TYPE Original Research PUBLISHED 27 October 2023 DOI 10.3389/fclim.2023.1235601 The development of a farmer OPEN ACCESS decision-making mind map to EDITED BY Ajay Bhave, inform climate services in Central University of Leeds, United Kingdom REVIEWED BY America Andrew John Dougill, University of Leeds, United Kingdom Gabriela Cruz, Universidad de la República, Uruguay Diana Giraldo1,2*, Graham Clarkson1, Peter Dorward1, Diana Ruiz, University of Cauca, Colombia Diego Obando3 and Julian Ramirez-Villegas4,5,6 *CORRESPONDENCE 1School of Agriculture, Policy and Development, University of Reading, Reading, United Kingdom, Diana Giraldo 2International Center for Tropical Agriculture, Cali, Colombia, 3International Center for Tropical d.c.giraldo@pgr.reading.ac.uk Agriculture, Tegucigalpa, Honduras, 4International Center for Tropical Agriculture (CIAT), c/o Bioversity International, Rome, Italy, 5Bioversity International, Rome, Italy, 6Plant Production Systems Group, RECEIVED 06 June 2023 Wageningen University & Research, Wageningen, Netherlands ACCEPTED 05 October 2023 PUBLISHED 27 October 2023 CITATION Giraldo D, Clarkson G, Dorward P, Obando D The growing complexity of the relationship between climate information and and Ramirez-Villegas J (2023) The agricultural decision-making necessitates the development of relevant and development of a farmer decision-making timely climate services for farmers. These services can eectively support risk mind map to inform climate services in Central America. Front. Clim. 5:1235601. management strategies in agriculture by fostering a comprehensive understanding doi: 10.3389/fclim.2023.1235601 of the intricacies involved in farmer decision-making dynamics. This paper COPYRIGHT addresses this critical gap by analyzing the drivers influencing decision-making © 2023 Giraldo, Clarkson, Dorward, Obando processes that shape adaptation strategies for staple grain and coee farming and Ramirez-Villegas. This is an open-access systems in Central America. The study answers the following research questions: article distributed under the terms of the Creative Commons Attribution License (CC BY). (i) Does the mind map tool eectively provide a holistic understanding of farmers’ The use, distribution or reproduction in other decision-making processes? (ii) How do Central American farmers make decisions forums is permitted, provided the original within their farm systems at multiple timescales? (iii) Which climate factors trigger author(s) and the copyright owner(s) are credited and that the original publication in this these decisions? Employing a combination of systematic literature review and a journal is cited, in accordance with accepted case study in Honduras, the study identifies 13 critical decisions farmers make academic practice. No use, distribution or throughout their crop cycle and their respective triggers. These decisions were reproduction is permitted which does not comply with these terms. grouped into three clusters (production, household, and environmental) and classified into lead-time categories (operational, tactical, and strategic). Findings reveal that farmers base their decisions regarding future climate expectations on their traditional knowledge, religious dates, and memories of recent past seasons’ rainfall patterns, and that one of the most significant factors influencing farmers’ decisions is food security shortages resulting from extreme events. For example, recent mid-summer droughts have led farmers to prioritize sowing beans over maize in the Primera season, while during the Postrera season, they face challenges due to excess rainfall and the hurricane season. We conclude that the mind map tool developed in this paper provides an eective and appropriate method and that the variation in farmers’ decision-making complexity across systems and landscapes presents a significant opportunity to design mind maps that span multiple timescales, facilitating the exploration of decision spaces. Farmers actively seek tailored weather and climate information while still valuing their existing experience and local knowledge, emphasizing the importance of integrating these elements into the development of climate services. KEYWORDS decision-making, climate services, risk management, dry corridor, systems thinking Frontiers inClimate 01 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 1. Introduction of higher-quality data (e.g., information and products) rather than to provide an integrated process for improved decision- Over the last century, prolonged droughts, shifting rainfall making (Lourenço et al., 2015; Findlater et al., 2021), involving patterns and extreme events have significantly impacted Central and encouraging users to make their own decisions based on the America, where more than two-thirds of the population depends analysis of information and their demands. Indeed, the generation on agriculture (Imbach et al., 2017). Climate variability and change, of output or product may not be as important as the process itself. with a variety of other local stressors, can motivate a shift in The definition of climate services, as provided by Findlater et al. strategy in farmers’ decision-making, such as planting a new crop, (2021), has shifted the focus from solely providing information to experimenting with a new variety, or migration of a household emphasizing the importance of understanding the processes behind member (Eakin et al., 2014). The literature recognizes that farmer decision-making, including who is involved and why and how decision-making is highly dynamic and complex, and is influenced decisions are made. This represents a significant paradigm shift in by social relations, individual experiences and their context (Soares the field of climate services. et al., 2018). Farmers are constantly making decisions about Here, we aim to first understand Central American farmer what, when, and where to plant, management practices and about decision-making and then explore how this understanding can be resource allocation to farm activities such as livestock and other integrated into the development of climate services. A “holistic livelihoods. Climate services can create opportunities to better picture” of farmers’ decision-making was created using the mind integrate local knowledge and scientific information into the map approach, which combined the results of a literature review decision-making process (Guido et al., 2021). Climate services are through a set of framing questions and a case study conducted defined as the processes that involve the production, translation, with farmers and crop experts in the field in Honduras. Our study transfer, and use of weather and climate information, all aimed aims to help fill the knowledge gap on farmers’ decision-making at enabling and informing effective decision-making (Born et al., in Central America by (i) documenting whether and how specific 2021). decisions are triggered by weather and climate variables; and (ii) In Central America, previous studies have analyzed farmers’ what weather and climate information are required to support responses to various climate-related changes including hurricanes decision-making by small-scale farmers who cultivate coffee, maize (Alayón-Gamboa et al., 2011; Cruz-Bello et al., 2011), El Niño and beans in Central America. We conclude by discussing the droughts (Ewbank et al., 2019), interannual climate variability implications of the results within the context of climate services (Eakin, 2000), and climate change (Harvey et al., 2017, 2018; for agriculture. Bielecki and Wingenbach, 2019; Gerlicz et al., 2019). Individually, these studies typically help identify what events affect farmers, and decisions are (or should be) made in response to such events, 2. Materials and methods with only limited attempts to establishing a link between the decisions and the broader spatio-temporal and socioeconomic 2.1. Study area context. There is thus a significant gap in the literature on (i) how Central American farmers make decisions within their farm Our study area is the Central American Dry Corridor system at multiple timescales, (ii) the climate factors that trigger (CADC)—a drought-prone area, mainly in Guatemala, El Salvador, those decisions, and (iii) how to map farmers’ decision-making Honduras, and Nicaragua (herein referred to as CA4 countries). dynamics together with their farming and support systems for Climate in the CADC is semi-arid, with two rainy seasons, divided climate services development. by a long dry season, and a mid-summer drought or canícula. To address these gaps, we chose a systems thinking approach to Variations in temperature and precipitation trigger severe droughts gain a more holistic understanding of farmers’ decision-making in and short dry spells, which impact farming systems and food Central America. Systems thinking can be classified under “hard” security (Alpízar et al., 2020). According to PRESANCA and the or “soft” approaches (Darnhofer et al., 2012; Rose et al., 2018). FAO (2011), there are 2.3 million small-scale farmers in the Central Hard approaches tend to rely onmathematical or economic models American Dry Corridor. The CA4 countries have two main small- based primarily on utility maximization outcomes (e.g., income, scale farming systems: basic grains (maize and beans) and small- cost-benefit, or highest yields) and are driven by assumptions that scale coffee production. Bouroncle et al. (2017) offer a review of farmers have full access to information (e.g., on seeds, soil, climate) agricultural statistics in the area and report that the most important andmake decisions on a single time frame (e.g., a production cycle), cash and subsistence crops in terms of cultivated area are maize thus simplifying assumptions of human behavior in the decision- (34%), coffee (16%), beans (14%), followed by sugar cane (8.4%), making process. Soft systems, on the other hand, view decision- rice (5.8%), and sorghum (4.9%). Figure 1 shows livelihood zones making with a focus on decisions as processes rather than just in the CADC, which integrate economic activities and farming a set of well-defined outcomes (Frisch and Clemen, 1994). They systems within each CA4 country (Grillo and Holt, 2009). allow more holistic enquiry and understanding (Singh et al., 2016), The landscapes in the livelihoods zone map include rain-fed and place emphasis on decision rules and social appraisals, mind coffee and basic grain production. Basic grains are produced under maps and ontologies, traditional ecological knowledge and adaptive the milpa system (Olson et al., 2012 and Hellin et al., 2017), pathways (Darnhofer et al., 2012). with average farm sizes ranging from 0.9 to 4.5 ha (Bokusheva Soft systems thinking has not been used so far to inform et al., 2012; Alpízar et al., 2020; Baumann et al., 2020). In the the development of climate services in Central America. Climate coffee (Coffea arabica L.) zones, namely GT11, SV02, HN05, and services are considered by many providers to be the delivery NI12, the annual rainfall ranges between 1,000 and 2,000mm, Frontiers inClimate 02 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 FIGURE 1 Livelihoods zone map for the productive systems of interest considered in the systematic literature review in El Salvador (SV), Nicaragua (NI), Guatemala (GT), and Honduras (HN) (data source: FEWS NET, 2007). while the temperature ranges between 12 and 28◦C. In the zones interrelated decisions at multiple timescales. The study thus starts of the subsistence grains maize (Zea mays) and beans (Phaseolus from understanding the full extent of farming system dynamics vulgaris), namely GT10, SV01, HN07, and NI03, the annual rainfall across time, to then create links at all relevant temporal scales ranges between 800 and 1,500mm, while the temperature ranges through the mind map. To this aim, agro-climatic calendars were between 21 and 30◦C. Mean household head age is 47.8 years and developed for the two systems in question (coffee and maize/beans) a mean household size of five to six members (Hellin et al., 2017; through a comprehensive literature review to identify the crop Dodd et al., 2020). Furthermore, household heads generally have cycles that are commonly used in the CA4. The calendars were a low level of formal education (i.e., have not completed primary then refined and validated through consultation with experts (field school) and limited access to technical support (Eakin et al., 2014). officers) and small-scale farmers, providing a better understanding According to FEWS NET (2007), the income sources in livelihood of the context. zones are from sales of crops (i.e., basic grains, coffee, and fruits), livestock, and firewood; migration to sugar cane and coffee areas for harvest seasons; and remittances. 2.3. The mind map tool A mind map is a tool for organizing ideas and identifying 2.2. Agro-climatic calendars thematic groups that show interconnections between ontologies— the distinction of different types of existing knowledge and A critical aspect of the decision-making is the timing of their elements, concepts and relations (Buzan and Buzan, 2006). decisions with respect to the productive cycle of the crops, and The mind map is a non-formal representation of ontologies the local agroecosystem dynamics. In the CA4 region, climate that can then evolve into a semi-formal (e.g., Unified Modeling services for agriculture that support farmer decision-making have Language—UML) or a more formal Ontology Web Language concentrated more on seasonal to decadal climate information (OWL) structure (Husáková and Bureš, 2020). Mind maps have through the Climate Outlook Forum (Garcia-Solera and Ramirez, been used to understand farmer decision-making in several sectors 2015) than on weather timescales (i.e., hours to days). However, this and countries including biodiversity conservation in Australia neglects the fact that the production systems involve a sequence of (Farmar-Bowers and Lane, 2009) and crop production in Ethiopia Frontiers inClimate 03 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 TABLE 1 Description of the competency questions (CQs) with examples. CQs Description Example DN The question that defines the decision made by the farmer What variety of maize do I plant? Early maturing variety? DT The type of decision (strategic, tactical, or operational) Tactical (medium-term) When do farmers make the decisions April The trigger event that results in the decision process being an action Prolonged drought, seed availability IN Information required to help make the decision Midsummer drought “canicula,” rainfall forecast, soil information DM The person responsible for making the decision The man and the woman in the household DN, decision name; DT, decision type; IN, information needs; DM, decision maker. (Kraaijvanger et al., 2016), Sri Lanka (Walisadeera et al., 2015), 2.4.1. Step 1–identification Nepal (Afzal and Kasi, 2019), and Thailand (Kawtrakul, 2012). Relevant literature was examined to identify common To define the purpose and scope of the mind map four vocabulary for the mind map through web searches with different Competency Questions (CQs, Walisadeera et al., 2015; Table 1) combinations of keywords connected with the AND-OR operators, were determined. This enabled the necessary information to be including “farmersm,” “decision-making,” “coffee,” “maize,” “bean,” obtained from the literature review as well as in the case study. “Central America,” “dry corridor,” and “climate services,” using The Decision Name (DN) encompasses the critical decisions that Google Scholar and Web of Science. Only peer-reviewed articles, farmers make in their crop cycle, ranging from why they plant their books, and dissertations published from 2000 to 2020 in English crops to whether they harvest for the market or consumption. The or Spanish were included in the review. Snowball sampling was Decision Type (DT) allowed us (i) to classify decisions into lead- employed to identify additional literature cited within the initial time categories, namely short-term operational decisions (days search. This process resulted in the identification of 74 articles in to weeks; e.g., land preparation), tactical medium-term decisions Central America. (months; e.g., crop selection), and strategic medium- to long- term decisions (a year or more; e.g., selection of irrigation system); (ii) to determine decision timing (e.g., the month or 2.4.2. Step 2–screening crop stage when a decision is made); and (iii) to identify Next, we assessed articles for inclusion based on their abstract trigger events (e.g., prolonged droughts) that influence farmers’ using three criteria (see Supplementary Table S1 for a list of the decision-making processes (Fountas et al., 2006; Hollinger, 2009; criteria). An essential criterion for inclusion was that each article Prokopy et al., 2013; Robert et al., 2016). The Information Needs involved collection of primary data in the field with farmers (IN) encompasses the information required for making decisions through surveys, interviews, or participatory approaches. The final (e.g., rainfall forecast). Lastly, the fourth CQ pertained to the list of 31 selected references that address used for the analysis are Decision Maker (DM), allowing us to understand the roles of shown in Supplementary Table S2. different household members in the decision-making processes (Rose et al., 2018). 2.4.3. Step 3–systematic analysis In this step, we first performed a descriptive analysis of the abstracts using word clouds in NVivo 12 (Zhou et al., 2016; see Supplementary Figure S1). The word clouds allowed analyzing the 2.4. Data collection and analysis frequency of certain words and are especially useful if one can identify some of the decisions for each system in the study area (e.g., We chose a systems thinking approach—mind map tool—to cultivars and diversification) as well as some factors that influence gain a more holistic understanding of farmers’ decision-making. such decisions (e.g., seasonal variability, hurricane, and coffee rust). This approach integrates a systematic literature review and a Next, the articles were classified and coded in nodes using NVivo mind mapping tool to better comprehend these processes through 12, requiring close reading and interpretation on the researcher’s qualitative analysis (Figure 2). The first three steps of the data part. In NVivo 12, a node refers to a collection of references that collection and analysis process (steps 1–3) relates to compiling deal with a specific topic and are used to group articles (Verdonck and systematizing the literature sources, whereas step 4 focuses et al., 2015). In this paper, each node represented a classification on building a first mind map, and then enriching it with case according to each CQs (Table 1), and certain paragraphs of an study information. To create the case study, we used qualitative article were assigned to a specific node. techniques (i.e., interviews, focus groups and observations) to increase the study’s internal validity, aiming to develop a holistic picture of the farmer’s decision-making. We applied the mind map 2.4.4. Step 4–the mind map to the main crops in the CA4 countries—maize, beans, and coffee, We constructed a first version of the mind map using three structuring the process along the four Competency Questions inputs from the literature review, (i) the key decisions that the (CQs) shown in Table 1. farmers make in their farm system, (ii) when those decisions are Frontiers inClimate 04 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 FIGURE 2 Schematic representation of the mind map process. made and who by, and (iii) the information and relations to the 3. Results weather and climate variables that trigger those decisions. The key decisions identified are the nodes in the mind map and the arrows 3.1. Key decisions made by farmers are the relations with the other CQs. After the first version of the mind map was completed, a case study was conducted with field Table 3 presents the maize/beans and coffee agro-climatic officers (n= 5) and farmers (n= 7) in 2021. The case study involved calendars based upon the literature review and the case study. nine interviews and three focus groups in which participants were Small-scale farmers in the CA4 region generally grow the first purposively and snowball-selected in Honduras due to their well- crop—maize—at the beginning of the rainy season (i.e., late known experience and knowledge of coffee and basic grain systems May or early June) and harvest it in October, locally called the (Table 2). No other characteristics were taken into account for the Primera season. By contrast, the second crop—beans—is planted selection. The snowball selection consisted of first identifying field regularly during the growing seasons of September–December and officers as participants, and then asking them to identify at least two December–March, locally called the Postrera and Apante seasons, farmers’ associations. For the farmers, they were asked to identify respectively (Hellin and Schrader, 2003 and Baumann et al., 2020; at least three individual farmers. This process was repeated three Ibáñez et al., 2022). In addition, the most frequently reported lean times until at least 15 participants were identified. We involved months—June, July, and August—coincide with the Mid-Summer these domain “experts” in the field to verify and address any gaps Drought (MSD, known as “canicula”), and are associated with a in the mind map. They provided advice/input on the following lack of income (Bacon et al., 2014). Maize and beans have an two aspects of the first version of the mind map: (i) the contents– approximate cycle length of 3–5 and 2–3 months, respectively. The decisions and (ii) structure–relations. The case study involved process is divided into four phenological phases, from planting semi-structured interviews and focus groups (∼2 h) for answering and germination to harvesting. Coffee production is divided into the CQs (see Supplementary material for the case study protocol). six phases, from germination and seedling to harvesting (Table 3). The case study involved open questions:When andwhy did you The exact length of the cycle and timing of the phases vary start planting beans/maize/coffee? When did you plant the crop? according to the variety, environmental conditions, and crop How has the crop been in recent years? Second, the interviewees management (Bacon et al., 2014). Moreover, as coffee is a perennial were asked to draw an agro-climatic calendar with the specific crop, the vegetative and reproductive growth phases may occur activities that they perform on their crop, how these activities simultaneously but in different plots on the same farm. The lifespan have been affected by weather and climate, and the role of family of a coffee plantation can be up to 30 years (Bunn et al., 2015). members in them. Finally, the decisions list identified from the As a result of the systematic literature review, relevant literature was used to ask the interviewees whether they identified information to answer the CQs were found. We identified the them as relevant and why. They were also asked whether any decisions that farmers make in their production systems, the decision was missing, as well as what information they would timing of these decisions, and the factors that influence them. require to make better decisions. The interviews and focus groups However, most articles have addressed only a particular decision were conducted in Spanish in July and August 2021. The case study without a holistic view of the farming system and the roles of protocol was approved by the University of Reading’s Research household members in the decision-making processes. A total Ethics Committee. The transcripts from the sets of interviews were of 13 decisions triggered by weather or climate events were coded and analyzed with NVivo 12 following the same process as found in the 31 articles from the CA4 countries (Table 4). for the systematic review. For the synthesis, a qualitative content The decisions were grouped into the following three clusters, analysis was conducted linked to the CQs. The mind maps were Cluster A: Production system, which comprised decisions related built using the Mindmaster tool (Edrawsoft, 2022). directly to maintaining or improving crop production; Cluster B: Frontiers inClimate 05 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 TABLE 2 Case study with key informant interviews and focus groups in Honduras. ID Farmer interviews ID Field ocers interviews Total FBean01 Bean farmer FOBean01 Bean field officer 9 FMaize02 Maize farmer FOMaize02 Maize field officer FCoffe01 Coffee farmer FOCoffee01 Coffee field officer FCoffe02 Coffee farmer FOCoffee02 Coffee field officer FCoffe03 Coffee farmer ID Farmer focus groups ID Field ocers focus groups Total FFGWA01 Women association FFGCA01 Coffee association 3 FFGFA02 Farmers association TABLE 3 Summary of the CA4 countries typical maize/beans and coee small-scale farmers seasonal calendars based upon the literature review and experts in the field. 1st rainy season MSD 2nd rainy season Maize/bean Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec stages Land preparation Planting and germination Vegetative growth Reproductive growth Harvesting Coee stages∗ Germination and ∗∗ seedling Vegetative growth Reproductive and dormancy Main flowering Fruit development Maturation and harvesting MSD, mid-summer drought. ∗Representation of the first and second phenological years. ∗∗The germination and seedling stage can take as long as 6 months. Maize Beans. Household strategies, which comprised decisions linked to family illustrates the holistic understanding that maize, beans, and coffee projects or collaborative networks for reducing vulnerabilities farmers have of their system, and how they make decisions within and maintaining or improving living standards; and Cluster it. The mind map enumerates every decision (e.g., A.1. What C: Environmental management, which comprised decisions crop can I plant?), the timing of the decision (long before the that allow farmers to adopt longer-term planning horizons to planting season begins) and associates it with the main factors sustain ecosystem services, preserve biodiversity, and enhance related to decision-making. These factors are influenced by both soil health. climatic (shaded boxes) and non-climatic variables (on a line). The shaded boxes present factors that are influenced by climatic variables (e.g., water availability, food security, and land slope). 3.2. Farmer decision space: the mind map Due to their dependency on climate variables, these factors are relevant for the development of climate services in the CA4 region. This section presents the results of the mind map regarding A detailed explanation of the mind map can be found in the the findings of the literature review and the case study. Figure 3 subsequent sections. Frontiers inClimate 06 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 TABLE 4 Key decisions made by maize, bean, and coee farmers found in the literature review. DT maize/bean decisions DT coee decisions (T) A.1—Crop choice (S) A.1—Crop choice (T) A.2—Variety Choice (S) A.2—Variety choice CLUSTER A (O) A.3—Planting date (T) A.3—Replanting (O) A.4—Land preparation (O) A.5—Pruning (O) A.6—Harvesting date (S) B.2—Diversification (T) B.1—Postharvest (O) B.3—Labor mobility CLUSTER B (T) B.2—Diversification (S) B.4—Migration (S) B.4—Migration (S) C.1—Ecosystem approach (S) C.1—Ecosystem approach (T) C.3—Soil conservation CLUSTER C (T) C.2—Quesungual system DT, decision type; O, operational (short-term) decision; T, tactical (medium-term) decision; S, strategic (medium-/long-term) decision. FIGURE 3 Farmer decision-making mind map to inform climate services in Central America as a result of a systematic review and the case study. The figure includes the timing of the decision in the context of Section 3.1. 3.2.1. Cluster A: production system FFGFA02 stated the following: “Planting staple crops allows us to The findings reveal that farmers’ decisions to plant maize or obtain the government bonus, which provides seeds”. However, need beans are influenced by household demands related to food security for an income has pushed farmers to start planting coffee. The shift and seed availability (Mendoza et al., 2017). Farmers association between maize and another crop (e.g., beans, sorghum) is triggered Frontiers inClimate 07 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 FIGURE 4 Farmer decision-making (operational, tactical and strategic) influenced by weather and climate information. The decisions (*) for maize and bean, and (**) for Coee. The decision without asterisks involves both production systems. See Supplementary Table S3 for descriptions of each climate information source. by the late arrival of the rains (Eakin, 2000). The slope of their of weather and climate information to inform this decision- land is why they decide to plant coffee over annual grain crops making process in Central America (Imbach et al., 2017). In (FOCoffee01). For staple crops, in some cases, the preference for the case study, the farmers (FBean01 and FMaize02) mentioned seed selection is due to culinary, tradition and cultural importance, that some peers traditionally sow on the same date for the and access to community-based grain banks (FFGWA01). For Primera season –Día de la Cruz 3rd May– waiting for the example, native or local varieties to make tortillas, tamales, and rains to begin. The first rains that fall early or mid-May trigger atole (a maize drink), are often consumed in almost every meal farmers to decide to plant. However, farmers risk planting and (van Etten, 2006; Hellin et al., 2017). The trigger events that lead to poor germination due to a false start to the rainy season; they choosing short-stature and fast-maturing maize varieties are crop have to replant with differences in height and maturity, creating lodging from high winds and drought risk from an extended mid- problems at harvest time (Baumann et al., 2020). Farmers also summer drought (Eakin, 2000). In the case study, coffee farmers consider alternative crops if an extreme event destroys their first FCoffe01–03 cited resistance to disease and pests (coffee rust), planting at a date that prevents replanting with maize (Eakin, heat and water stress tolerance, and higher yields are the primary 2000). For coffee, a shade-grown coffee plantation lasts ∼30 reasons for selecting suitable varieties. Additionally, the use of low years, but on a sun-grown plantation with intensive production stature coffee varieties allows for higher spacing and facilitates would have to be renewed more frequently (Bunn et al., pruning (Eakin et al., 2006). 2015). Replacing susceptible varieties with resistant varieties will This study emphasizes the importance of planting date trigger the renewal decision for coffee. For example, field officer selection as the most critical operational decision. Despite its FOCoffee02 stated the following: “After the impact of rust in the significance, there is limited evidence regarding the utilization 2011/12 season, coffee production recovered through the renovation Frontiers inClimate 08 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 of production areas with improved varieties that were tolerant have space on their farms with patios to dry the grains in the sun to rust.” for storage (for 4 months or more) in silos (Bokusheva et al., 2012). During case study data collection, it was observed that some These stored grains are used mainly during periods of high prices small-scale farmers in Honduras prepare the land and sow on or food shortages in the community. the same day, predominantly using herbicides and machetes for In this study, on-farm diversification is considered a strategic weed clearance (Eash et al., 2019). The soil moisture levels that decision made by households to adapt to climate variability accompanies the first rainfall triggers farmers to make the decision and change. In the literature review, it has been identified as a to prepare their land; moreover, the timing of input management factor in managing food security risks, with diversified livelihoods also depends on the rainfall and temperatures. In addition, farmers generally being more food secure (Gerlicz et al., 2019; Hellin alter the landscape by creating terraces and furrows to take et al., 2019; Dodd et al., 2020). In the case study, the respondents advantage of rainfall run-off in areas where erosion is high, or mentioned home gardens, fruit trees, coffee agroforestry systems flooding is frequent (Eakin, 2000). However, land tenure affects and timber as enterprises that can improve household income how farmers manage their plots influencing their willingness to and food security and buffer environmental effects—high or low invest in sustainable land management practices (Mendoza et al., temperatures, strong winds, and heavy rains. However, in some 2017). In the focus group, the farmers association (FFGFA02) cases farmers are unwilling to engage in crop diversification due mentioned have access to inputs at a reasonable price through to problems associated with new pests and diseases and knowledge rural banks—cajas rurales—or waiting to receive a bonus from gaps in understanding which crops it would be best to diversify the government to avoid the risk of losing crops. Furthermore, associated with growing coffee (Bielecki and Wingenbach, 2019). bean farmer FBean01 stated the following: “preparing organic The findings of this study reveal that families complement fertilizers is cheaper but takes time, and we need training on how and finance farm production with family members finding to prepare them.” employment in temporary or seasonal labor (i.e., collection and Coffee pruning is an operational decision made once a few processing during harvest season), generating strong mobility weeks before the beginning of the coffee harvest season and after it within and between the CA4 countries. The cash obtained ends (Cerda et al., 2020). In the focus group, the coffee association during the coffee harvest is used to (i) meet the food needs FFGCA01 mentioned that the rainy seasons trigger them to decide of households, mainly during the food shortage season due to to prune regularly to increase yields, ensure free entry of light, extreme events; and (ii) the purchase of inputs for planting staple and rejuvenate the coffee plants. However, in times of crisis, crops in future seasons (Bacon et al., 2014). But migration can households reduce the time andmoney that they dedicate for coffee also be permanent, triggered by loss of harvest, bad prices for maintenance practices such as weeding, pruning, and fertilization farmers and permanent deterioration in the standard of living (Eakin et al., 2006). Finally, a successful staple harvest is essential of the staple grains and coffee families. The households with for food availability in the family and selling the remainder in local permanent migrants are more vulnerable to food insecurity due markets (Baumann et al., 2020). A forecast of a prolonged mid- to the reduced family labor available, such as for replanting summer drought or an extreme event (e.g., hailstorm or hurricane) crops or rebuilding farm infrastructure following extreme events can affect the harvest of a crop, ending in total loss if the farmer (Tucker et al., 2010; Ibáñez et al., 2022). However, remittances does not make the correct decision of when to harvest. Mendoza from migrants could offset these negative impacts of reduced et al. (2017) reported that maize growers base their harvest dates family labor (Davis and Lopez-Carr, 2014; Alpízar et al., 2020). on key calendar dates (e.g., after All Saints’ Day, celebrated on According to the U.S. Census Bureau 2019 the contribution November 1st) or moon phases, with a full moon considered to of Immigrants to the United States from the CA4 Countries result in much tougher grain. Coffee farmers FCoffe02–03 stated are: El Salvador (37%), Guatemala (29%), Honduras (19%), and the following: “If there is rain, the coffee ripens quickly, but when it Nicaragua (7%). is heavy rain and excessive sun the next day, then the coffee suffers and burns, and we have ripe but black coffee berries.” 3.2.3. Cluster C: environmental management In Central America, linkages exist between extreme weather 3.2.2. Cluster B: households strategies events, climate change, and land-cover change. In the focus We found that weather and other factors heavily influence post- groups, the coffee association FFGCA01 mentioned shift from harvest decisions for basic grains. Upon harvest, households must coffee to sugarcane due to market and climatic stressors, along evaluate grain availability and decide whether to store it for later with migration—partially propelled by the coffee crisis—that marketing or consumption, or to consume and sell their entire impacts alternative crop viability and land use in coffee-growing harvest. Extreme events that affect crop productivity can cause food areas. Furthermore, financial incentives encourage reforestation shortages and hunger spells (Alpízar et al., 2020), and preclude of marginal agricultural land and safeguarding of forested areas farmers from storing grain. In northern Nicaragua, a majority of against conversion into farmland (Tucker et al., 2010). The farmers buy a portion of their grains in the market (Bacon et al., shift in farming practices in Central America is evident as the 2014). In the case study, bean and maize officers (FOBean01 and region transitions from predominantly cultivating annual crops FOMaize02) mentioned that the price fluctuates due to extreme to prioritizing planted trees, with cereals playing a subsidiary weather and climate events and that this, among other factors, role (Gerlicz et al., 2019; Alpízar et al., 2020). This shift not triggers decisions to store the grains. Farmers mentioned that they only addresses land use challenges but also bolsters the resilience Frontiers inClimate 09 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 of farming households when faced with extreme weather events week), current weather, short-term forecasts (3–5 days), (Harvey et al., 2018). and/or sub-seasonal (2–4 weeks) forecasts. Short-term Converting slash and burn into slash and mulch called the information and early warning systems allow relatively Quesungual—agroforestry system is gaining importance in Central rapid feedback and learning (Griggs et al., 2021). The America to increase soil fertility (Schnetzer, 2018). In the case production cycle of staple grains and coffee highly depends study, the women association (FFGWA01) mentioned having on rainfall patterns. For example, when the rains will environment conservation objectives in mind when deciding to start informs when land preparation should commence. stop burning residues. On moderate to steep slopes, contour Additionally, germination and flowering are triggered by the and terrace planting of coffee is necessary, as they are practical first rains of the rainy season. However, in the case study, measures that limit soil surface erosion, water retention capacity the farmers mentioned that shorter Primera season (first and loss of organic matter (Harvey et al., 2017). In the case rains) and extended mid-summer droughts in recent years study, the field officers (FOCoffee01–02) mentioned that farmers are precluding the sowing of maize, in favor of beans. Thus, who experienced severe rainfall or hurricane impacts started the mid-summer drought, which coincides with the maize establishing soil conservation techniques. flowering and grain-filling phases poses significantly limits In recent years, the “climate-friendly” certification is gaining small-scale farmers in CA4 countries (Baumann et al., 2020). recognition in the coffee sector, offering a price premium to farmers By contrast, in the Postrera season (second rains), farmers who implement favorable climate adaptation and mitigation are affected by excess rain and the hurricane season. Strong practices based on ecosystem services’ conservation, restoration winds lead to lodging and grains falling, and torrential rainfall and sustainable management (Eakin et al., 2014). Ecosystem-based in the Eta and Iota hurricanes in 2020, for instance, brought adaptation (EbA) is also a way to enhance farm management caused substantial damage to coffee plantations, and “milpas” with environmental outcomes. EbA includes planting live fences, (intercropping of maize and bean) were lost entirely due to creating barriers to animal movement, and providing animal landslides (Pons et al., 2021). fodder, firewood, timber and fruits (Harvey et al., 2017). Many • Tactical decisions support planning actions that depend other farm-level practices have external benefits when implemented upon farmer perceptions of the past season, climatological at the landscape scale, such as helping retain moisture and regulate information, seasonal forecasts (3–6 months) and interannual the temperature of the soil. However, improved management with variability (i.e., El Niño, La Niña, and neutral conditions) environmental outcomes in mind can be limited by (i) the lack to minimize food insecurity risk and maximize annual of formal property rights that precludes farmers from longer-term farm profits. Tactical planning involves decisions such as planning and from making more ambitious investments in their crop and variety choice for staple systems, postharvest, lands; and (ii) lack of family labor due to out-migration as a barrier soil conservation, diversification, and implementation to implement new practices (Kearney et al., 2019; Alpízar et al., of agroforestry systems that impact different stages of 2020). production. For these, farmers need access to historical climate information, seasonal rainfall, and drought forecasts. Agroforestry systems and soil conservation strategies play a crucial role in mitigating the effects of droughts. These 3.3. Needs and demands of climate tactical decisions, well-adapted to the region’s dry and hilly information conditions, are renowned for their resilience to climate change, as they help conserve water, maintain soil health, and Farmer engagement in the early stages of the development support biodiversity. of climate services can help identify variables or meteorological • Strategic decisions require advance planning based on events of interest. It can also help determine lead times of medium- to long-term information (interannual up to 10 information, formats and translation tools, and capacity gaps to years and multiple decades). For example, a 3-year drought enable use. Figure 4 shows the weather and climate information (2014–2016) in the dry Pacific region of Central America needs, along with the required variables (in colors), identified in resulted in 1.6 million people becoming food insecure and the literature review and the case study for each type of farmer’s 3.5 million requiring humanitarian assistance (FAO, 2016). decision. These decisions are categorized base on the timescale that In the case study was difficult for the extension officers and influences them (for further details, see Supplementary Table S3). farmers to anticipate responses due to complexity of long- Operational and tactical decisions are made based on known or term planning, and to the uncertainty of any available climate predicted conditions, and strategic decisions are based on plausible information at those timescales. However, long-term climate conditions or scenarios. The graph shows that extreme events and scenarios have been shown useful to determine suitable rainfall data appear to be the most required information that could cropping zones (Bunn et al., 2015). Strategic planning is support the farmers’ decisions across different time scales in Central especially useful for coffee, which is a perennial crop. Varietal America. However, variation exists between types of decisions in choices, diversification, full exposition or agroforestry, and terms of what information is most useful. migration are some of the decisions that can use mid- to long-term climate projections. The most useful information • Operational decisions impacting farmers’ day-to-day at this timescale includes historical climate data to identify fieldwork are based on local knowledge (bioindicators, any existing trends (climate change) or lack thereof, as well observation), recent past weather conditions (e.g., days to as to examine the current frequency of events (e.g., droughts) Frontiers inClimate 10 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 over extended periods, projections of drought frequency traditional knowledge, religious dates, and memories of near past and intensity, rainfall and temperature, and changes in the seasons’ rainfall. However, in light of this study, climate services frequency of ENSO and extreme events. must increase the understanding of the usefulness of weather and climate information. For example, short-term information is helpful for operational decisions that are continually adjusted 4. Discussion in the next few days (i.e., apply inputs, land preparation and management). On the contrary, a rainfall forecast for the next Climate Services can be a powerful way to better integrate local few days would be inadequate to decide on a crop or variety knowledge and scientific information into the decision-making for planting. However, the weather forecast may be adequate to process in Central America. According to Born et al. (2021), “as determine the planting window, whereas a climate forecast for the complex as farmer decision-making for climate risk management entire season appears not to be (Guido et al., 2020). This study might be, understanding the farmer decision space allows for facilitated an understanding of the usefulness of specific weather identifying potentially useful information and gaps in information and climate information, as well as the potential applications for provision”. This paper used a mind map approach to gain a deeper operational, tactical, and strategic farmer decisions. understanding of how farmers make decisions within their farm This paper makes several contributions to the design and systems in the CA4 region at various timescales, as a holistic system. implementation of climate services in Central America for small- It identifies the factors that influence these decisions at the farmer scale farmers in staple and coffee systems. First, the climate services level and discusses the approach’s limitations and opportunities. developers must recognize that many coffee and basic grain small- This analysis carries significant implications for the development scale farmers already actively demand tailored weather and climate of climate services in Central America. The results reveal that information without leaving aside their existing experience and (i) the mind map approach facilitated and provided a holistic local knowledge. Second, the fact that farmer’s decision-making understanding of the farmer’s decision-making. The approach complexity varies across systems and landscapes represents a was flexible enough to involve literature review and field data in significant opportunity to design cross-time scale climate services. the various stages of the development, whereby farmer decision- Third, the results also indicate the need to enhance climate literacy making processes can be presented in a mind map diagram, which among farmers, enabling them to better incorporate and demand is more understandable for the non-modelers and, thus, enhances relevant information. This improvement will empower farmers to farmer’s discussion, (ii) 13 critical decisions were identified that determine which tools and knowledge are most valuable for their farmers make in their crop cycle and their triggers, allowing specific situations. to group them into three clusters (production, household and environmental) and classify the decisions into lead-time categories (operational, tactical and strategic) and (iii) explored the role of the weather and climate information in the maize, bean and coffee 4.1. Limitations and future research and production systems involving a sequence of interrelated decisions action at multiple timescales, where one of the most important factors that trigger the decisions of farmers is the food security shortages due to With the development of the mind map it was possible to extreme events in Central America. identify gaps and provide recommendations for providing climate The findings highlight distinct considerations in comparison services in Central America based on evidence from the literature to more extensively studied regions, particularly Africa. Evidence review and a range of qualitative data (i.e., interviews, focus groups from Africa reveals that climate services for agriculture have and observation). However, this research does not address how brought about significant changes in how farmers access and farmers can access weather and climate information to support utilize climate information, influencing decision-making (Guido their decision-making. This implies that farmers may be utilizing et al., 2020; Born et al., 2021). These studies emphasize the data from various sources that this study was unable to account importance of integrating short-term actions with long-term for. This paper acknowledges that the impacts of using climate resilience-building efforts. In Central America, marked differences and weather information cannot be isolated from other variables, in farming systems, decision-making processes, socioeconomic such as price fluctuations, migration caravans, or government contexts, trade agreements, and non-climatic constraints compared incentives. Despite the study’s novelty, only a semi-structured to Africa play a pivotal role. These disparities underscore specific interviews and focus groups with a small sample in Honduras challenges that shape farmers’ decisions and strategies in the region. were conducted with a relatively small number of participants Historical factors like land tenure disputes and civil conflicts as a case study. Future studies could involve a larger sample influence decision-making. Additionally, access to credit and of farmers and extension officers from various staple and coffee financial services through rural banks plays a crucial role, further zones in Central America to gain deeper insights into the usability highlighting the unique challenges faced in this region. of climate and weather information services for on-the-ground The mind map approach encouraged dialogue between farmers decision-making. Additionally, research is needed on gender, and agriculture experts in a two-way communication helping set youth, and social inclusion in climate services considering the roles opportunities and gaps in the early design of climate services. of different household members in decision-making and differing Farmers have developed strategies to decide what, when, and access to information. where to plant. The results suggest that many farmers in CA4 The mind map was the first non-formal representation of base the decisions of their future expectations of climate on their ontologies applied in Central America to better understand Frontiers inClimate 11 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 farmer decision-making that could evolve into a more formal Author contributions Ontology Web Language, establishing a decision support system to help the process of co-production into the climate services DG: conceptualization, methodology, formal analysis, development. However, as technologies emerge, it is important to writing—original draft, writing—review and editing. GC, PD, DO, consider the integration of traditional knowledge with new sources and JR-V: Writing—review and editing. All authors contributed to of information (e.g., advice from extension officers, seasonal the article and approved the submitted version. forecasts, early warning systems, and agroclimatic calendars) to foster innovation for decision-making. This study could help the Acknowledgments farmers adjust their decision-making to operate time-efficiently and avoid extreme climatic events during sensitive growing phases. This work was supported by the Resilient Central America However, more efforts should bemade to improve farmers’ capacity (ResCA) project, led by the Nature Conservancy under a grant and skills toward using weather/climate information in farm from the United States Department of State. We also acknowledge management decisions, ensuring agricultural cropping systems’ support from the Climate Change, Agriculture and Food Security future adaptability and profitability. (CCAFS), under the project Agroclimas (http://bit.ly/2i3V0Nh). CCAFS was carried out with support from CGIAR Trust Fund 5. Conclusion Donors and through bilateral funding agreements. For details, please visit https://ccafs.cgiar.org/donors. We also acknowledge the support of the AgriLAC Resiliente One CGIAR Initiative (2022– By examining the farmer decision-making mind map 2024). within their system, understanding the factors that trigger those decisions, and identifying the weather and climate information required, along with the challenges faced by small-scale farmers Conflict of interest in Central America, regional governments, in collaboration with donors, researchers, and the private sector, can effectively The authors declare that the research was conducted in the support small agricultural producers in implementing climate absence of any commercial or financial relationships that could be change adaptation measures. For this, small-scale farmers construed as a potential conflict of interest. require tailored climate services with technical assistance and financial and legislative support to implement the appropriate Publisher’s note adaptation measures for their production systems and their local context. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the Data availability statement reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or The raw data supporting the conclusions of this endorsed by the publisher. article will be made available by the authors, without undue reservation. Author disclaimer Ethics statement The views expressed in this paper cannot be taken to reflect the official opinions of these organizations. The studies involving humans were approved by Research Ethics Committee, University of Reading. The Supplementary material studies were conducted in accordance with the local legislation and institutional requirements. The participants The Supplementary Material for this article can be found provided their written informed consent to participate in online at: https://www.frontiersin.org/articles/10.3389/fclim.2023. this study. 1235601/full#supplementary-material References Afzal, H., and Kasi, M. K. (2019). “Ontology-based knowledge modeling for rice farmer households in Central America: recurrent versus extreme weather-driven crop production,” in 2019 7th International Conference on Future Internet of Things and events. Reg. Environ. Change 20, 22. doi: 10.1007/s10113-020-01592-y Cloud (FiCloud). Presented at the 2019 7th International Conference on Future Internet Bacon, C. M., Sundstrom, W. A., Flores Gómez, M. E., Ernesto Méndez, of Things and Cloud (FiCloud) (Istanbul), 343–350. V., Santos, R., Goldoftas, B., et al. (2014). Explaining the ‘hungry farmer Alayón-Gamboa, J. A., Ku-Vera, J. C., and de Campeche, S. F. (2011). Vulnerability paradox’: smallholders and fair trade cooperatives navigate seasonality and change of smallholder agriculture in Calakmul, Campeche, Mexico. Indian J. Tradit. Knowl. in Nicaragua’s corn and coffee markets. Glob. Environ. Change 25, 133–149. 10, 125–132. doi: 10.1016/j.gloenvcha.2014.02.005 Alpízar, F., Saborío-Rodríguez,M.,Martínez-Rodríguez,M. R., Viguera, B., Vignola, Baumann, M. D., Zimmerer, K. S., and van Etten, J. (2020). Participatory R., Capitán, T., et al. (2020). Determinants of food insecurity among smallholder seed projects and agroecological landscape knowledge in Central Frontiers inClimate 12 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 America. Int. J. Agric. Sustain. 18, 300–318. doi: 10.1080/14735903.2020. Frisch, D., and Clemen, R. T. (1994). Beyond expected utility: rethinking behavioral 1775930 decision research. Psychol. Bull. 116, 46. doi: 10.1037/0033-2909.116.1.46 Bielecki, C. D., and Wingenbach, G. (2019). Using a livelihoods framework to Garcia-Solera, I., and Ramirez, P. (2015). Central America’s seasonal climate outlook analyze farmer identity and decision making during the Central American coffee leaf forum. Clim. Serv. 8. rust outbreak: implications for addressing climate change and crop diversification. Gerlicz, A., Méndez, V. E., Conner, D., Baker, D., and Christel, D. (2019). Use and Agroecol. Sustain. Food Syst. 43, 457–480. doi: 10.1080/21683565.2019.1566191 perceptions of alternative economic activities among smallholder coffee farmers in Bokusheva, R., Finger, R., Fischler, M., Berlin, R., Marín, Y., Pérez, F., et al. (2012). Huehuetenango and El Quiché departments in Guatemala.Agroecol. Sustain. Food Syst. Factors determining the adoption and impact of a postharvest storage technology. Food 43, 310–328. doi: 10.1080/21683565.2018.1532480 Sec. 4, 279–293. doi: 10.1007/s12571-012-0184-1 Griggs, D., Stafford-Smith, M., Warrilow, D., Street, R., Vera, C., Scobie, M., et al. Born, L., Prager, S., Ramirez-Villegas, J., and Imbach, P. (2021). A global meta- (2021). Use of weather and climate information essential for SDG implementation.Nat. analysis of climate services and decision-making in agriculture. Clim. Serv. 22, 100231. Rev. Earth Environ. 2, 2–4. doi: 10.1038/s43017-020-00126-8 doi: 10.1016/j.cliser.2021.100231 Grillo, J., and Holt, J. (2009). Application of the Livelihood Zone Maps and Profiles Bouroncle, C., Imbach, P., Rodríguez-Sánchez, B., Medellín, C., Martinez-Valle, for Food Security Analysis and Early Warning. Washington, DC: USAID FEWSNET. A., and Läderach, P. (2017). Mapping climate change adaptive capacity and Guido, Z., Lopus, S., Waldman, K., Hannah, C., Zimmer, A., Krell, N., et al. vulnerability of smallholder agricultural livelihoods in Central America: ranking and (2021). Perceived links between climate change and weather forecast accuracy: descriptive approaches to support adaptation strategies. Clim. Change 141, 123–137. new barriers to tools for agricultural decision-making. Clim. Change 168, 9. doi: 10.1007/s10584-016-1792-0 doi: 10.1007/s10584-021-03207-9 Bunn, C., Läderach, P., Ovalle Rivera, O., and Kirschke, D. (2015). A bitter cup: Guido, Z., Zimmer, A., Lopus, S., Hannah, C., Gower, D., Waldman, K., climate change profile of global production of Arabica and Robusta coffee. Clim. et al. (2020). Farmer forecasts: impacts of seasonal rainfall expectations on Change 129, 89–101. doi: 10.1007/s10584-014-1306-x agricultural decision-making in Sub-Saharan Africa. Clim. Risk Manag. 30, 100247. Buzan, T., and Buzan, B. (2006). The Mind Map Book. Pearson Education. New doi: 10.1016/j.crm.2020.100247 York, NY: Penguin Group. Harvey, C. A., Martínez-Rodríguez, M. R., Cárdenas,’, J. M., Avelino, J., Rapidel, B., Cerda, R., Avelino, J., Harvey, C. A., Gary, C., Tixier, P., and Allinne, C. Vignola, R., et al. (2017). The use of Bieby smallholder farmers in Central America. (2020). Coffee agroforestry systems capable of reducing disease-induced yield and Agric. Ecosyst. Environ. 246, 279–290. doi: 10.1016/j.agee.2017.04.018 economic losses while providing multiple ecosystem services. Crop Prot. 134, 105149. Harvey, C. A., Saborio-Rodríguez, M., Martinez-Rodríguez, M. R., Viguera, B., doi: 10.1016/j.cropro.2020.105149 Chain-Guadarrama, A., Vignola, R., et al. (2018). Climate change impacts and Cruz-Bello, G. M., Eakin, H., Morales, H., and Barrera, J. F. (2011). Linking multi- adaptation among smallholder farmers in Central America. Agric. Food Sec. 7, 57. temporal analysis and community consultation to evaluate the response to the impact doi: 10.1186/s40066-018-0209-x of Hurricane Stan in coffee areas of Chiapas, Mexico. Nat. Hazards 58, 103–116. Hellin, J., Cox, R., and López-Ridaura, S. (2017). Maize diversity, market access, and doi: 10.1007/s11069-010-9652-0 poverty reduction in the western highlands of Guatemala. Mt. Res. Dev. 37, 188–197. Darnhofer, I., Gibbon, D., Dedieu, B. (eds.) (2012). Farming Systems Research into doi: 10.1659/MRD-JOURNAL-D-16-00065.1 the 21st Century: The New Dynamic. Netherlands: Springer. Hellin, J., and Schrader, K. (2003). The case against direct incentives and the Davis, J., and Lopez-Carr, D. (2014). Migration, remittances and smallholder search for alternative approaches to better landmanagement in Central America.Agric. decision-making: implications for land use and livelihood change in Central America. Ecosyst. Environ. 99, 61–81. doi: 10.1016/S0167-8809(03)00149-X Land Use Policy 36, 319–329. doi: 10.1016/j.landusepol.2013.09.001 Hellin, J. J., Lopez-Ridaura, S., Sonder, K., Camacho Villa, T. C., and Gardeazabal Dodd, W., Gómez Cerna, M., Orellena, P., Humphries, S., Sadoine, M. L., Zombré, Monsalue, A. (2019). A Guide to Scaling Soil and Water Conservation in the Western D., et al. (2020). Factors associated with seasonal food insecurity among small-scale Highlands of Guatemala. Mexico: CIMMYT. subsistence farming households in rural Honduras. Int. J. Environ. Res. Public Health Hollinger, S. (2009). Meteorological forecasting for agricultural production. Syst. 17, 706. doi: 10.3390/ijerph17030706 Anal. Model. Food Agric. 1, 397–409. Eakin, H. (2000). Smallholder maize production and climatic risk: a case study from Husáková, M., and Bureš, V. (2020). Formal ontologies in information systems Mexico. Clim. Change 45, 19–36. doi: 10.1023/A:1005628631627 development: a systematic review. Information 11, 66. doi: 10.3390/info11020066 Eakin, H., Tucker, C., and Castellanos, E. (2006). Responding to the coffee crisis: a Ibáñez, A. M., Quigua, J., Romero, J., and Velásquez, A. (2022). Responses pilot study of farmers’ adaptations in Mexico, Guatemala and Honduras. Geogr. J. 172, to Temperature Shocks: Labor Markets and Migration Decisions in El Salvador. 156–171. doi: 10.1111/j.1475-4959.2006.00195.x Washington, DC: World Bank. Eakin, H., Tucker, C. M., Castellanos, E., Diaz-Porras, R., Barrera, J. F., andMorales, Imbach, P., Beardsley, M., Bouroncle, C., Medellin, C., Läderach, P., Hidalgo, H. (2014). Adaptation in a multi-stressor environment: perceptions and responses to H., et al. (2017). Climate change, ecosystems and smallholder agriculture in climatic and economic risks by coffee growers in Mesoamerica. Environ. Dev. Sustain. Central America: an introduction to the special issue. Clim. Change 141, 1–12. 16, 123–139. doi: 10.1007/s10668-013-9466-9 doi: 10.1007/s10584-017-1920-5 Eash, L., Fonte, S. J., Sonder, K., Honsdorf, N., Schmidt, A., Govaerts, B., Kawtrakul, A. (2012). Ontology engineering and knowledge services for agriculture et al. (2019). Factors contributing to maize and bean yield gaps in Central domain. J. Integr. Agric. 11, 741–751. doi: 10.1016/S2095-3119(12)60063-X America vary with site and agroecological conditions. J. Agric. Sci. 157, 300–317. doi: 10.1017/S0021859619000571 Kearney, S. P., Fonte, S. J., García, E., Siles, P., Chan, K. M. A., and Smukler, S. M. (2019). Evaluating ecosystem service trade-offs and synergies from Edrawsoft (2022). EdrawMind (formerly MindMaster), a Versatile Mind Mapping slash-and-mulch agroforestry systems in El Salvador. Ecol. Indic. 105, 264–278. Tool. Edrawsoft. Available online at: https://www.edrawsoft.com/edrawmind/ doi: 10.1016/j.ecolind.2017.08.032 (accessed June 16, 2022). Kraaijvanger, R., Almekinders, C. J. M., and Veldkamp, A. (2016). Identifying Ewbank, R., Perez, C., Cornish, H., Worku, M., and Woldetsadik, S. (2019). crop productivity constraints and opportunities using focus group discussions: a Building resilience to El Niño-related drought: experiences in early warning and case study with farmers from Tigray. NJAS Wagening J. Life Sci. 78, 139–151. early action from Nicaragua and Ethiopia. Disasters 43, S345–S367. doi: 10.1111/disa. doi: 10.1016/j.njas.2016.05.007 12340 Lourenço, T. C., Swart, R., Goosen, H., and Street, R. (2015). The rise of demand- FAO (2016). Dry Corridor - Situation Report June 2016: FAO in Emergencies. driven climate services. Nat. Clim. Change 6, 13–14. doi: 10.1038/nclimate2836 Available online at: http://www.fao.org/emergencies/resources/documents/resources- detail/en/c/422097/ (accessed January 14, 2020). Mendoza, J. R., Sabillón, L., Martinez, W., Campabadal, C., Hallen-Adams, H. E., and Bianchini, A. (2017). Traditional maize post-harvest management Farmar-Bowers, Q., and Lane, R. (2009). Understanding farmers’ strategic decision- practices amongst smallholder farmers in Guatemala. J. Stored Prod. Res. 71, 14–21. making processes and the implications for biodiversity conservation policy. J. Environ. doi: 10.1016/j.jspr.2016.12.007 Manage. 90, 1135–1144. doi: 10.1016/j.jenvman.2008.05.002 Olson,M. B., Morris, K. S., andMéndez, V. E. (2012). Cultivation ofmaize landraces FEWS NET (2007). Livelihoods | Famine Early Warning Systems Network. Available by small-scale shade coffee farmers in western El Salvador. Agric. Syst. 111, 63–74. online at: https://fews.net/livelihoods (accessed January 31, 2020). doi: 10.1016/j.agsy.2012.05.005 Findlater, K., Webber, S., Kandlikar, M., and Donner, S. (2021). Climate services Pons, D., Muñoz, Á. G., Meléndez, L. M., Chocooj, M., Gómez, R., Chourio, X., promise better decisions but mainly focus on better data. Nat. Clim. Change 11, et al. (2021). A coffee yield next-generation forecast system for rain-fed plantations: 731–737. doi: 10.1038/s41558-021-01125-3 the case of the Samalá Watershed in Guatemala. Weather Forecast. 36, 2021–2038. doi: 10.1175/WAF-D-20-0133.1 Fountas, S., Wulfsohn, D., Blackmore, B. S., Jacobsen, H. L., and Pedersen, S. M. (2006). A model of decision-making and information flows for information- PRESANCA and the FAO (2011). PRESANCA and the FAO. Centroamérica intensive agriculture. Agric. Syst. 87, 192–210. doi: 10.1016/j.agsy.2004. en Cifras - Datos de Seguridad Alimentaria Nutricional y Agricultura Familiar. 12.003 Agronoticias: Agriculture News from Latin America and the Caribbean. Food and Frontiers inClimate 13 frontiersin.org Giraldo et al. 10.3389/fclim.2023.1235601 Agriculture Organization of the United Nations Available online at: https://www.fao. Soares, M. B., Daly, M., and Dessai, S. (2018). Assessing the value of seasonal org/in-action/agronoticias/detail/en/c/492666/ (accessed January 17, 2022). climate forecasts for decision-making. Wiley Interdiscip. Rev. Clim. Change 9, e523. doi: 10.1002/wcc.523 Prokopy, L. S., Haigh, T., Mase, A. S., Angel, J., Hart, C., Knutson, C., et al. (2013). Agricultural advisors: a receptive audience for weather and climate information? Tucker, C. M., Eakin, H., and Castellanos, E. J. (2010). Perceptions of risk and Weather Clim. Soc. 5, 162–167. doi: 10.1175/WCAS-D-12-00036.1 adaptation: coffee producers, market shocks, and extreme weather in Central America and Mexico. Glob. Environ. Change 20, 23–32. doi: 10.1016/j.gloenvcha.2009.07.006 Robert, M., Thomas, A., and Bergez, J.-E. (2016). Processes of adaptation in farm decision-making models. A review. Agron. Sustain. Dev. 36, 64. van Etten, J. (2006). Changes in farmers’ knowledge of maize diversity in highland doi: 10.1007/s13593-016-0402-x Guatemala, 1927/37-2004. J. Ethnobiol. Ethnomedi. 2, 12. doi: 10.1186/1746-4269-2-12 Rose, D. C., Keating, C., and Morris, C. (2018). Understanding How to Influence Verdonck, M., Gailly, F., de Cesare, S., and Poels, G. (2015). Ontology-driven Farmers’ Decision-Making Behaviour: A Social Science Literature Review. Agriculture conceptual modeling: a systematic literature mapping and review. Appl. Ontol. 10, and Horticulture Development Board. UEA Consulting Ltd. 197–227. doi: 10.3233/AO-150154 Schnetzer, J. (2018). Quesungual and Kuxur Rum: Ancestral Agroforestry Systems in Walisadeera, A. I., Ginige, A., and Wikramanayake, G. N. (2015). User the Dry Corridor of Central America. FAO, Subregional Office for Mesoamerica. centered ontology for Sri Lankan farmers. Ecol. Inform. 26, 140–150. doi: 10.1016/j.ecoinf.2014.07.008 Singh, C., Dorward, P., and Osbahr, H. (2016). Developing a holistic approach to the analysis of farmer decision-making: implications for adaptation Zhou, Z., Goh, Y. M., and Shen, L. (2016). Overview and analysis of ontology policy and practice in developing countries. Land Use Policy 59, 329–343. studies supporting development of the construction industry. J. Comput. Civ. Eng. 30, doi: 10.1016/j.landusepol.2016.06.041 04016026. doi: 10.1061/(ASCE)CP.1943-5487.0000594 Frontiers inClimate 14 frontiersin.org