Natural distribution

The mamey sapote in Mexico, according to the databases consulted, is distributed on the Pacific coast, from southern Sinaloa to Chiapas, and on the Gulf of Mexico coast, from Tamaulipas to Tabasco, as well as on the Yucatán peninsula, Morelos, southern Guanajuato, the southern part of the state of Mexico, southern Puebla, eastern San Luis Potosí, and northern Hidalgo and Querétaro (Figure 2). This location coincides with tropical and subtropical zones of Mexico. Its greatest probability of adaptation is mainly found in Veracruz, Campeche, Quintana Roo and Yucatán, where its probable center of origin is reported (^{Pennington & Sarukhán-Kermez, 2005}). This information corresponds to that previously proposed by ^{Villegas-Monter and Granados-Friely (2012)}, who recorded the presence of this species between 30 and 1,050 masl.

Figure 2 Map of the estimated natural distribution of mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) in Mexico. 

Eco-climatic characterization

The accessions considered in this study were divided into three climatic groups (Figure 3) that presented different distributions and climates, possibly associated with three different gene pools, which should be considered within the breeding programs. Under different environmental conditions, populations may have different genes of adaptation (^{Dobzhansky, 1970}). One possible practical implication of this information in breeding programs is the development of hybrid varieties by considering different gene pools, since they are highly likely to obtain heterosis by crossing (^{Wright, 1978}). For their part, the climatograms of the three groups (Figure 4a, 4b, 4c) showed that these have different climatic patterns.

Figure 3: Dendrogram of passport data of mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) using climatic variables. 

Figure 4 Climatograms of the three eco-climatic groups as potential eco-regions of mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) distribution in Mexico (a, b, c) and comparison of mean temperature (d), precipitation (e) and the differential between the maximum and minimum temperatures (f) of the three climatic groups.  

Group 1 was found on average at 115.2 masl, and had the smallest difference between the maximum and minimum temperature of the three groups and the lowest precipitation. Its annual rainfall accumulation was 1,586.9 mm and its average annual temperature was 23.5 °C, with an extreme minimum and maximum of 18.9 and 31.1 °C, respectively. The driest month had 33.8 mm of rainfall and the wettest 244 mm (Figure 4a). According to the Köppen classification modified by ^{García (2004)}, these characteristics correspond to those of a warm subhumid climate with summer rains (Aw).

Group 2 was located on average at 235.3 masl and had the highest precipitation of the three groups. Its annual rainfall accumulation was 2,131.6 mm and its annual average temperature was 24.9 °C, having as an extreme minimum and maximum 16.6 and 32.8 °C, respectively. The driest month had 49.5 mm of rainfall and the wettest 358.5 mm (Figure 4b). According to the Köppen classification modified by ^{García (2004)}, these characteristics belong to those of a warm humid climate with summer rains (Am).

Group 3 was characterized by being found at 1,060.6 masl and by the lowest temperature of the three groups. Its annual rainfall accumulation was 1,605.9 mm and its average annual temperature was 20.8 °C, with its minimum and maximum being 12.5 and 28.8 °C, respectively. The driest month had 23.3 mm of rainfall and the wettest 281 mm (Figure 4c). According to the Köppen classification modified by ^{García (2004)}, these are the characteristics of a semi-warm humid climate with summer rains [A(C)(m)].

When comparing the climatic factors, Group 3 presented the lowest mean temperature (Figure 4d), Group 2 the highest precipitation (Figure 4e) and Group 1 the smallest difference between the minimum and maximum temperature (Figure 4f). This information constitutes an important reference for the selection of specific genotypes for each climatic group; this is due to the benefit represented by its satisfactory adaptation and competitiveness, especially considering the presence of local materials in each of the eco-climatic regions where mamey sapote is distributed (^{Zagaja, 1988}).

Guayaba (Psidium guajava L.) has two gene pools, one tropical and one subtropical (^{Cázares-Sánchez et al., 2010}), and it has been observed that tropical materials are more susceptible to frost (^{Mondragón-Jacobo, Toriz-Ahumada, & Guzmán-Maldonado, 2010}). ^{Rajan, Yadava, Kumar, and Saxena (2007)} indicate that improving crops requires a reserve of genes that are naturally protected in the genetic diversity of each species. In this sense, cultivars may not have a good adaptation to more than one of the eco-climatic regions (^{Cázares-Sánchez et al., 2010}; ^{Núñez-Colín & Goytia-Jiménez, 2009}). Therefore, identifying and selecting specific mamey sapote materials for each eco-climatic region will enable finding desirable genes to incorporate into the cultivars, including those related to resistance to pests, diseases and abiotic factors, as well as to improving fruit quality and productivity (^{Callahan, 2003}; ^{Quamme & Stushnoff, 1988}; ^{Sistrunk & Moore, 1988}).

The genetic hybridization method may be an option for mamey sapote. This is due to the fact that this species is allogamous (^{Villarreal-Fuentes et al., 2015}) and by interbreeding individuals with different genotypes, due to their adaptation to the environment, they would be expected to produce hybrids with high heterosis (^{Bringhurst, 1988}; ^{Hansche, 1988}; ^{Wright, 1978}). This requires that the breeding programs be established in the three regions found in the eco-climatic characterization to optimize the response to the selection of the candidate genotypes to be cultivars in Mexico. ^{Balerdi and Crane (2015)} report various selections suitable for Florida conditions, which should be tested in different regions of Mexico, especially in the distribution areas feasible for Group 1 since most were selected in the Caribbean.

Group 1 is mainly found on the Yucatán peninsula, with some atypical points in Veracruz, Guerrero and Sinaloa (Figure 5). Group 2 covers the coast of the Gulf of Mexico in Veracruz, Tabasco, Campeche and Chiapas, as well as the Pacific coast with a segmented distribution in Jalisco, Colima, Michoacán, Guerrero, Oaxaca and Chiapas (Figure 6). Group 3, which covers the largest area, comprises mainly the northern Gulf of Mexico (Tamaulipas) and the entire Pacific region, from southern Sonora to Chiapas, including some points in Baja California Sur (Figure 7). In addition, Group 3 showed a probable distribution in the central north of the country (Durango, Coahuila and Zacatecas), as well as in southwestern Guanajuato, Morelos, southern Puebla, eastern San Luís Potosí, and northern Querétaro and Hidalgo (Figure 7).

Figure 5 Estimated distribution map of Group 1 (climate Aw) of mamey sapote (Pouteria sapota [Jacq. ] H. E. Moore & Steam) in Mexico. 

Figure 6 Estimated distribution map of Group 2 (climate Am) of mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) in Mexico.  

Figure 7. Estimated distribution map of Group 3 [climate A(C)(W) of mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) in Mexico. 

This suggests that it is necessary to generate and select specific cultivars for each eco-climatic region (^{Cázares-Sánchez et al., 2010}; ^{Rajan et al., 2007}; ^{Zagaja, 1988}), since the obtaining of ideal cultivars in one of the regions does not guarantee their adaptability in the others (^{Quamme & Stushnoff, 1988}). To confirm the inferred existence of different gene pools, according to the groups found by environmental characteristics, morphological and molecular characterization studies are required. It would be advisable to carry out an ethnobotanical study to determine the role of human beings in the distribution of mamey sapote in the three eco-climatic regions found in this work.

Modeling of current potential zones and with the climate change scenario

The Eco-Crop model considered that the mamey sapote growth cycle is 365 days, which is why it shows no lethargy. Additionally, some of the conditions found for this species with Eco-Crop include: extreme temperatures of 15 and 36 °C, optimum minimum and maximum temperatures of 24 and 30 °C, respectively, extreme rainfall patterns of 800 and 4,000 mm, optimum minimum and maximum rainfall patterns of 2,000 and 3,300 mm, respectively, and death temperature in the early stages of 0 °C. Using these data, the following adaptation zones were modeled: excellent, very good, good, marginal, very marginal and unsuitable.

In the CPZ maps it was found that the regions with excellent adaptation, according to the 14 regions described by ^{Morrone (2005)}, are located on the Mexican Pacific coast, along the Gulf of Mexico, on the Yucatan peninsula and at some isolated points in the Balsas River basin (Figure 8). Specifically, the mamey sapote is found in Tabasco and the adjoining part of Campeche, on the coast of Chiapas, several regions on the coast of Guerrero and Jalisco, the west coast of Oaxaca and in some areas in southern Veracruz (Figure 8).

Left: the biogeographic zones of Mexico; right: the political division of Mexico.

Figure 8 Eco-Crop model for mamey sapote (Pouteria sapota [Jacq.] H. E. Moore & Steam) in Mexico, with current data. 

In the FPZ there is an enlargement of the regions with excellent adaptation (Figure 9), mainly in the Balsas River basin and on the Gulf of Mexico. Based on the scenario projected to 2050, it was found that the potential states for producing mamey sapote, according to their degree of adaptation, will be: Tabasco, Chiapas, Veracruz, Guerrero, Campeche, Jalisco, Oaxaca, Nayarit, Quintana Roo, Colima, Michoacán, Morelos, Puebla, San Luis Potosí and Hidalgo (Figure 9). That is, there is a possibility of good to excellent adaptation for the cultivation of mamey sapote in 15 of Mexico’s 32 states, several of them included in the official statistics of 2014 (^{SIAP, 2015}). The current production zones could be expanded to other potential regions described in this paper as FPZ (Figure 9), which could strengthen the cultivation of mamey sapote in Mexico.

Left: the biogeographic zones of Mexico; right: the political division of Mexico.

Figure 9 Eco-Crop model for mamey sapote (Pouteria sapota [Jacq. ] H. E. Moore & Steam) in Mexico, with simulated climate change.