cognitive maps. Value of

ecosystem services determined through interview analysis is also provided.

Farmer Years Organic Connections (C) Variables (V) C:V Cognitive Map

Density Ecosystem Services Valuation (%)

1 6 24 15 1.6 0.8 2.45

2 18 24 16 1.5 0.75 2.19

3 18 30 17 1.76 0.88 1.93

4 18 21 14 1.5 0.75 3.71

5 15 30 20 1.5 0.75 5.44

6 14 29 18 1.61 0.81 7.57

7 22 29 18 1.61 0.81 7.82

8 20 23 17 1.35 0.68 2.45

9 15 30 18 1.67 0.83 5.32

10 18 34 21 1.62 0.81 3.00

11 18 21 15 1.4 0.7 3.01

42

Figure 4.1 A sample cognitive map demonstrating a farmer’s perception of ecosystem services (in green text), disservices (red text), and

key factors that affect decision making and control of weeds (blue text). Solid lines represent positive relationships whereas dashed lines

represent negative relationships.

43

Table 4.8 Highest domain and centrality variable derived from cognitive map analysis using

Decision Explorer (Banxia Software Ltd. 2014). Highest domain variables are concepts with

most in and out linkages to other concepts in the cognitive maps. Highest centrality variables are

concepts with the highest number of direct and indirect links to other variables.

Farmer Highest domain variables Highest centrality variables

1 healthy coffee plants, organic material healthy coffee plants, soil nutrients

2 organic material soil moisture

3 healthy coffee plants, organic material healthy coffee plants, soil moisture

4 organic material healthy coffee plants, organic material

5 healthy coffee plants, organic material healthy coffee plants, erosion control

6 healthy coffee plants, organic material healthy coffee plants, organic material

7 healthy coffee plants, organic material soil moisture

8 healthy coffee plants, organic material healthy coffee plants, soil nutrients

9 healthy coffee plants, organic material healthy coffee plants, soil nutrients

10 organic material erosion control

11 organic material healthy coffee plants, organic material

44

groundcover, “buena cobertura,” or “buena hierba” when they played an important role such as

soil coverage, or nutrient cycling for instance species within the Fabaceae, Commelinaceae or

Convolvulaceae family. Farmers perceived herbaceous species that were neither good nor bad as

“regular”, or neutral plants which did not harm their crop but did not provide a noticeable

service, such as species within the Amaranthaceae family. Principal component analysis assessed

the multivariate relationship across morphological traits, where PCA axis 1 explained 30.8% of

the total variation (Fig 4.2). Qualitatively, herbaceous community categories of good, neutral,

and bad broadly overlap with these data indicating traits along the axis 1 align with HC

categories. Based on this, herbaceous species with high heights and high LDMC were considered

bad by farmers, whereas herbaceous species with increased SLA and LNC were considered

good, species with high LCC were considered neutral by farmers.

In order to capture and rank participants’ perception of ecosystem services in organic coffee

production, the percentage of time discussing services in the interview was operationalized as a

metric of ecosystem services. Time allocated to discussing ecosystem services showed a nearly

four-fold increase among participants ranging from 1.9% to 7.6% of informational interviews

(Table 4.7). Participants’ value of ecosystem services was varied across gender and was not

significantly related to management practices (weeding intensity, farm size, number of years

organic and shade tree management).

4.5 Social and ecological predictors of herbaceous community composition and farm health

Using stepwise regression, farmer attributes that predicted functional diversity of the herbaceous

community were identified (Table 4.9). This analysis determined that cognitive map connection-

to-variable ratio was the best predictor for functional richness (model r2=0.151; p=0.012). With a

positive coefficient, this indicates that an increase in connection-to-variable ratio was positively

associated with functional richness. Connection-to-variable ratio was also the significant farmer

attribute in predicting functional evenness of the herbaceous community (model r2=0.173;

45

Figure 4.2 Principal component analysis of the community weighted mean values for four leaf-

level traits (specific leaf area, leaf dry-matter content, leaf nitrogen concentration, leaf carbon

concentration) and one whole-plant trait (plant height) of 39 herbaceous species found in 45

organic plots across the Turrialba region. Colours represent farmer perspective on herbaceous

species as being good, neutral or bad within their organic coffee agroforestry farms. Ellipses

correspond to 95% confidence ellipses for community weighted mean values for herbaceous

species sampled in this study.

46

Table 4.9 Stepwise and multiple regression model analysis to determine the farmer attributes that best predict indices of herbaceous community functional diversity. Indices that have been log-transformed are marked with an asterisk (*). Parameter estimates and p-values are shown for parameters retained in most parsimonious AIC-selected model. Parameters in bold are significant (p<0.05) in a multiple regression analysis. AIC values for full model and most parsimonious model are presented and ΔAIC values representing the difference between the two. Full model was of the form: functional diversity response ~ farmer attributes [years organic (yorg) +value of ecosystem services (ves) + cognitive map connection-to-variable ratio (cv) + cognitive map density (density)]

Model AIC-retained

parameters

Coefficient (p- value)

FullAIC

AIC ΔAIC Model r2 (p-value)

Starting Equation

Most parsimonious

equation FRic* (Intercept) -2.77

(0.121) -33.02 -40.51 3.93 0.151

(0.012) FRic*~yorg+ves +cv +density

FRic*~ cv

cv 30.51

(0.039)

density -57.77

(0.057)

FEve (Intercept) 1.70 (<0.001)

-117.1 -119.77 2.67 0.172 (0.009)

FEve~ yorg + ves + cv +

density

FEve ~ cv

cv -11.27

(0.048)

density 21.04

(0.070)

FDiv* (Intercept) -0.12 (<0.001)

-180.4 -184.62 4.18 N/A FDiv*~ yorg+ ves+ cv+ density

FDiv*~ 1

FDQ (Intercept)

3.00

(<0.001)

-14.42

-14.42

0.00

0.423

(<0.001)

FDQ~yorg +ves+cv +

density

FDQ ~ ves + cv + density

yorg -0.075

(0.108)

ves -0.188

(0.027)

cv -88.27

(<0.001)

density 177.74

(<0.001)

47

p=0.009) but was significantly negatively associated. There was no measured farmer attribute

that predicted functional divergence of the herbaceous community. However, farmer value of

ecosystem services, cognitive map density and connection-to-variable ratio were all significant

predictors of FDQ of the herbaceous community (model r2=0.423; p<0.001). Stepwise regression

found that FDQ declined with increased farmer value of ecosystem services and connection-to-

variable ratio, however FDQ increased with increased cognitive map density.

48

Chapter 5 - Discussion

5.1 Unrecorded diversity in organic agroforestry systems

Organic agriculture has been reported to increase biodiversity in agricultural landscapes in

comparison to conventional management (Hyvönen & Salonen 2002; Nascimbene et al. 2012).

Previous research has determined that organic coffee agroforestry systems have greater shade-

tree taxonomic and functional diversity than conventional coffee production methods (Jose 2009;

Haggar et al. 2011; Toledo & Moguel 2012). This study provides insight into the previously

undocumented diversity of the herbaceous community in coffee agroforestry systems. Results

from this research project found that herbaceous communities not only increase species richness

but also functional diversity within coffee agroforestry systems, which has been linked to

efficient resource use (Storkey & Neve 2018), nutrient cycling (Méndez et al. 2010; Tully et al.

2013; Isbell et al. 2017) and resilience to disturbance (Norgrove & Beck 2016) in other

agroecosystems.

The herbaceous community in these organic coffee agroforestry systems was both taxonomically

and functionally diverse compared to coffee in monoculture or simplified high input agroforestry

systems (Rossi et al. 2011). Overall, there were 39 herbaceous species from 20 taxonomic groups

found in plots across this study. The species identified, however, only represent a portion of the

more than 200 herbaceous species commonly found in coffee farms in the region (Laurito et al.

2016). Within organic coffee agroforestry systems, the herbaceous community contributed

substantially to plant species richness where the only other source of diversity is coffee plants

and canopy shade-trees, typically composed of one to five species (Rossi et al. 2011; Gagliardi et

al. 2015; Cerda et al. 2017a). The taxonomic and functional diversity of the herbaceous

community within organic coffee systems supports the growing body of research that organic

agriculture provides an alternative to the perpetual decline in biodiversity within industrial

agriculture systems (Jones et al. 2018; Storkey & Neve 2018). While herbaceous community

diversity has been documented in temperate and Mediterranean climates (see Gaba et al. 2016),

my findings contribute some of the first insights into herbaceous community diversity in tropical

agroecosystems.

49

Interestingly, across all farms, high taxonomic and functional diversity of the herbaceous

community had no measurable negative impact on coffee yield under current weeding intensities

and shade-tree management practices. This may be because many herbaceous species possess

either different resource acquisition traits (Smith et al. 2009) or nutrient acquisition segregation

(Storkey & Neve 2018) than those of the crop. As a result, organic coffee agroforestry systems

can support a high diversity of herbaceous species without declining yield (Rossi et al. 2011).

The species within the herbaceous community in organic coffee agroforestry systems have a

wide range of leaf functional traits. Broad-leaf species such as Hydrocotyle bowlesioides had

high SLA and LNC values. These species were considered beneficial ground covers and green

manure by farmers, a relationship supported by similar research in Latin America (Rossi et al.

2011) and the Caribbean (Damour et al. 2014). Grasses and sedges in this study, such as

Brachiaria platyphylla and Cyperus tenuis had low LNC, SLA and high LDMC values. These

species are difficult to control due to their tough leaves (Labrada 1997; Pérez-Harguindeguy et

al. 2013) and are less beneficial to soil as they can have slower decomposition rates than leaves

with higher LNC and SLA. On a plot-scale, the mass ratio hypothesis explains that the most

abundant species will determine ecosystem function (Grime 1998; Díaz et al. 2007). For

example, farms with an abundance of Physalis angulata, a broad-leaf species, may have high

levels of photosynthesis (Grime 1998; Wright et al. 2004) and nutrient cycling as chemical

(CWMLNC) and structural traits (CWMSLA) that facilitate rapid decomposition will be dominant

in these plots. Conversely, farms with Paspalum fasciculatum, a grass species, as the dominant

species will likely have lower rates of decomposition as the CWMLNC and CWMSLA is low.

Overall, functional dissimilarity (FDQ) was the most significant functional diversity index of the

impact of management on herbaceous community resource use efficiency and complementarity,

as explored in the next section. Based on the large body of literature linking leaf functional traits

to ecosystem processes (Díaz & Cabido 2001; Botta-Dukát 2005; Violle et al. 2007; Cadotte et

al. 2011) one could expect that when the herbaceous community functional dissimilarity is high,

resource-use strategies will vary and lead to higher overall function.

50

5.2 Farm management and attributes strongly predict herbaceous community diversity

The farms in this study that were below the regional mean size had higher herbaceous

community CWMLNC and higher herbaceous community FDQ compared to larger farms. Since

leaf nitrogen concentration of plant communities is one of the strongest driving forces for

decomposition (Bakker et al. 2011), small farms would be expected to have enhanced nutrient

cycling, soil health and microbial populations in the soil (Yadvinder-Singh et al. 2005).

Integrating organic material with rapid N-mineralization potential, such as the herbaceous

community with high CWMLNC, promotes closed nutrient cycles (Tully & Ryals 2017) and

reduce inputs costs for farmers (Kilian et al. 2006). This has far-reaching implications as

inorganic nitrogen fertilizer is the single largest source of energy use in agricultural production

(Camargo et al. 2013). Smaller farms in this study also had herbaceous communities with higher

functional dissimilarity (FDQ), which is comprised of species diversity and distinctness

(Shimatani 2001; Lepš et al. 2006), compared to larger farms. Higher levels of functional

dissimilarity suggest that the herbaceous community will have more efficient and, often,

complementary use of soil water and nutrients (Loreau 1998; Holzwarth et al. 2015). Studies

from temperate landscapes have found that smaller sized fields have a positive effect on diversity

of herbaceous plants (Belfrage et al. 2015; Fahrig et al. 2015) and this study indicates that

functional dissimilarity could support diverse herbaceous communities with lower competition

for resources within tropical agroforestry systems.

While all farms in this study fell within the global standard of smallholder farms (World Bank

2003; Conway 2011; Graeub et al. 2016; Lowder et al. 2016), the decrease in CWMLNC and FDQ

on larger farms posits that as private foreign investments in agro-industry promote larger-scale

farms in Latin America (Chavarría 2015), plot-level nutrient cycles and resource-use efficiency

may decrease. This may because on smaller farms, farmers have more time to physically engage

with their crops and herbaceous communities, which may affect a farmer’s perception of

functional traits and subsequent management responses (Ntshangase et al. 2018). In the Central

Valley of Costa Rica, Isaac et al. (2018) found that on very small farms, farmers’ perception of

leaf nutrient requirements better matched documented leaf nutritional deficiencies of coffee

leaves. My research raises caution that larger-sized farms may see a reduction in previously

51

undocumented herbaceous community CWMLNC and FDQ which indicates less nutrient cycling

and greater herbaceous species similarity and competition for resources.

This study found that canopy openness and light dynamics also influence herbaceous community

functional diversity within coffee agroforestry systems. In this study, medium canopy openness

(20-30%) promoted the highest levels of herbaceous community FDQ and CWMLNC and medium

levels of FEve. The herbaceous community with canopy openness above 30% had lower levels of

FDQ and therefore likely more resource competition (Loreau 1998; Holzwarth et al. 2015) and

lower levels of CWMLNC which indicates a decreased level of decomposition and nutrient cycling

(Yadvinder-Singh et al. 2005). This finding is supported by studies in temperate forests where

canopy openness has been found to affect niche space differentiation in the herbaceous

community (Mason et al. 2013; Mouillot et al. 2013). In a clear-cut temperate forest, full canopy

openness resulted in lower levels of herbaceous species functional diversity as compared to

herbaceous communities within forests stands (Janēcek et al. 2013). The type of sunlight may be

an important factor as well. A recent study into herbaceous plant diversity within Taiwanese

conifer plantations found that herbaceous species diversity was significantly negatively

correlated with direct sunlight, but not with indirect sunlight (Liu et al. 2015). Shade trees have

been well documented for their beneficial role including supporting coffee yield stability (Staver

et al. 2001; DaMatta 2004), nutrient cycling (Beer et al. 1998; Cerda et al. 2017a) and pest

suppression (Cerdán et al. 2012) in coffee agroforestry systems. One farmer in this study

highlighted the benefits of shade-tree coverage, stating that “we needed to start managing shade

to help with weeds, so started to plant trees. We have found many advantages. We weed less and

have more tree products to harvest” (personal correspondence June 2018).

Some level of management to facilitate light entry into the forest floor is suggested for the

promotion of herbaceous communities with greater CWMLNC within temperate forests

(Kusumoto et al. 2015). Canopy openness and thus light dynamics are promoted in local

agroforestry guides within Costa Rica (Montagnini et al. 2015), but a specific level of canopy

openness is not suggested. This research suggests a canopy openness of 20-30% results in higher

levels of herbaceous community FDQ, FEve, and CWMLNC, which indicates higher nutrient

cycling and reduced levels of competition. However, this study was limited to exploring canopy

52

openness only; and recognizes that shade-tree architecture and pruning can influence light

transmission greatly (Beer et al. 1998) and pruning residues can also suppress herbaceous

community growth (Staver et al. 2001). To understand the interaction between shade-tree traits

and the herbaceous community, future studies should account for indirect and direct light

transmission, shade-tree root traits and canopy architecture.

Weeding intensity did have a significant effect on herbaceous community diversity. However,

the varied results suggest that the operationalization of farmer identified weeding intensities may

not be effective in capturing actual on-farm practices. The findings that low and high weeding

intensities resulted in the highest FDQ and CWMLDMC suggest that there may be other factors

within weeding intensity levels. One potential factor is that all high weeding intensity farms were

managed by women who, stated that “women never learn how to use a machete” (personal

correspondence, June 2018), and therefore outsource weeding to paid labour. This high weeding

intensity by outside labourers could be a result of low engagement with plants on the farm and

therefore a coarser scale weeding approach (Bellamy 2011; Maharani et al. 2018), than farmers

who have constant engagement with managing their herbaceous communities. Interestingly,

weeding intensity did not affect coffee yield nor coffee leaf rust, therefore, there may be

opportunities women farm owners to decrease labour costs by hiring weeding labour less

frequency. Overall, however, more research into the impact of weeding disturbance on

herbaceous community should be conducted in a controlled environment.

5.3 The role of farmer perception in driving diversity

All participants in this study indicated that ecological and health values were the main motivators

for their conversion to organic coffee production but varied in their specific reasoning to convert.

One participant mentioned that his primary reason for transition to organic production his

enjoyment of a “greener” life, saying “me gusta lo verde.” Another farmer was motivated to

leave conventional agriculture after developing persistent stomach pains from using Paraquat

herbicide. Others were driven to convert to organic agriculture by family and spiritual

motivations such as, “to take care of God’s garden.” The diversity of motivations for converting

to organic is an indication that while farmers implemented similar organic guidelines and

53

practices, their convictions and backgrounds are different. Such differences may inform a

diversity of connections to their farms (Kaufman & Mock 2014) and, therefore, perception of

their herbaceous community.

All farmers in this study placed value on the herbaceous community’s ecosystem service

provisioning (ranging from 1.9% to 7.8%), which outweighed their perception of ecosystem

disservices of the herbaceous community (ranging from 0.7% to 3.0%). One farmer shared his

admiration for the weeds in his herbaceous community saying that, “every plant has a role”

(personal correspondence, May 2018). Farmers placed the most value on soil health benefits of

the herbaceous community including “refreshing the soil/providing soil moisture” and

“providing soil nutrients” such as “nitrogen fixation,” and the herbaceous community’s overall

role supporting “soil health” and “soil erosion control.” Based on cognitive map analysis,

organic material was the highest domain variable in 100% of farmer interviews, meaning that

thematically soil organic matter was the most connected concept. This signifies that chopping

weeds, pruning shade trees and leaving material to decompose is an important value held by

farmers and can inform management (Lescourret et al. 2015), such as weeding practices and

shade tree management. Moreover, all farmers placed higher emphasis on soil health and organic

matter than coffee yield, which may be indicative of their role as land stewards (Jose 2009).

Farmer perception of the herbaceous community within their farm was related to herbaceous

community diversity. Farmer cognitive map connection-to-variable ratio, an indicator of

perceived interconnectedness between farm management variables (Isaac et al. 2009), was an

important predictor for the functional dissimilarity, functional evenness and functional richness

of their herbaceous community. Stepwise and multiple regression model analysis determined that

cognitive map connection-to-variable ratio was positively related to herbaceous community

functional richness, and negatively related to functional evenness and functional dissimilarity.

Moreover, farmer value of ecosystem service was a significant negative predictor of functional

dissimilarity. This result may be because farmers with high herbaceous community functional

richness and therefore species richness (Mason et al. 2005) saw more linkages between their

herbaceous community diversity and ecosystem services (Swift et al. 2004; Garcia et al. 2018)

than farmers with high herbaceous community functional evenness and/or dissimilarity. Recent

54

research (Isaac et al. 2018), and the PCA analysis in this study, indicate that farmers do have a

strong sense of crop and herbaceous community functional traits. However, functional diversity

indices remain understudied (Ronchi & Silva 2006) and underutilized in practice. The gap in

farmer knowledge about herbaceous community functional diversity and management indicates

that more research and dissemination is needed, particularly around how herbaceous community

functional dissimilarity can support resource-use efficiency while fostering biodiversity

(Karadimou 2016).

Although weed management is perceived to be one of the largest barriers for farmers to shift to

organic production (Lyngbæk et al. 2001), the Organic and Sustainable Producers Association

(APOYA) network has not yet had the capacity to provide workshops on organic weed

management (personal correspondence, July 2018). All participants in this study had attended

workshops on other organic coffee production practices such as shade-tree intercropping, but

none had attended a workshop focused specifically on herbaceous community management,

diversity and ecosystem function. The lack of a centralized information on the herbaceous

community is likely the reason that farmer perceptions of ecosystem services and disservices of

the herbaceous community varied greatly across farms. However, the strong social relationships

reinforced by the APOYA network allows for the diffusion of information and adoption of new

techniques (Rogers 2003). With this strong base, workshops on organic herbaceous community

management will be necessary to communicate practices that support soil health, coffee health

and reduce farmer labour (Valencia et al. 2015).

5.4 The potential for agroecosystems service provisioning

Ecosystems services of the herbaceous community observed in this study included increased soil

moisture, soil carbon and stable coffee yield. Farmers also described additional ecosystems

services provided by the herbaceous community that were not measured in this study but are

supported by other research within coffee agroforestry systems. These ecosystem services

included erosion control (Meylan et al. 2013), pollination (Gordon et al. 2007), fostering

biodiversity (Perfecto et al. 2014) and providing food for animals and medicine for human health

(Soto-Pinto et al. 2002). One farmer mentioned the nutrient benefits from his herbaceous

55

community stating that “weeds are the cheapest manure you can find,” (personal correspondence,

May 2018).

This study found key relationships between soil moisture and the herbaceous community:

significant positive relationship between herbaceous community functional evenness and soil

moisture and between functional dissimilarity and soil moisture. These findings are important as

soil moisture and water management are critical to coffee health and yield (Chemura 2014),

particularly as rainfalls in the region become increasingly unreliable (DaMatta 2004; Isaac et al.

2018) due to climate change (IPCC 2013). Coffee leaf rust was found to be negatively correlated

with herbaceous community functional dissimilarity. Overall, these results support research that

soil moisture is positively related to decomposition of organic matter, soil carbon, soil nitrogen

cycling, coffee yield (El-Kader et al. 2010) and the suppression of coffee leaf rust (Zambolim et

al. 1997).

The growth of the herbaceous community and continual integration of organic material can

increase soil water storage (Minasny & Mcbratney 2018), as well as soil microbial biomass

(Ghimire et al. 2017) and support long-term carbon sequestration (Karlen et al. 1994; Ghimire et

al. 2017). Since one gram of soil carbon equates to 3.66 grams of CO2 in the atmosphere

(Poeplau & Don 2015), the herbaceous community should be considered in Costa Rica’s climate

change goals of becoming the world’s first carbon neutral country by 2021 (Defrenet et al.

2016). Costa Rica’s elaborate payment for ecosystem service program may be able to encourage

practices that foster ecosystem services (Pagiola 2008; Saadun et al. 2018). Currently, Costa

Rica’s payment for ecosystem service program supports farmers who plant and conserve forests

that contribute to ecosystem services including the mitigation of greenhouse gases, water

cycling, biodiversity conservation and provision of scenic beauty (FONAFIFO 2000; 2005;

Pagiola 2008). Recent research has shown that this program disproportionately supports large

landowners and that there is an urgent need to better support smallholder farmers in gaining

more equitable access to payment opportunities (Lansing 2017). This study suggests that the

herbaceous community is a previously undocumented source of diversity and ecosystem service

provisioning and should be considered in future development of Costa Rica’s payment for

ecosystem service program.

56

Herbaceous community functional diversity had no measurable impact on coffee yield. While

yield in organic coffee agroforestry systems is generally lower than conventional systems

(Toledo & Moguel 2012), studies support the finding that organic systems can have higher weed

diversity while maintaining yields (Rossi et al. 2011). Furthermore, herbaceous community

diversity may contribute to more stable yields year-to-year (Yachi & Loreau 1999; Isbell et al.

2017), which can help farmers with income planning and financial management. In this study,

age and variety of coffee varied between farms, which may influence yield per plant, and

therefore age and variety of coffee should be held constant in future studies.

Higher herbaceous community biomass and coffee leaf rust were positively correlated, which

may be due to decreasing airflow around coffee plants which can foster disease (Arneson 2000).

This indicates that some optimal level of weeding is necessary to reduce disease. Farmers were

well aware of this, as one farmer noted that “tall weeds can heat up the coffee plants and spread

disease” (personal correspondence, June 18 2018). Management practices including low to

medium weeding intensity, 20-30% canopy openness and further education around the role of

functional dissimilarity within herbaceous communities, farmers can encourage the reduction in

ecosystem disservice associated with the herbaceous community (Filho et al. 2013; Power et al.

2010). Exploring ways to include herbaceous community ecosystem service provisioning may

promote management practices that foster herbaceous community soil carbon, soil nutrient

provisioning and biodiversity.

57

Chapter 6 - Conclusion 6.1 Conclusions

While the herbaceous community is ubiquitous in organic coffee agroecosystems, it has been

rarely studied and has never been looked at through a functional trait lens. My thesis aims to

provide insights into how farmer perceptions and management practices impact the functional

diversity of the herbaceous community and explore the subsequent ecosystem (dis)services the

herbaceous community provides in coffee agroforestry systems.

As I hypothesized, the herbaceous community varied across farms and management practices.

This study found that the herbaceous community is an important, yet previously undocumented,

source of diversity within coffee agroforestry plots, which are typically composed of only one to

five species (Rossi et al. 2011; Gagliardi et al. 2015; Cerda et al. 2017a). This study found that

the herbaceous community contributes to coffee system functional diversity and influences

ecosystem functions including plant-soil feedbacks and nutrient cycling (Garnier & Navas 2012).

Moreover, interviews demonstrated that similar to recent research for coffee leaf traits in the

region (Isaac et al. 2018), the organic coffee farmers in the Central Valley of Costa Rica are

aware of herbaceous community functional traits. Farmers in this study preferred herbaceous

species that had high SLA and LNC as these, often broadleaf, species are easier to control and

have higher nutrients and decomposition rates. Farmers perceived tall herbaceous species with

low SLA and LNC, and high LDMC to be undesirable, as they are often thicker and more

difficult to control and decompose more slowly. This study found that these traits can be

favoured on the plot-level through management practices.

This research found that size of farm and canopy openness influenced both single-trait and multi-

trait herbaceous community functional diversity. Plots below the average regional size of farm

fostered an herbaceous community with higher CWMLNC and FDQ. This research cautions that as

farm-size increases, farmers will have less time to engage with their plants which could result in

less functionally dissimilar herbaceous communities and therefore less resource-use efficiency

within the herbaceous community. This research also found canopy light management affected

the herbaceous community. Farmers who fostered 20-30% canopy openness through planting of

58

diverse shade-trees and annual pruning schedules (Montagnini et al. 2015) supported herbaceous

communities with higher functional evenness, FDQ and CWMLNC, which promotes resource-use

efficiency and higher nutrient cycling. This research also found that while farmer identified

weeding intensities may not be effective in capturing actual on-farm practices, weeding less than

five times per year saves labour costs and can foster an herbaceous community with lower

CWMLDMC, which is easier to mechanically chop, and supports higher resource-use efficiency

(Karadimou 2016). This may be particularly relevant for farmers who hire external labour to

weed their plots frequently, as higher levels of weeding did not result in increased coffee yield or

soil health.

While all farmers in this study believed that the herbaceous community provided multiple

ecosystem services, they were not uniform in their values. This is likely due to farmers’ values

being created from each participant’s individual experience, rather than a collective network.

Interviews and cognitive mapping indicate that there are opportunities to shift farmer practices.

Every farmer in this study said they would be interested in workshops on herbaceous community

management and the local organic network said that they would be very happy to share research

from any herbaceous community studies in the region. Workshops would be important spaces for

organic farmers in the region to share their values of ecosystem services and to build a

knowledge base of herbaceous community functional diversity and management to enhance

sustainable farm development (Isaac et al. 2009; Valencia et al. 2015).

Beyond increasing biodiversity and eliminating chemical herbicides, this study demonstrated

appropriate management practices of the herbaceous community can foster ecosystem services

including increased soil moisture, soil carbon and stable coffee yields. These benefits,

particularly the important role that soil carbon can play in the sequestration of atmospheric CO2

(Poeplau & Don 2015) are especially relevant as Costa Rica aims to become the world’s first

carbon neutral country (Defrenet et al. 2016). The role of the herbaceous community in

ecosystem service provisioning should be considered for future growth of Costa Rica’s payment

for ecosystem service program. The provisioning of services within small plots of land could

inform a more accurate and equitable program.

59

This study aimed to provide initial insight into the herbaceous community functional diversity

within coffee agroforestry systems. However, as herbaceous communities are one of the largest

perceived barriers for farmers to transition to organic agriculture in the region, it is necessary to

continue research into the herbaceous community, particularly from a functional trait lens to

connect these communities to ecosystem functions and processes.

6.2 Areas of future research

This study focused on the herbaceous communities in organic coffee farms, however,

conventional systems also have a present herbaceous community between sprays (Rossi et al.

2011). Research from orchards in Spain demonstrates that the herbaceous community within

plots that have chemical weed control were as high as plots without herbicide use (Mas et al.

2007). Research into the herbaceous community before and after herbicide application would

provide insight into the functional diversity of herbaceous communities within conventional

coffee systems. While organic systems provide significant biodiversity and conservation benefits

in comparison to coffee in monoculture (Toledo & Moguel 2012; Perfecto et al. 2014), future

research into the ecosystem services of herbaceous communities within conventional plots could

support more sustainable practices, such as the reduction of herbicide use, within conventional

farms.

While my research provides some of the first insights into links between herbaceous community

functional traits and ecosystem services within organic coffee systems, future studies with coffee

variety, coffee age, fertilizer application and time since weeding held constant will help to

improve the accuracy of measuring ecosystem services provided by herbaceous communities.

Furthermore, this study looked at the above-ground functional traits of the herbaceous

community but did not explore the below-ground functional traits. I recommend that a future

study explore the root functional traits of the herbaceous community to understand how

herbaceous species respond to available resources. As well, more research into shade tree light

dynamics and canopy architecture will provide useful insight to the role of canopy management

of the herbaceous community. Overall, further research will provide insight into how farmers can

better manage their herbaceous community for complementarity and resource-use efficiency.

60

References Abramoff, M. D., Magalhães, P. J., & Ram, S. J. (2004). Image processing with ImageJ.

Biophotonics Interantional. 11(7), 36-43. Retrieved September 6, 2018 from http://dspace. library.uu.nl/handle/1874/204900

Allinne, C., Savary, S., & Avelino, J. (2016). Delicate balance between pest and disease injuries yield performance, and other ecosystem services in the complex coffee-based systems of Costa Rica. Agriculture Ecosystems and Environment, 22(2), 1–12. Doi:10.1016/2016.02.001

Altieri, M. (2009). Agroecology, Small Farms, and Food Sovereignty. Monthly Review, 61(3), 102-113. Doi:1 0.14452/MR-061-03-2009-07_8

Arneson, P.A. (2000). Coffee rust. The Plant Health Instructor. Updated 2011. Retrieved March 2, 2018 from http://www.apsnet.org/edcenter/intropp/lessons/fungi/Basidiomycetes/ Pages/CoffeeRust.aspx

Atangana, A., Khasa, D., Chang, S. & Degrande, A. (2014). Definitions And Classification Of Agroforestry Systems. In Tropical Agroforestry, 35–47. Dordrecht: Springer Netherlands.

Audebert, L. (2011). Productivité aérienne du café agroforestier : effets de l’ombrage et de l’âge des rejets. Retrieved July 2018 from http://orton.catie.ac.cr/repdoc/A11074f/A11074f.pdf

Avelino, J., Cristancho, M., Georgiou, S., Imbach, P., Aguilar, L., Bornemann, G.…Morales, C. (2015). The coffee rust crises in Colombia and Central America (2008–2013): impacts, plausible causes and proposed solutions. Food Security, 7(2), 303–321. Doi:10.1007/s12571-015-0446-9

Ayalew, T. (2014). Characterization of Organic Coffee Production, Certification and Marketing Systems: Ethiopia as a Main Indicator: A Review. Asian Journal of Agricultural Research, 8(4), 170-180. Doi:10.3923/ajar.2014.170.180

Bakker, M. A., Carreño-Rocabado, G. & Poorter, L. (2011). Leaf economics traits predict litter decomposition of tropical plants and differ among land use types. Functional Ecology, 25(3), 473–483. Doi:10.1111/j.1365-2435.2010.01802.x

Balmford, A., Carey, P., Kapos, V. Manica, A. Rodrigues, A.S., Scharlemann, J.P.& Green, R. E. (2009). Capturing the many dimensions of threat: comment on Salafsky et al. Conservation Biology, 23(2), 482–487. Doi:10.1111/j.1523-1739.2009.01196.x

Bandeira, F. P. S. D. E. F., Blanco, J. L. & Toledo, V. M. (2002). Tzotzil Maya Ethnoecology: Landscape Perception and Management As a Basis for Coffee Agroforest Design. Journal of Ethnobiology, 22(2), 247–272. Retreived January 2019 from https://ethnobiology.org/sites/ default/files/pdfs/JoE/22-2/Bandeiraetal2002.pd

Banxia Software Ltd. (2014). Decision Explorer Student Version 3.2.3. Cumbria, UK: Kendal. Beer, J., Muschler, R., Kass, D. & Somarriba, E.(1998). Shade management in coffee and cacao

plantations. Agroforestry Systems, 38(1-3), 139-164. Doi:10.1023/A:1005956528316 Belfrage, K., Björklund, J. & Salomonsson, L. (2015). Effects of Farm Size and On-Farm

Landscape Heterogeneity on Biodiversity—Case Study of Twelve Farms in a Swedish Landscape. Agroecology and Sustainable Food Systems, 39(2), 170–188. Doi:10.1080/21683565.2014.967437

61

Bellamy, A. S. (2011). Weed control practices on Costa Rican coffee farms: Is herbicide use necessary for small-scale producers? Agriculture and Human Values, 28(2), 167–177. Doi:10.1007/s10460-010-9261-2

Bertrand B., Marraccini P., Villain L., Breitler J.C. & Etienne H. (2016). Healthy Tropical Plants to Mitigate the Impact of Climate Change—As Exemplified in Coffee. In: Torquebiau E. (Eds) Climate Change and Agriculture Worldwide. Dordrecht: Springer.

Botta-Dukát, Z. (2005). Rao’s quadratic entropy as a measure of functional diversity based on multiple traits. Journal of Vegetation Science, 16(5), 533–540. Doi:10.1111/j.1654-1103.2005.tb02393.x

Bryman, A. (2012). Social Research Methods Fourth Edition. New York: Oxford University Press.

Butterfield, B. J. & Suding, K. N. (2013). Single-trait functional indices outperform multi-trait indices in linking environmental gradients and ecosystem services in a complex landscape. Journal of Ecology, 101(1), 9–17. Doi:10.1111/1365-2745.12013

Cadger K, Quaicoo A.K, Dawoe D. & Isaac, M.E. (2016) Development interventions and agriculture adaptation: a social network analysis of farmer knowledge transfer in Ghana. Agriculture, 6(3), 32-46. Doi:10. 3390/agriculture 6030032

Cadotte, M.W., Carscadden, K. & Mirotchnick, N. (2011) Beyond species: functional diversity and the maintenance of ecological processes and services. Journal of Applied Ecology, 48(5), 1079–1087. Doi:10.1111/j.1365-2664.2011.02048.x

Cadotte, M.W. (2017).Functional traits explain ecosystem function through opposing mechanisms. Ecology Letters, 20,989–996. Doi:10.1111/ele.12796

Camargo G.G.T., Ryan M.G.& Richard T.M. (2013). Energy use and greenhouse gas emissions from crop production using the farm energy analysis tool. Bioscience 63,263–273. Doi:10.1525/bio.2013.63.4.6

Carmona, C. P., Rota, C., Azcárate, F. M. & Peco, B. (2015). More for less: Sampling strategies of plant functional traits across local environmental gradients. Functional Ecology, 29(4), 579–588. Doi:10.1111/1365-2435.12366

Celette, F., Findeling, A., & Gary, C. (2009). Competition for nitrogen in an unfertilized intercropping system: The case of an association of grapevine and grass cover in a Mediterranean climate. European Journal of Agronomy, 30(1), 41-51. Doi:10.1016/j.eja.2008.07.003

Cerda, R., Allinne, C., Gary, C., Tixier, P., Harvey, C. A….Avelino, J. (2017a). Effects of shade, altitude and management on multiple ecosystem services in coffee agroecosystems. European Journal of Agronomy, 82, 308–319. Doi:10.1016/j.eja.2016.09.019

Cerda, R., Avelino, J., Gary, C., Tixier, P., Lechevallier, E. & Allinne, C. (2017b). Primary and secondary yield losses caused by pests and diseases: Assessment and modeling in coffee. PLoS ONE, 12(1), 1–17. Doi: 10.1371/journal.pone.0169133

Cerdán, C. R., Rebolledo, M. C., Soto, G., Rapidel, B. & Sinclair, F. L. (2012). Local knowledge of impacts of tree cover on ecosystem services in smallholder coffee production systems. Agricultural Systems, 110, 119–130. Doi:10.1016/j.agsy.2012.03.014

62

Chavarría, H.(Ed). (2015). The Outlook for Agriculture and Rural Development in the Americas: A Perspective on Latin America and the Caribbean 2015-2016. San José, Costa Rica: ECLAC, FAO, IICA.

Chemura, A. (2014). The growth response of coffee (Coffea arabica L ) plants to organic manure, inorganic fertilizers and integrated soil fertility management under different irrigation water supply levels. International Journal of Recycling of Organic Waste in Agriculture,3(2), 59-70. Doi:10.1007/s40093-014-0059-x

Christen, B., Kjeldsen, C., Dalgaard, T. & Martin-Ortega, J. (2015). Can fuzzy cognitive mapping help in agricultural policy design and communication? Land Use Policy, 45, 64–75. Doi: 10.1016/j.landusepol.2015.01.001

Centro de Investigaciones Agronomicas (CIA). (2016). Mapa digital de suelos de Costa Rica. In: Centro De Investigaciones Agronómicas. Universidad de Costa Rica. Retrieved 15 April 2017 http://www.cia.ucr.ac.cr/?page id=139.

Collins, S.L., Carpenter, S.R., Swinton, S.M., Orenstein, D.E., Childers, D.L…Kaye, J. P. (2010). An integrated conceptual framework for long-term social–ecological research. Frontiers in Ecology and the Environment, 9(6), 351–357. Doi:10.1890/100068

Conroy, M. E., Murray, D. L. & P. M. Rosset. (1996). A cautionary tale: failed U.S. development policy in Central America. Boulder, CO: Lynne Rienner Publisher

Conway, G. (2011). On being a smallholder. IFAD conference on new directions for smallholder agriculture. Rome: IFAD. Retrieved January 2018 from http://www.ifad.org/events /agriculture/doc/papers/conway.pdf.

Cordell, D., Drangert, J. & White, S. (2009). The story of phosphorus: Global food security and food for thought. Global Environmental Change, 19(2), 292-305. Doi:10.1016/j.gloenvcha.2008.10.009

Cornelissen, J.H.C., Lavorel, S., Garnier, E., Díaz, S., Buchmann, N….Gurvich, D.E. (2003) A handbook of protocols for standardised and easy measurement of plant functional traits worldwide. Australian Journal of Botany, 51, 335–380. Doi: 10.1071/BT02124

DaMatta, F.M. (2004). Ecophysiological constraints on the production of shaded and unshaded coffee: a review. Field Crops Research, 86(2-3), 99–114. Doi:10.1016/j.fcr.2003.09.001

Damour, G., Dorel, M., Quoc, H. T., Meynard, C. & Risède, J. M. (2014). A trait-based characterization of cover plants to assess their potential to provide a set of ecological services in banana cropping systems. European Journal of Agronomy, 52, 218–228. Doi:10.1016/j.eja.2013.09.004

De Beenhouwer, M., Aerts, R. & Honnay, O. (2013). A global meta-analysis of the biodiversity and ecosystem service benefits of coffee and cacao agroforestry. Agriculture, Ecosystems and Environment, 175, 1–7. Doi:10.1016/j.agee.2013.05.003

Defrenet, E., Roupsard, O., Van Den Meersche, K., Charbonnier, F., Pérez-Molina, J. P., Khac, E., Prieto, I.,…Jourdan, C. (2016). Root biomass, turnover and net primary productivity of a coffee agroforestry system in Costa Rica: Effects of soil depth, shade trees, distance to row and coffee age. Annals of Botany, 118(4), 833–851. Doi.10.1093/aob/mcw153

63

Delignette-Muller, M.L.,& Dutang, C. (2015). fitdistrplus: An R Package for Fitting Distributions. Journal of Statistical Software, 64(4), 1-34. Retrieved September 12, 2018 http://www.jstatsoft.org/v64/i04/.

Derroire, G., Powers, J. S., Hulshof, C. M., Varela, L. E. C. & Healey, J. R. (2018). Contrasting patterns of leaf trait variation among and within species during tropical dry forest succession in Costa Rica. Scientific Reports,8(285), 1–11. Doi:10.1038/s41598-017-18525-1

Díaz, S. & Cabido, M. (2001). Vive la difference: plant functional diversity matters to ecosystem processes. Trends in Ecology & Evolution, 16(11), 646–655. Doi: 10.1016/S0169-5347(01)02283-2

Díaz, S., Lavorel, S., de Bello, F., Quetier, F., Grigulis, K., & Robson, T.M. (2007). Incorporating plant functional diversity effects in ecosystem service assessments. Proceedings of the National Academy of Sciences of the United States of America. 104: 20684-20689

Dickinson, A. K. (2017). Intraspecific Trait Variation in Cacao Agroecosystems: Influence of Local Conditions and Cultivars, and Role in Local Knowledge Systems (Unpublished dissertation thesis). University of Toronto, Toronto, Canada.

Dodouras, S. & James, P. (2007). Fuzzy cognitive mapping to appraise complex situations. Journal of Environmental Planning and Management, 50(6), 823–852. Doi:10.1080/09640560701610578

Dudley, N. & Alexander, S. (2017). Agriculture and biodiversity: a review. Biodiversity, 18(2-3), 45-49. Doi:10.1080/14888386.2017.1351892

eco-LOGICA. (2017). Certificado de Conformidad con Normas Organicas. Costa Rica: MAG. Eden, C. (2004). Analyzing cognitive maps to help structure issues or problems. European

Journal of Operational Research, 159(3), 673–686. Doi:10.1016/S0377-2217(03)00431-4

El-Kader A, Shaaban S, A. & El-Fattah M. (2010). Effect of irrigation levels and organic compost on okra plants (Abelmos- chus esculentus L.) grown in sandy calcareous soil. Agriculture and Biology Journal of North America, 1(3):225–231. Doi: 10.1007/s40093-014-0059-x

Fahrig, L., Girard, J., Duro, D., Pasher, J., Smith, A., Javorek, S., King, D.…Tischendorf, L. (2015). Farmlands with smaller crop fields have higher within-field biodiversity. Agriculture, Ecosystems and Environment, 200, 219–234. Doi:10.1016/j.agee.2014.11.018

Filho, d. M. V., E., Andrade, R. & Sanchez, L. (2013). Manejo Integrale Hierbas en Cafetales Guía Illustrativa. Utz Certified, Costa Rica. Retrieved May 12, 2018 from https://ra-training-library.s3.amazonaws.com/manejo-integral-hierbas-utz-red.pdf

Foley, J.A., Ramankutty, N., Brauman, K.A., Cassidy, E.S., Gerber, J.S….Zaks, D.P.M.(2011). Solutions for a cultivated planet. Nature, 478, 337-342. 10.1038/nature10452

Fondo Nacional de Financiamiento Forestal (FONAFIFO). (2000). El desarollo del sistema de pago de servicios ambientales en Costa Rica. San José, Costa Rica: FONAFIFO.

64

Fondo Nacional de Financiamiento Forestal (FONAFIFO).(2005). FONAFIFO: Más de una década de acción. San José, Costa Rica: FONAFIFO.

Fondo Nacional de Financiamiento Forestal (FONAFIFO). (2018). Pago de Servicios Ambientales. Retrieved 16 December 2018 http://www.fonafifo.go.cr/es/servicios/pago-de-servicios-ambientales/##pilares

Frazer, G.W., Canham, C. & Lertzman, K. (1999). Gap Light Analyzer (GLA), Version 2.0: Imaging software to extract canopy structure and gap light transmission indices from true-colour fisheye photographs, user manual and program documentation. Burnaby, British Columbia: Simon Fraser University, Burnaby, British Columbia, and New York: The Institute of Ecosystem Studies, Millbrook.

Gaba, S., Reboud, X. & Fried, G. (2016). Agroecology and conservation of weed diversity in agricultural lands. Botany Letters, 163(4), 351–354. Doi: 10.1080/23818107.2016.1236290

Gagliardi, S., Martin, A. R., Filho, E. de M. V., Rapidel, B., & Isaac, M. E. (2015). Intraspecific leaf economic trait variation partially explains coffee performance across agroforestry management regimes. Agriculture, Ecosystems and Environment, 200, 151–160. Doi:10.1016/j.agee.2014.11.014

Garcia, L., Celette, F., Gary, C., Ripoche, A., & Valdés-Gómez, H. (2018). Agriculture, Ecosystems and Environment Management of service crops for the provision of ecosystem services in vineyards : A review. Agriculture, Ecosystems and Environment, 251, 158–170.Doi:10.1016/j.agee.2017.09.030

Garnier, E., Cortez, J., Billes, G., Navas, M.L., Roumet, C., Debussche, M., Laurent, G., …Toussaint, J-P. (2004). Plant functional markers capture ecosystem properties during secondary succession. Ecology, 85(9), 2630–2637. Doi/10.1890/03-0799

Garnier, E. & Navas, M.L. (2012). A trait-based approach to comparative functional plant ecology: concepts, methods and applications for agroecology. A review. Agronomy for Sustainable Development, 32(2), 365–399. Doi:10.1007/s13593-011-0036-y

Gerowitt, B. (1997). Practical Use of Economic Thresholds for Weeds. In Labrada, R (Ed.). In Proceedings of the Expert Consultation on Weed Ecology and Management Conferences (8–13). Rome, Italy: Food and Agriculture Organization.

Ghimire, B., Ghimire, R., Vanleeuwen, D., & Mesbah, A. (2017). Cover Crop Residue Amount and Quality Effects on Soil Organic Carbon Mineralization. Sustainability, 9(12), 2316-2330. Doi:10.3390/su9122316

Goodwin, C. & Hertiage, J. (1990). Conversation Analysis. Annual Review of Anthropology, 19, 283-307.

Gordon, C., Manson, R., Sundberg, J. & Cruz-Angón, A. (2007). Biodiversity, profitability, and vegetation structure in a Mexican coffee agroecosystem. Agriculture, Ecosystems and Environment, 118(1–4), 256–266. Doi:10.1016/j.agee.2006.05.023

Graeub, B. E., Chappell, M. J. & Wittman, H. (2016). The State of Family Farms in the World. World Development, 87, 1–15. Doi:10.1016/j.worlddev.2015.05.012

Gray S.A., Zanre E., & Gray S.R.J. (2014). Fuzzy Cognitive Maps as Representations of Mental Models and Group Beliefs. In: Papageorgiou E. (Eds.) Fuzzy Cognitive Maps for Applied

65

Sciences and Engineering. Intelligent Systems Reference Library, vol 54. Berlin: Springer.

Green, J. M. (2014). Current state of herbicides in herbicide-resistant crops. Pest Management Science, 70(9), 1351–1357. Doi:10.1002/ps.3727

Greiner, R. (2015). Motivations and attitudes influence farmers’ willingness to participate in biodiversity conservation contracts. Agricultural Systems, 137, 154–165 Doi:10.1016/j.agsy.2015.04.005

Grime. J.P. (1998). Benefits of plant diversity to ecosystems: immediate, filter and founder effects. Journal of Ecology, 86(6), 902-910. Doi:10.1046/j.1365-2745.1998.00306.x

Hage, P.& Harary, F. (1983). Structural Models in Anthropology. New York: Oxford University Press.

Haggar, J., Barrios, M., Bolaños, M., Merlo, M., Moraga, P., Munguia, R., Ponce, A.,…Filho, d. M. F. V. (2011). Coffee agroecosystem performance under full sun, shade, conventional and organic management regimes in Central America. Agroforestry Systems, 82(3), 285-30. Doi: 10.1007/s10457-011-9392-5

Halbrendt, J., Gray, S. A., Crow, S., Radovich, T., Kimura, A. H. & Bahadur, B. (2014). Differences in farmer and expert beliefs and the perceived impacts of conservation agriculture. Global Environmental Change, 28, 50–62. Doi:10.1016/j.gloenvcha.2014.05.001

Hartwig, N. L. & Ammon, H. U. (2002). Cover crops and living mulches. Weed Science, 50(6):688–699.Doi: 10.1614/0043-1745(2002)050[0688:AIACCA]2.0.CO;2

Hoffman, J. (2014) The World Atlas of Coffee: From Beans to Brewing – Coffees Explored, Explained and Enjoyed. London: James Hoffman Octopus Publishing Group.

Holzwarth, F., Rüger, N. & Wirth, C. (2015). Taking a closer look: Disentangling effects of functional diversity on ecosystem functions with a trait-based model across hierarchy and time. Royal Society Open Science, 2(3). Doi: 10.1098/rsos.140541

Hyvönen, T. & Salonen, J.(2002). Weed species diversity and community composition in cropping practices at two intensity levels – a six-year experiment. Journal of Plant Ecology, 159(1), 73-81. Doi:10.1023/A:1015580722191

IFOAM. (2009). The IFOAM Norms for Organic Production and Processing Version 2005. Nürnberg, Germany: IFOAM.

Ingram, J., Gaskell, P., Mills, J., & Short, C. (2013). Incorporating agri-environment schemes into farm development pathways: a temporal analysis of farmer motivations. Land Use Policy, 31, 267–279. Doi: 10.1016/j.landusepol.2012.07.007

International Coffee Organization (ICO). (2016). World Coffee Production. Retrieved March 12, 2018 http://www.ico.org/prices/po-production.pdf.

International Panel on Climate Change (IPCC). (2013). Working Group I Contribution to the IPCC Fifth Assessment Report, Climate Change 2013. Geneva, Switzerland: The Physical Science Basis, United Nations Environmental Program.

66

Isaac, M. E., Dawoe, E., & Sieciechowicz, K. (2009). Assessing local knowledge use in agroforestry management with cognitive maps. Environmental Management, 43(6), 1321–1329. Doi:10.1007/s00267-008-9201-8

Isaac, M. E. (2012). Agricultural information exchange and organizational ties: The effect of network topology on managing agrodiversity. Agricultural Systems, 109, 9–15. Doi: 10.1016/j.agsy.2012.01.011

Isaac, M. E., & Matous, P. (2017). Social network ties predict land use diversity and land use change: a case study in Ghana. Regional Environmental Change, 17(6), 1823–1833. Doi: 10.1007/s10113-017-1151-3

Isaac, M. E., Martin, A. R., de Melo Virginio Filho, E., Rapidel, B., Roupsard, O., & Van den Meersche, K. (2017). Intraspecific Trait Variation and Coordination: Root and Leaf Economics Spectra in Coffee across Environmental Gradients. Frontiers in Plant Science, 8(July), 1–13. Doi:10.3389/fpls.2017.01196

Isaac, M. E., Cerda, R., Rapidel, B., Martin, A. R., Dickinson, A. K., & Sibelet, N. (2018). Farmer perception and utilization of leaf functional traits in managing agroecosystems. Journal of Applied Ecology, 55(1), 69–80. Doi:10.1111/1365-2664.13027

Isaacs, K. B., Snapp, S. S., Kelly, J. D. & Chung, K. R. (2016). Farmer knowledge identifies a competitive bean ideotype for maize–bean intercrop systems in Rwanda. Agriculture & Food Security, 5 (1), 15-33. Doi:10.1186/s40066-016-0062-8

Isakson, S. R. (2011). Market provisioning and the conservation of crop biodiversity: An analysis of peasant livelihoods and maize diversity in the Guatemalan Highlands. World Development, 39(8), 1444–1459. Doi:10.1016/j.worlddev.2010.12.015

Isakson, S. R. (2014). Maize Diversity and the Political Economy of Agrarian Restructuring in Guatemala. Journal of Agrarian Change, 14(3), 347-379. Doi: 10.1111/joac.12023

Isbell, F., Adler, P. R., Eisenhauer, N., Fornara, D., Kimmel, K., Kremen, C., … Polley, H. W. (2017). Benefits of increasing plant diversity in sustainable agroecosystems. Journal of Ecology, 105(4), 871–879. Doi:10.1111/1365-2745.12789

Janēcek, S., de Bello, F., Hornık, J., Bartōs, M., Černÿ T… Klimešová, J. (2013) Effects of land-use changes on plant functional and taxonomic diversity along a productivity gradient in wet meadows. Journal of Vegetation Science, 24(5), 898–909. Doi:10.1111/jvs.12012

Jones, K. R., Venter, O., Fuller, R. A., Allan, J, Maxwell, S.L., Negret, P.J. & Watson, J.E.M. (2018). One-third of global protected land is under intense human pressure. Science, 360(6390), 788-791. Doi: 10.1126/science.aap9565

Jones, N., Ross, H., Lynam, T., Perez, P., & Leitch, A. (2011). Mental models: An interdisciplinary synthesis of theory and methods. Ecology and Society, 16(1), 46. Retrieved November 2018 from https://www.ecologyandsociety.org/vol16/iss1/art46/

Jose, S. (2009). Agroforestry for ecosystem services and environmental benefits: An overview. Agroforestry Systems, 76(1), 1–10. Doi:10.1007/s10457-009-9229-7

Kazakou, E., Fried, G., Richarte, J., Gimenez, O., Violle, C., & Metay, A. (2016). A plant trait-based response-and-effect framework to assess vineyard inter-row soil management. Botany Letters, 163(4), 373–388. Doi:10.1080/23818107.2016.1232205

67

Karadimou, E. K., Kallimanis, A. S., Tsiripidis, I. & Dimopoulos, P. (2016). Functional diversity exhibits a diverse relationship with area, even a decreasing one. Scientific Reports, 6, 1–9. Doi:10.1038/srep35420

Karlen, D.L.,Wollenhaupt, N.C. Erbach, D.C., Berry, E.C. Swan, J.B. Eash, N.S. & Jordahl, J.L. Crop residue effects on soil quality following 10-years of no-till corn. (1994). Soil and Tillage Research, 31(2-3), 149–167. Doi:10.1016/0167-1987(94)90077

Kaufman, A. H., & Mock, J. (2014). Cultivating Greater Well-being: The Benefits Thai Organic Farmers Experience from Adopting Buddhist Eco-spirituality. Journal of Agricultural and Environmental Ethics, 27(6), 871–893. Doi:10.1007/s10806-014-9500-4

Kaye, J. P., & Quemada, M. (2017). Using cover crops to mitigate and adapt to climate change. A review. Agronomy for Sustainable Development. 3(4), 4-21. Doi:10.1007/s13593-016-0410-x

Keenan, R. J. (2017). Costa Rica: lessons from 30 years of forest and environmental policy, Revista Forestal Mesoamericana Kurú, 28(1-3) 10–13, Doi:10.18845/rfmk.v12i28.2094

Kilian, B., Jones, C., Pratt, L. & Villalobos, A. (2006). Is sustainable agriculture a viable strategy to improve farm income in Central America? A case study on coffee. Journal of Business Research, 59(3), 322–330. Doi: 10.1016/j.jbusres.2005.09.015

Kusumoto, B., Shiono, T., Miyoshi, M., Maeshiro, R. & Fujii, S. (2015). Functional response of plant communities to clearcutting : management impacts differ between forest vegetation zones. Journal of Applied Ecology, 52, 171–180. Doi:10.1111/1365-2664.12367

Labrada, R. (1997). Problems related to the development of weed management in the developing world. In Proceedings of the Expert Consultation on Weed Ecology and Management Conferences (8–13). Rome, Italy: Food and Agriculture Organization.

Laliberté, E., Legendre, P. & Shipl, B. (2015). Measuring functional diversity (FD) from multiple traits, and other tools for functional ecology. Retrieved October 2, 2019 from https://cran.r-project.org/web/packages/FD/FD.pdf

Lansing, D. M. (2017). Understanding Smallholder Participation in Payments for Ecosystem Services: the Case of Costa Rica. Human Ecology, 45(1), 77–87. Doi: 10.1007/s10745-016-9886-x

Laurito, V. N, Vindas, P. E. S., & Abarca, R. M. (2016). Hierbas y Arbustos: Comunes en Cafetales y Otros Cultivos: Guía para Su Identificacion. San José, Costa Rica:Herbario Juvenal Valerio Rodríguez.

Lavorel, S. & Garnier, E. (2002) Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Functional Ecology, 16(5), 545–556. Doi: 10.1046/j.1365-2435.2002.00664.x

Lēps, J., de Bello, F., Lavorel, S., & Berman, S. (2006). Quantifying and interpreted funcional diversity of natural communities. Preslia, 78, 481–501. Retrieved October 2018 from botanika.bf.jcu.cz/suspa/pdf/P064CLep.pdf

Lescourret, F., Magda, D., Richard, G., Adam-Blondon, A. F., Bardy, M., Baudry, J., … Soussana, J. F. (2015). A social-ecological approach to managing multiple agro-ecosystem services. Current Opinion in Environmental Sustainability, 14, 68–75. Doi:10.1016/j.cosust.2015.04.001

68

Liu, T., Lin, K., Vadeboncoeur, M. A., Chen, M., Huang, M., & Lin, T. (2015). Understorey plant community and light availability in conifer plantations and natural hardwood forests in Taiwan, Applied Vegetation Science, 18, 591–602. Doi:10.1111/avsc.12178

Loreau M. (1998). Separating sampling and other effects in biodiversity experiments. Oikos, 82, 600–602. Doi:10.2307/3546381

Lowder, S. K., Skoet, J., & Raney, T. (2016). The Number, Size, and Distribution of Farms, Smallholder Farms, and Family Farms Worldwide. World Development, 87, 16–29. Doi:10.1016/j.worlddev.2015.10.041

Lyngbæk, A.E., Muschler, R. & Sinclair, F.L. (2001). Productivity and profitability of multistrata organic versus conventional coffee farms in Costa Rica. Agroforestry Systems, 53(2), 205–213. Doi:10.1023/A:1013332722014

Maharani, C. D., Moeliono, M., Wong, G. Y., Brockhaus, M., Carmenta, R. & Kallio, M. (2018). Development and equity: A gendered inquiry in a swidden landscape. Forest Policy and Economics, 101, 120–128. Doi:10.1016/j.forpol.2018.11.002

Martin, A. R. & Isaac, M. E. (2015). Plant functional traits in agroecosystems: A blueprint for research. Journal of Applied Ecology, 52(6), 1425–1435. Doi:10.1111/1365-2664.12526

Martin, A. R., Rapidel, B., Roupsard, O., Van den Meersche, K., de Melo Virginio Filho, E., Barrios, M., & Isaac, M. E. (2017). Intraspecific trait variation across multiple scales: the leaf economics spectrum in coffee. Functional Ecology, 31(3), 604–612. https://doi.org/10.1111/1365-2435.12790

Martins, B.H., Araujo-Junior, C.F., Miyazawa, M., Vieira, K.M. & Milori, D.M.B.P. (2015). Soil organic matter quality and weed diversity in coffee plantation area submitted to weed control and cover crops management. Soil and Tillage Research, 153, 169–174. Doi:10.1016/j.still.2015.06.00

Mas, T.M, Poggio, S.L., Verdu, & A.M.C. (2007). Weed community structure of mandarin orchards under conventional and integrated management in northern Spain. Agriculture Ecosystems and Environment, 119(4), 305–310. Doi:10.1016/j.agee.2006.07.016

Mason, N. W. H., MacGillivray, K., Steel, J. B. &Wilson, B.W. (2003). An index of functional diversity. Journal of Vegetation Science, 14(4), 571-578. Doi: /10.1111/j.16541103.2003.tb02184.x

Mason, N. W. H., D. Mouillot, W. G. Lee, & J. B. Wilson. (2005). Functional richness, functional evenness and functional divergence: the primary components of functional diversity. Oikos, 111(1),112-118. Doi: 10.1111/j.0030-1299.2005.13886.x

Mason, N.W.H., de Bello, F., Mouillot, D., Pavoine, S. & Dray, S. (2013) A guide for using functional diversity indices to reveal changes in assembly processes along ecological gradients. Journal of Vegetation Science, 24(5), 794–806.Doi:10.1111/jvs.12013

Matos, F. S., Wolfgramm, R., Gonçalves, F. V., Cavatte, P. C., Ventrella, M. C., & DaMatta, F. M. (2009). Phenotypic plasticity in response to light in the coffee tree. Environmental and Experimental Botany, 67(2), 421–427. Doi:10.1016/j.envexpbot.2009.06.018

McGill, B.J., Enquist, B.J., Weiher, E. & Westoby, M. (2006). Rebuilding community ecology from functional traits. Trends in Ecology & Evolution, 21(4),178–185. Doi:10.1016/j.tree.2006.02.002

69

Méndez V.E., Bacon C.M., Olson M., Morris K.S., & Shattuck A. (2010). Agrobiodiversity and shade coffee smallholder livelihoods: A review and synthesis of ten years of research in Central America. Professional Geographer, 62(3), 357–376. Doi:10.1080/00330124.2010.483638

MEA. (2005). Millenium ecosystem assessment. In Ecosystems and human well-being: biodiversity synthesis. Washington, DC: World Resources Institute.

Meylan, L., A. Merot, C. Gary, & B. Rapidel. (2013). Combining a typology and a conceptual model of cropping system to explore the diversity of relationships between ecosystem services: The case of erosion control in coffee-based agroforestry systems in Costa Rica. Agricultural Systems, 118, 52-64. Doi:10.1016/j.agsy.2013.02.002

Minasny, B. & Mcbratney, B. M. (2018). Limited effect of organic matter on soil available water capacity. European Journal of Soil Science,69(1), 39–47. Doi:10.1111/ejss.12475

Montagnini, F. Sombarria, E., Murgueitio, E., Fassola, H. & Eibl, B. (2015). Sistemas Agroforestales Funciones Productivas, Socioeconómicas y Ambientales. Cali, Colombia: CIPAV and Turrialba, Costa Rica; Centro Agronómico de Investigación y Enseñanza.

Mora, A. & Beer, J. (2013). Geostatistical modeling of the spatial variability of coffee fine roots under Erythrina shade trees and contrasting soil management. Agroforestry Systems, 87(2), 365-376. Doi:10.1007/s10457-012-9557-x

Morlat, R., & A. Jacquet. (2003). Grapevine root system and soil characteristics in a vineyard maintained long- term with or without interrow sward. American Journal of Enology and Viticulture 54 (1): 1–7. Retrieved February 2019 from http://www.ajevonline.org/content/54/1/1

Moses, M., Johnson, E.S., Anger, W.K., Burse, V. W., Horstman, S. W… Zahm, S.H. (1993). Environmental equity and pesticide exposure. Toxicology and Industrial Health, 9(5), 913– 959. Doi:10.1177/074823379300900512

Mouillot, D., Graham, N.A.J., Villéger, S., Mason, N.W.H. & Bellwood, D.R. (2013). A functional approach reveals community responses to disturbances. Trends in Ecology and Evolution, 28(3), 167–177. DOI:10.1016/j.tree.2012.10.004

Myers, J. P., Antoniou, M. N., Blumberg, B., Carroll, L., Colborn, T., Everett, L. G… Benbrook, C.M. (2016). Concerns over use of glyphosate-based herbicides and risks associated with exposures: A consensus statement. Environmental Health, 15(1), 1–13. Doi:10.1186/s12940-016-0117-0

Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A. & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403 (6772), 853–858. Doi: 10.1038/35002501

Naeem, S., L. J. Thompson, S. P. Lawler, J. H. Lawton, & R. M. Woodfin. (1994). Declining biodiversity can alter the performance of ecosystems. Nature, 368,734-737. Doi: 10.1038/368734a0

Nascimbene, J., Marini, L. & Paoletti, M.G.(2012). Organic farming benefits local plant diversity in vineyard farms located in intensive agricultural landscapes. Environmental Management, 49 (5),1054-60. Doi10.1007/s00267-012-9834-5.

70

National Research Council. (1989). Alternative agriculture. Washington, DC: National Academy Press.

Naturalba. (2018). Our Certifications. Retrieved 14 November 2018 http://www.naturalba.net/ en/content/13-our-certifications

Nestel, D., & Altieri, M. A. (1992). The weed community of Mexican coffee agroecosystems: Effect of management upon plant biomass and species composition. Acta Oecologica, 13(6), 715–726.

Nestel, D. (1995). Coffee in Mexico: international market, agricultural landscape and ecology. Ecological Economics, 15(2),165-178. Doi:10.1016/0921-8009(95)00041-0

Newbold, T., Hudson L.N., Hill S.L., Contu, S., Lysenko, I., Senior, R., Börger, L… Purvis A. (2015). Global effects of land use on local terrestrial biodiversity. Nature, 520(7545), 45–50. DOI:10.1038/nature14324.

Njoroge, J. M. (1994). Weeds and weed control in coffee. Experimental Agriculture, 30(4), 421–429. Doi:10.1017/S0014479700024662

Nkoa, R., Owen, M. D. K., & Swanton, C. J. (2015). Weed Abundance, Distribution, Diversity, and Community Analyses. Weed Science, 63, 64–90. Doi:10.1614/WS-D-13-00075.

Norgrove, L., & Beck, J. (2016). Biodiversity Function and Resilience in Tropical Agroforestry Systems Including Shifting Cultivation. Current Forestry Reports, 2(1), 62–80. Doi:10.1007/s40725-016-0032-1

Novara, A., Gristina, L., Saladino, S., Santoro, A. & Cerdà, A. (2011). Soil erosion assessment on tillage and alternative soil managements in a Sicilian vineyard. Soil and Tillage Research, 117:140–147. Doi: 10.1016/j.still.2011.09.007

Ntshangase, N.L., Muroyiwa, B. & Sibanda, M. Farmers’ Perceptions and Factors Influencing the Adoption of No-Till Conservation Agriculture by Small-Scale Farmers in Zashuke, KwaZulu-Natal Province. (2018). Sustainability,10 (2), 555. Doi:10.3390/su10020555

Oksanen, J., Blanchet, G.F., Kindt, R., Legendre, P., Minchin, P.R….Wagner, H., (2016). Vegan: Community Ecology Package. Retrieved September 2018 from https://cran.r-project.org/web/packages/vegan/.

Özesmi, U., & Özesmi, S. L. (2004). Ecological models based on people’s knowledge: A multi-step fuzzy cognitive mapping approach. Ecological Modelling, 176(1–2), 43–64. Doi:10.1016/j.ecolmodel.2003.10.027

Pagiola, S. (2008). Payments for environmental services in Costa Rica. Ecological Economics, 65(4), 712–724.Doi:10.1016/j.ecolecon.2007.07.033

Pérez-Harguindeguy, N., Garnier, E., Lavorel, S., Poorter, H., Jaureguiberry, P., Cornwell, W.K., Craine, J.M…Cornelissen, J.H.C. (2013). New Handbook for standardized measurement of plant functional traits worldwide. Australian Journal of Botany, 61(34), 167–234. Doi:10.1071/BT12225

Perfecto I., Rice R., Greenberg M. & Van der Voort M.E. (1996). Shade coffee: a disappearing refuge for biodiversity. BioScience, 46(8), 598–608. Doi: 10.2307/1312989.

71

Perfecto, I., Vandermeer, J., & Philpott, S. M. (2014). Complex Ecological Interactions in the Coffee Agroecosystem. Annual Review of Ecology, Evolution, and Systematics, 45(1), 137–158. Doi:10.1146/annurev-ecolsys-120213-091923

Poeplau C. & Don, A. (2015). Carbon sequestration in agricultural soils via cultivation of cover crops—a meta-analysis. Agriculture, Ecosystems and Environment, 200, 33–41. Doi:10.1016/j.agee.2014.10.024

Power, A. G. (2010). Ecosystem services and agriculture: Tradeoffs and synergies. Philosophical Transactions of the Royal Society B, 365(1554), 2959–2971. Doi:10.1098/rstb.2010.0143

Pretty, J.N. (1995). Participatory learning for sustainable agriculture. World Development, 23(8), 1247-1263. Doi: 10.1016/0305-750X(95)00046-F

Raedeke A.H. & Rikoon J.S. (1997). Temporal and spatial dimensions of knowledge: implications for sustainable agriculture. Agriculture and Human Values, 14(2), 145–158. Doi: 10.1023/A:1007346929150

Rao, C.R. 1982. Diversity and dissimilarity coefficients: a unified approach. Theoretical Population Biology, 21(1), 24-43. Doi: 10.1016/0040-5809(82)90004-1

Relyea, R.A. (2005). The impact of insecticides and herbicides on the biodiversity and productivity of aquatic communities. Ecological Applications, 15(2), 618–627. Doi:10.1890/03-5342

Rogers, E.M. (2003). Diffusion of Innovations, 5th Edition. New York: Free Press. Ronchi, C. P. & Silva, A.A. (2006). Effects of weed species competition on the growth of young

coffee plants. Planta Daninha, 24(3),415-423. Doi:10.1590/S0100-83582006000300001. Rossi, E., Montagnini, F., & Filho, d. M. V., E. (2011). Effects of management practices on

coffee productivity and herbaceous species diversity in agroforestry systems in Costa Rica. In Montagnini, F., Francesconi, W. & Rossi, E. (Eds.). Agroforestry as a tool for landscape restoration (115-132). New York: Nova Science Publishers.

Ruthenberg, H. (1971). Farming Systems in the Tropics. Oxford: Clarendon Press.

Ryan, M. R., Smith, R. G., Mortensen, D. A., Teasdale, J. R., Curran, W. S., Seidel, R., & Shumway, D. L. (2009). Weed–crop competition relationships differ between organic and conventional cropping systems. Weed research, 49(6), 572-580. Doi: 10.1111/j.1365-3180.2009.00736.x

Saadun, N., Lim, E. A. L., Firdaus, M.A., Esa, S.M., Ngu, F...Azhar, B.(2018). Socio-ecological perspectives of engaging smallholders in environmental- friendly palm oil certification schemes. Land Use Policy, 72, 333-340. Doi: 10.1016/j.landusepol.2017.12.05

Sandhu, H. S., Wratten, S. D., & Cullen, R. (2010). Organic agriculture and ecosystem services. Environmental Sciences and Policy, 13, 1–7. Doi:10.1016/j.envsci.2009.11.002

Sarno, J.L., Afandi, T. Adachi, Y. Oki, M. Senge, & Watanabe, A. (2004). Effect of weed management in coffee plantation on soil chemical properties. Nutrient Cycling in Agroecosystems, 69(1), 1-4. Doi:10.1023/B:FRES.0000

Schleuter, A. D., Daufresne, M., Massol, F., & Argillier, C. (2010). A user’s guide to functional diversity indices. Ecological Monographs, 80(3), 469–484. Doi:10.1890/08-2225.1

72

Shackleton, C.M., Ruwanza, S., Sinasson Sanni, G. K., Bennett, S., De Lacy, P., Modipa, R., Mtati, N., & Sachikonye, M. (2016). Unpacking Pandora’s Box: Understanding and Categorising Ecosystem Disservices for Environmental Management and Human Wellbeing. Ecosystems, 19(4), 587–600. Doi:10.1007/s10021-015-9952-z.

Shimatani, K. (2001). On the measurement of species diversity incorporating species differences. Oikos, 93,135-147. Doi: 10.1034/j.1600-0706.2001.930115.x

Sick, D. (1999). Farmers of the golden bean: Costa Rican households and the global coffee economy. Dekalb, IL: Northern Illinois University Press.

Smith RG, Mortensen DA, & Ryan M.R. (2009). A new hypothesis for the functional role of diversity in mediating resource pools and weed-crop competition in agroecosystems. Weed Research, 50, 37–48. Doi:1111/j.1365-3180.2009.00745.x

Soto-Pinto, L., Perfecto I., Castillo-Hernández J., & Caballero-Nieto J. (2000). Shade effect on coffee production at the northern Tzeltal zone of the state of Chiapas, Mexico. Agriculture, Ecosystems and Environment, 80(1-2), 61–69. Doi: 10.1016/S0167-8809(00)00134-1

Soto-Pinto, L., Perfecto, I. & Caballero-Nieto, J. (2002). Shade over coffee: its effects on berry borer, leaf rust and spontaneous herbs in Chiapas, Mexico. Agroforestry Systems, 55(1), 37–45. Doi: 10.1023/A:102026670

Staver, C., Guharay, F., Monterroso, D., & Muschler, R.G. (2001). Designing pest suppressive multistrata perennial crop systems: shade grown coffee in Central America. Agroforestry Systems, 53 (2), 151–170. Doi:10.1023/A:1013372403359

Stier, A. C., Samhouri, J. F., Gray, S., Martone, R. G., Mach, M. E., Halpern, B. S., … Levin, P. S. (2017). Integrating Expert Perceptions into Food Web Conservation and Management. Conservation Letters, 10, 67–76. Doi:10.1111/conl.12245

Stirling, E.J, C. Filardi, A. Toomey, A. Sigouin, E. Betley, N. Gazit,…Jupiter, S.D. (2017). Biocultural approaches to well-being and sustainability indicators across scales. Nature Ecology & Evolution, 1, 1798–1806. Doi: 10.1038/s41559-017-0349-6.

Storkey, J. & Neve, P. (2018). What good is weed diversity? Weed Research, 58, 239–243. Doi:10.1111/wre.12310

Swift, M. J., Izac, A.-M. N., & van Noordwijk, M. (2004). Biodiversity and ecosystem services in agricultural landscapes—are we asking the right questions? Agriculture, Ecosystems & Environment, 104(1), 113–134. Doi:10.1016/j.agee.2004.01.013

Talhinhas, P, Batista, D., Diniz, I., Veiira, A., Silva, D….Silva, M. d. C. (2017). The coffee leaf rust pathogen Hemileia vastatrix : one and a half centuries around the tropics. Molecular Plant Pathology, 18(8), 1039-1051. Doi:10.1111/mpp.12512

Terry, P. J, Matthews, G., & Boonman, J.G. (1984). A Guide to Weed Control in East African Crops. Nairobi, Kenya: Kenya Literature Bureau

Tilman, D., Clark, M., Williams, D. R., Kimmel, K., Polasky, S., & Packer, C. (2017). Future threats to biodiversity and pathways to their prevention. Nature, 546(7656), 73–81. Doi:10.1038/nature22900

73

Toledo, V.M., & Moguel, P. (2012). Coffee and Sustainability: The Multiple Values of Traditional Shaded Coffee. Journal of Sustainable Agriculture, 36(3), 353–377. Doi: 10.1080/10440046.2011.583719

Tolman, E. C.(1948). Cognitive maps in rats and in men. Psychological Review, 55(4), 189– 208. Doi10.10372Fh0061626

Tscharntke, T., Clough, Y., Bhagwat, S.A., Buchori, D., Faust, H., Hertel, D.,…Wanger, T.C. (2011). Multifunctional shade-tree management in tropical agroforestry landscapes - a review. Journal of Applied Ecology, 48, 619–629. Doi: 10.1111/j.1365-2664.2010.01939.x.

Tully, K. L., Wood, S. A. & Lawrence, D. (2013). Fertilizer type and species composition affect leachate nutrient concentrations in coffee agroecosystems. Agroforestry Systems, 87(5), 1083–1100. Doi:10.1007/s10457-013-9622-0

Tully, K., & Ryals, R. (2017). Nutrient cycling in agroecosystems: Balancing food and environmental objectives. Agroecology and Sustainable Food Systems, 41(7), 761–798. Doi:10.1080/21683565.2017.1336149

United Stated Environmental Protection Agency (US EPA). 1997. R.E.D. Facts: Paraquat Dichloride. EPA-738-F-96-018. Washington DC: United States Environmental Protection Agency. Retrieved November 18, 2018 from https://archive.epa.gov/pesticides/reregistration/ web/pdf/0262fact.pdf

Valencia, V., West, P., Sterling, E. J., García-Barrios, L. & Naeem, S. (2015). The use of farmers’ knowledge in coffee agroforestry management: implications for the conservation of tree biodiversity. Ecosphere, 6(7), 1-17. Doi:10.1890/ES14-00428.

Van Dusen, E. M., & Taylor, J. E.(2005). Missing markets and crop diversity: evidence from Mexico. Environment and Development Economics, 10(4), 513-531.

van Winsen, F., de Mey, Y., Lauwers, L., Van Passel, S., Vancauteren, M. & Wauters, E. (2013). Cognitive mapping: A method to elucidate and present farmers’ risk perception. Agricultural Systems, 122, 42–52. Doi:10.1016/j.agsy.2013.08.003

Vandermeer, J. & Perfecto, I. (2015). Coffee Agroecology: A New Approach to Understanding Agricultural Biodiversity, Ecosystem Services and Sustainable Development. New York: Routledge.

Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I. & Garnier, E. (2007). Let the concept of trait be functional! Oikos, 116, 882–892. Doi: 10.1111/j.0030-1299.2007.15559.x

Wardle, D. A., Jonsson, M., Bansal, S., Bardgett, R. D., Gundale, M. J. & Metcalfe, D. B. (2012). Linking vegetation change, carbon sequestration and biodiversity: Insights from island ecosystems in a long-term natural experiment. Journal of Ecology, 100(1), 16–30. Doi:10.1111/j.1365-2745.2011.01907.x

Westoby, M. & Wright, I.J. (2006) Land-plant ecology on the basis of functional traits. Trends in Ecology & Evolution, 21(5), 261–268. Doi:10.1016/j.tree.2006.02.004

Willer, H. & Lernoud, J. (Eds.). (2017) The World of Organic Agriculture. Statistics and Emerging Trends. Frick and Bonn: FiBL and IFOAM – Organics International.

74

Woodbury, P. B., Kemanian, A. R., Jacobson, M. & Langholtz, M. (2017). Biomass and Bioenergy Improving water quality in the Chesapeake Bay using payments for ecosystem services for perennial biomass for bioenergy and biofuel production. Biomass and Bioenergy. Doi:10.1016/j.biombioe.2017.01.024

Woods, J., Williams, A., Hughes, J. K., Black, M. & Murphy, R. (2010). Energy and the food system. Philosophical Transactions of the Royal Society B: Biological Science, 365(1554), 2991-3006. Doi:10.1098/rstb.2010.0172

World Bank (2003). Reaching the rural poor: A renewed strategy for rural development. Washington, DC: World Bank. Retrieved January 2019 from http://documents. worldbank.org/curated/en/2003/08/7036682/ reaching-rural-poor-renewed-strategy-rural-development

World Coffee Research. (2016). Las Variedades de Café de Mesoamérica y el Caribe. College Station, Texas: World Coffee Research.

Wright, I. J., Reich, P. B., Westoby, M., Ackerly, D. D., Baruch, Z., Bongers, F., … Gulias, J. (2004). The worldwide leaf economics spectrum. Nature, 12, 821–827. Doi:10.1007/sll466-009-0028-z

Yachi, S., Loreau M. (1999). Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. PNAS, 96(4), 1463–8. DOI:10.1073/pnas.96.4.1463

Yadvinder-Singh, Bijay-Singh, & Timsina, J. (2005). Crop Residue Management for Nutrient Cycling and Improving Soil Productivity in Rice-Based Cropping Systems in the Tropics. Advances in Agronomy, 85, 269–407. Doi:10.1016/S0065-2113(04)85006-5

Zambolim L, Vale, F.X.R., Pereira, A.A.; Chaves, G.M. (1997). Café (Coffea arabica L.): controle de doenças - doenças causadas por fungos, bactérias e vírus. In: Vale, F.X.r & Zambolim (Eds.) Controle de doenças de plantas: grandes culturas. p.83-180. Viçosa: Universidade Federal de Viçosa,

Zhang, W., Ricketts, T. H., Kremen, C., Carney, K. & Swinton, S. M. (2007). Ecosystem services and dis-services to agriculture. Ecological Economics, 64(2), 253–260. Doi:10.1016/j. ecolecon.2007.02.024

75

Appendices

Appendix A- List of all herbaceous species

All herbaceous species found in organic coffee agroforestry plots (n=45) in this study, listed in order of frequency. Range and mean functional traits are given.

Species name Family name Region of Origin Frequency

Height Range (cm)

Mean Height (cm)

LDMC Range (mg

g-1)

Mean LDMC (mg g-1)

SLA Range (mm2 mg-1)

Mean SLA (mm2 mg-1)

LNC Range (mg g-1)

Mean LNC

(mg g-1)

LCC Range (mg g-1)

Mean LCC

(mg g-1)

Commelina diffusa Commelinaceae Asia 36 8-65 37.57 38.34-623.74 184.61 7.20-

69.56 33.57 23.44-95.44 37.82 353.10-

630.10 428.88

Brachiaria platyphylla Poacea North America 26 19-103 53.75 31.99-579.54 233.03 6.97-

42.89 29.41 24.27-95.91 37.61 244.8-

569 421.35

Hydrocotyle mexicana Araliaceae North America/Mexico 20 9 - 26 8.33 36.41 -

883.33 173.29 7.52-70.78 35.90 27.85-

67.59 36.65 362.90-659.10 428.11

Pseudelephantopus spicatus Asteraceae Central America 18 9-33 16.50 25.14-

263.73 173.63 6.59-28.34 18.36 23.68-

45.40 34.71 378.10-428.30 404.58

Cyperus tenuis Cyperaceae Central America 15 13-55 41.53 103.75-784 358.17 1.43-55.09 28.12 13.07-

47.00 28.12 348.80-435.10 420.13

Physalis angulata Solanaceae Central America 12 13-78 36.29 117.89-390.16 192.09 11.52-

41.79 27.78 33.99-57.13 41.31 382.60-

464.50 412.30

Drymaria cordata Caryophyllaceae Central America 11 8 - 29 15.00 110.47-238.81 158.67 5.87-

84.62 46.24 26.40-53.16 33.31 403.30-

574.80 407.40

Cyathula achyrantoides Amaranthaceae Africa 8 12-43 21.5 167.83-238.17 206.12 3.41-

42.15 25.01 32.63-39.94 35.78 390.20-

448.90 407.45

Cyathula prostrata Amaranthaceae Africa 6 8-44 24.5 93.75-313.11 156.62 11.51-

53.26 36.77 37.00-46.01 40.98 388.40-

525.80 427.17

Hydrocotyle bowlesiodes Araliaceae North America/Mexico 5 1 - 27 9.00 116.40-

201.07 174.60 5-56.19 8.18 26.96-49.85 38.80 399.60-

426.60 416.70

Laportea aestuans Urticaceae Africa 5 19 - 49 32.40 177.11-300.57 234.44 37.62-

59.63 47.45 27.55-48.93 40.05 363.50-

446.90 396.98

76

Digitaria sanguinalis Poaceae Europe 4 12-61 49.50 160.56-265.27 219.07 28.76-

38.11 29.28 21.83-31.05 25.86 415.80-

427.90 421.85

Oplismenus burmanii Poaceae Africa 4 14-35 22.75 75.73-677.22 315.87 28.28-

51.22 39.97 30.44 - 45.03 37.50 422-

435.50 431.68

Alysicarpus vaginalis Fabaceae Asia/Africa 3 26-43 32.33 155.41-177.94 163.20 31.20-

41.10 37.24 33.05-55.4 47.86 434.70-

464 450.07

Crassocephalum crepidioides Asteraceae Africa/Australia 3 16-62 32.00 32.46-

236.99 119.31 27.59-37.84 42.82 37.56-

51.54 42.82 416-439.30 430.10

Cyperus luzulae Cyperaceae Central America 3 41-78 54.33 286.22-343.83 323.13 11.68-

23.57 15.66 19.64-27.16 22.32 417.9-

435.3 426.47

Pteridium aquilinum Dennstaedtiaceae Central America 3 18-28 21.33 65.41-178.18 132.90 4.73-

75.52 40.47 33.39-42.80 37.75 409-

443.10 421.73

Spermacoce latifolia Rubiaceae Central America 3 26-43 36.00 185.56-272.19 217.20 28.19-

42.89 33.51 29.15-46.17 36.38 308.90-

395.60 363.43

Colocasia esculenta Araceae India 2 6 - 26 16.00 139.92-183.54 161.73 31.01-

31.54 31.27 34.58-36.68 35.63 393.70-

419.30 406.50

Cyprus ferax Cyperaceae Central America 2 35 35.00 192.31-250.00 221.15 23.46-

36.65 30.05 19.45-44.71 32.08 420.70-

422.70 421.70

Lobelia xalapensis Campanulaceae North America 2 53-66 59.50 204.92-256.41 230.61 23.57-

36.57 30.07 35.78-36.92 36.35 426.70-

427.10 426.90

Oxalis corniculata Oxalidaceae Asia 2 13-19 16.00 129.13-179.35 154.24 33.40-

44.40 38.90 30.80-35.07 32.94 418.30-

437.20 427.75

Paspalum fasciculatum Poaceae Africa 2 35-138 86.50 236.92-313.45 275.18 14.95-

20.09 17.54 21.41-26.85 24.13 409.80-

416.60 413.20

Pilea hyalina Urticaceae South America 2 66-76 71.00 191.50-243.11 217.30 25.35-

50.62 37.98 48.09-49.52 48.81 349.70-

391.90 370.80

Pteridium Dennstaedtiaceae Central America 2 27-72 50.00 179.49-239.46 209.43 18.92-

27.64 23.28 28.21-39.31 33.76 424.00-

448.40 436.20

Solanum torvum Solanaceae Central America 2 21 21.00 112.97-148.63 148.63 25.26-

34.75 30.00 39.47-39.68 39.58 412.50-

438.90 425.70

Youngia japonica Asteraceae Asia 2 13-16 14.50 93.83-110.99 102.41 42.24-

55 48.62 39.50-127.60 83.10 203.90-

428.80 316.35

77

Calyptocarpus wendlandii Asteraceae Mexico 1 44 44.00 217.01 217.01 28.62 28.62 35.92 35.92 382.30 382.30

Clerodendrum viscosum Lamiaceae Asia 1 14 14.00 95.10 95.10 41.48 41.48 37.40 37.40 428.00 428.00

Convallaria majalis Asparagaceae Europe 1 17 17.00 176.41 176.41 16.78 16.78 28.13 28.13 443.00 443.00

Emilia fosbergii Asteraceae Central America 1 48. 48.00 108.06 108.06 31.43 31.43 42.04 42.04 433.30 433.30

Erythrina poeppigiana Fabaceae South America 1 10 10.00 97.77 97.77 59.30 59.30 60.60 60.60 471.20 471.20

Musa Musaceae Southeast Asia 1 38 38.00 103.37 103.37 32.87 32.87 38.59 38.59 404.10 404.10

Phyllanthus niruri Phyllanthaceae India 1 43 43.00 500.00 500.00 52.08 52.08 22.99 22.99 448.60 448.60

Spanathe paniculata Apiaceae Central America 1 22 22.00 170.56 170.56 42.80 42.80 42.41 42.41 526.20 526.20

Tithonia rotunfolia Asteraceae Central America 1 33 33.00 192.64 192.64 30.52 30.52 38.39 38.39 392.80 392.80

Unidentified.A Unidentified Unidentified 1 87 87.00 328.26 328.26 25.56 25.56 41.00 41.00 424.50 424.50

Unidentified.B Unidentified Unidentified 1 21 21.00 111.08 111.08 36.02 36.02 39.60 39.60 401.50 401.50

78

Appendix B - Interview Questions

Interview questions used in all farmer interviews, approved by University of Toronto Human Research Ethics Program. Italicized questions were prompts used to support further elaboration on questions from farmers.

Introduction Questions Preguntas Iniciales Name Nombre Age Edad

Name of Region/Location Nombre de la Región

Please, tell me the story of your farm. How long has it been in coffee production? Were

there any crops here before? Any other details?

Por favor, explícame la historia de su finca. ¿Cuánto tiempo ha pasado en la producción de café? ¿Hubo alguna cosecha aquí antes?

¿Algún otro detalle?

How long have you been a farmer? ¿Cuánto tiempo ha sido agricultor?

When and how did you decide to transition to organic production? What was that process

like?

¿Cuándo y cómo decidió cambiar al producción orgánica? ¿Cómo fue ese

proceso?

When did you join the APOYA network? Are you a part of any other farmer networks here?

¿Cuándo se unió a la red APOYA? ¿Es usted parte de alguna otra red de agricultores

aquí?

Have you noticed a difference when not using herbicides on soils?

¿Notó una diferencia cuando no usa herbicidas en los suelos?

What is the size of your farm? ¿Cuál es el tamaño de su finca?

Herbaceous Community Questions Preguntas sobre la Comunidad Herbáces

Do you do any weeding? If so, how (chop, turn soil, mulch, burn)?

¿Hace usted la chapea? ¿Si es así, como (usa machete, moto guaraña, girar la tierra,

abono, quemar)? Walk me through your weeding process.

How much time does it take? Who performs this task? Do you weed the whole farm at

once?

¿Puede llevarme a través de tu proceso de la chapea? ¿Cuánto tiempo toma? ¿Quién

realiza esta tarea? ¿Chapea toda la finca a la vez?

What are the key triggers that to encourage you to weed? Is there a maximum threshold of

herbaceous cover before you weed?

¿Cuáles son los factores desencadenantes clave que lo alientan a la chapea? ¿Hay un

umbral máximo de cobertura herbácea antes de desherbar?

79

How often do you weed? Is it something scheduled, or is it triggered by a maximum

threshold?

¿Con qué frecuencia chapea usted? ¿Es algo programado, o algo que hace como resultado

de alcanzar el umbral máximo? Are there specific months or times of year

that you weed more than others? ¿Hay meses o épocas específicas del año en

las que chapea más que otras? What do you do with the debris? Do you leave

plant material on the ground? ¿Qué hace con el material orgánico? ¿Lo deja

sobre la tierra? Are there seasons or years when you have not

weeded? Did you notice differences in the coffee (yield/quality) in those years?

¿Hay temporadas o años en los que no ha limpiado? ¿Notó diferencias en el café (rendimiento / calidad) en esos años?

When do you see a large growth of weeds? After rains? Fertilization? Harvest?

¿Cuándo vea un gran crecimiento de monte? ¿Después de las lluvias? ¿Fertilización?

¿Cosecha?

Do other management practices have an effect on weeds, i.e. pruning?

¿Piensa que otras prácticas de gestión un efecto sobre el monte? (por ejemplo, la

poda)?

Species-specific Questions

Preguntas Específica de la Especies

How do you know this is a bad weed? How do you know is this a good weed? What

impacts do they have on the coffee plant?

¿Cómo sabe que esta es una mala hierba? ¿Cómo sabe que esta es una buena hierba? ¿Qué impacto tienen en la planta de café?

What are you looking for in herbaceous species that you keep (i.e. pollinator habitat,

nitrogen fixer)?

¿Qué está buscando en especies herbáceas que conserva (es decir, hábitat de

polinizadores, fijador de nitrógeno)? Are there some you do not consider weeds? If

so, why not? ¿Hay alguno que no piensa son malas

hierbas? Si es así, ¿por qué no?

Ecosystem Service What are the most important services that

your herbaceous community provides to your farm out of the following?

improved soil health

erosion control soil moisture regulation soil nutrient additions

improved biological diversity medicine

pollination farm beauty pest control

Servicios ecosistémicos

¿Cuáles son los servicios más importantes que su comunidad herbácea brinda a su finca de

los siguientes?

mejor salud del suelo control de la erosión

regulación de la humedad del suelo adiciones de nutrientes del suelo diversidad biológica mejorada

medicina polinización

belleza de la granja control de plagas

80

disease control food for your family food for your animals

control de enfermedades comida para su familia

comida para sus animales

Other management practices

Otras prácticas de gestión Do you use fertilizers (compost or inorganic

fertilizers)? Where did you learn to make them?

¿Utiliza fertilizantes (compost o fertilizantes inorgánicos)? ¿Dónde aprendiste a

prepararlos?

Any other farm management practices that affect the soil?

¿Alguna otra práctica de manejo que afecte el

suelo?

General Questions:

Preguntas generales

What challenges and benefits do you have as an organic farmer?

Can you provide any other information on your practices for maintaining successful

coffee production?

¿Qué desafíos y beneficios tiene usted como agricultor orgánico?

¿Puede proporcionar otra información sobre sus prácticas para mantener una producción

de café exitosa?

Do you share information with different farmers/ What sorts of tools or resource are available for you to get information? What

would be useful to you in the future?

¿Comparte información con diferentes

agricultores / Qué tipo de herramientas o recursos están disponibles para que usted

obtenga información? ¿Qué te sería útil en el futuro?