Population dynamics of seed and seedlings of Albizia procera (Roxb.) in Mizoram, India

Albizia procera (Roxb.) (Family Fabaceae) is a fast growing, medium sized tree with an open canopy that can reach up to a height of 30 m^1,2. It is native to India, America, Pakistan and Australia³ and found in tropical countries such as Myanmar, Thailand, Bangladesh, Malaysia, Laos, Cambodia, Vietnam⁴, Nepal, Indonesia, China, Andaman, Kenya, South Africa and Uganda^1,5. Due to its varied adaptability and economic importance, A. procera is being planted in agroforestry systems including tea gardens, reforestation, afforestation and social forestry programmes for restoration of degraded lands^6,7. Besides, the species is reported to have high medicinal properties^8,9. Additionally, it is extensively used for other purposes, such as its leaves are consumed as a vegetable² and its bark used as a fish poison^5,10. These extensive uses can impact the species status and reduce its distribution in its native area.

Seed dynamics and seedling establishment are critical life phases of a tree species^11,12. Understanding these processes is crucial to recognize species recruitment and directional change in ecosystem restoration¹¹. In contrast, seed production is considered a critical bottleneck of tree life cycle¹³ and its variation may affect regeneration and numerous biological phenomena including interaction between plants and animals, vegetation dynamics and nutrient cycle¹⁴. It may be limited by various external factors such as adverse climate, pollination failure, predation of flowers, fruits and leaves, wind speed etc¹⁵, and internal factors such as genetic condition, age and size of the tree^16,17. These factors could lead to seed production reduction and significantly impact seed dynamics and seedling recruitment of a given species.

Seed dispersal is another important link in the reproductive cycle of a tree from the end of the adult plant to the beginning of the new plant. The distance to which the seeds get dispersed has important ecological significance. For example, seed dispersal can avoid high density-dependent mortality in close proximity to the parent plants¹⁸, while it can facilitate maintaining species diversity and may induce spatial accumulation of seeds and seedlings in pioneer trees¹⁹. Soil seed banks, on the other hand, are influenced by seed production and dispersal and various governing factors that affect seed germination^12,20. The fate of the seeds after dispersal from the parent plants can be influenced by several microclimatic and biotic factors²¹ and in turn primarily determine the future population structure. It is therefore of paramount importance to have a complete understanding of the population dynamics of a species from seed production to seedling recruitment. However, there are very limited studies in trees that deal with population dynamics of seeds and seedlings.

A review of literature reveals that studies on seeds dynamics of A. procera is lacking or very limited. A. procera is a pioneer fast growing legume tree having hard seed coat and this property of seeds results in reduced germination in wild^22,23 which may cause seedling population mosaic or heterogeneity in natural conditions^24,25. The results have shown that seed pre-treatments of A. procera^6,23,26, can bring seedling population homogeneity in nursery and plantations⁷. We hypothesized that seed production of A. procera is influenced by both intrinsic traits (e.g. tree size, and crown structure) and extrinsic environmental factors, and seedling success is modulated by microclimatic variation. Accordingly, the study aimed to quantify interannual variation in seed production and its relationship with tree characteristics, assess seed dispersal patterns and post-dispersal fate (germination, disappearance, and seed bank formation), and evaluate seedling growth and survival in relation to soil temperature, moisture, and humidity. To achieve these objectives, we conducted field experiments for A. procera species to address the following questions: (a) is there variation in seed production between years and among individual trees?, (b) how do the tree traits affect its seed production?, (c) do the seed dispersal distance influence seed germination?, and (d) which microclimatic variables affect the seedling growth the most?. The findings provide essential insights into population dynamics of the A. procera and can inform strategies for monitoring growth and restoring degraded forests.

Study area

The study was carried out at Mizoram University (MZU) campus (23°39’52”-23°48’43″N and 92°39’49”-. 92°46’39″E) (Fig. 1), located 15 km away from Aizawl, the capital city of Mizoram having elevations ranging from 300 m to 880 m asl. MZU campus encompasses roughly 980 acres of land, and harbors tropical wet evergreen forests, small biodiversity park and protected forested water catchment reserve in the north. Several streams flow through the campus^27,28,29. The forest included diversified plants species, where Wapongnungsang et al.²⁹ reported 384 plants belonging to 290 genera and 107 families and many grass species²⁷. The area receives an average annual rainfall of approximately 1,850 mm, influenced heavily by the southwest monsoon, with most precipitation occurring between May and September. The mean annual temperature is about 21.6 °C, with summer highs reaching 30 °C and winter lows ranging between 10 and 12 °C (Meteorological data of Mizoram, Aizawl, 2023).

Data collection

Seed production

Seed production of A. procera was estimated from randomly selected 15 individual trees between 2022 and 2024. However, for each tree, the number of branches (B), secondary branches (sub-branches (SB)), tertiary branches (sub-sub-branches (SSB)), and inflorescences (INF) per sub-sub-branches were recorded to calculate total seed produced per plant following formula (1). The seed were estimated during November and December of each year before seed maturation and fall¹⁶. Additionally, in each tree, the mean number of inflorescences per SSB was determined from random 5-SSB, while the average number of fruits per inflorescences was calculated from a random selection of 10 inflorescences. Conversely, the mean number of seeds per fruit were estimated from fifteen fruits (Fig. 2A) and the number of seeds per kilogram was estimated using the average weight of 25 seeds (Fig. 2B) following the formula (2).

Seed dispersal

To study the seed dispersal, five individual fruiting trees were marked in the forest stand following the method suggested by Khan et al.³⁰. The five trees selected for dispersal studies were a subset of initial 15 trees, chosen based on their health, canopy structure, accessibility, and adequate spacing (> 100 m) to avoid seed overlap. This smaller number was selected to allow intensive monitoring of seed-fall and dispersal pattern around each individual tree. Each selected individual tree base was considered as a center, of which concentric circles of 2.5 m circular increments were marked on the ground around the mother tree, extending outside the crown radius as deliberated by Sahoo and Lalfakawma¹⁷. The first circle had a radius of a minimum five meters and maximum was 25 m. In the meantime, the seeds that fell under the tree crown were not considered as dispersed seeds. However, each selected individual tree was visited at three-days interval over 8-weeks during seed-fall season (Fig. 2C). All seeds collected within each marked circle were counted separately and the seeds damaged during dispersal were excluded from the analysis.

Fate of seed populations in the soil

The fate of seed population in the soil was assessed, where the number of seeds disappearing (fraction) during the seed fall period was studied. Five seed traps sizing (1 m x 1 m with 30 cm depth) were placed randomly on the ground under the mother tree crown at the beginning of seeds fall (Fig. 3A). The seed traps were visited every five-days until completed seed shedding following the method described by Sahoo and Lalfakawma¹⁷ and de Sá Dechoum, et al.³¹. Simultaneously, during each visit, all seeds in the traps were counted, and seeds damaged (due to insects and rodents) were separated from undamaged one. However, the difference between total produced seed and undamaged seeds that fell under the individual mother tree was estimated as the fraction of seeds loss during the seed fall period¹⁷.

To estimate the germination of healthy seeds after fall and/or dispersal, the fate of undamaged seeds after dispersal was studied by sowing 20 seeds in plot sizing (1 m x 1 m) under the five sampled individual trees (Fig. 3B). Additionally, seed fate was assessed in relation to distance from the mother tree. At each distance e.g. (5, 15, 25, and 35 from the mother tree), plots sizing (1 m x 1 m each) were established (n = 3 replicates per distance). Twenty seeds were placed per plot (total 60 seeds per distance), and seed fate was recorded weekly for three months. The number of seeds that germinated, disappeared (e.g. translocated or consumed by dispersal agents) or rotted was recorded, and germination was defined as emergence of radicle visible through the seed coat.

Soil seed bank

The study on soil seed soil bank was conducted at the end of rainy season following the method described by Souza et al.³². Four sites were selected using stratified sampling²⁰, and from each site, five soil samples from area of (5 × 5 cm and 5 cm depth) were collected and bulked to estimate buried seed density. From each of the bulked sample, 100 g of soil was weighed and washed gently using a jet of water (Fig. 2D) following the procedure and method described by Padonou et al.²⁰ to recover the seeds. The number of seeds found in the sample was then extrapolated to 1 m x 1 m area.

Seedling growth performance in natural conditions and its relation to microclimate variables

To evaluate seedling growth performance and its relation to microclimatic variables, 10 quadrats (1 m ×1 m) were laid randomly in the forest near A. procera species stand. All seedlings recruited from the dispersed seeds within quadrates were monitored for 14 months. The seedlings height and stem collar diameter (SCD) were measured at two-weeks interval; seedlings height was measured using 30 cm ruler, while SCD measured using digital Vernier caliper (150 mm). Monthly seedling growth increments were calculated along with relative growth. Seedling mortality and microclimatic variables such as soil temperature, soil moisture, and humidity were recorded monthly throughout the study period.

In contrast, microclimate data were collected in each study plot, where soil temperature was measured using a stainless-steel dial thermometer (Model ST-9283B, 0.1 °C accuracy), soil moisture using digital soil moisture meter (Kelway HB-2), while relative humidity using portable digital hygrometer (HTC-1). One set of sensors was installed per plot (10 total). Sensors were positioned in the center of the (1 m x 1 m) seedling quadrats at 5 cm soil depth. Moreover, measurements were taken three days in a week during morning and afternoon, and averaged to monthly means for correlation with seedling growth increments.

Population flux of A. procera

The life cycle of A. procera species is computed from the average of seed production, seed dispersal and fate of seeds during seed-fall and post-seed fall, and soil seed bank over three years (2022–2024), as well as seedling growth, survival, and mortality for the period of 14 months. The mean seed production from 15 trees was estimated, along with seed dispersal and seed damage. Consequently, the percentage of seed disappeared was calculated from the fate of undamaged seed at varying distances from the mother trees, which also included seed germinated and contribution to the soil seed bank. Moreover, seedling survival and mortality were monitored over 14 months to determine overall seedling survival rates.

Data analysis

All field data collected on seed production, dispersal, fate of seed population, and seedling growth were compiled and organized into meaningful table for analysis. Data were summarized and expressed as mean and standard deviation (SD). The mean seed production for each year was computed using formula (1), and seed germinated and disappeared was calculated using formulas (3 and 4 respectively). While the spatial distribution of seeds (under the mother tree and at varying distances) was first determined by calculating the percentage of seeds in each location, this percentage was then used to allocate the annual mean seed production across the same spatial categories, yielding the estimated number of seeds at each distance as studied by researcher³³. The relationship between seed production and tree characteristics such as DBH, number of branches, number of inflorescences, crown height, crown diameter was examined. In contrast, post-hoc Fisher LSD, Tukey, and One-way ANOVA test (Sig. 0.05) were used to examine the variation of seed production between years and among individual trees. Conversely, seedlings’ growth (height and SCD) were converted to monthly basis, and growth increment was calculated. Seedling growth rates were correlated with microclimate variables such as soil moisture, temperature, and humidity. In the meantime, post-hoc Tukey (Sig. 0.05) was used to examine the variation in monthly seedling growth increments during study period. One-way ANOVA (Sig. 0.05) was used to assess the variation in rainfall, temperature, and humidity between the study years. The seed population flux was computed as accumulative mean of seed production, dispersal, germination, disappearance, soil seed bank, and survival rates. All data analysis were performed using SPSS, Version 22.0, Jamovi (Version 2.5.6)³⁴, OriginPro 2025 and Microsoft Excel (Version 365).

Formulas and equations

All abbreviations in formulas are defined as full form in (Table 1).

Seed production, seed dispersal, and soil seed bank

The mean seed production of A. procera significantly differed among individual trees (F = 4.63, P < 0.0001) (Fig. 4) and between years (F = 12.09, P < 0.0001) (Fig. 5). The mean seed production per individual tree was 145,352 seed/tree in 2022, dropping sharply to 43,607 in 2023, and 41,490 in 2024 (Table 2). The variation in seed production across the years can be driven from variation in number of inflorescences (F = 12.16, P < 0.0001) and number of fruits per inflorescences (F = 8.92, P < 0.000) (Table 2). Additionally, these declines correspond with notable interannual (Table 2) and monthly (Fig. 6) variation in rainfall, mean temperature, and humidity. In contrast, the result suggests that the pronounced decline in seed output was likely driven by climatic stress during the flowering and fruit set. The mean seed weight was 0.039 g resulting estimated 25,654 seed/kg (Table 2). A strong correlation was observed between mean seed production/tree and tree traits such as DBH, number of branches, inflorescences, fruits, crown height, and crown diameter (Table 3). The seed soil bank was relatively low across the years, without showing discernable variation among sample plots and years (Table 2).

The number of seeds per unit area decreased as the dispersal distance from the mother tree increased showing a negative relation (reverse J-shaped curve) between seed density and dispersal distance across the years (Fig. 7). The distance to which the seeds got dispersed and the dispersed seed density varied significantly between years (F = 13.46, P < 0.0001, Table 4). A significant proportion of the total seeds produced (55.38% in 2022, 54.25% in 2023 and 52.92 in 2024) fell directly under the mother tree (Fig. 8) and the dispersed seed density gradually decreased with increased dispersal distance and this was true for all the years (Fig. 8). Further, the maximum distance to which the seeds were found dispersed was 22.50 m from the mother trees (Fig. 8).

Fate of seed populations in the soil

During the seed-fall period, more than 97% of the seeds that fell under the mother tree was found undamaged and viable when observed year-wise variation (Table 5). However, during post-seed-fall, the fraction of seed disappeared was 56.6% in 2022 while it decreased to 48% in 2023, and 49.80% in 2024. Similarly, the seed germination rates of the sown seeds were quite moderate (39.00%, 47.90% and 45.40% in 2022, 2023, and 2024 respectively) and seeds those got rotten were minimal (Table 5).

Seed disappearance and germination in relation to distance from the mother tree

The seed disappearance due to predation was significantly (F = 9.61, P < 0.001) higher around the mother trees and it decreased gradually as the dispersal distance increased from the mother trees, a trend commonly observed in all the studied years. Our results showed that a very high proportion (82.33%) of the sown seeds disappeared within a 5 m circle while the disappeared seeds drastically reduced (67.33%) at a distance of 35 m from the mother tree (Fig. 9).

Seedlings dynamics in the forest and its relation to microclimatic variables

During 14-months period, the seedlings of A. procera attained an average height and SCD of 22.50 cm and 3.25 mm respectively (Fig. 10). The species showed significant change in its growth increment with time (F = 3.29, P < 0.001 for seedling height, and F = 8.36, P < 0.001 for seedling collar diameter) (Fig. 11). The soil temperature and soil moisture showed seasonal variation; the highest soil temperature was recorded in June (26.86 °C), and soil moisture in July (70.30%) (Fig. 12). Among the microclimatic variables, soil moisture showed the strongest positive correlation with height growth and collar diameter (Table 6). The other parameters (relative humidity and soil temperature) too showed positive relationship with the seedling growth increment (height and collar diameter), however, soil pH was negatively related to these seedling attributes (Table 6).

Population flux of A. procera

The population flux of A. procera integrates seed production, seed dispersal, seed soil bank, and seedlings recruitment over three years (2022–2024). Based on 15-sampled trees, the mean seed production of A. procera over three years was 76,816 seeds/tree. Of these 97.63% seeds fell beneath the mother tree and dispersed at varying distances. However, during post-seed-fall, we found that 73.44%, 25.50% and 1.06% of the felled seeds got disappeared, germinated and stored in soil bank respectively. Among the germinated seeds, only 46.07% of seeds developed into seedlings that survived till the end of the experiment (Fig. 13). In the meantime, the seedling continues with vigorous growth (height and diameter) during the rainy season while dropping the leaves and stop shoot growth during dry season (Fig. 14).

Seed production

Seed production of A. procera showed marked interannual variation, with a pronounced peak in 2022 followed by significant declines in 2023–2024. Similar variation in seed production across years have been reported in other trees species^{11,14,16,17,35,36,37}. However, such variability is typical of tropical legumes and can be linked to fluctuations in rainfall and temperature that affect flowering and fruiting set. In contrast, the seed production variation was obviously related to the fruit loading of the species and the number of fruits bearing species in a given year. Several authors have suggested that seed production in a species will be influenced by several intrinsic and extrinsic factors during the flowering and seed setting period. For example, Iralu et al.¹² reported that the annual rainfall acts as a limited factor for seed production and seedling survival while^16,17,38 explained that seed production of tree species can be influenced by a variety of factors such as availability of resources, pollination failure, predation on flower, fruits, climatic condition, plant age and size. We observed a strong correlation between mean seed production and tree characteristics such as DBH, number of branches, number of inflorescences, and crown cover. On the other hand, larger trees with wider canopies tended to produce more seeds; however, this relationship must be interpreted cautiously because estimated production was driven from tree traits. Direct seed count in a validation subset are recommended to refine predictive models, and similarly used by several researchers^{11,13,14,16,36}. In the meantime, Khan et al.³⁰ observed that dominant trees species with large crowns, which receive a lot of light, tend to produce an optimum number of seeds. Conversely, the higher seed production occurring in warmer year, meanwhile increased temperature negatively affects seedling establishment³⁹.

Seed dispersal

Seed dispersal is the movement of seed from the mother tree by help of different dispersal agents. For successful tree regeneration, it is important that the seeds should disperse to a safe location where they can germinate, survive and translate into mature plants. Thus, it determines the seeds distribution and trees during natural regeneration process which can be influenced by several factors (e.g. biological and environmental). In A. procera, seed dispersal occurred through a combination of wind dispersal and gravity. We found seed dispersal declined exponentially with distance, confirming that most seeds remain beneath or near the parent crown. A significant variation was observed in seed dispersal across the study years, this variation is mainly driven by seed production. We found the maximum seed dispersed up to 22.50 m from the mother tree, which indicated that the species dispersed its seed via explosion/gravity, while a small fraction of seeds was transported to distance places by wind. The limited seed dispersal pattern may increase density-dependent mortality and seed predation under the mother trees, a similar pattern reported for other Fabaceae species^40,41. These findings suggest that the seeds with capsules tend to be exposed to secondary dispersal where wind moves it to another location. Seed dispersal, however, was highly restricted in this species, and for successful regeneration, the seeds need to be transported to far off places. The restricted seed dispersal by gravity nevertheless possesses some challenges for this species as it leads to overcrowding and competition for resources among the closely spaced individuals upon germination. To avoid for this challenge, the seeds of this species have wings/appendages which help them to be carried out and disperse to new locations. Several studies observed that seeds from same mother tree with varying seed mass are influenced by dispersal distance and differ significantly among tree species sharing the same dispersal mode^41,42,43,44. Similar findings of decline the density of seed with increased dispersal distance from the mother trees have been observed for other trees species^17,45. In contrast, Nathan et al.⁴⁶ argued that long-dispersal distance is more common in open terrestrial landscapes and driven by migratory animals and wind. While Chen et al.³⁵ and Kasi and Ramasubbu⁴⁷ indicated that tree species that dispersed their seed by gravity are aggregated around the parent tree.

Soil seed bank

Despite high seed production, only a small proportion (≈ 1%) contributed to the persistent soil seed bank. Most seeds either germinated shortly after dispersal or disappeared due to predation and decay. Conversely, soil seed-bank contribution suggests that A. procera relies primarily on current-year’s recruitment rather than long-term soil storage. However, soil seed bank nevertheless plays a crucial role in regeneration of tree species, and its size is determined by seed dispersal and seed characteristics, and its ability to remain viable during the unfavourable conditions. Berihun et al.⁴⁸ reported that seed bank can serve as a “memory” of past plant communities by containing seed from previous years and enhancing future plant communities. Meanwhile, the prevailing microclimate such as low temperature and moisture can have a bearing on soil seed bank by limiting germination⁴⁹. The relatively low rates of seed germination under natural conditions in this species reveal that a larger fraction of the seeds remains viable-dormant in soil for a longer period, contributing to the resilience and persistent in its natural habitats. The size of the soil seed bank of a species in a given time is related to seed inputs (through seed rain) and seed outputs (through germination, predation and other losses) which are influenced by several factors including environmental and anthropogenics^16,17,33,37. Though predation is reported to reduce the seed soil bank in forest floor and may affect survival and mortality^12,50, the small viable seed bank can ensure the species’ survival and persistent in the nature.

Seed disappearance and germination in relation to distance from the mother tree

Seed disappearance in the present study was closely related to distance from the mother tree. The seeds that escaped or dispersed far off from the mother trees were found to be less predated than those which fell near the mother trees. Higher rate of predation and low survival near the mother tree were obviously due to density-dependent competition and predator preference, as also has been reported by several other workers^16,17,33. Additionally, our results revealed that with increased distance from the mother tree the germination increased while seed disappearance decreased, in conformity with Souza et al.⁵¹. We observed that low seed germination compared to the fractions of the seeds that disappeared or got predated after seed fall in their natural habitats at varying distances from the mother trees. Majority of seeds that still remained in capsule could not transform into successful seedlings due to unfavourable environmental conditions (Fig. 15), which significantly reduce the seed germination and seedling recruitment in the forest. In contract, it reported that the ability of plant propagules to reach microhabitats with the adequate conditions for seed germination and establishment of sapling will have direct effect on the plant’s fitness⁵². The results clearly demonstrate that the seeds will have a better chance of survival if they are dispersed far away from the parent plants and get favorable germination conditions.

Seedling dynamics and microclimate effects

Seedling growth and survival are often influenced by a combination of environmental factors such as soil temperature, moisture, light availability, and relative humidity. In the present study, A. procera seedlings reached an average height of 22.5 cm and SCD of 3.25 mm over study period and survival after germination was below 50% indicating the sensitivity of the seedling during early growth to various environmental stresses. This find support as the seedling growth of the species was positively correlated with soil moisture, soil temperature, and relative humidity, indicating that water availability is a key driver of recruitment success. Seasonal declines in soil moisture during the dry period sharply reduced the seedling survival, underscoring the vulnerability of young seedlings to drought. Similar findings are reported by Musa and Sahoo²⁶, who reported that moisture availability and temperature significantly affect seedling performance of tropical species while Bebre et al.⁵³ stated that multiple environmental factors in forests influence the seedling growth such as light, temperature, soil moisture, litter depth, intra and interspecific competition for various resources. Consequently, Greenwood et al.⁵⁴ and Wieser et al.⁵⁵ observed that higher soil temperature and moderate temperatures can improve physiological processes such as photosynthesis and nutrient uptake, leading to improved seedling establishment and growth. We found that seedling mortality during early growth was higher compared to their survival, a result that find support from the studies reported by Johnson et al.⁵⁶. These findings further underscore the importance of the favorable microclimates during this critical stage of seedling development and can drastically affect survival rate among species⁵⁷. In the meantime, the presence of canopy gaps, for instance, has been shown to provide improved conditions for seedling recruitment and survival due to increased light availability and moderated competition, and similar findings have been observed by several researchers^58,59. While Awal⁶⁰ and Kharuk et al.⁶¹ reported that soil temperature not only affects plant growth directly but also regulates microbial activity and nutrient cycling, which indirectly supports seedling vigor. Our study results confirm that microclimatic conditions particularly the soil temperature and moisture at the forest floor significantly influenced the seedling dynamics of A. procera.

Study limitations and future directions

This study excludes determining the viable-dormant seed fraction and association of various types of seed dormancy in the soil seed bank which would have provided better clues in understanding the role of the small-sized soil seed bank in regeneration of this fast-growing species in natural habitats. The reportedly under-estimation of dispersed seed density might have occurred due to limited seed trap coverage and secondary removal by predators. Incorporating camera monitoring or automatic traps could have enhanced better accuracy in estimation of various fractions of seeds during post seed-fall. Extending this study across climatic gradients and disturbance regimes would provide a more comprehensive understanding of A. procera regeneration ecology.

Seed production and limited dispersal significantly constrain the natural regeneration of A. procera. Although the species produces abundant seeds, most fell beneath the mother tree canopies, where predation and low moisture reduced germination and survival. The low persistent soil seed bank ensure the regeneration of this species in its natural landscape, however, high mortality of seedling during early growth stage highlights a strong bottleneck in its seedling survival. To enhance the regeneration and restoration success, management should focus on assisted seed dispersal, moisture retention, and partial shade maintenance to improve seedling establishment.