Content area
Climate change will have a permanent impact on the Mesoamerican agricultural sector. Current crops, such as shade coffee that is grown in middle-elevation areas, are already showing signs of climatic stress and may not secure agricultural subsistence. Therefore, the first stages of crop diversification are being observed in countries such as Nicaragua, where the migration of new crops like non-shade cocoa may lead to a reorganisation of ecological and social structure. Diversification is an already undergoing process whose underlying motivations and decision-making are not yet fully understood. This study analyses subjacent motivations and contexts that lead to the potential incorporation of cocoa crops in present-day Nicaraguan coffee farms. To achieve that, three main motivations were identified: climatic, economic, and governmental. An econometric analysis was performed over the variables that affect farmers’ motivations and decisions to analyse first this decision-making process and, second to understand how social and climatic evolution over the next decades will impact the context under which agricultural output is shaped.
Introduction
Climate change impacts are expected to hit harder developing countries, among other reasons, due to their lower capacity to adapt (IPCC, 2022a). Food security, water supply, and agricultural production will be some of the most important troubles to be faced by countries that are already dealing with important challenges. Farmers in developing countries already face problems arising from diminished productivity rates caused by a lack of access to extension services, credit, and quality agrarian inputs. This exacerbated vulnerability is expected in poor countries regardless of their climatological characteristics (Cline, 2008).
Poor households with coffee farms represent one of the vulnerable segments of these countries’ populations, as they heavily rely on crops for income, due to limited access to other sources. Many small producers have already observed some early effects of climate change overwhelming their response capacity (Panhuysen and Pierrot, 2014). Coffee crop productivity and its adequacy in context of climate change have been extensively analysed for the short-term (van Vuuren et al. 2007; Olesen et al. 2011; Anwar et al. 2013; IPCC, 2022b; Trnka et al. 2014). Forecasts for coffee-producing countries show scenarios of high uncertainty originating from the expected effects of climate change. This will increase the impacts of pests and diseases, which imply a shrinking productivity and a decreasing quality, as well as increases in production costs, and therefore, small producers will be negatively affected. In the case of Central America, more concretely Nicaragua, climate change has the potential of reducing crop suitability by 40% (Läderach et al. 2010; Glenn et al. 2013). In the long term, it must be noted that impacts are expected to rise. Reductions in quality and yields are expected, accompanied by a rise in production costs. As a direct implication of this new state, drastic reductions in smallholders’ incomes will occur. Poor households with small plantations and high dependence on their yield, will be the most vulnerable; some of them have already seen their bearing capacity overwhelmed (Panhuysen and Pierrot, 2014).
Recognising that climate change generates negative impacts on agricultural output has spawned a desire to increase resilience in agricultural systems. A rational and efficient method of improving resilience may relay in higher diversification of agricultural crops (Lin, 2011). This might serve as an incentive for farmers to incline for strategies that increase resistance while generating economic profits. Cocoa cultivation has been proposed as an alternative to coffee production. Cocoa tree is a sylvatic plant that is known to be sensitive to drought, though quantitative information about the hydric relationship of cultivated plants is scarce (Carr and Lockwood, 2011). Cocoa has played a fundamental role in wood conservation and biodiversity, both in a positive and a negative way. Calculations for the time from 2001–2015 show that in some regions of Brazil, Peru, and Ecuador cocoa was responsible for up to 6% of the deforestation (Goldman et al. 2020). Environmental risk in the cocoa value chain is also caused by the use of pesticides and non-organic fertilisers (Maddela et al. 2020). Many pesticides banned in Europe are still available in America, and the scale of the problem is beginning to be recognised (Vryzas et al. 2020). Nevertheless, it should not be forgotten that cocoa has also been an important factor in the agricultural transformation of wood. Moreover, cocoa’s shade offers a valuable habitat for flora and fauna in woods belonging to agricultural landscapes (Ruf and Schroth, 2004).
Moreover, an emerging body of literature, most notably the seminal work of Dimova and Gbakou (2013), examines the transition from food to cash crops among West African farmers in the context of climate-induced agricultural change in developing countries. Research suggests that cocoa is a promising alternative to coffee in climate-vulnerable areas, with the potential to cover 85% of vulnerable coffee-growing areas below 400 m altitude and 53% in the 400–700 m range (de Sousa et al. 2019). Several studies have reported on this strategy, particularly in countries such as Nicaragua, Honduras, and El Salvador, where the threat of rising temperatures is a serious challenge (Läderach et al. 2010).
The transition to cocoa has already been crucial in other parts of the world; for example, cocoa is already the main export product in several West African countries, which together represent over 70% of the total world production (Afoakwa, 2016). Climate change and improved cocoa prices in the international market have forced the expansion of agricultural lands in these areas and the reduction of natural forest lands. In Mesoamerica, where traditionally shade-grown coffee is beginning to be replaced by extensive monoculture systems, it could generate significant environmental problems (Fernández-Manjarrés et al. 2018).
The need for adaptation of agroforestry systems could be compounded by a significant increase in global demand, with a projected 100% increase by 2050. Worldwide, between 5 and 6 million people work in small-scale agriculture, cultivating over 7 million hectares and contributing a significant portion of their family incomes. According to the World Trade Organisation (WTO), cocoa product exports generate between 5 and 6 million euros annually, and the use of cocoa and cocoa paste for the production of chocolate and cosmetics enables a larger and more equitable market (Guiltinan, 2007).
Decisions about addressing these challenges will significantly affect tropical forests and wildlife species in cocoa-producing countries (Bisseleua et al. 2009). Current trends favour unsustainable, environmentally less conscious practices, primarily focused on meeting consumer demand (Slomkowski, 2005).
On the other side of the balance, sustainable agriculture and rural development’s success will depend greatly on the involvement of different sectors, such as rural populations, governments, the private sector, and international cooperation. The response to climate change impacts will require multi-scale action. This means that even when dealing with local impacts, all rural, national, and global agents must take action, especially where vulnerable populations are involved. When considering rural response, we must also note that this must be oriented by research in order to generate adequate measures for adaptation and mitigation that consider newly developed scenarios. Participatory agricultural research has been defined as the collaboration of farmers and scholars in agricultural research and development. There is a need to explore the climatic, market, and institutional aspects that coffee producers could take into account when dealing with the possibility of introducing cocoa production into their economies.
This work has the aim of analysing the factors taken into account by smallholders when deciding if they switch from coffee to cocoa agriculture. In order to analyse this issue, we performed an econometric analysis of both subjective and objective determinants influencing the decision of changing or not the crop type. Through a multivariate probit differentiating climatic, economic, and public motivations, we studied in the first stage which factors modulate farmers’ choices. In the second phase, we studied how farmers’ perceptions will be affected by climatic and social changes under different scenarios. Different indicators for climate change were included, alongside information about producers’ vulnerability, the percentage of damaged plants in the last decade in incidents that could be related to climate change, water scarcity, price and production cost awareness, and vulnerability indicators.
Context
Crop diversification as adaption strategy in developing countries
Developing countries will need to undertake significant adaptive efforts during the following decades, due to the predictably higher impacts of climate change in their territories and their lower ability to adapt (IPCC, 2022b). Greater vulnerability is expected in poorer countries regardless of their geographical and climatological condition (Cline, 2008). Within developing countries, poor households bear an important share of the burden associated with climate change impacts. Coffee farmers with small plantations lie within this vulnerable segment. Though common perception attributes climate change impacts to the near future, smallholders in developing countries are already seeing their response capacity surpassed by current conditions (Panhuysen and Pierrot, 2014).
The impacts of climate change over coffee (Coffea arabica L.) productivity in the short-term have been thoroughly explored (IPCC, 2022b, Olesen et al. 2011, Trnka et al. 2014, Anwar et al. 2013, Moss et al. 2010, Van Vuuren et al. 2007), though scenarios show a high level of uncertainty. Among the risks prone to be exacerbated by climatological changes, the increase of pests and diseases occupies an important place. Such problems are expected to cause decreases not only in crop productivity and quality but also an increase in costs (Schroth et al. 2009; Baca et al. 2014). This situation has generated an interest in promoting resilient agricultural alternatives to obtain socially and ecologically sustainable agricultural systems.
Crop diversification has been proposed as a tool to achieve a more resilient and sustainable agriculture (Lin, 2011), due to the predictably strong demand and crop adaptability. The inclusion of cocoa trees (Theobroma cacao L.) for crop diversification is one of the mechanisms already in practice by some farms in Mesoamerica. The introduction of cocoa often implies not substituting coffee plants but introducing a mixture of both. In other cases, cocoa serves as a way of giving use to degraded soils and areas where coffee has become a less viable crop. In both cases, cocoa has already been used in certain areas of Mesoamerican region to deal with both present and future impacts of climate change. However, climate change adaptation by introducing cocoa in one way or another can be a threat to the ecosystem when primary forest areas are substituted by cocoa monocultures (Rice and Greenberg, 2000, Ruf, 2011) or when shaded coffee plantations with native trees are replaced by mono-specific open sun cultures. Therefore, it is necessary to achieve a deeper understanding of the visions and goals that local farmers shared, in a context where private benefits from different implementation methods remain unclear (Ruf, 2011).
Role of participatory research
There is a high potential for establishing links between research and the agricultural sector. The establishment of these links in order to promote agricultural research and development lies within what has been denominated as participatory agricultural research (Lin, 2011). Rural response to climate change may benefit from research output, which can influence smallholders’ decision-making processes in order to increase their production and, therefore, living standards. The relation is bidirectional, as research can also benefit from farmers’ expertise to have a wider context and produce more accurate results. The participatory research, in this case, is supported by discrepancies between recommendations made by researchers and implementation of cocoa agriculture in Africa, where persistence of monocultures does not seem rational from the perspective of scientific literature describing higher potential profits coming from shade cocoa (Ruf, 2011).
Materials and methods
Developing a behavioural framework for diversification
The way in which individuals respond to external influences depends on a series of variables most of them non-observable. However, this “black box”-like decision-making process may be analysed through the set of variables that are suspected to be modulating the interest outcome, in this case, the decision of whether or not to implement agricultural diversification based on the incorporation of cocoa plantations in coffee farms.
This study analyses the farmers’ perceptions and expectations to determine which factors will be determinant for the decision to introduce cocoa in order to adapt to the changing climatological and socioeconomic conditions in rural Nicaragua. These perceptions have been modelled to produce a framework that shows which factors affect in which way farmers’ decision-making process. Economic, social, and climate factors all interact with the decision to implement agricultural adaptation strategies based on cocoa diversification. Three main types of drivers were identified: economic, climatic, and social. Economic drivers, such as crop prices or production costs may influence subsistence-level farms precisely to an increased vulnerability caused by poverty (Fig. 1).
Fig. 1 [Images not available. See PDF.]
Transition conceptual framework for crop diversification.
RCPs correspond to Representative Concentration Pathways, and SSPs correspond to Shared Socioeconomic Pathways. Source: Own elaboration.
In this framework, external factors lie at a starting point. These factors do not affect behaviour in a direct way, but they give form to the set of beliefs and experiences that create the set of knowledge and values shaping the individual decision-making process. Once the filter that will be applied to future stimuli is set, decision-making depends not only on the conditional factors—such as global changes—that trigger behaviour, but also on the interpretation of them made by the farmer, which is influenced by their human capacity and values (ie. knowledge, experience, risk perception, etc).
While it is clear that the model analyses the effects of external variables in farmers’ decision-making process and that this perspective is the main factor for the social component of research, it must also be taken into account that farmer behaviour also has an external impact on the environment, both ecological and social. Maladaptation is a common problem in several contexts, and the cocoa transitions offer a challenge for future environment and climate, in cases where transition is made without environmental criteria. It is, therefore, necessary to take into account how aggregated behaviours can impact the environment, which would set a feedback effect over initial triggering factors, not only climatic but also socio-economic.
Climatic factors are vital in any agricultural decision-making process, but changes and uncertainties caused by climate change may increase the weight of these factors in risk-averse individuals. Shared socioeconomic pathway (SSP) scenarios have been taken as a reference combining these perspectives (O’Neill et al. 2014, O’Neill et al. 2017, Delink et al. 2017). Factors related to social development, such as education, poverty level, and support from authorities, have also been considered relevant for the decision-making process. Farmers’ adaptive capacity is closely related both to these social factors and to the magnitude and nature of the changes in climate.
Diversification may also be dependent on the evolution of socioeconomic and climatological circumstances both in a global and local sense. In order to illustrate this range of possibilities, different socioeconomic (SSP) and climatic (RCP) scenarios have to be taken into account, when analysing the evolution of the context that influences outcomes of smallholders’ decision-making process (O’Neill et al. 2014; O’Neill et al. 2017). Within this analysis, three scenarios have been taken into account: a baseline scenario, based on present conditions; an optimistic scenario, which departs from the more optimistic SSP1 scenario, associated with a 4.5 W/m2 radiative forcing as described by the RCP 4.5 scenario; and the more pessimistic SSP3 scenario, linked to an RCP 8.5 based climatic context.
Data collection
The study database includes data from two sources: a survey of smallholder farmers in the Nicaraguan departments of Jinotega and Estelí, and secondary data from various sources. Climatic data such as temperature, rainfall, and humidity were obtained from the Nicaraguan Institute of Territorial Research (INETER) and integrated using the geographical location of the surveyed farms (Fries et al. 2012). In addition, social vulnerability data from the National Institute of Information for Development (INIDE) were incorporated into the farm-level survey. In particular, smallholder coffee producers in Jinotega and Estelí are segmented into municipalities (26 for Jinotega and 112 for Estelí). INIDE provides regional vulnerability indicators at the municipal level (INIDE Instituto Nacional de Información de Desarrollo (2008a); INIDE Instituto Nacional de Información de Desarrollo (2008b)), which were matched with the survey data.
The Nicaraguan Ministry of Agriculture and Forestry (MAGFOR) supported the survey, conducted in the northern volcanic highlands, where most coffee is produced, between February and March 2016. We used a stratified random sampling method, with proportions based on the departmental population, to determine the sample size (Scheaffer et al. 2012):
1
where is the proportion of farmers that hold the characteristics of interest. We can substitute = 0.5 to obtain a conservative sample size. ; , with B being a bound on the error sampling. N is the population. With a population of 1624 smallholder coffee producers, the required sample size was determined to be 209 farmers to be interviewed, with a margin of error of 6.33% and a confidence level of 95%. Figure 2 shows the spatial distribution of the surveyed farms, all located in the representative locations of Jinotega and Estelí, in the north-central region of Nicaragua.Fig. 2 [Images not available. See PDF.]
Sampled farms in Jinotega and Estelí, Nicaragua.
Note: dots represent the surveyed farms. Source: own elaboration.
Table 1 summarises all relevant variables used for the study, as well as descriptive statistics linked with them. It includes both subjective and objective measurements, such as Note: dots represent the surveyed farms. Source: own elaborationproduction factors, water requirement, percentage of plants presenting climate-induced damages, precipitation, and temperatures—which include measures for both dry and wet semesters. This information was complemented with the subjective views given by participants over issues such as cocoa’s prices and costs. This analysis includes also indicators for vulnerability, such as education and households in a situation of extreme poverty (INIDE Instituto Nacional de Información de Desarrollo (2008a); INIDE Instituto Nacional de Información de Desarrollo (2008b)). These descriptive statistics include averages and standard deviations for quantitative data and frequencies for qualitative variables. The estimated regressions combined a set of human capital variables (labour, %Damaged plants and Training courses) and three sets of specific conditional factors grouped into climate and water variables (high water use, average temperature during the rainy season—may to october, average temperature during the dry season—november to april, and soil humidity), economy related factors (awareness about prices and costs of cocoa production), and variables related to social development (education and poverty households).
Table 1. Descriptive statistics of the variables (mean and standard deviation for the quantitative variables and frequency for qualitative variables).
Variables | Name | Unit | Mean | Std dev | Source |
|---|---|---|---|---|---|
Decision-making for transition | Climatic change | 0 = No 1 = Yes | 17.3 82.7 | survey | |
Economic reasons | 0 = No 1 = Yes | 62.7 37.4 | survey | ||
Government support | 0 = No 1 = Yes | 94.9 5.0 | survey | ||
Human capital | Labour | Number | 12.2 | 11.0 | survey |
%Damaged plants | Number | 4.1 | 3.2 | survey | |
Training courses | 0 = No 1 = Yes | 47.4 52.6 | survey | ||
Climate and water | Water use | 0 = No 1 = Yes | 47.9 52.1 | survey | |
T rainy season | Number | 23.5 | 1.8 | INETER | |
T dry season | Number | 22.5 | 1.8 | INETER | |
Humidity | Number | 78.1 | 3.6 | INETER | |
Economy | Prices information (cocoa) | 0 = No 1 = Yes | 69.4 30.6 | survey | |
Costs information (cocoa) | 0 = No 1 = Yes | 80.4 19.6 | survey | ||
Social Development | Education | Number | 32.5 | 8.9 | INIDE |
Poverty households | Number | 445.8 | 1.1 | INIDE |
This data shows that 82.7% of coffee producers would consider switching to cocoa trees because of climate change-related impacts, 37.4% would have in mind purely economic reasons, and that for 5% of the government aid. An average plantation has 12 workers and has seen a 4.1% of its plants damaged by climate-related issues in the last 10 years. 30.6% of coffee farmers have information about cocoa’s market prices and 19.6% of them are aware of the production costs.
Econometric model for farmers’ perception
The econometric model that summarises the theoretical analysis presented so far includes as interdependent variables (Yij) the main reasons for introducing cocoa in coffee farms (climatic, economic and governmental support). The econometric procedure used to jointly estimate the interrelated equations is the multivariate probit model (Cappellari and Jenkins, 2003; Greene, 2018); this model was selected from the intuition that farmers are more likely to change for a mix of reasons than for a single one. We consider two main sets of explanatory variables to evaluate the reasons for adaptation: human capital, which is common to all the alternatives (X), and specific variables, which are specific to each particular dependent variable (W). The model is specified as follows:
2
where i = 1, …, N are farmers, j = 1, …, J are reasons for changing coffee for coca, 1[·] is the indicator function that shows the reason j why the farmer i would change the coffee for cocoa. Xi and Wij are vectors of variables that collect farmers’ characteristics which may be common (X) or not (W) in the specifications of equations; β and γ are parameters to be estimated; and εij are the error terms distributed as a with the variances equal to one and also the model allows for correlation between unobservable information from equations (ρjk).Simulated probabilities for global change scenarios
Probability of diversification to cocoa in Nicaragua has been simulated considering global change scenarios. The joint probability of observing types of answers for changing through cocoa comes from a J-variate standard Normal distribution:
3
The global change scenarios consider two sources of drivers: climate and socioeconomic scenarios. Our aim is to compare the present baseline with a sustainable development alternative and a business-as-usual context for the future. For that purpose, simulations have considered the following scenarios: (i) a baseline based on the current data; (ii) climate projections for RCP 4.5 associated with social conditions in SSP1 (low challenges for adaptation and mitigation), and (iii) climate projections for RCP 8.5 associated to socioeconomic conditions in SSP3 (high challenges for adaptation and mitigation) (O’Neill et al. 2014). IIASA global change projections (Riahi et al. 2017) and more specifically the data projected for Central American region (for temperature, precipitation1 and GDP2) have been used as input data for calculating the simulated probabilities. Soil humidity has been calculated from the temperature and precipitation projections, and the reduction in poverty households has been considered as an inverse proportion to the increase in GDP.
Results
Drivers for crop diversification: from coffee to cocoa
The regression run explains the relationship among different variables and the probability of farmers answering yes to the question of whether each of the three proposed factors would affect their decision to switch crops from coffee to cocoa, being the factors climatic, economic, or the existence of government support. As stated previously, regressions combined a set of human capital variables and three sets of specific conditional factors grouped into climate and water, economy, and social development factors.
It is shown in Table 2 whether each of the variables found impacts the response probability in a positive or negative way. As for human capital variables, their impact varies in both sign and significance levels for all equations, while specific variables obtain higher levels of significance.
Table 2. Results obtained from the regression.
Climatic change | Economic reasons | Government support | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Variables | Coef | Std. Err. | Coef | Std. Err. | Coef | Std. Err. | ||||
Human capital | Labour | 0.034 | (0.015) | ** | −0.013 | (0.008) | −0.011 | (0.014) | ||
%Damaged plants | 0.295 | (0.066) | *** | −0.143 | (0.034) | *** | 0.199 | (0.073) | *** | |
Training courses | −0.107 | (0.277) | −0.267 | (0.195) | 0.893 | (0.403) | ** | |||
Climate and water | Water use | 0.609 | (0.265) | ** | ||||||
T rainy season | 1.010 | (0.506) | ** | |||||||
T dry season | −1.156 | (0.523) | ** | |||||||
Humidity | −0.262 | (0.117) | ** | |||||||
Economy | Prices inf (cocoa) | 0.355 | (0.213) | * | ||||||
Costs inf (cocoa) | 1.053 | (0.251) | *** | |||||||
Social development | Education | −0.069 | (0.032) | ** | ||||||
Poverty households | 0.002 | (0.001) | *** | |||||||
Intercept | 22.359 | (11.333) | ** | 0.206 | (0.233) | −2.095 | (0.925) | ** | ||
ρ21 | −0.492 | (0.125) | *** | |||||||
ρ31 | −0.379 | (0.195) | ** | |||||||
ρ32 | 0.855 | (0.109) | *** | |||||||
Log likelihood | −201.579 | |||||||||
LR test: χ2(17) | 78.970 | *** | ||||||||
Obs. | 209 | |||||||||
(***) significant coefficient at 1%; (**) significant coefficient at 5%; (*) significant coefficient at 10%.
Among the variables affecting the idea that climate would be a reason for switching crops, “labour”, which refers to the number of workers at each farm, is positively and significantly correlated to dependent variable. The percentage of damaged plants also significantly increases the probability of farmers answering positively, which is a result consistent with the intuition that costs caused by climatic variability would favour farmers’ interest in adapting to more resistant crops. Whether or not the farmer has received specific training courses was not found to be significantly related to the result.
The specific variables affecting climate and water as a trigger for crop change allude to four related issues such as water use, average temperatures for rainy and dry seasons and humidity. Pressures over the water supply positively affect this trigger. This result is significant at the 95% confidence level and also corresponds with the intuition that worse hydro-climatic conditions are linked to a positive response. Dry season average temperature and wet season average temperature offer results similar in magnitude and significance but of opposite sign. While higher temperatures in the rainy season increase the probability of a transition, higher dry season temperatures decrease it. Finally, higher humidity has a negative impact on the dependent variable, a result also significant at the 95% confidence level. The correlation between losses caused by environmental and climatic conditions and the prearrangement of adaptation measures against climate change indicates that not only those who have suffered the early consequences of climate change are aware of the risks, but, parallelly, those who have not yet been affected may present a lower degree of awareness and preparedness.
Fewer variables offered significant results for the question of whether the economic pressures would be important when facing the decision to switch crops. Among the transversal variables, the percentage of damaged plants was found to be significantly correlated with the answer to this question. This relation was negative, i.e., the higher the amounts of plants lost, the lower the probability of a positive answer for this question. The number of labourers and training was not found to be significantly related to the dependent variable. Both specific variables related to market and economic issues were found to be significant in the relation. Knowledge of the prices and costs associated with cocoa was positively related to the variable, implying that the more the knowledge of the market conditions, the higher the chance for taking economic and market conditions into account when considering a change in cocoa production.
Again, the percentage of damaged plants was found relevant when questioning farmers on whether government support would be a relevant issue when deciding on using a new crop as a way for adaptation. As with the first equation, this variable was positively related to the outcome. The reception of training courses was also found to be positively related to the result, while the quantity of people working at the plantation was not.
Specific variables affecting this response were also found significant. Education was positively related to the outcome. The number of households under the poverty line in the municipality was also found to be positively related to the probability of answering yes to this question.
Figure 3 shows the multivariate probit marginal effects following Mullahy’s work (Mullahy, 2017).
Fig. 3 [Images not available. See PDF.]
Marginal effects obtained from the regressions for alleging climatic (1), economic (2), and social (3) motivations.
Note: (***) significant coefficient at 1%; (**) significant coefficient at 5%; (*) significant coefficient at 10%.
Climate is a relevant issue when taking the decision of introducing cocoa into their plantations (Fig. 3, left panel). Variables such as the percentage of plants damaged in previous years due to droughts or water requirements appear as relevant factors. Climatic variables also appear to be relevant in this sense. Farmers that have already started to experience climate change related impacts are therefore more prone to care about this phenomenon. While there is an extended idea that climate change impacts will affect future generations, individuals in different areas are already suffering early consequences. The existing correlation between farmers who have experienced such consequences and those showing concern about adaptation to climate change indicates that measures of adaptation may be constrained and held back by those who expect climate change to be an issue not affecting them.
Knowledge of the prices and costs of cocoa increased the probability of a positive answer regarding the choice of cocoa as an attractive adaptation option. The second equation (Fig. 3, central panel) links its own set of variables to the probability of farmers answering that economic variables would be important in their decision to incorporate cocoa into their crops. In this case, the amount of damaged plants had the opposite effect on the probability of giving a positive answer. Observed plant damage drove the farmers’ main decisions with an inverse sign than in the previous equation explaining climatic motivations.
Previous plant damage also appears to predispose farmers to accept more easily government programs. The third equation (Fig. 3, right panel) was related to the probability of farmers answering positively to the question of whether government aid would be relevant in their decision-making process. Again, the amount of damaged plants in the precious decade appears to be a relevant factor affecting positively the variable. In this case, the implications are not as straightforward as in the previous case but seem to indicate that farmers who suffer worse consequences of climatic adversity are more disposed to follow government incentive policies. This represents a challenge for government programs, as their efforts should be precisely targeted to those not predisposed to take action by themselves. As for social variables, poverty increases the odds of considering government support, while education levels have the opposite effect.
In addition, we wanted to analyse whether farmers who were more informed about cocoa production costs and prices might be more likely to consider switching to cocoa when faced with significant damage to their farms. To this end, we introduced interaction effects between specific economic factors (awareness of cocoa prices and production costs) and the percentage of damaged plants. A negative and significant coefficient was obtained when combining the cost of cocoa production and the percentage of damaged plants. The results obtained are presented in Table 3.
Table 3. Multiplicative scenario: results obtained from the regression.
Climatic change | Economic reasons | Government support | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Variables | Coef | Std. Err. | Coef | Std. Err. | Coef | Std. Err. | ||||
Human capital | Labour | 0.033 | (0.015) | ** | −0.009 | (0.009) | −0.016 | (0.018) | ||
%Damaged plants | 0.290 | (0.066) | *** | −0.088 | (0.038) | ** | 0.198 | (0.069) | *** | |
Training courses | −0.124 | (0.279) | −0.271 | (0.198) | 0.784 | (0.383) | ** | |||
Climate and water | Water use | 0.639 | (0.267) | ** | ||||||
T rainy season | 1.039 | (0.512) | ** | |||||||
T dry season | −1.183 | (0.530) | ** | |||||||
Humidity | −0.259 | (0.117) | ** | |||||||
Economy | Prices inf (cocoa) | 0.812 | (0.435) | * | ||||||
Costs inf (cocoa) | 1.956 | (0.578) | *** | |||||||
Prices inf*%Damaged plants | −0.114 | (0.084) | ||||||||
Cost inf*%Damaged plant | −0.176 | (0.095) | * | |||||||
Social development | Education | −0.064 | (0.031) | ** | ||||||
Poverty households | 0.002 | (0.001) | *** | |||||||
Intercept | 22.120 | (11.352) | * | −0.032 | (0.248) | −2.003 | (0.893) | ** | ||
ρ21 | −0.480 | (0.127) | *** | |||||||
ρ31 | −0.376 | (0.203) | * | |||||||
ρ32 | 0.866 | (0.103) | *** | |||||||
Log likelihood | −197.410 | |||||||||
LR test: χ2(19) | 81.520 | *** | ||||||||
Obs. | 209 | |||||||||
(***) significant coefficient at 1%; (**) significant coefficient at 5%; (*) significant coefficient at 10%.
This result showed that farmers who had suffered the loss of plants, and were unaware of the production costs associated with cocoa cultivation, were more willing to switch to cocoa production. It is also in line with other findings where adaptation in vulnerable communities has consisted of unplanned actions by farmers in response to multiple simultaneous sources of change (Eakin et al. 2014).
Another issue that could have been of interest in this study would have been the possible influence of the age of the coffee plant as an explanation for the switch to cocoa, due to its greater exposure to disease, but unfortunately, no such information was available. However, related research in the same geographical area suggests that meteorological factors played a predominant role in the coffee rust epidemic (Villarreyna et al. 2020).
Pressures in coffee production as drivers for introducing cocoa
Figure 4 shows how predicted perceptions were distributed among smallholders. All factors considered, modelled probability for farmers considering each of the three proposed motives to introduce cocoa, there is a clear propensity towards considering climatic change as a reason. Though both variability and probability of alleging economic reasons are higher than those associated with government support, neither of both gets close to the 0.5 probability level. Adaptation to climate change may, therefore, be signalled as the main reason behind the decision to switch crops from coffee to cocoa or take similar adaptation measures based on the introduction of cocoa plants. The fact that purely economic reasons do not show similar levels of modelled probability may be explained by climatic risk aversion, as diversifying into cocoa may offer more stable incomes in uncertain climatic conditions. While economic reasons presented a high variance skewed towards low probabilities, government support was generally associated with low probabilities, often close to 0.
Fig. 4 [Images not available. See PDF.]
Probabilities predicted by the model for the three main drivers: climatic change, economic factors, and government support.
Note: The central line in each box represents the median, the boxes indicate the interquartile range (IQR), and the whiskers extend to 1.5 times the IQR. Outliers are shown as individual points. Source: own elaboration.
Figure 5 shows how the amount of plants lost in the previous 10 years impacts the probability of each answer given by farmers. We observe that, while farmers will generally take climatic and ecological reasons into account, they are more likely to take them as a relevant factor when their losses in the past decade are higher. The probability of considering economic reasons as a reason for the change in crop type behaves in a different way, as it diminishes from a probability slightly above 0.4 to values near 0.1 when the percentage of lost plants in the previous decade gets near the 10% line. Finally, farmers focusing on government support are more present among those that have lost more plants, though numbers are low at most points. So it can be concluded that the risk perception is here heavily influenced by present damages occurring. This figure shows us that while awareness is extended over the majority of farmers, the more rapid increase in climate change-related responses rises fast when damaged plants appear to then stabilise when reaching the highest level of probability. This again leaves space for the possibility that short-term planning may cause suboptimal adaptation strategies in Nicaraguan agricultural production. Adaptation plans based on short-term planning may lead not only to lower effectiveness levels, but even to environmental degradation and similar negative externalities, which can fit the definition of maladaptation made by Barnett and O’Neill (2010).
Fig. 5 [Images not available. See PDF.]
Main drivers for crop substitution and modelized behaviour against the amount of damaged plants in the previous decade.
Cocoa as an adaptation option
Figure 6 shows how climate change will affect the probability of farmers across Nicaragua to change from coffee to cocoa under three climate change scenarios: (a) Baseline scenario, which represents present climatic conditions; (b) RCP 4.5 scenario; (c) RCP 8.5 scenario. Under the baseline scenario it can be seen that high probabilities for crop change are restricted to the driest areas in the north-west highlands, while central and eastern Nicaragua, as well as most of the west coast present lower probabilities. Under conditions related to the RCP 4.5 scenario, which presents a reduction of carbon emissions, higher probabilities of change expand to most of the country. More humid mountain and coastal areas in eastern Nicaragua retain lower probabilities, but the impact of climate change is notorious even in the most optimistic scenario. Under RCP 8.5 or business as usual scenario, probability for change is further expanded. Lower probabilities remain just in the restricted areas of the southern zone of Nicaragua’s east coast. Moreover, probabilities increase all over the rest of the country, and reach levels over 0.9 in most of Nicaraguan geography.
Fig. 6 [Images not available. See PDF.]
Probability of farmers in Nicaragua transitioning from coffee to cocoa under different climate scenarios.
Color gradients indicate probability levels, with green representing lower probabilities and red indicating higher probabilities of transition. a Baseline scenario; b RCP 4.5 scenario; c RCP 8.5 scenario. Source: own elaboration.
Even a medium-strength climate change scenario would suggest that most of the territory in Nicaragua would undergo a spread of cocoa-based culture either in mix stands or alone. Our results are simulated for three different scenarios (Baseline, RCP 4.5 and RCP 8.5) depicted in Fig. 6. Scenarios were constructed through the use of averages of values obtained from CCSRNIES, CSIRO, ECHAM4, HADCM3 and NCAR PCM models. The first map (Fig. 6a) shows probabilities for change based on present conditions that correlate highly with the mountainous areas of Nicaragua that are concentrated towards the north and where coffee is planted mainly as this crop prefers the cooler conditions of middle-elevation mountains. Cultivated lands of the Northern mountain region show the highest probabilities for the introduction of cocoa plantations. On the other hand, the Atlantic coast shows lower probability levels. Under an optimistic climate change scenario (RCP 4.5, Fig. 6b), the year 2050 shows an overall increase on probability levels in most of the country. The map shows that most of the country presents probabilities for implementation of cocoa over the 0.9 threshold. Again, the southern Atlantic coastal area presents the lowest probability levels. Finally, the more pessimistic RCP 8.5 scenario (Fig. 6c) shows these low-probability Atlantic areas diminished even more in extension. Again, most of the country shows probabilities for change over 0.9, with the main difference being the reduction of the areas with lower associated probabilities. Under the most pessimistic scenario, most of the country will have made a transition from coffee to cocoa as probabilities for change are over 0.9 except for the lowland areas in the Atlantic part of the country.
Water scarcity is one of the main drivers behind the decisions, according to the studied data. Both humidity and capacity to obtain enough water for plantations were found significantly correlated to farmers’ probability of changing crops due to climatic reasons. Humidity levels and implementation of cocoa under different scenarios increase very fast with even moderate losses of humidity (Fig. 7). Each dot represents the humidity conditions that each Nicaraguan province can expect under different scenarios as related to the average probability for implementation of the introduction of cocoa crops in local farmlands. The figure shows a general decrease in humidity levels in the country, more pronounced under the RCP 8.5 scenario conditions. This tendency is highly correlated with the behaviour of the probability related. Under climate change scenarios probability for adoption of cocoa is higher across the country, though it achieves near-one levels at the RCP 8.5 scenario.
Fig. 7 [Images not available. See PDF.]
Probability of transition as a response to climate change driven humidity changes.
Note: Markers indicate different climate scenarios: black triangles for the baseline, red circles for RCP 8.5 (high emissions scenario) and green circles for RCP 4.5 (medium emissions scenario). Source: own elaboration.
While similar magnitude increases that occur within seasons have small impact levels on probabilities, changes affecting the rainy season temperature averages can produce higher pressure over farmers (Fig. 8). Overall variations in implementation probability of cocoa under the baseline scenario as well as changes due to temperature variations associated to mitigation efforts (RCP 4.5) and business as usual (RCP 8.5) climate scenarios show that changes in temperatures may affect the probability of cocoa crops being introduced.
Fig. 8 [Images not available. See PDF.]
Probabilities predicted by the model for temperature variability under different scenarios.
Note: shades of grey indicate the probability of transition, with darker areas representing higher probabilities. The labelled points correspond to different climate scenarios: “Baseline” represents current conditions, while “RCP 4.5” and “RCP 8.5” indicate medium and high emission scenarios, respectively. Source: own elaboration.
Discussion and conclusions
Determining the processes behind important decisions, such as the introduction of new crops on a farm, is a hard task, but due to the potential impacts on the Mesoamerican ecosystem, it is necessary to reach this knowledge. Farmers’ decisions may have both positive and negative impacts on the environmental equilibrium of the area, an equilibrium that is already being affected by climate change. Knowing the incentives and concerns of regional farmers is therefore compulsory in order to design policies that can tackle the future challenges of the region.
The livelihoods of smallholders may be severely affected by climate change in developing countries such as Nicaragua. High dependence on agriculture posts an increased vulnerability to changes in climate and the ecosystem. Cocoa may also help in this sense, providing more reliable rents in such areas. This could reduce the need for alternative risk reduction schemes, such as insurance schemes, whose effectiveness depends on diverse factors such as access to capital, asymmetric information, or risk assessment capabilities (Skees et al. 2008).
This study presents the results regarding perceptions of Nicaraguan farmers, trying to determine the main variables behind the decision of introducing cocoa crops as a measure to adapt to climate change. According to these perceptions and a series of variables specific to each farm it can be stated that there is evidence signalling crop diversification and change as a method to deal with consequences of climate change.
While the introduction of cocoa is itself an adaptation mechanism for changing environmental conditions, this change may suppose an ecosystemic change by itself. Changes in the composition of crops such as coffee and cocoa in a biodiverse ecosystem may have several impacts. Agricultural systems and techniques play an important role at this point, as impacts may have both positive and negative effects on such environments.
Past experiences in West African agroecosystems show the possible threats that cocoa implantation may bring (Ruf, 2011; Vaast and Somarriba, 2014). Though cocoa agroforestry may help environmental sustainability by promotion of biodiversity (Rice and Greenberg, 2000), implantation systems based on clearing primary forest lands and the adoption of monocultures may imply the environmental costs. Adaptation strategies focused on short-term planning that increase either social or environmental vulnerabilities are regarded as maladaptation (Barnett and O’Neill, 2010) and therefore to be avoided.
Climate change will have a notorious impact on Nicaraguan smallholders’ livelihoods. Their adaptive response has not only impacts on their own welfare but also on the country’s economy and environment. The potential introduction of cocoa plantations as a substitute or complement for coffee seems a preferred way to adapt to changing climatic conditions, as it offers a more reliable source of income for farmers that have their plantations as the main source of family income, often the only one. Agricultural policy in Mesoamerica will have to deal with any possible trade-offs between agricultural well-being and environmental protection if previously forested areas are cleared for cocoa, as happened in Africa. There are several issues that will have to be considered.
The first one is to respond to farmers’ and other agents’ need for insurance. Risk aversion may lead to suboptimal outcomes due to arising externalities such as maladaptation and loss of economic potential. Mechanised cocoa monocultures may be an example of maladaptation, as the spread of this technique has been linked with biodiversity loss (Lin, 2011), though it can be a profitable and relatively secure option unless it depends on irrigation. Agricultural insurance may also be considered a way of ensuring income stability for low-income households (Kimengsi and Azibo, 2015), though their impact may be ambiguous. Considerations such as implementation costs, moral hazard and their uncertain effect on environmental externalities (Kimengsi and Azibo, 2015) make them an option to be more thoroughly analysed. Strategies that reduce risk may imply several benefits. The triple dividend of resilience framework (Tanner et al. 2015), which describes the capacity resilience of policies to avoid targeted risks, stimulate investment and economic activity and produce co-benefits, results adequate to describe potential positive outcomes to be obtained through careful policies.
The second issue is to promote environmentally sustainable practices that help maintain biodiversity in the Mesoamerican region. In this case, crop diversification may offer a high potential. Combined plantations and agroforest techniques may allow for an optimal implantation of cocoa-based diversification in the context of climate change adaptation. Still, this process must be overseen in order to verify if this is the preferred way undertaken by farmers and create incentive systems in case it is not. Typically, agroforestry systems buffer climate variability by maintaining a microclimate in the plantations but require larger areas to produce the same amount of produce as intensive plantations (see below).
A third point to be taken into account is the lesser role of market motivations. This would imply a more inelastic but increased supply of Nicaraguan cocoa in the future. This is relevant due to the foreseen increase in European cocoa demand in the coming years. On the other side of production, European markets should expect lower amounts of Nicaraguan coffee, due to the decreased suitability for coffee cultivation (Baca et al. 2014) in the Nicaraguan territory. Whether this trend extends to other Mesoamerican countries lies out of the scope of this research, but the possibility that global trends in coffee and cocoa production rely on supply-demand equilibrium or on supply restraints will probably be a relevant question for both producers in less developed countries and consumers all over the world. An extension of this implies that if production responds to climatic instead of market forces, policies targeting market failures such as environmental externalities will be constrained, and incentives such as Pigouvian taxes or subventions will have little or no effectiveness in these markets.
Furthermore, while global markets offer crucial opportunities for Nicaraguan agricultural exports, conflicts in distant regions can disrupt the dynamics of global trade, leading to unpredictable market fluctuations and supply chain interruptions. Geopolitical tensions can exacerbate climate-driven risks and deteriorate the capacity to respond through better governance or economic resiliency mechanisms. For a more comprehensive exploration in a subsequent paper, it would be beneficial to integrate geopolitical considerations with climatic, economic, and governmental dimensions.
Acknowledgements
We would like to acknowledge the Spain’s Carolina Foundation for funding the research. Also the institutions that facilitated the survey: the Nicaraguan Institute of Territorial Studies–Instituto Nicaragüense de Estudios Territoriales (INETER), the Ministry of Agriculture and Farming - Ministerio de Agricultura y Ganadería (MAGFOR) in the departments of Estelí and Jinotega and their team of pollster, as well as SOPPEXCCA R.L, (Agrupation of Farming and Services Cooperatives, Jinotega).
Author contributions
All authors contributed equally to this work.
Data availability
The data set will be available under request.
Competing interests
The authors declare no competing interests.
Ethical approval
The Ethics Committee for Biomedical Research (CEIB) at the Faculty of Medical Sciences, UNAN-León (FWA00004523/IRB00003342), approved the entire study on 8 April 2017. The applied survey complies with ethical standards and does not harm the health or rights of the farmers, as established by Law No. 423, approved on 14 March 2002, by the National Assembly of Nicaragua.
Informed consent
The consent for this study was obtained orally because the surveys were conducted in the field with farmers in rural areas of Nicaragua. Given the context, written consent or recordings were not feasible. Instead, the purpose of the study was explained, ensuring that participation was voluntary, The interview only commenced after oral consent had been given. No payment or other incentives were offered in this research. Participants were fully informed about how their anonymity would be protected, the purpose of the research, how their data would be used, and any potential risks of participation.
Database from IIASA International Institute for Applied Systems Analysis (2017a).
2Database from ; IIASA International Institute for Applied Systems Analysis (2017b).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
Afoakwa EO (2016) Chocolate science and technology, 2nd edn. Blackwell Wiley, Oxford
Anwar, MR; Liu, DL; Macadam, I; Kelly, G. Adapting agriculture to climate change: a review. Theor Appl Climatol; 2013; 113, pp. 225-245.
Baca, M; Läderach, P; Haggar, J; Schroth, G; Ovalle, O. An integrated framework for assessing vulnerability to climate change and developing adaptation strategies for coffee growing families in mesoamerica. PLoS One; 2014; 9, e88463.
Barnett, J; O’Neill, S. Maladaptation. Glob Environ Chang; 2010; 20, pp. 211-213. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2009.11.004]
Bisseleua, DHB; Missoup, AD; Vidal, S. Biodiversity conservation, ecosystem functioning, and economic incentives under cocoa agroforestry intensification. Conserv Biol; 2009; 23, pp. 1176-1184.
Cappellari, L; Jenkins, SP. Multivariate probit regression using simulated maximum likelihood. Stata J; 2003; 3, pp. 278-294. [DOI: https://dx.doi.org/10.1177/1536867X0300300305]
Carr, MKV; Lockwood, G. The water relations and irrigation requirements of cocoa (Theobroma Cacao L.): a review. Exp Agric; 2011; 47, pp. 653-676. [DOI: https://dx.doi.org/10.1017/S0014479711000421]
Cline, WR. Global warming and agriculture. Finance Dev; 2008; 45, pp. 23-27.
de Sousa, K; van Zonneveld, M; Holmgren, M; Kindt, R; Ordoñez, JC. The future of coffee and cocoa agroforestry in a warmer Mesoamerica. Sci Rep.; 2019; 9,
Dellink, R; Chateau, J; Lanzi, E; Magné, B. Long-term economic growth projections in the shared socioeconomic pathways. Glob Environ Chang; 2017; 42, pp. 200-214. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2015.06.004]
Dimova, R; Gbakou, M. The global food crisis: disaster, opportunity or non-event? household level evidence from Côte d’Ivoire. World Dev; 2013; 46, pp. 185-196. [DOI: https://dx.doi.org/10.1016/j.worlddev.2013.02.007]
Eakin, H; Tucker, CM; Castellanos, E; Diaz-Porras, R; Barrera, JF; Morales, H. Adaptation in a multi-stressor environment: perceptions and responses to climatic and economic risks by coffee growers in Mesoamerica. Environ Dev Sustain; 2014; 16, pp. 123-139. [DOI: https://dx.doi.org/10.1007/s10668-013-9466-9]
Fernández-Manjarrés, J; Roturier, S; Bilhaut, AG. The emergence of the social-ecological restoration concept. Restor Ecol; 2018; 26, pp. 404-410. [DOI: https://dx.doi.org/10.1111/rec.12685]
Fries, A; Rollenbeck, R; Nauß, T; Peters, T; Bendix, J. Near surface air humidity in a megadiverse Andean mountain ecosystem of southern Ecuador and its regionalization. Agric Meteorol; 2012; 152, pp. 17-30. [DOI: https://dx.doi.org/10.1016/j.agrformet.2011.08.004]
Glenn DM, Kim SH, Ramirez-Villegas J, Läderach P (2013) Response of perennial horticultural crops to climate change. In: Janick J (ed) Horticultural Reviews. John Wiley & Sons, Inc., Hoboken, New Jersey, USA, p 47–130
Goldman E, Weisse M, Harris N, Schneider, M (2020). Estimating the role of seven commodities in agriculture-linked deforestation: Oil palm, soy, cattle, wood fiber, cocoa, coffee, and rubber. Technical Note, World Resources Institute. https://doi.org/10.46830/writn.na.00001
Greene WH (2018) Econometric Analysis, 8th ed. Prentice Hall, Essex
Guiltinan MJ (2007) Cacao. In: Pua E, Davey M (eds) Biotechnology in agriculture and forestry - transgenic crops VI. Springer-Verlag, Berlin Heidelberg, pp 1–30
IIASA (International Institute for Applied Systems Analysis) (2017a) RCP Database (version 2.0.5). http://tntcat.iiasa.ac.at/RcpDb/dsd?Action=htmlpage&page=compare. Accessed Sept 2023
IIASA (International Institute for Applied Systems Analysis) (2017b) SSP Database (Shared Socioeconomic Pathways)— (version 2.0). https://tntcat.iiasa.ac.at/SspDb/dsd?Action=htmlpage&page=welcome. Accessed Sept 2023
INIDE (Instituto Nacional de Información de Desarrollo) (2008a) Estelí En Cifras. Ed INIDE, Managua, Nicaragua
INIDE (Instituto Nacional de Información de Desarrollo) (2008b) Jinotega En Cifras. Ed INIDE, Managua, Nicaragua
IPCC (2022a) (Intergovernmental Panel on climate Change). Summary for policymakers. Pörtner HO, Roberts DC, Poloczanska ES, Mintenbeck K, Tignor M, Alegría A, Craig M, Langsdorf S, Löschke S, Möller V, Okem A (eds) Climate Change 2022: Impacts, Adaptation and Vulnerability. Cambridge University Press, Cambridge, UK and New York, p 3–33
IPCC (2022b) (Intergovernmental Panel on Climate Change). Climate change 2022: impacts, adaptation and vulnerability. Pörtner HO, Roberts DC, Tignor M, Poloczanska ES, Mintenbeck K, Alegría A, Craig M, Langsdorf S, Löschke S, Möller V, Okem A, Rama B (eds.), Cambridge University Press, Cambridge, UK, and New York
Kimengsi, JN; Azibo, BR. How prepared are Cameroon’s cocoa farmers for climate insurance? Evidence from the South West region of Cameroon. Procedia Environ Sci; 2015; 29, pp. 117-118. [DOI: https://dx.doi.org/10.1016/j.proenv.2015.07.196]
Läderach P, Haggar J, Lau C, Eitzinger A, Ovalle O, Baca M, Jarvis A, Lundy M (2010) Mesoamerican coffee: building a climate change adaptation strategy. CIAT Policy Brief no. 2. Cali
Lin, BB. Resilience in agriculture through crop diversification: adaptive management for environmental change. Bioscience; 2011; 61, pp. 183-193. [DOI: https://dx.doi.org/10.1525/bio.2011.61.3.4]
Maddela, NR; Kakarla, D; García, LC; Chakraborty, S; Venkateswarlu, K; Megharaj, M. Cocoa-laden cadmium threatens human health and cacao economy: a critical view. Sci Total Environ; 2020; 720, 137645.
Moss, RH; Edmonds, JA; Hibbard, KA; Manning, MR; Rose, SK; van Vuuren, DP; Carter, TR; Emori, S; Kainuma, M; Kram, T; Meehl, GA; Mitchell, JFB; Nakicenovic, N; Riahi, K; Smith, SJ; Stouffer, RJ; Thomson, AM; Weyant, JP; Wilbanks, TJ. The next generation of scenarios for climate change research and assessment. Nature; 2010; 463, pp. 747-756.
Mullahy, J. Marginal effects in multivariate probit models. Empir Econ; 2017; 52, pp. 447-461. [DOI: https://dx.doi.org/10.1007/s00181-016-1090-8]
O’Neill, BC; Kriegler, E; Riahi, K; Ebi, KL; Hallegatte, S; Carter, TR; Mathur, R; Van Vuuren, DP. A new scenario framework for climate change research: the concept of shared socioeconomic pathways. Clim Change; 2014; 122, pp. 387-400.
O’Neill, BC; Kriegler, E; Ebi, KL; Kemp-Benedict, E; Riahi, K; Rothman, DS; van Ruijven, BJ; van Vuuren, DP; Birkmann, J; Kok, K; Levy, M; Solecki, W. The roads ahead: Narratives for shared socioeconomic pathways describing world futures in the 21st century. Glob Environ Chang; 2017; 42, pp. 169-180. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2015.01.004]
Olesen, JE; Trnka, M; Kersebaum, KC; Skjelvåg, AO; Seguin, B; Peltonen-Sainio, P; Rossi, F; Kozyra, J; Micale, F. Impacts and adaptation of European crop production systems to climate change. Eur J Agron; 2011; 34, pp. 96-112. [DOI: https://dx.doi.org/10.1016/j.eja.2010.11.003]
Panhuysen S, Pierrot J (2014) Coffee barometer 2014. Ed. Hivos, International Union for Conservation of Nature (IUCN NL), Oxfam Novib, Solidaridad and World Wide Fund For Nature (WWF). Den Hague
Riahi, K; van Vuuren, DP; Kriegler, E; Edmondsd, J; O’Neill, BC; Fujimori, S; Bauer, N; Calvin, K; Dellink, R; Fricko, O; Lutz, W; Popp, A; Crespo Cuaresma, J; Samir, KC; Leimbach, M; Jiang, L; Kram, T; Rao, S; Emmerling, J; Ebi, K; Hasegawa, T; Havlik, P; Humpenöder, F; Da Silva, LA; Smith, S; Stehfest, E; Bosetti, V; Eom, J; Gernaat, D; Masui, T; Rogelj, J; Strefler, J; Drouet, L; Krey, V; Luderer, G; Harmsen, M; Takahashi, K; Baumstark, L; Doelman, JC; Kainuma, M; Klimont, Z; Marangoni, G; Lotze-Campen, H; Obersteiner, M; Tabeau, A; Tavoni, M. The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: an overview. Glob Environ Chang; 2017; 42, pp. 153-168. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2016.05.009]
Rice, RA; Greenberg, R. Cacao cultivation and the conservation of biological diversity. Ambio; 2000; 29, pp. 167-173.
Ruf F, Schroth G (2004) Chocolate forests and monocultures: a historical review of cocoa growing and its conflicting role in tropical deforestation and forest conservation. In: Schroth G, Da Fonseca GAB, Harvey CA, et al. (eds) Agroforestry and biodiversity conservation in tropical landscapes. Island Press, Washington DC, p 107–134
Ruf, F. The myth of complex cocoa agroforests: the case of Ghana. Hum Ecol; 2011; 39, pp. 373-388. [DOI: https://dx.doi.org/10.1007/s10745-011-9392-0]
Scheaffer RL, Mendenhall W, Ott RL, Gerow KG (2012). Elementary survey sampling. 7th edn. Cengage Learning, Boston
Schroth, G; Laderach, P; Dempewolf, J; Philpott, S; Haggar, J; Eakin, H; Castillejos, T; Moreno, JG; Pinto, LS; Hernandez, R; Eitzinger, A; Ramirez-Villegas, J. Towards a climate change adaptation strategy for coffee communities and ecosystems in the Sierra Madre de Chiapas, Mexico. Mitig Adapt Strateg Glob Chang; 2009; 14, pp. 605-625. [DOI: https://dx.doi.org/10.1007/s11027-009-9186-5]
Skees, JR; Barnett, BJ; Collier, B (2008) Agricultural insurance background and context for climate adaptation discussions; Paris
Slomkowski, K. Chocolate’s dark side. E Environ Mag; 2005; 16, pp. 33-33.
Tanner T, Swenja S, Emily W, Reid R, Rentschler JE, Rajput S (2015) The triple dividend of resilience: realising development goals through the multiple benefits of disaster risk management. Global Facility for Disaster Reduction and Recovery (GFDRR) at the World Bank and Overseas Development Institute (ODI), London
Trnka, M; Rötter, RP; Ruiz-Ramos, M; Kersebaum, KC; Olesen, JE; Žalud, Z; Semenov, MA. Adverse weather conditions for European wheat production will become more frequent with climate change. Nat Clim Chang; 2014; 4, pp. 637-643.
Vaast, P; Somarriba, E. Trade-offs between crop intensification and ecosystem services: the role of agroforestry in cocoa cultivation. Agrofor Syst; 2014; 88, pp. 947-956. [DOI: https://dx.doi.org/10.1007/s10457-014-9762-x]
Van Vuuren, DP; Lucas, PL; Hilderink, H. Downscaling drivers of global environmental change: enabling use of global SRES scenarios at the national and grid levels. Glob Environ Chang; 2007; 17, pp. 114-130. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2006.04.004]
Villarreyna, R; Barrios, M; Vílchez, S; Cerda, R; Vignola, R; Avelino, J. Economic constraints as drivers of coffee rust epidemics in Nicaragua. Crop Prot; 2020; 127, 104980. [DOI: https://dx.doi.org/10.1016/j.cropro.2019.104980]
Vryzas, Z; Ramwell, C; Sans, C. Pesticide prioritization approaches and limitations in environmental monitoring studies: from Europe to Latin America and the Caribbean. Environ Int; 2020; 143, 105917.
Copyright Palgrave Macmillan Dec 2025