Introduction

In the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, 2022) it is noted that climate change results from the increase of greenhouse gas (GHG) caused by anthropogenic activities, which are generating severe impacts on agricultural systems (Phillips et al., 2017; Timoteo et al., 2016). This is due to changes in rainfall patterns, droughts, floods and the geographic redistribution of pests and diseases (Food and Agriculture Organization of the United Nations [FAO], 2016).

In Peru, 47.9 % of GHG emissions come from land use and land-use change, as well as forestry, and 13.5 % from agriculture (Ministerio del Ambiente [MINAM], 2023). In response to this issue, the Peruvia governnment has proposed reducing GHG emissions by 40 % by 2030 (Actualidad Ambiental, 2020). One strategy to achieve this target is the implementation of sustainable production systems, such as agroforestry systems (AFS).

AFS are a viable option for achieving resilience and connectivity in landscapes facing increasing human pressure (Doblas-Miranda et al., 2014; Grosrenaud et al., 2021). AFS contribute to mitigating the effects of climate change (Schroth et al., 2016), because the associations of crops with tree species increase carbon stocks (De Stefano & Jacobson, 2018), biodiversity (Torralba et al., 2016), water infiltration and retention, as well as soil fertility (Awazi & Tchamba, 2019; Villa et al., 2020).

Coffee AFS stand out as important carbon sinks. According to Hergoulac'h et al. (2012), coffee AFS store more carbon than monocultures, with 25.2 Mg·ha^-1 and 9.8 Mg·ha^-1, respectively. Ehrenbergerová et al. (2016) reported that coffee AFS combined with Pinus spp. store up to 177.5 Mg·ha^-1, whereas coffee monocultures retain 99.7 Mg·ha^-1. Espinoza-Domínguez et al. (2012) estimated that coffee AFS associated with pink cedar (Acrocarpus fraxinifolius Wight & Arn.), macadamia (Macadamia tetraphylla L. A. S. Johnson), and banana (Musa paradisiaca L.) sequester an average of 110 Mg·ha^-1 in aboveground biomass.

In Peru, the Junín region is one of the main coffee-growing areas, accounting for 16 % of national coffee production (United States Department of Agriculture [USDA], 2024). According to the National Coffee Board of Peru (2019), between 2016 and 2019, losses were reported nationwide due to low coffee prices and high production costs, which reached 2.54 USD·kg^-1, resulting in losses of up to 0.84 USD·kg^-1 in some cases. In addition, 80 % of coffee farms in Peru use traditional technology, which contributes to low yields, with a maximum of 15 quintals per hectare annually (León Carrasco, 2020). Given this situation, low yields and volatile coffee prices generate economic losses for farmers, who are forced to expand their plantations, leading to deforestation in tropical forests. Moreover, reduced income leads to migration to urban areas and contributes to an increase in regional poverty, reaching 16.21 % (Instituto Nacional de Estadística e Informática [INEI], 2020). As an alternative to this problem, the economic valuation of carbon sequestration in coffee AFS is proposed through the opportunity cost (OC) method, within a payment for ecosystem services program, to increase producer’s income.

In Latin America, public incentive programs have been developed, which recognize ecosystem services and generate additional income for farmers. Among the most notable are the National Forestry Financing Fund (FONAFIFO) in Costa Rica and the Incentives Program for Small Holders of Forest Lands (PINPEP) in Guatemala (National Forest Institute [INAB], 2025), which offer financial compensation for environmental conservation. In Brazil, the Suruí Carbon Project was a pioneer in the scale of carbon credits certified under the REDD+ mechanism, promoting sustainable AFS (Charchalac Santay, 2012). In Peru, the Alto Mayo Protected Forest Conservation Project has generated more than 4 million carbon credits and has prevented the deforestation of 6 000 ha, demonstrating the feasibility of valuing the environmental services provided by forests (Conservación Internacional, 2025).

In this context and given the importance of studies that economically quantify the environmental contribution of coffee AFS, this research aims to compare carbon sequestration in AFS with that of monocultures and to estimate the economic value of this environmental service as a mechanism to increase income from coffee AFS. The hypothesis is that coffee AFS have greater carbon-sequestration potential and higher economic viability compared with monocultures, due to their capacity to generate additional income through payment-for-ecosystem-services schemes. This hypothesis is supported by previous studies that demonstrate the potential of AFS to combine environmental and economic benefits (Canal Daza & Andrade Castañeda, 2019; De Stefano & Jacobson, 2018; Funk et al., 2019).

Materials and Methods

Study area

The study area is located in the sub-watershed of the Toro and Garou rivers in Chanchamayo, Junín, at elevations ranging from 600 to 2 500 m. The area comprises two life zones: pluvial montane forest and very humid lower montane forest. Temperature ranges from 16 to 24.5 °C, mean annual precipitation of 3 142 ± 398.6 mm and a potential evapotranspiration of 1 347.6 ± 30.6 mm·yr^-1.

The agroforestry and monoculture soils have a loam texture. According to reported literature, agroforestry soils have 24.6 % moisture retention and 2.2 % organic matter, similar to those of monocultures (Noriega-Puglisevich & Eckhardt, 2022).

Land use identification

A simple random stratification was conducted using field data and satellite imagery. Based on land cover and vegetation type (herbaceous, shrub, and tree), a land use map was generated. For this purpose, land use trajectories, 2019 satellite imagery (Sentinel at 10 m), and the Normalized Difference Vegetation Index (NDVI) were employed (Table 1). The images were analyzed using ArcGIS 10.7 (Environmental Systems Research Institute [ESRI], 2019).

Table 1. Land use identification in Chanchamayo, Junín, Peru.

Estimation of carbon stocks

Plot establishment

Following the methodology of Phillips et al. (2016), rectangular plots measuring 20 x 50 m were randomly established, and all trees with DBH greater than 30 cm were measured. Within each plot, three subplots measuring 10 x 10 m were set up to measure trees with DBH between 5 and 30 cm, as well as the height and stem diameter of coffee plants measured 15 cm above the ground. Figure 1 shows the spatial distribution of the 15 plots corresponding to coffee AFS and the 17 plots of monocultures included in the study. Tables 2 and 3 present the characteristics of the identified AFS and monocultures.

Table 2. Coffee agroforestry systems (AFS) in the study area in Chanchamayo, Junín, Peru.

Table 3. Description of the monocultures identified in the study area in Chanchamayo, Junín, Peru.

Calculation of aboveground biomass in coffee AFS and monocultures

Aboveground biomass was estimated using a non-destructive method employing allometric equations (Table 4). These equations are recommended for montane forests because they were developed under similar climate and precipitation conditions. After calculating the aboveground biomass (kg), it was converted to total carbon (Mg·ha^-1), assuming that carbon corresponds to 50 % of the biomass (Rafdinal et al., 2021; Tak & Kakde, 2020).

Table 4. Allometric equations used to estimate aboveground biomass (AGB, kg) of shade trees and monocultures.

Estimation of land-use profitability

The productive characterization was based on 102 surveys applied to households selected from the National Directory of Population Centers of INEI (2018) between May 2021 and January 2022 (Figure 2). The sample size was determined using the formula for known finite populations (Spiegel & Stephens, 2009).

The producers surveyed are smallholder farmers with plots of 2-3 ha and limited access to financing and technology, which influences their production decisions. According to the information collected, 64 % establish coffee plantations using slash-and-burn methods, 9 % through selective thinning, and 27 % acquire already established coffee plots. Additionally, 85 % use uncertified seeds and mainly sell coffee as parchment.

Estimation of opportunity cost

Profitability per hectare was calculated for the 2019-2020 agricultural season. Total costs include expenses related to planting, harvesting, and post-harvest activities.

In this study, gross and net income were estimated exclusively for coffee AFS. Fixed costs were not included due to their high variability among producers and the difficulty of standardizing them in the context of family farming, following the approach proposed by Bentes et al. (2019). Gross income was calculated from the values obtained from the sale of coffee and other crops on the farm and in nearby markets. Net income (NI) was determined using the gross margin resulting from the difference between income and variable production costs (Bentes et al., 2019).

Where,

p = price

q = cost

FC = Fixed cost

Vc = Variable cost

Opportunity cost (OC) is defined as the foregone opportunities related to the optimal use of certain economic resources (Moreno-Sánchez, 2012; Watson et al., 2018). The OC was calculated as the economic value that farmers in the Toro River sub-watershed would receive for not abandoning sustainable agriculture (coffee AFS) and for protecting forest areas.

Economic value per tCO2eq

The economic value per ton of carbon dioxide equivalent (tCO₂eq) was estimated by multiplying the carbon storage results by the equivalence factor of 3.67. As references, the prices per tCO₂eq suggested by Funk et al. (2019) at 20.83 USD, Sohngen et al. (2008) at 27.25 USD and SENDECO₂ (2023) at 89.45 USD were used. The conversion factor of 3.67, recommended by the IPCC (2006) is used to convert tons of carbon to tCO₂eq based on the ratio of their molecular weights (44/12).

The price per tCO₂eq on the SENDECO₂ platform has shown variations between 2008 and 2023. In 2008, the price was 22.02 USD per tCO₂eq, maintaining a downward trend until 2017. From that year onward, prices began to rise progressively, reaching 89.43 USD in 2023 (Figure 3).

Statistical Analysis

Statistically significant differences were determined for the following three aspects: (1) carbon stored between coffee AFS and monocultures, (2) carbon stored among the types of coffee AFS, and (3) the average opportunity cost among these systems.

Since the analysis of carbon stored between AFS and monocultures included more than 50 observations, the normality assumption was verified using the Lilliefors test and the Anderson-Darling test, the latter having greater statistical power. For the analysis of carbon stored among types of AFS and of the average OC, whose groups had sample sizes below 50 observations, the Shapiro-Wilk test was used to assess normality and Bartlett’s test to verify homogeneity of variances.

In all cases, the non-parametric Kruskal-Wallis test (P ≤ 0.05) was applied to determine whether statistically significant differences existed between the medians. Statistical analyses were performed using RStudio, version 4.2.3 (R Core Team, 2023).

Results and Discussion

Carbon storage in coffee AFS and monocultures

Coffee AFS store an average of 62.67 ± 63.6 Mg·ha^-1 of carbon, showing statistically significant differences (P < 0.05) compared to monocultures, which store 6.25 ± 7.33 Mg·ha^-1. Figure 4 shows the average values of carbon stored in aboveground biomass by system type, with error bars representing the standard deviation. The results demonstrate that AFS have a greater carbon storage potential and far exceed monocultures, due to the tree component, which stores between 47 and 50 % of carbon in wood and 39 to 40 % in leaves (Salgado-Mora et al., 2018).

This trend is evidenced in several studies. For example, in Central America and Colombia, Van Rikxoort et al. (2014) report that polycultures associated with Andean walnut (Juglans neotropica Diels) and inga tree (Inga spp.) store 42.5 Mg·ha^-1, whereas monocultures store only 10.5 Mg·ha^-1. Also in Colombia, Canal Daza y Andrade Castañeda (2019) indicate that coffee AFS associated with walnut (L. punicifolia) and banana (M. paradisiaca) store 18.03 Mg·ha^-1 compared to 1.9 Mg·ha^-1 in monocultures. In Costa Rica, Hergoulac´h et al. (2012) found that coffee AFS with inga tree store 25.2 Mg·ha^-1 compared to 9.8 Mg·ha^-1 in monocultures.

Carbon storage by floristic arrangement in coffee AFS

No statistically significant differences were found in carbon storage among the floristic arrangements of the AFS (P > 0.05). However, there was a general trend toward higher carbon storage in systems with greater floristic diversity. This is evident in AFS 5 (coffee + inga tree + fruit species + timber species) and AFS 6 (coffee + inga tree + fruit species + two timber species), which showed the highest carbon content, at 78.45 ± 38.66 Mg·ha^-1 and 81.45 ± 17.44 Mg·ha^-1, respectively (Figure 5).

This trend is consistent with the results of Solis et al. (2020) in Colombia, who reported higher carbon storage in AFS composed of coffee and multiple tree species, including nine and twelve timber and fruit species, reaching values of up to 189 Mg·ha^-1.

The lowest carbon value was reported for AFS 4 (54.65 ± 28.22 Mg·ha^-1) composed of coffee, inga tree and a single timber species. This result is similar to that obtained by Andrade et al. (2014) in Colombia, who reported 36.7 Mg·ha^-1 for coffee AFS with only one timber species (walnut, L. punicifolia). In contrast, the study of Ehrenbergerová et al. (2016) in Peru (Villa Rica) reported higher carbon values in coffee-pine AFS and coffee-eucalyptus AFS, at 77.5 Mg·ha^-1 and 162.3 Mg·ha^-1, respectively. These values exceed those obtained in the present study by 41.82 %, because the authors evaluated AFS with higher tree densities (between 124 and 472 trees·ha^-1). In contrast, in the present study, AFS with timber species had only one to two individuals per plot, resulting in a total of 18 trees across the entire study area, corresponding to a lower average density of 7.5 trees·ha^-1 (based on a total evaluated area of 2-3 ha).

On the other hand, AFS 1 (63.69 ± 33.26 Mg·ha^-1), despite being composed only of coffee and inga tree (Inga spp.), contains carbon above the average of the AFS. The inga tree has a high density (306 trees corresponding to an average density of 127.5 trees·ha^-1) and, therefore has a greater carbon storage potential compared to fast-growing species such as pine or eucalyptus (López-Fernández et al., 2023). In addition, AFS 1 is the most common arrangement in the study area, as coffee farmers consider it a natural fertilizer. The root nodules of the inga tree (a leguminous species) promote the formation of mycorrhizae and consequently facilitate nutrient recycling, especially nitrogen (Dilas-Jiménez & Mugruza-Vassallo, 2020). Furthermore, the inga tree grows rapidly, and its leaf litter produces a considerable amount of organic matter that helps retain soil moisture (León et al., 2016).

The results highlight the effectiveness of coffee AFS in carbon sequestration, in addition to providing various ecosystem benefits. The combination of tree species improves water infiltration and retention, reduces runoff, and stabilizes the microclimate (Villarreyna et al., 2020). Meanwhile, the leaf litter and organic residues enrich the soil’s organic matter and enhance the efficiency of nutrient cycles, including nitrogen and phosporus (Alegre et al., 2017; Alvez & Alayon Luaces, 2020; Dilas-Jiménez & Mugruza-Vassallo, 2020; Navas Panadero et al., 2020; Villa et al., 2020). AFS create microhabitats and foster biological diversity (Canal Daza & Andrade Castañeda, 2019; Vera, 2017), which translates into increased populations of soil macro- and microorganisms, including Collembola, Pseudomonas spp., Bradyrhizobium spp. and earthworms (Vera, 2017). AFS play a crucial role in controlling plant pests and diseases (Tamayo Ortiz & Alegre Orihuela, 2022; Villarreyna et al., 2020), act as biological corridors, and facilitate connectivity among remnant habitats within the landscape (Salazar et al., 2018). All these benefits strengthen agroecosystem resilience to climate change and contribute to greater long-term productive sustainability (Altieri et al., 2015).

Analysis of economic value of carbon sequestration in AFS

The average production cost in coffee AFS is 319.31 USD·ha^-1·yr^-1 compared to 219.37 USD·ha^-1·yr^-1 in monocultures, representing a 31.29 % difference. This variation is due to the additional investments farmers make in fertilizers and in pest-resistant seed or seedlings. In terms of income, coffee sales in AFS are 33 % lower than in monocultures, influenced by price variability in recent years. In Junín, the average coffee price ranged from 1.35 USD·kg^-1 in 2018 to 2.19 USD·kg^-1 in the second quarter of 2023 (Ministerio de Desarrollo Agrario y Riego [MIDAGRI], 2021, 2023). Figure 6 shows the analysis of the average price of coffee grown under chacra conditions (USD·kg^-1) at the national level and in Junín.

Estimation of opportunity cost

OC in the Toro River sub-watershed is interpreted as the compensation or payment that should be offered to the coffee farm when choosing AFS over monoculture practices. This average OC was calculated at 1 774.17 USD·ha^-1·yr^-1.

According to the analysis, no significant differences were reported among the average OC values of the evaluated AFS (P > 0.05). AFS 5 recorded the highest OC at 2 240.46 USD·ha^-1·yr^-1, while AFS 6 had the lowest OC at 500.12 USD·ha^-1·yr^-1. In AFS 1, the OC was 2 231.82 USD·ha^-1·yr^-1, even though it is composed solely of Inga spp. (Table 5).

Table 5. Average opportunity cost (OC) for each coffee agroforestry system (AFS) in Chanchamayo, Junín, Peru.

It is important to note that income is not directly related to the floristic composition of the AFS. Moreover, the sale of by-products from the system is minimal or primarily intended for household consumption (e.g., avocado and banana), while the tree species mainly provide shade for the coffee crop. No system for timber harvesting or rotation has yet been established in these AFS; therefore, income from wood sales is not included.

Economic value of carbon storage in coffee AFS

The economic value per tCO₂eq exceeds the recorded opportunity costs for all agroforestry systems (Table 6). When the lowest price of 20.83 USD is considered, the average economic value is 2.74 times higher than the average opportunity cost. For intermediate prices of 25.83 USD, this increase rises to 3.4 times, while at a price of 27.25 USD, the value is 3.6 times the OC. Finally, with the highest Price of 89.45 USD, the economic value surpasses the opportunity cost substantially, reaching a level 16.9 times greater.

Table 6. Opportunity cost (OC) and economic value of CO₂eq for each agroforestry system (AFS) evaluated in Chanchamayo, Junín, Peru.

The results indicate that the average net income from the sale of other crops is 33 % higher than that obtained from coffee AFS. Additionally, fluctuations in coffee prices over the past several years (2018-2023), combined with unfavorable offers from intermediaries, have led coffee growers to choose for the cultivation of more profitable crops. For this reason, the commercialization of carbon credits emerges as an economically attractive option for coffee producers.

Funk et al. (2019) indicate that a price of 20.83 USD per tCO₂eq would be optimal for discouraging unsustainable activities such as deforestation. This price is similar to the base price of 21.45 USD (20 EUR, exchange rate for February 2024: 1 EUR = 1.07 USD) used by the Duch bank Rabobank, which has linked smallholder farmers in the San Martín region of Peru with international carbon markets through its Agroforestry Carbon Removal Units for the Organic Restoration of Nature (ACORN) platform, in collaboration with the “Asómbrate” initiative of the non-governmental organization Solidad Network (Solidaridad Network, 2023). However, although the carbon market may be an attractive option for diversifying the income of coffee producers, it is essential to consider fluctuation in carbon prices. These variations can be substantially high, up to 33.79 % within a single year, as evidenced in 2021 according to SENDECO₂ data, when the average price closed at 57.36 USD per tCO₂eq·ha^-1, notably different from the 2023 average (89.45 USD per tCO₂eq·ha^-1). Consequently, farmers participating in such Payment for Environmental Services (PES) scheme must be aware of the inherent volatility of carbon market prices, which represents a significant risk.

In Latin America, several PES programs, such as those implemented in Costa Rica (Zamora-Cristales et al., 2022), Mexico (Secretaría de Bienestar, 2020, 2025) and Guatemala (INAB, 2025), have successfully compensated farmers for the conservation of trees in AFS. These programs establish criteria such as the minimum area allocated to AFS and the required number of trees per hectare. Mexico’s “Sembrando Vida” program is particularly relevant to the study region in Junín, as it provides support to smallholders in the amount of 201 USD per month for every 2.5 ha (Secretaría de Bienestar, 2020; Zamora-Cristales et al., 2022). The relevance of this initiative to the research area is based on its consistency with the characteristics of farmers in the study region, whose plots average 2.38 ha. By comparison, the requirements of the programs in Costa Rica and Guatemala are not feasible under the conditions of the study area in Junín. Costa Rica requires between 500 and 10 000 trees per farm, with a payment of 1.55 USD per tree (Oficina Nacional Forestal [ONF], 2024; Zamora-Cristales et al., 2022). In Guatemala, the PROBOSQUE program provide payments of 1 206 USD·ha^-1, requiring a density of 120 trees per hectare, at least 60 of which must be high-value timber species (INAB, 2025). In contrast, local coffee growers have AFS with fewer than 200 trees per hectare, many of which are remnants of primary forests.

Although coffee AFS demonstrates high potential for carbon sequestration and income generation through PES, this study has certain limitations. It is a short-term assessment, which does not allow for evaluating long-term sustainability, as factors such as species growth, natural regeneration, and the evolution of the carbon market may alter the results (Pérez-Portilla & Geissert-Kientz, 2006). In addition, specific management practices employed by coffee farmers, which influence both productivity and carbon sequestration, were not considered (Obando & Obando, 2024).

Despite these limitations, the study’s findings can guide public policies and incentive programs that recognize the environmental benefits of AFS, as long as they are tailored to local conditions and accompanied by technical and financial support.

Finally, evidence that coffee AFS store significantly more carbon than monocultures supports their promotion as a strategy to reduce pressure on forests and encourage sustainable agriculture practices. Quantifying the carbo capture and its economic valuation enables the integration of these systems into national climate change mitigation strategies, such as Peru’s Nationally Determined Contributions (NDC) and REDD+ programs. This information is critical for designing more effective PES schemes with the potential to improve the economic well-being of smallholders and strengthen productive sustainability in rural areas.

Conclusions

Coffee agroforestry systems (AFS) store 80 % more carbon in their aboveground biomass compared to monocultures. Although coffee AFS show no significant differences in overall carbon storage, systems with higher floristic diversity were observed to retain more carbon. The average opportunity cost associated with coffee production in AFS is 1 774.17 USD·ha^-1·yr^-1, however, the implementation of payment for environmental services schemes could increase this value by 3.6 to 16.9 times. This would provide additional income to farmers and promote sustainable practices. Promoting sustainable agriculture requires intersectoral collaboration, policies that provide support to smallholders, and an integrated approach that considers economic, social, and environmental aspects. Nevertheless, a major limitation to adopting AFS is their higher implementation and maintenance costs compared to monocultures, which may pose a barrier for smallholders with limited resources.