Location

The six forest ejidos are located in the Othón P. Blanco and Bacalar municipalities, in the state of Quintana Roo (Figure 1). The information in this study comes from the databases associated with the timber and non-timber forest management programs, called Unified Timber Forestry Procedures, which were presented, submitted and validated in 2016, by six ejidos to the State Forestry Council of Quintana Roo, for authorization by the Ministry of the Environment and Natural Resources.

Figure 1 Location of the six forest ejidos: 1) Guadalupe Victoria, Bacalar; 2) Otilio Montaño, Bacalar; 3) Francisco J. Mújica, Bacalar; 4) Manuel Ávila Camacho, Bacalar; 5) Río Escondido, Bacalar; 6) Othón P. Blanco, Veracruz. 

Plot selection

Ten 500 m² circular plots were chosen per ejido, in which vegetation samplings were taken with mensuration data of the tree specimens from 7.5 cm in diameter at a height of 1.30 m, for the use of timber and non-timber forest resources, with a 283D/5m Forestry Suppliers measuring tape. The information was confirmed by the State Forestry Council of the State of Quintana Roo, through the Forest Management Technical Committee, in order to issue the technical opinion on the authorization process for timber forest exploitation carried out by Semarnat and Conafor.

Additionally, the extensive documents of the Unified Timber Forestry Procedures were reviewed, and their subsequent verification with the Register and History of Agrarian Nuclei and the Diario Oficial de la Federación (https://phina.ran.gob.mx). In addition, a debugging and updating of the taxonomic identities of the tree species, was carried out through The Plant List (www.theplantlist.org), The Global Biodiversity Information Facility (www.gbif.org) and the Herbarium of the Center of Scientific Research of Yucatán, A. C. (https://www.cicy.mx/sitios/flora%20digital/indice_busqueda.php) databases.

Floristic composition, diversity and relative importance value (VIR)

Species richness was estimated with the taxonomic identity data, while the basimetric area per species was determined with the diameter; the Relative Importance Values (VIR, for its acronym in Spanish) were obtained by the sum of the relative abundance, frequency and basimetric area (AB), for each of the 14 sites analyzed, based on standardized formulas (^{Magurran, 2004}) in spreadsheets.

Statistical analysis

The age of the forests of the six ejidos was considered to be eight, taken from the time elapsed from Dean hurricane in 2007, the closest disturbance, since there are no previous reports of anthropic extraction or disturbances. The data of the analyzed plots of the six forest ejidos were compared to other forests in the south of the Yucatán Peninsula: Laguna Om and Xhazil with 20 and over 50 years, respectively, without use, (^{Macario, 2003}); Calakmul, from 7 to 10 years and more than 20 years without use (^{Aguirre et al., 2018}) and from the Calakmul Biosphere Reserve: km 20 and Ramonal, 10 years without use. The normality test (Shapiro-Wilks) was performed at the abundance values for the 14 sites. A nonparametric test was performed using the Kruskal-Wallis test by intervals, as the normality assumptions were not met. Combinations were compared for the 14 sites using Tukey's paired tests. In addition, a Bray-Curtis similarity analysis was made and the results were presented in a dendrogram for the 14 sites, using the PAST program 3.1 version (^{Hammer, 2001}). Statistical tests were performed with the SigmaStat 4.0 program (^{Systat Software, 2016}).

Floristic composition, richness and diversity of species

For the 60 plots in the six forest ejidos analyzed, the total record amounted for 2 119 individuals, which belong to 93 species, 72 genera and 29 families (Table 1). The families with greatest richness were Fabaceae, Sapotaceae and Euphorbiaceae with 23, 9 and 6 % of species, respectively, and the most abundant were Gymnanthes lucida Sw., Manilkara zapota (L.) P.Royen and Bursera simaruba (L.) Sarg., which make up to 7.7, 6.8 y 4.9 % of the total number of individuals.

From 93 species found in the total number of the analyzed plots of the six ejidos, only six out of 11 are called tropical common (in bold), that is, that they are potential harvesters; they were included between those which had a better representation based upon the Relative Importance Value (VIR): Lysiloma latisiliquum (L.) Benth, Haematoxylon campechianum L. , Gymnanthes lucida, Manilkara zapota, Terminalia buceras (L.) C. Wright, Coccoloba spicata Lundell , Bursera simaruba, Metopium brownei (Jacq.) Urb., Pouteria reticulata (Engl.) Eyma , Vitex gaumeri Greenm. and Brosimum alicastrum Sw. (Figure 2). From half of the total number of individuals, four species were added to the 11 formerly mentioned in order to highlight the importance of a small number of those that are dominant in the six forest ejidos: Piscidia piscipula (L.) Sarg., Caesalpinia gaumeri Greenm., Caesalpinia pulcherrima (L.) Sw and Thouinia paucidentata Radlk. From these 15 species, seven are dominant in Francisco Mújica, Otilio Montaño, Guadalupe Victoria; five in Manuel Ávila Camacho, Río Escondido; three in Veracruz.

MA = Manuel Ávila Camacho; FM = Francisco Mújica; OM = Otilio Montaño; RE = Río Escondido; VE = Veracruz; GV = Guadalupe Victoria. Frecuencia = Frequency; Abundancia = Abundance; Área basal = Basimetric area.

Figure 2 Relative Importance Value (VIR) for the five tree species with greater representation in the six forest ejidos. 

The species with highest accumulated VIR for the six ejidos were Manilkara zapota, Metopium brownei and Bursera simaruba. However, when designing the abundance interval curves for each ejido (^{Ramírez-Marcial et al., 2017}), and with a small number of individuals, the species with greatest relative importance was Gymanthes lucida, which contributed with 11, 20 and 14 % of the individuals ratio in three sites (Figure 3), with similar percentages to Lysiloma latisiliquum, (13 %), Coccoloba spicata (12 %) and Manilkara zapota (17 %). In contrast, from the precios species such as Cedrela odorata, only one individual was recorded, while Swetenia macrophylla had 18 in total. When analyzing and comparing the number of samples, the only one that statistically differed was Veracruz (p < 0.05) in Otilio Montaño, while, based upon VIRs the statistical difference was also significant between Veracruz with Otilio Montaño (p < 0.05) and Guadalupe Victoria, in the same magnitude (p < 0.05).

MA = Manuel Ávila Camacho, FM = Francisco Mújica, OM = Otilio Montaño, RE = Río Escondido, VE = Veracruz, GV = Guadalupe Victoria. Número de individuos = Number of individuals.

Figure 3 Abundance interval curves of the six ejidos, which indicate the species with the greatest abundance for each ejido. 

The number of species in the present study and of eight analyzed sites, was 156, from 21 (Ramonal) to 70 (Xhazil 20 years). These results are within the parameters of wealth and dominance with other forest ejido locations in Quintana Roo, where 168 species have been registered, in an interval of 93 to 117 per site, with a dominance of Pouteria reticulata, Manilkara zapota and Bursera simarouba (^{Negreros et al., 2014}). The sites with the greatest statistical differences were Laguna Om and Xhazil, 20 years old, compared to six and seven, respectively (p <0.05). On the other hand, the percentage of similarity between the 14 sites ranged from 13 % (Veracruz and Ramonal) to 75 % (Otilio Montaño and Francisco Mújica), with a general average between sites of 41 %. The sites with the least similarity were Ramonal (26 %) and Km 20, and those with the greatest similarity were Otilio Montaño and Francisco Mújica (48 %) compared to the other sites.

The similarity tree (Figure 4) shows the formation of two groups at the 0.35 value (that is, 35 % similar) which are associated with their geographic location west of the Yucatán Peninsula (km 20 and Ramonal) with respect to the others. Two groups are formed in the middle of the similarity tree, consistent with the harvesting and incidence of hurricanes: the first, made up of sites near the Calakmul Biosphere Reserve and the other by the six forest ejidos of the state of Quintana Roo.

Km 20 = Km 20; RAM = Ramonal; CKM1 = Calakmul 7-10 years; LO 20 = Laguna Om 20 years; Xh 20 = Xhazil 20 years; LO 50 = Laguna Om 50 years; Xh 50 = Xhazil 50 years; CKM2 = Calakmul 20 years; MA = Manuel Ávila Camacho; FM = Francisco Mújica; OM = Otilio Montaño; RE = Río Escondido; VE = Veracruz; GV = Guadalupe Victoria; Similitud = Similarity

Figure 4 Bray-Curtis similitude dendogram for the 14 evergreen tropical forests analyzed. 

The values of abundance and relative importance among the six ejidos, from the time elapsed since major hurricanes such as Dean (2007) occurred, suggests that, in addition to having been affected by these natural disturbances, they have been subject to anthropogenic influences; they dominate as pioneer species, typical of early successional stages (Lysiloma latisiliquum), in-between such as Coccoloba spicata and Gymnanthes lucida (^{Mize and Negreros, 2007}), to late species (Manilkara zapota), probably associated with the opening of natural gaps or due to selective extraction (^{Dickinson et al., 2000}) and that they coexist with persistent dominant species (Bursera simarouba) and persistent non-dominant species (Coccoloba spicata), which can survive disturbances of different kinds (^{Román-Dañobeytia et al., 2014}).

Most of the taxa used in timber forestry are differentially light-tolerant (^{Vester and Calmé 2003}), so the gap dynamics derived from large-scale events (hurricanes and fires), and not selective extraction, have an effect on other species in sites under logging in Quintana Roo (^{Whigham et al, 1999}). The differential abundance of the species, as an indicator of anthropic disturbance by forest extraction, has favored plant communities dominated by species such as Metopium brownei (Jacq.) Urb or Brosimum alicastrum Sw., due to the greater availability of light and an uneven establishment of the species, as verified in Xhazil, Quintana Roo (^{Macario et al., 1995}). This has been demonstrated by the proportion of the damage, its association with the density of wood (^{Vandecar et al., 2011}) and the history of use of the sites on different tree species, in several ejidos in the center of Quintana Roo, affected by Hurricane Dean in 2007 (^{McGroddy et al., 2013}).

According to the results of the present study, in the Manuel Ávila Camacho ejido, 90 km from the Quintana Roo coast, Lysiloma latisiliquum was the most abundant species with the lowest wood density. In contrast, Gymnanthes lucida, with a higher wood density, was dominant in three of the six ejidos, but is located 130 km inland from the entry site of Hurricane Dean, which affected the six ejidos in which it worked. Therefore, it is likely that the distance between sites associated with the distribution gradient of plant associations in the Yucatán Peninsula (^{Islebe et al., 2015}), the history of exploitation of ejidos (^{McGroddy et al., 2013}), the onslaught of anthropic or natural disturbances (^{de Jong, 2013}; ^{Sánchez-Sánchez et al., 2015}), as well as the intrinsic characteristics of tree species, operate simultaneously (^{Pat-Aké et al., 2018}) on the composition and structure of these jungles. These abundance and dominance of species, not considered profitable for commercial use, are used for firewood, medical applications and construction material, as indicated by ^{Negreros and Mize (1993)}, ^{Negreros and Hall (2000)} in other forests at the center of the state of Quintana Roo.