All vital rates showed positive relationships with the height of individuals and negative relationships with successional age, with the former having larger effects than the latter. This means that in general the environmental modifications in the forest understorey associated to the successional development of forest structure (Lebrija-Trejos et al., 2011) are detrimental to the demographic success of M. acantholoba , although not all vital processes are affected with the same intensity. These results are consistent with previous observations on the performance of this species in the successional stands (Lebrija-Trejos et al., 2008; Romero, 2014).

Among all vital processes, survival presents a more marked change as succession unfolds. In the initial years of the process, survival probability is close to 1 for all individual sizes, but in subsequent years this probability decreases rapidly, especially for smaller individuals. Lebrija-Trejos et al. (2011) described the environmental factors characteristic of the early stages of secondary succession, such as high soil and air temperatures, as well as low soil water availability. M. acantholoba displays different functional attributes which confer it better performance under these conditions, such as high wood density, low wood water content, narrow leaves, small leaflets, etc. (Romero et al., 2020a, 2020b). Within the guild of pioneer species encompassing few species that are capable of growing under these conditions (Lebrija-Trejos et al., 2010), M. acantholoba stands out as the only species commonly attaining large population sizes (Rozendaal et al., 2016), which is the case in our study. However, as these conditions continue to change in response to the structural community development, other species enter the community which appear to be better competitors for light (Lebrija-Trejos et al., 2011; Dalmaso, Marques, Higuchi, Zwiener & Marques, 2020: Saenz-Pedroza et al., 2020); ultimately, this process of community taxonomic enrichment may be the major driver of the decrease in M. acantholoba ’s survival, thus reducing population size. This possibility is supported by the fact that, at advanced successional ages, the survival probability decreased mainly for small-sized individuals, as observed elsewhere (Saenz-Pedroza et al., 2020). This can be related to a decrease in recruitment, since at this stage light availability has largely decreased in the understory (Lebrija-Trejos et al., 2011), causing a reduction in the production of resprouts and the establishment of newly recruited individuals (Maza-Villalobos, Balvanera & Martínez-Ramos, 2011). As succession unfolds, survival decreases even for medium to large sized organisms, supporting the idea that the mechanisms affecting survival are related to the height of individuals, so that competition for light could be a very relevant factor, as it is directly linked to this feature (Matsuo et al., 2021). Future analyses of light gradients through the canopies of this and other tropical dry forests may confirm this possibility.

Many pioneer species have high individual growth rates in the early years of succession (Galia Selaya, Oomen, Netten, Werger &, Anten, 2008). This growth, reflected in basal area increments, may be the main cause of the increase of biomass in pioneer species, even over fertility (Rozendaal and Chazdon, 2015). In some systems the growth rate decreases as succession advances, presumably due to changes in nutrient availability (Berger, Adams, Grimm &, Hildenbrandt, 2006). However, in the case of M. acantholoba , the growth rate remained relatively constant with successional age, and it was higher for small individuals than for large ones (Fig. 3B). This constant growth rate suggests that mechanisms potentially affecting the survival of individual trees, including decreased light availability, do not affect their growth. Regardless of the cause of the lack of variation in growth rates over time, this process does not appear to be responsible for the changes in basal area along succession; rather, an increase in the number of individuals should explain these changes (Muñoz et al., 2021), which is consistent with the increase in individual density also observed. Thus, individual tree growth may be discarded as a major driver of population dynamics of M. acantholoba over succession.

Like other vital processes, fecundity was higher for larger individuals and decreased with successional age. Against initial expectations, seed production had very low values from the beginning and did not show an increase in early successional stages. Within the processes involving fecundity, the probability of establishment of individuals recruited from seed was particularly low, since the proportion of seeds that succeed to germinate and establish is extremely low (< 0.0001). In other pioneer species this early stage of the life cycle has already been identified as the bottleneck of their populations (Álvarez-Bullya & Martínez-Ramos, 1992; Martínez-Ramos et al. 2021), mainly due to decreased light availability in the understory (Maza-Villalobos, Balvarena &, Martínez-Ramos, 2011). Thus, the effect of low fecundity added to a decreased survival gradually causes a reduction in population size, since more individuals die than establish. Other studies have found that resprouting in pioneer species may be a more important component of population maintenance than fecundity (Dietze and Clark, 2008). Therefore, due to its low values, we conclude that fecundity can hardly drive the population growth or biomass increase that occurs in M . acantholoba populations during the early phases of succession.