Study area

The study considered regions in the state of Campeche, in southeastern Mexico, where Brosimum alicastrum grows, according to the records of the National Forest and Soil Inventory (INFyS) of the National Forest Commission (^{Conafor, 2014}). These are characterized by a medium sub-evergreen forest cover and high evergreen forest, with clay soils; rainfall ranging between 600 and 4 000 mm, with low water periods of three to seven months; average annual temperature from 18 to 27 ºC, and altitudes in a range from 20 to 1 000 m. The main economic activities are livestock breeding and forestry, in addition to the cultivation of corn, beans, squash, and sugarcane (^{Inegi, 2015}). It should be noted that, during the dry season, the main livestock fed is obtained from the Ramón tree.

Data collection

For the sampling of producers, the localities to be analyzed were first selected, based on the identification of the region with the highest abundance of the species of interest in the state of Campeche, according to the georeferenced data on point distribution from the National Forest and Soil Inventory (^{Conafor, 2014}) and the use of Geographic Information Systems; the observed richness (abundance) and potential distribution (probability of occurrence) were determined using the Maximum Entropy algorithm. The software used was DivaGis v7.5 (^{DivaGis, 2017}), with a pixel size (resolution) of 30 s (1 km²). Both techniques were applied according to the methodology described by ^{Hijmans et al. (2005)}.

The observed richness was obtained through a richness analysis, and the potential distribution was obtained by associating the point distribution with the soil and climate variables of Campeche (^{WORLDCLIM, 2019}) which delimited areas with different probabilities of presence for Brosimum alicastrum, based on its soil and climate requirements. This information made it possible to rule out areas where the taxon is not naturally distributed.

Once the region with the greatest abundance of Ramón trees was determined, eight locations were selected. The selection criteria were: accessibility, travel distance, and the presence of roads. Between September and November 2018, the inhabitants of the selected localities were surveyed, depending on their availability, references in the region, development of some activity in the agroforestry sector, and the presence of B. alicastrum individuals in their plots.

The survey included four sections: 1) Control data: name of the respondent, age in years, locality, schooling in years, occupation and land tenure: private, ejido, and communal; 2) Knowledge about the Ramón tree, with questions about the local name and the common uses in the region; 3) Socioeconomic aspects of the respondents: number of hectares of their plots, activities that they carry out in their plots: agricultural, livestock or forestry; family income from the activity in Mexican pesos; importance of the ramón tree in the region, and, if they commercialize the species: what part of the tree is sold and to what market it caters: regional, national or international; and 4) Desire to reconvert their plots to Ramón tree plantations: limitations for their establishment and cost at which they would sell the kilogram of tree seed.

In addition, a specialist from each of the following institutions was interviewed: Universidad Autónoma Chapingo (Autonomous University of Chapingo, UACH), Colegio de Postgraduados (Graduate College, Colpos), Universidad Autónoma de Yucatán (Autonomous University of Yucatán, UADY), and Centro de Investigación Científica de Yucatán (Scientific Research Center of Yucatan, CICY). The interview guidelines considered aspects of economic, social and environmental constraints to the establishment of Ramón tree plantations in Campeche. The format for the producer survey and the specialist interview was based on the guide for the design of surveys on the adoption of agricultural technologies proposed by ^{CIMMYT (1993)}.

Data analysis

The data collected in the survey of producers were captured in a spreadsheet of the R software, and the descriptive (frequencies) and multivariate statistics (Principal Component Analysis, PCA) were made; producers were characterized socio-economically according to their activity, and the factors associated with the reconversion to ramón tree plantations were determined.

Socio-economic characterization of producers by activity

Agriculture of cyclic crops, mainly corn and beans, was the activity to which 49.47 % (94) of the producers surveyed were dedicated; the remaining 50.53 % (96) corresponded to the livestock (30.53 %), beekeeping (11.05 %) and forestry (8.95 %) sectors. The common names with which the species is denominated are Ramón (80.42 %); Oox, Mayan word, (7.46 %); and Ojuch, Mayan word, or its variant ojite (2.12 %).

Registered uses were as fodder (91.57 %, 174 respondents), lumber or fuel (5.78 %, 11 respondents, and 2.72 %, five respondents, respectively) do not have any use for it. However, its use as fodder is presented as a secondary alternative in periods of low rainfall as indicated by ^{Rojas-Schroeder et al. (2017)} in southeastern Mexico, who argue that Brosimum alicastrum is only used in periods of scarcity of traditional fodder (grasses). Therefore, the Ramón tree does not represent a primary input in the region, so that only 4.74 % (9 producers) mentioned marketing the seed for flour production in a regional market.

Table 1 shows that the average age of producers varies between 48 and 50 years, with similar levels of study not exceeding high school. ^{Aguilar-Gallegos et al. (2015)} cite age as the main condition for conversion among tropical crop producers in southeast Mexico. However, in the case documented herein, the producers that showed greater availability to reconversion were those with private land tenure; this fact is explained by some specialists in the subject due to the freedom of decision that the producers have on their plots.

Factors that condition the conversion to ramón tree plantations by activity

In the PCA, the first component (Prin1) grouped the variables of age (continuous variable), schooling (continuous), land tenure (1: Communal, 2: Ejido, and 3: Private), number of hectares owned by the producer (continuous) and economic income that represents the activity currently carried out (continuous) explains 38.03 % of the total variance of the data. The second component (Prin2) integrated the variables: Importance of Ramón in the region (1: Very low, 2: Low, 3: Medium, 4: High and 5: Very high), price at which it would sell the fruit of the ramón tree (continuous) and adoption of plantations (0: No and 1: Yes) explains 29.06 %. Both components explain 67.09 % of the total variance of the data.

^{Flores-Trejo et al. (2016)} report that, among rambután producers in Soconusco (Chiapas), age, schooling and income are all factors that limit the conversion to new production systems. The study of B. alicastrum grouped these factors under Prin1, in which also, land tenure was a limiting factor (^{Morett-Sánchez and Cosío-Ruiz, 2017}). Prin2 grouped the variables related to the willingness to reconvert, which, according to ^{Escoto et al. (2007)}, is directly linked to the profitability that the producer expects to obtain with the new production system.

In order to establish which are the limiting and readiness factors that most condition the establishment of plantations, the eigenvectors of the first three components that account for 86.63 % of the total variance of the data were analyzed. Maximum values were extracted and re-categorized by quality (Very low ENT#091;0.000 - 0.200ENT#093;, Low (0.200 - 0.400ENT#093;, Medium (0.400 - 0.600ENT#093;, High (0.600 - 0.800ENT#093;, and Very high (0.800 - 1.000ENT#093;) (^{Pérez-Guel et al., 2016}) (Table 2).

It was statistically confirmed in PCA that the factor that most limits the reconversion to ramón tree plantations in Campeche is land tenure. According to ^{Romero-Navarrete (2015)}, one of the main obstacles for the development of the Mexican countryside is the lack of legal certainty regarding the ownership of the plots; hence, the producers who presented more willingness to reconvert their production areas were those who have more freedom to make decisions regarding their plots (^{Morett-Sánchez and Cosío-Ruiz, 2017}).

The PCA also established that the sales price per kilogram of seed ($23.00 MN in average) was the factor that most incentivizes the producers surveyed to reconvert their production areas to Ramón tree plantations. In certain regions of Veracruz, ^{Almaguer-Vargas and Ayala-Garay (2014)} proved that the substitution of perennial species plantations for those of cyclic species improves the profitability for the producers.

Based on the dispersion of the eigenvalues for each of the 190 surveyed producers, they were grouped and differentiated by activity; the factor that most limits reconversion (X axis: land tenure) with the willingness to reconvert (Y axis) (Figure 2). As may be seen, the agricultural sector presented the greatest willingness, followed by livestock breeding, forestry and beekeeping, in a hierarchical order.

Prin1 = Land tenure (reference scale from left to right: communal-ejido-private); Prin2 = Willingness to reconvert (reference scale from bottom to top: very low-low-medium-high-very high).

Figure 2 Limiting factors (Prin1) and willingness (Prin2) to reconvert with Ramón plantations, according to economic activity developed in the production plots.  

Livestock producers mentioned that their priority is not to abandon the activity, but to achieve silvopastoral systems that will generate a more economically accessible feed for livestock. In this regard, ^{Vergara-Yoisura et al. (2014)} consider that commercial ramón tree plantations can satisfy the demand of the livestock sector and reduce the import of grass.

^{Flores-Trejo et al. (2016)} recorded in regions of Soconusco, Chiapas that producers of cyclical crops (agricultural sector) are more willing to reconvert their production areas in order to improve the profitability of their plots. However, ^{Pérez and Terrón (2004)} point out that, in the process of conversion to perennial species, the agricultural sector is the most vulnerable, with respect to livestock and forestry; this is due to the fact that it is the sector that requires most investment and training; therefore, ^{García et al. (2011)} indicate that technical advice and training substantially improve the time of adaptation to new production systems.

Finally, Table 3 summarizes the factors that the surveyed producers perceive as the main limitations for the reconversion to Ramón tree plantations. In addition to land tenure (32.11 %), they also included as determining factors: the lack of technological packages for the management and production of the species (24.74 %), access to the producing areas where B. alicastrum individuals are currently located (22.63 %), and the difficulty to collect the fruits of the tree (14.21 %).

In this regard, ^{Vergara-Yoisura et al. (2014)} document that the problems of access to production areas and post-harvest management of Brosimum alicastrum are solved by creating a specific technological package for the species. For their part, ^{Monge and Hartwich (2008)} postulate that technical advice and training, complemented by the interaction and trust that is generated between producers and trainers, help to minimize the time period required to adapt to new production systems; in order to do this, it is necessary to go through the process of recognizing, knowing, collaborating, cooperating and associating (^{Rovere, 1999}).