Interactive effects of fallow age and Chromolaena odorata invasion on tree regeneration in Central African agricultural landscapes

Interactive effects of fallow age and Chromolaena odorata invasion on tree regeneration in Central African agricultural landscapes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Interactive effects of fallow age and Chromolaena odorata invasion on tree regeneration in Central African agricultural landscapes Jean Pierre Azenge, Justin N'Dja Kassi, Paxie W. Chirwa This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9203304/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Natural regeneration of trees outside forests (TOF) is a key process sustaining biodiversity and ecosystem functions in tropical agricultural landscapes, yet it is increasingly shaped by interactions between succession and biological invasion. This study quantified the combined effects of fallow age and Chromolaena odorata cover on tree regeneration in agricultural fallows of Mongala Province (Democratic Republic of the Congo). Seedling inventories were conducted around 300 seed trees across gradients of fallow age (0–3, 3–6, > 6 years) and invasion intensity (0–25% to > 75% cover). Seedling density, spatial patterns, and size structure were analysed using zero-inflated negative binomial models. Seedling density declined significantly with increasing C. odorata cover, particularly in early fallows, where predicted densities decreased from 1051 to 270 ind. ha⁻¹ (≈ 74% reduction). This effect weakened along the successional gradient and became negligible in older fallows, indicating a strong interaction between succession and invasion. Invasion effects were non-linear, with sharp declines occurring beyond intermediate levels of cover, suggesting threshold responses and increased probability of recruitment failure. Species responses varied markedly: Erythrophleum suaveolens and Pycnanthus angolensis showed strong declines, whereas Petersianthus macrocarpus increased under high invasion, indicating species-specific ecological filtering. Invasion also reduced spatial heterogeneity and limited progression to larger seedling size classes. These findings highlight that regeneration is governed by context-dependent interactions between succession and invasion intensity, with early fallows representing both a window of opportunity and a phase of high vulnerability. Accounting for non-linear invasion effects is critical for understanding forest recovery and designing targeted management strategies in tropical agricultural landscapes. Tree regeneration Chromolaena odorata Fallow age Invasion intensity Non-linear effects Ecological filtering Tropical agricultural landscapes Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Across tropical Africa, agricultural expansion and shifting cultivation have profoundly transformed forest landscapes, resulting in extensive mosaics of croplands, fallows, and remnant tree cover (Curtis et al. 2018 ; Mpanda et al. 2021 ). In these human-modified systems, trees outside forests (TOF) play a crucial ecological and socio-economic role by maintaining biodiversity, supporting ecosystem services, and contributing to forest recovery processes (FAO 2016 ; Zomer et al. 2016 ). In particular, natural regeneration within agricultural fallows represents a major pathway for the persistence and renewal of woody species in landscapes where access to intact forests is increasingly limited (Curtis et al. 2018 ; Loubota Panzou et al. 2024 ). Natural regeneration in fallow systems is shaped by multiple interacting drivers, including land-use history, successional stage, seed availability, and biotic interactions (Chazdon et al. 2009 ; Mwampamba and Schwartz 2011 ). Among these factors, fallow age is widely recognised as a key determinant of regeneration dynamics, as it influences light availability, soil conditions, and vegetation structure (Chazdon 2014 ; Poorter et al. 2021 ). Early successional stages may favour seedling establishment due to higher light availability, whereas later stages may enhance survival through improved microclimatic buffering and soil development. However, the relationship between fallow age and regeneration is often non-linear and species-dependent, reflecting differences in ecological strategies among tree species (Poorter et al. 2019 ; Rozendaal et al. 2019 ). In tropical agricultural landscapes, these successional processes are increasingly altered by invasive plant species (Pyšek et al. 2020 ). Among them, Chromolaena odorata (L.) R.M. King & H. Rob. is one of the most aggressive invasive shrubs in tropical Africa and is now widespread in post-cultivation fallows (Rai and Singh 2024 ). Its rapid growth, dense canopy, and potential allelopathic effects can reduce light availability, limit seedling establishment, and modify competitive interactions (Kato-Noguchi and Kato 2023 ; Juru et al. 2024 ). While C. odorata is often associated with reduced tree regeneration, its effects are not uniformly negative. In some contexts, dense shrub cover may alter competitive dynamics or microenvironmental conditions in ways that influence seedling recruitment differently across species and successional stages (Gbètoho et al. 2018 ; Hardy et al. 2025 ). Despite increasing attention to the role of TOF in landscape resilience, the combined effects of fallow age and invasive shrub cover on tree regeneration remain insufficiently understood, particularly in Central African agricultural systems (Reed et al. 2016 ; Chazdon et al. 2020 ). Most studies have examined these drivers independently, overlooking potential interactions between successional stage and invasion intensity. In addition, species-specific responses to these interacting factors remain poorly documented, limiting our ability to predict regeneration trajectories and design context-adapted management strategies. In the Democratic Republic of Congo (DRC), and particularly in Mongala Province, shifting cultivation dominates rural land use and generates extensive networks of agricultural fallows where TOF constitute the main source of natural forest regeneration (FAO 2016 ; Loubota Panzou et al. 2024 ). These landscapes provide an appropriate context for analysing how regeneration patterns vary across successional stages and invasion gradients. In such systems, understanding both the magnitude and variability of regeneration responses is essential for informing sustainable land management and restoration-oriented practices. Against this background, this study investigates the natural regeneration of five dominant TOF species ( Petersianthus macrocarpus (P. Beauv.) Liben, Pycnanthus angolensis (Welw.) Warb., Ricinodendron heudelotii (Baill.) Heckel, Erythrophleum suaveolens (Guill. & Perr.) Brenan, and Piptadeniastrum africanum (Hook.f.) Brenan) across agricultural fallows differing in age and C. odorata cover in Mongala Province, Democratic Republic of the Congo. Specifically, we aimed to: (i) quantify the effects of fallow age and invasion level on seedling density; (ii) assess whether the effect of C. odorata varies across successional stages; (iii) analyse species-specific responses to these interacting drivers; and (iv) evaluate how these factors influence the spatial and structural patterns of regeneration. We hypothesised that: (1) seedling density decreases with increasing C. odorata cover; (2) this negative effect is strongest in early fallows and weakens with increasing fallow age; (3) species exhibit contrasting responses to invasion and succession; and (4) both spatial patterns of recruitment and seedling size structure are jointly influenced by distance from seed trees, fallow age, and invasion level. By explicitly testing these hypotheses, this study provides a quantitative assessment of how successional dynamics and biological invasion interact to shape tree regeneration in fallow-based tropical landscapes, and contributes to a better understanding of forest recovery processes outside forests. 2. Materials and methods 2.1. Study area The study was conducted in Mongala Province, located in the north-western part of the Democratic Republic of the Congo within the central Congo Basin (Fig. 1 ). The province is characterised by lowland tropical landscapes dominated by moist evergreen and semi-evergreen forests interspersed with agricultural fields and fallows resulting from shifting cultivation. The climate is humid tropical, with relatively stable high temperatures throughout the year and mean annual rainfall generally exceeding 1,500 mm (Azenge et al. 2025 ). Rainfall is distributed across a long rainy season and a shorter dry season, creating favourable conditions for continuous plant growth and rapid vegetation recovery following agricultural abandonment. Vegetation is largely composed of dense tropical rainforest formations, although extensive areas have been converted into agricultural mosaics. Shifting cultivation is the dominant land-use system, with fields typically cultivated for one to two years before being abandoned to fallow. Fallow duration varies widely, resulting in a heterogeneous patchwork of fallows of different ages that constitute key sites for the regeneration of trees outside forests. Soils are predominantly highly weathered tropical soils with low to moderate fertility. Repeated cultivation can further reduce nutrient availability, making fallow periods essential for restoring soil structure and fertility. However, vegetation recovery during fallow stages is often constrained by competition with herbaceous and shrub species, particularly invasive plants (Diyarzola and Bernard 2024 ). C. odorata is among the most widespread invasive species in agricultural fallows of Mongala Province, where it forms dense shrub layers in recently abandoned fields. 2.2. Data collection Natural regeneration of TOF was assessed through a field-based seedling inventory conducted in agricultural fallows across Mongala Province, (DRC). The study focused on five dominant TOF species ( P. macrocarpus , P. angolensis , R. heudelotii , E. suaveolens and P. africanum ) and aimed to quantify how regeneration varies across gradients of fallow age, C. odorata cover, and distance from seed trees. The sampling design was structured to capture variability in both successional stage and invasion intensity while ensuring spatial representativeness. Fifteen villages were selected across the three administrative territories of Mongala Province to reflect the range of agroecological and land-use conditions in the study area. Within each village, agricultural fallows were identified in collaboration with local farmers to determine fallow age and recent land-use history. Fallow age was classified into three categories (0–3 years, 3–6 years and > 6 years), representing early, intermediate and advanced stages of post-cultivation succession. The cover of C. odorata was visually estimated around each seed tree and assigned to four ordinal classes (C1: 0–25%, C2: 25–50%, C3: 50–75% and C4: >75%), capturing increasing levels of invasion. For each of the five species, 60 mature seed trees were selected as focal sampling units, resulting in a total of 300 seed trees. Seed trees were distributed across fallow age classes and invasion levels to ensure that all combinations of successional stage and C. odorata cover were represented at the landscape scale. While minor imbalances occurred at the village level due to local availability constraints, the overall design remained well balanced across treatments. Around each seed tree, seedling abundance was assessed using four rectangular transects oriented along the cardinal directions. Each transect was 40 m long and 3 m wide, corresponding to the average radius of cultivated fields, and was subdivided into eight segments of 5 m. This design allowed the spatial distribution of seedlings to be captured along a gradient of distance from the seed tree while standardising sampling effort across sites. All seedlings encountered within transects were counted and assigned to one of three height classes (0–30 cm, 30–50 cm and > 50 cm), providing information on both recruitment and size structure. Seedling counts were subsequently aggregated at the seed-tree level prior to statistical analysis, so that each seed tree represented an independent sampling unit associated with a specific combination of fallow age, invasion level and local environmental conditions. This aggregation reduced the risk of pseudoreplication associated with multiple transects and ensured consistency with the analytical framework used in the study. 2.3. Data processing and analysis 2.3.1. Data preparation All statistical analyses were conducted using the R statistical environment (R version 4.5.3; R Core Team, 2025). The dataset comprised seedling counts recorded around individual seed trees across combinations of fallow age, C. odorata cover, species identity, and distance from the seed tree. Seedling density was expressed as the number of individuals per hectare (ind. ha⁻¹). Total seedling density was calculated as the sum of individuals across all height classes. Additional variables described seedling structure by three height classes (0–30 cm, 30–50 cm, and > 50 cm). Explanatory variables were defined as follows: Fallow age : three ordered classes (0–3 years, 3–6 years, > 6 years); C. odorata cover : four ordered levels (C1: 0–25%, C2: 25–50%, C3: 50–75%, C4: >75%); Species : five focal tree species ( E. suaveolens , P. macrocarpus , R. heudelotii , P. angolensis , and P. africanum ). Distance from the seed tree was treated as a continuous variable. Prior to analysis, data were checked for consistency, missing values, and outliers. Distributions of response variables were examined using histograms and boxplots (see Figure S1 ). Seedling density exhibited strong right-skewness, high variance, and a large proportion of zero counts, indicating overdispersion and zero inflation typical of ecological count data. 2.3.2. Modelling of seedling density To analyse the effects of fallow age, C. odorata cover, species identity, and distance on seedling density, we fitted zero-inflated negative binomial (ZINB) models using the glmmTMB package. This modelling framework was selected to simultaneously account for: (i) overdispersion in count data, and (ii) excess zeros arising from ecological processes such as recruitment failure. The final model structure was: \( \text{Seedling density}\sim \text{species}\times \text{fallow age}\times \text{cover}+\text{ns(distance, df = 3)}\) with a zero-inflation component specified as: \( \text{Zero inflation}\sim \text{fallow age}+\text{cover}\) Distance from the seed tree was modelled using a natural spline (df = 3) to capture non-linear spatial patterns of seed dispersal and establishment. Model coefficients were interpreted on the log scale and exponentiated to obtain rate ratios. Estimated marginal means (EMMs) and associated confidence intervals were computed using the emmeans package and used to generate predicted response curves and interaction plots. Pairwise comparisons among C. odorata cover levels within each fallow age class and species were performed using Tukey-adjusted contrasts to control for multiple testing. Results were presented as: a simplified summary table in the main text (Table 2 ), and a full set of contrasts in the Supplementary Material (Table S2). 2.3.3. Spatial patterns of seedling recruitment Spatial variation in seedling density relative to distance from the seed tree was analysed within the ZINB modelling framework described above. The inclusion of a spline function of distance allowed flexible modelling of non-linear dispersal and establishment patterns. Predicted seedling densities were generated across continuous gradients of distance and invasion levels for each fallow age class and species. These predictions were visualised using heatmaps, facilitating the interpretation of interactions between spatial processes, successional stage, and invasion intensity. 2.3.4. Analysis of seedling size structure To characterise regeneration dynamics beyond total density, seedling populations were analysed by height class. Relative proportions of each size class were calculated as: \( {p}_{i}=\frac{\text{density of size class }i}{\text{total seedling density}}\) Variation in size-class composition across fallow age and invasion gradients was analysed descriptively and visualised using stacked bar charts. In addition, ternary plots were used to represent the relative contribution of each size class across environmental gradients, providing an integrated view of regeneration structure (Fig. S5). 2.3.5. Model diagnostics and validation Model adequacy was assessed using simulation-based diagnostics implemented in the DHARMa package. Diagnostic procedures included tests of residual uniformity, dispersion, zero inflation, and outliers. Results indicated that the zero-inflated negative binomial model adequately captured the excess of zero observations (zero-inflation test: non-significant) and did not exhibit problematic outliers. Although dispersion and uniformity tests indicated statistically significant deviations, visual inspection of residual plots suggested only moderate departures from model assumptions, which are common in large ecological count datasets. Diagnostic results supported the adequacy of the model despite minor deviations typical of large ecological datasets (Fig. S2, Fig. S3; Text S2). 2.3.6. Data visualisation All figures were produced using the ggplot2 package, with additional extensions ( ggdist , patchwork , and ggtern ) for advanced visualisation. Figures were designed to maximise interpretability and facilitate comparison across species, fallow age classes, invasion levels, and spatial gradients. 3. Results 3.1. Dataset structure and distribution of observations The dataset followed a fully balanced factorial design combining fallow age (0–3, 3–6, > 6 years), Chromolaena odorata cover (C1-C4), and five tree species, resulting in an equal number of observations per treatment combination (n = 160). This balanced structure ensured robust estimation of main and interaction effects while minimising potential biases associated with unequal sampling effort (Fig. 2 ). 3.2. Effects of fallow age and invasion level on total seedling density Seedling density varied markedly across both fallow age and Chromolaena odorata cover gradients (Table 1 ; Fig. 3 ). The zero-inflated negative binomial (ZINB) model revealed significant effects of fallow age, invasion level, and their interaction (all p < 0.001), indicating that the impact of invasion depends strongly on successional stage. Predicted seedling densities declined sharply with increasing invasion intensity, particularly in early fallows (0–3 years). Model-based estimates showed a decrease from 1051 ind. ha⁻¹ under low invasion (C1) to 270 ind. ha⁻¹ under high invasion (C4), corresponding to a reduction of approximately 74%. This strong decline was consistent across descriptive statistics, which showed a reduction in mean density from 1343 to 203 ind. ha⁻¹ across the same gradient. Table 1 Descriptive statistics of seedling density (individuals ha⁻¹) across fallow age and Chromolaena odorata cover gradients. Fallow age C. odorata cover n Mean SD Median Q1 Q3 Min Max 0–3 years C1 800 1343 1185 1270 197 2235 0 5264 C2 800 771 673 646 192 1192 0 3394 C3 800 324 240 304 117 488 0 1565 C4 800 203 208 162 0 291 0 1279 3–6 years C1 800 932 867 772 168 1524 0 6141 C2 800 803 774 586 201 1210 0 3749 C3 800 621 772 324 100 757 0 3542 C4 800 668 832 242 91 1060 0 3481 > 6 years C1 800 758 704 496 194 1222 0 3226 C2 800 760 651 590 266 1092 0 2811 C3 800 706 659 539 134 1070 0 3077 C4 800 753 701 615 94 1244 0 2779 In intermediate fallows (3–6 years), invasion effects remained significant but were less pronounced, with predicted densities declining from 784 ind. ha⁻¹ (C1) to 454 ind. ha⁻¹ (C4), representing a reduction of approximately 42%. In contrast, in older fallows (> 6 years), predicted densities were relatively stable across invasion levels (698–760 ind. ha⁻¹), indicating a marked attenuation of invasion effects. The zero-inflation component of the model further indicated that the probability of structural zeros increased under high invasion levels and in early fallows, suggesting that invasion not only reduces seedling density but also increases the likelihood of recruitment failure. 3.3. Species-specific responses to fallow age and invasion gradients Species responses to invasion and succession varied significantly, as indicated by a strong three-way interaction between species, fallow age, and invasion level (p < 0.001). Predicted marginal means revealed contrasting response patterns among species (Fig. 4 ; Table S4). E. suaveolens exhibited the strongest negative response to invasion, with predicted densities declining from 1454 ind. ha⁻¹ under low invasion (C1) to 149 ind. ha⁻¹ under high invasion (C4) in early fallows, corresponding to a reduction of approximately 90%. Similarly, Pycnanthus angolensis showed pronounced declines across invasion gradients, particularly in early and intermediate fallows. P. africanum showed a moderate response, with significant declines under high invasion in early and intermediate fallows, but weaker and non-significant responses in older fallows. In contrast, P. macrocarpus exhibited a positive response to invasion, with predicted densities increasing from 203 ind. ha⁻¹ (C1) to 396 ind. ha⁻¹ (C4) in early fallows, and up to 1090 ind. ha⁻¹ under high invasion in older fallows, representing more than a fourfold increase. R. heudelotii showed relatively stable responses across invasion gradients, particularly in intermediate and older fallows, where differences among invasion levels were small and non-significant. These results indicate that C. odorata acts as a strong ecological filter, suppressing regeneration in some species while favouring others, thereby generating substantial interspecific variability in regeneration dynamics. 3.4. Spatial patterns of seedling recruitment Seedling density decreased non-linearly with increasing distance from seed trees, as captured by the spline term included in the ZINB model (Fig. 6 ). This decline was steepest in early fallows, indicating strong dispersal limitation during initial successional stages. Predicted densities were highest near seed trees and declined rapidly within the first tens of metres, after which the decline became more gradual. In intermediate and older fallows, spatial gradients were less pronounced, suggesting more diffuse recruitment patterns. Invasion intensity significantly modified these spatial patterns. High C. odorata cover reduced seedling density across all distances and flattened spatial gradients, resulting in more homogeneous distributions (Fig. 7 ). This indicates that invasion weakens distance-dependent recruitment processes and reduces spatial heterogeneity in seedling establishment. Overall, these results demonstrate that regeneration patterns are jointly structured by dispersal limitation and environmental filtering, with invasion affecting both the magnitude and spatial configuration of recruitment. 3.5. Changes in regeneration structure across invasion gradients Seedling size structure varied systematically across invasion gradients and fallow age classes (Fig. 8 ). Across all conditions, the smallest size class (0–30 cm) dominated seedling populations. However, the proportion of larger seedlings (> 50 cm) declined consistently with increasing invasion intensity. Under high invasion levels (C3-C4), the relative contribution of larger size classes was substantially reduced compared to low invasion conditions, indicating a shift towards younger developmental stages. This pattern suggests that C. odorata not only limits seedling recruitment but also constrains growth and progression to later developmental stages, potentially affecting long-term regeneration trajectories. 3.6. Pairwise comparisons of invasion effects Pairwise comparisons based on estimated marginal means confirmed significant differences in seedling density across invasion levels for most species (Table 2 ; Table S2). High invasion levels (C3 and C4) were generally associated with significantly lower seedling densities compared to low invasion (C1), particularly in early fallows. For example, in Erythrophleum suaveolens, rate ratios decreased from 0.53 (C2 vs C1) to 0.06 (C4 vs C1) in early fallows, indicating a progressive and substantial decline in seedling density with increasing invasion intensity. Table 2 Pairwise comparisons of seedling density across invasion levels within fallow age classes (rate ratios from negative binomial models). Species Fallow age C2 vs C1 C3 vs C1 C4 vs C1 E. suaveolens 0–3 years 0.53*** 0.22*** 0.06*** 3–6 years 0.60*** 0.30*** 0.12*** > 6 years 0.70*** 0.39*** 0.15*** P. africanum 0–3 years 0.73* 0.28*** 0.19*** 3–6 years 0.81 ns 0.21*** 0.17*** > 6 years 0.91 ns 0.61*** 0.63*** P. angolensis 0–3 years 0.50*** 0.16*** 0.13*** 3–6 years 0.76 ns 0.26*** 0.16*** > 6 years 0.92 ns 1.00 ns 0.97 ns P. macrocarpus 0–3 years 1.44 ns 3.36*** 2.60*** 3–6 years 1.51* 3.72*** 7.32*** > 6 years 3.76*** 5.15*** 8.51*** R. heudelotii 0–3 years 0.52*** 0.13*** 0.08*** 3–6 years 1.00 ns 1.04 ns 1.07 ns > 6 years 1.07 ns 1.00 ns 1.01 ns Note : Values represent rate ratios relative to low invasion (C1). Values 1 indicate an increase. Significance levels: *** p < 0.001; * p < 0.05; ns = not significant . In contrast, P. macrocarpus showed the opposite pattern, with rate ratios exceeding 1 under moderate to high invasion levels and reaching up to 8.51 in older fallows, indicating a strong positive response to invasion. Other species, such as R. heudelotii , showed no significant differences among invasion levels in intermediate and older fallows, confirming species-specific variability in sensitivity to invasion. Overall, these pairwise comparisons reinforce the results of the main model and highlight the contrasting responses of species to invasion across successional stages. 4. Discussion 4.1. Successional stage as a major driver of regeneration dynamics This study demonstrates that fallow age is a major determinant of tree regeneration dynamics in Central African agricultural landscapes, but its effect is strongly contingent on invasion intensity. While seedling density was highest in early fallows under low invasion (≈ 1050 ind. ha⁻¹), this advantage was rapidly eroded under high C. odorata cover, with densities declining to ≈ 270 ind. ha⁻¹, representing a reduction of more than 70%. This result refines the classical view that early successional stages favour regeneration due to higher light availability and reduced structural competition (Poorter et al. 2021 ). Our findings show that this advantage is conditional, and can be overridden by strong competitive exclusion from invasive species. In contrast, older fallows (> 6 years) maintained relatively stable regeneration levels across invasion gradients (≈ 700–760 ind. ha⁻¹), suggesting that successional development buffers the negative effects of invasion. This pattern indicates that regeneration dynamics are not simply driven by successional stage, but by interactions between succession and biotic filtering processes. These findings support a more nuanced view of succession, where early stages represent both a window of opportunity for recruitment and a period of high vulnerability to invasion, while later stages provide greater ecological resistance to competitive exclusion, consistent with increasing biotic resistance and structural complexity along successional gradients (Chazdon et al. 2020 ; Poorter et al. 2021 ). 4.2. Strong but non-linear effects of Chromolaena odorata on regeneration The results provide strong evidence that increasing C. odorata cover reduces seedling density, but this effect is clearly non-linear and threshold-dependent. Across species, major declines occurred primarily between low (C1) and high invasion levels (C3-C4), with reductions exceeding 70–90% in early fallows for several species. This pattern suggests the existence of ecological thresholds, beyond which invasion shifts from a moderate influence to a dominant limiting factor for regeneration. Below these thresholds, regeneration may still occur, albeit at reduced levels; above them, dense shrub cover likely imposes severe constraints through light limitation, space pre-emption, and possibly allelopathic effects (Kato-Noguchi and Kato 2023 ). Importantly, the zero-inflation component of the model showed that invasion increased the probability of structural zeros, particularly in early fallows. This indicates that invasion does not merely reduce seedling abundance but also increases the likelihood of complete recruitment failure, a critical but often overlooked mechanism in regeneration studies. These results extend previous findings by demonstrating that invasion impacts are not gradual but may involve non-linear transitions in regeneration success, reinforcing the need to explicitly account for invasion intensity rather than presence alone, as ecological impacts of invasive species are often strongly dependent on their abundance and may exhibit threshold responses (Gbètoho et al. 2018 ; Pyšek et al. 2020 ). 4.3. Interactive effects of succession and invasion: a context-dependent process A central contribution of this study is the demonstration that the impact of C. odorata is strongly modulated by successional stage. The negative effects of invasion were most pronounced in early fallows, where density reductions exceeded 70%, but became negligible in older fallows. This interaction highlights that early successional stages constitute a critical ecological bottleneck, where recruitment is both highly active and highly sensitive to environmental filtering. In such conditions, invasive shrubs can effectively suppress regeneration and potentially alter successional trajectories. In contrast, the reduced impact of invasion in older fallows suggests that increasing vegetation complexity, canopy development, and microclimatic buffering enhance ecosystem resistance to invasion. This aligns with the concept of biotic resistance, whereby more developed plant communities limit the establishment and impact of invasive species (Chazdon et al. 2020 ). Overall, these findings demonstrate that invasion effects cannot be generalised across successional gradients, but instead emerge from dynamic interactions between ecosystem development and competitive processes. 4.4. Species-specific responses and ecological filtering The strong interspecific variation observed in this study confirms that C. odorata acts as a selective ecological filter shaping regeneration patterns. Species such as E. suaveolens experienced drastic declines (> 90%) under high invasion in early fallows, whereas P. macrocarpus showed a marked increase in density, with values exceeding four times those observed under low invasion. These contrasting responses suggest that invasion favours species with traits adapted to shaded and competitive environments, while excluding more light-demanding or competition-sensitive species. This pattern is consistent with trait-based community assembly theory, where environmental filters select species according to functional traits (Funk et al. 2008 ; Kraft et al. 2015 ). The relatively stable response of R. heudelotii across invasion gradients further illustrates that some species may be tolerant to a wide range of environmental conditions, contributing to their persistence in disturbed landscapes. Such species-specific responses have important implications for long-term forest composition. By selectively favouring certain species, invasion may lead to functional and compositional shifts, potentially reducing diversity and altering ecosystem functioning even when overall seedling abundance remains high. 4.5. Spatial constraints on regeneration and their interaction with invasion The observed decline in seedling density with increasing distance from seed trees confirms the central role of dispersal limitation in structuring regeneration patterns. This effect was particularly strong in early fallows, where recruitment was highly concentrated near seed sources, indicating limited dispersal distances and strong spatial clustering. However, invasion significantly modified these spatial dynamics. High C. odorata cover reduced seedling density across all distances and flattened spatial gradients, leading to more homogeneous distributions. This suggests that invasion weakens distance-dependent recruitment processes, likely by imposing strong environmental constraints that override dispersal-driven patterns. These findings highlight the joint role of dispersal limitation and environmental filtering in shaping regeneration, and demonstrate that invasive species can alter not only the magnitude but also the spatial structure of recruitment. Such alterations in spatial patterns may have cascading effects on community assembly and spatial heterogeneity, which are key components of ecosystem resilience and biodiversity maintenance. 4.6. Implications for regeneration dynamics and management The results of this study have important implications for the management of agricultural fallows and forest restoration in Central Africa. First, the high regeneration potential observed in early fallows under low invasion confirms their importance as key regeneration niches. However, the strong decline in seedling density under high invasion (up to 70–90%) indicates that these systems are highly vulnerable to competitive exclusion during early successional stages. Second, the identification of strong invasion effects in early fallows suggests that management interventions should prioritise early-stage control of C. odorata , before dense stands become established. Preventing invasion during this critical window is likely to be more effective than attempting to restore regeneration after invasion thresholds have been exceeded. Third, the marked species-specific responses observed highlight the need for context-dependent management strategies. While some species are strongly suppressed by invasion, others may tolerate or even benefit from altered environmental conditions. Management approaches should therefore consider species traits and ecological strategies rather than applying uniform interventions. Finally, the observed effects of invasion on both spatial patterns and size structure indicate that invasion can influence not only recruitment but also subsequent growth and community development. This underscores the importance of integrating multiple dimensions of regeneration (density, structure, spatial patterns) in restoration planning. 5. Conclusion This study provides a robust quantitative assessment of how successional dynamics and biological invasion jointly shape tree regeneration in agricultural fallows of Central Africa. By integrating field data with zero-inflated modelling approaches, we demonstrate that fallow age and Chromolaena odorata cover interact strongly to determine not only seedling density, but also the spatial structure and developmental trajectory of regeneration. Our results show that early fallows constitute a critical phase for regeneration, characterised by high recruitment potential under low invasion but also high vulnerability to competitive exclusion. Under high invasion levels, seedling density declined by more than 70%, and the probability of recruitment failure increased substantially, highlighting the sensitivity of early successional stages to biotic constraints. In contrast, older fallows exhibited more stable regeneration across invasion gradients, suggesting increasing ecological resistance with successional development. Importantly, the effects of C. odorata were not gradual but strongly non-linear, with marked declines in regeneration occurring beyond intermediate levels of invasion. This threshold-like behaviour indicates that invasion intensity is a key determinant of ecological impact and should be explicitly considered when assessing regeneration processes and designing management interventions. The study further reveals pronounced species-specific responses to invasion. While several species experienced strong declines in seedling density under high invasion, others showed neutral or even positive responses, indicating that invasion acts as a selective ecological filter shaping regeneration trajectories and potentially altering long-term community composition. Beyond overall density patterns, invasion was also found to modify spatial recruitment processes and seedling size structure, reducing distance-dependent gradients and limiting the progression of seedlings to larger size classes. These findings demonstrate that invasion affects multiple dimensions of regeneration, with potential cascading effects on forest structure and recovery dynamics. From a management perspective, our results emphasise the importance of early intervention. Preventing the establishment of dense Chromolaena odorata stands during early successional stages appears critical to maintaining regeneration potential. At the same time, the strong interspecific variability observed suggests that management strategies should be adapted to ecological context and species-specific responses rather than relying on uniform control approaches. Overall, this study highlights that regeneration in human-modified tropical landscapes is governed by complex and context-dependent interactions between succession, invasion intensity, dispersal processes, and species traits. By demonstrating the importance of non-linear invasion effects and successional context, it contributes to a more mechanistic understanding of forest recovery processes outside forests and provides a foundation for more targeted and effective restoration strategies. Declarations Author Contribution Jean Pierre Azenge conceptualised the study, designed the methodology, coordinated all research data collection and analysis interpretation, and drafted the manuscript. Paxie W. Chirwa and Justin N’Dja Kassi provided crucial academic supervision throughout the study, offering substantial intellectual inputs and critical revisions to the manuscript. All authors critically reviewed and approved the final version of the manuscript submitted for publication. 6. Acknowledgements We thank the Regional Scholarship and Innovation Fund (RSIF) for the scholarship that made this research possible. References Azenge JP, Wassila IS, Kassi JN, Chirwa PW (2025) Diversity and ethnobotanical use-value of trees outside forests on the agricultural landscape of the Mongala Province, Democratic Republic of Congo. Agrofor Syst 99:1–20. https://doi.org/10.1007/s10457-025-01332-3 Chazdon RL (2014) Second Growth The Promise of Tropical Forest Regeneration in an Age of Deforestation, Th e Unive. Th e University of Chicago, Chicago and London Chazdon RL, Harvey CA, Komar O et al (2009) Beyond reserves: A research agenda for conserving biodiversity in human-modified tropical landscapes. Biotropica 41:142–153. https://doi.org/10.1111/j.1744-7429.2008.00471.x Chazdon RL, Lindenmayer D, Guariguata MR et al (2020) Fostering natural forest regeneration on former agricultural land through economic and policy interventions. Environ Res Lett 15. https://doi.org/10.1088/1748-9326/ab79e6 Curtis PG, Slay CM, Harris NL et al (2018) Classifying drivers of global forest loss. Sci (80-) 361:1108–1111 Diyarzola J, Bernard C (2024) Diagnostic agraire de la province de la Mongala en RDC. Enabel-RDC, Kinshasa, République Démocratique du Congo FAO (2016) Trees, forests and land use in drylands: the first global assessment. Rome Funk JL, Cleland EE, Suding KN, Zavaleta ES (2008) Restoration through reassembly: plant traits and invasion resistance. Trends Ecol Evol 23:695–703. https://doi.org/10.1016/j.tree.2008.07.013 Gbètoho AJ, Kingbo A, Gnanguènon-guéssè D et al (2018) Impacts of Chromolaena odorata on native trees’ regeneration in the Lama secondary forests in Benin, West Africa. Bois Forets des Trop 338:5–14 Hardy NG, Kuebbing SE, Duguid MC et al (2025) Non-native invasive plants in tropical dry forests: a global review of presence, impacts, and management. Restor Ecol 33. https://doi.org/10.1111/rec.14288 Juru VN, Ndam LM, Tatah BN, Fonge BA (2024) Rhizospheric soil chemical properties and microbial response to a gradient of Chromolaena odorata(L) invasion in the Mount Cameroon Region. PLoS ONE 19:1–24. https://doi.org/10.1371/journal.pone.0312199 Kato-Noguchi H, Kato M (2023) Evolution of the Secondary Metabolites in Invasive Plant Species Chromolaena odorata for the Defense and Allelopathic Functions. Plants 12:6–12. https://doi.org/10.3390/plants12030521 Kraft NJB, Adler PB, Godoy O et al (2015) Community assembly, coexistence and the environmental filtering metaphor. Funct Ecol 29:592–599. https://doi.org/10.1111/1365-2435.12345 Loubota Panzou GJ, Mankessi F, Tsiba Ngambou FC et al (2024) Natural forest regeneration over a fallow age chronosequence in central African moist forests. Afr J Ecol 62:1–11. https://doi.org/10.1111/aje.13255 Mpanda M, Kashindye A, Aynekulu E et al (2021) Forests, farms, and fallows: The dynamics of tree cover transition in the southern part of the uluguru mountains, tanzania. Land 10. https://doi.org/10.3390/land10060571 Mwampamba TH, Schwartz MW (2011) The effects of cultivation history on forest recovery in fallows in the Eastern Arc Mountain, Tanzania. Ecol Manage 261:1042–1052. https://doi.org/10.1016/j.foreco.2010.12.026 Poorter L, Craven D, Jakovac CC et al (2021) Multidimensional tropical forest recovery. Sci (80-) 374:1370–1376 Poorter L, Rozendaal D, Bongers F et al (2019) Wet and dry tropical forests show opposite successional pathways in wood density but converge over time. Nat Ecol Evol 3:928–934. https://doi.org/10.1038/s41559-019-0882-6 Pyšek P, Hulme PE, Simberloff D et al (2020) Scientists’ warning on invasive alien species. Biol Rev 95:1511–1534. https://doi.org/10.1111/brv.12627 Rai PK, Singh JS (2024) Ecological insights and environmental threats of invasive alien plant Chromolaena odorata: Prospects for sustainable management. Weed Biol Manag 24:15–37. https://doi.org/10.1111/wbm.12286 Reed J, Van Vianen J, Deakin EL et al (2016) Integrated landscape approaches to managing social and environmental issues in the tropics: learning from the past to guide the future. Glob Chang Biol 22:2540–2554. https://doi.org/10.1111/gcb.13284 Rozendaal DMA, Bongers F, Aide TM et al (2019) Biodiversity recovery of Neotropical secondary forests. Sci Adv. https://doi.org/10.1126/sciadv.aau3114 . 5: Zomer RJ, Neufeldt H, Xu J et al (2016) Global Tree Cover and Biomass Carbon on Agricultural Land: The contribution of agroforestry to global and national carbon budgets. Sci Rep 6:1–12. https://doi.org/10.1038/srep29987 Azenge JP, Wassila IS, Kassi JN, Chirwa PW (2025) Diversity and ethnobotanical use-value of trees outside forests on the agricultural landscape of the Mongala Province, Democratic Republic of Congo. Agrofor Syst 99:1–20. https://doi.org/10.1007/s10457-025-01332-3 Chazdon RL (2014) Second Growth The Promise of Tropical Forest Regeneration in an Age of Deforestation, Th e Unive. Th e University of Chicago, Chicago and London Chazdon RL, Harvey CA, Komar O et al (2009) Beyond reserves: A research agenda for conserving biodiversity in human-modified tropical landscapes. Biotropica 41:142–153. https://doi.org/10.1111/j.1744-7429.2008.00471.x Chazdon RL, Lindenmayer D, Guariguata MR et al (2020) Fostering natural forest regeneration on former agricultural land through economic and policy interventions. Environ Res Lett 15. https://doi.org/10.1088/1748-9326/ab79e6 Curtis PG, Slay CM, Harris NL et al (2018) Classifying drivers of global forest loss. Sci (80-) 361:1108–1111 Diyarzola J, Bernard C (2024) Diagnostic agraire de la province de la Mongala en RDC. Enabel-RDC, Kinshasa, République Démocratique du Congo FAO (2016) Trees, forests and land use in drylands: the first global assessment. Rome Funk JL, Cleland EE, Suding KN, Zavaleta ES (2008) Restoration through reassembly: plant traits and invasion resistance. Trends Ecol Evol 23:695–703. https://doi.org/10.1016/j.tree.2008.07.013 Gbètoho AJ, Kingbo A, Gnanguènon-guéssè D et al (2018) Impacts of Chromolaena odorata on native trees’ regeneration in the Lama secondary forests in Benin, West Africa. Bois Forets des Trop 338:5–14 Hardy NG, Kuebbing SE, Duguid MC et al (2025) Non-native invasive plants in tropical dry forests: a global review of presence, impacts, and management. Restor Ecol 33. https://doi.org/10.1111/rec.14288 Juru VN, Ndam LM, Tatah BN, Fonge BA (2024) Rhizospheric soil chemical properties and microbial response to a gradient of Chromolaena odorata(L) invasion in the Mount Cameroon Region. PLoS ONE 19:1–24. https://doi.org/10.1371/journal.pone.0312199 Kato-Noguchi H, Kato M (2023) Evolution of the Secondary Metabolites in Invasive Plant Species Chromolaena odorata for the Defense and Allelopathic Functions. Plants 12:6–12. https://doi.org/10.3390/plants12030521 Kraft NJB, Adler PB, Godoy O et al (2015) Community assembly, coexistence and the environmental filtering metaphor. Funct Ecol 29:592–599. https://doi.org/10.1111/1365-2435.12345 Loubota Panzou GJ, Mankessi F, Tsiba Ngambou FC et al (2024) Natural forest regeneration over a fallow age chronosequence in central African moist forests. Afr J Ecol 62:1–11. https://doi.org/10.1111/aje.13255 Mpanda M, Kashindye A, Aynekulu E et al (2021) Forests, farms, and fallows: The dynamics of tree cover transition in the southern part of the uluguru mountains, tanzania. Land 10. https://doi.org/10.3390/land10060571 Mwampamba TH, Schwartz MW (2011) The effects of cultivation history on forest recovery in fallows in the Eastern Arc Mountain, Tanzania. Ecol Manage 261:1042–1052. https://doi.org/10.1016/j.foreco.2010.12.026 Poorter L, Craven D, Jakovac CC et al (2021) Multidimensional tropical forest recovery. Sci (80-) 374:1370–1376 Poorter L, Rozendaal D, Bongers F et al (2019) Wet and dry tropical forests show opposite successional pathways in wood density but converge over time. Nat Ecol Evol 3:928–934. https://doi.org/10.1038/s41559-019-0882-6 Pyšek P, Hulme PE, Simberloff D et al (2020) Scientists’ warning on invasive alien species. Biol Rev 95:1511–1534. https://doi.org/10.1111/brv.12627 Rai PK, Singh JS (2024) Ecological insights and environmental threats of invasive alien plant Chromolaena odorata: Prospects for sustainable management. Weed Biol Manag 24:15–37. https://doi.org/10.1111/wbm.12286 Reed J, Van Vianen J, Deakin EL et al (2016) Integrated landscape approaches to managing social and environmental issues in the tropics: learning from the past to guide the future. Glob Chang Biol 22:2540–2554. https://doi.org/10.1111/gcb.13284 Rozendaal DMA, Bongers F, Aide TM et al (2019) Biodiversity recovery of Neotropical secondary forests. Sci Adv. https://doi.org/10.1126/sciadv.aau3114 . 5: Zomer RJ, Neufeldt H, Xu J et al (2016) Global Tree Cover and Biomass Carbon on Agricultural Land: The contribution of agroforestry to global and national carbon budgets. Sci Rep 6:1–12. https://doi.org/10.1038/srep29987 Additional Declarations No competing interests reported. Supplementary Files SupplementarymaterialAzengeetal23032026.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9203304","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":611773303,"identity":"1d0cc7cf-5481-48b2-8201-9cf45ac6757b","order_by":0,"name":"Jean Pierre Azenge","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9klEQVRIiWNgGAWjYDACCQbGA3DOByBmYyeshQGuhXEGSAszKVqYecAkAR38s5sfHPjZZmfPL91+8bHNr23yfMwMjB8+5uCx5M4xg4O9bcnMknPOFBvn9t02bGNmYJacuQ23FgOJBIMDvG3MbAY3ctKkc3tuMwK1sDHz4tWS/uHg37Z6HvsbOem/LXtu2xOhJcfgMG/bYQmg3mPMDD9uJxLUInEjp+CwzLnjBkAGs2Rvw+3kNmbGZrx+4Z+RvvHhm7JqeyDj4Ycff27bzm9vPvjhIx4tYMDIBiJ5DBgY28DcBgLqQeAPiGB/AGWMglEwCkbBKEAFACAGUoOE+8hsAAAAAElFTkSuQmCC","orcid":"","institution":"University of Pretoria","correspondingAuthor":true,"prefix":"","firstName":"Jean","middleName":"Pierre","lastName":"Azenge","suffix":""},{"id":611773306,"identity":"e37aa6d0-6815-4aba-8c10-3d083ffa715a","order_by":1,"name":"Justin N'Dja Kassi","email":"","orcid":"","institution":"Université Félix Houphouët-Boigny","correspondingAuthor":false,"prefix":"","firstName":"Justin","middleName":"N'Dja","lastName":"Kassi","suffix":""},{"id":611773311,"identity":"664d1e39-b01d-45c0-a432-bd491952b0d8","order_by":2,"name":"Paxie W. Chirwa","email":"","orcid":"","institution":"University of Pretoria","correspondingAuthor":false,"prefix":"","firstName":"Paxie","middleName":"W.","lastName":"Chirwa","suffix":""}],"badges":[],"createdAt":"2026-03-23 17:23:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9203304/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9203304/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106388409,"identity":"9babbec8-ce08-4e7d-a732-fc36caeb0f5d","added_by":"auto","created_at":"2026-04-08 06:43:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":196214,"visible":true,"origin":"","legend":"\u003cp\u003eLocation of surveyed villages and regeneration inventory sites in Mongala Province, Democratic Republic of the Congo.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/21e190bf50a495b2bcd4fb77.png"},{"id":106388418,"identity":"406144ee-7637-4b7c-bac0-6bfe8f407cef","added_by":"auto","created_at":"2026-04-08 06:43:35","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":18257,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmap of sample size distribution across species, fallow age classes, and Chromolaena odorata cover levels.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/aa41b9a11768cd4dfd3c81ca.png"},{"id":106388385,"identity":"d6fc7097-75b9-44ef-a9a2-88bc07d55baf","added_by":"auto","created_at":"2026-04-08 06:43:17","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":98962,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted seedling density across Chromolaena odorata cover levels and fallow age classes based on zero-inflated negative binomial models (mean ± 95% confidence intervals).\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/bce9b9a12e66d7ea1c949d04.png"},{"id":106388382,"identity":"0fef937b-0027-470e-923f-5320aaf79e1d","added_by":"auto","created_at":"2026-04-08 06:43:15","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":122440,"visible":true,"origin":"","legend":"\u003cp\u003eSpecies-specific predicted seedling density across Chromolaena odorata cover levels and fallow age classes derived from zero-inflated negative binomial models (mean ± 95% confidence intervals).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/d9f4e364df7aa3c7f1e3b465.png"},{"id":106388339,"identity":"5acf8d34-04bd-4af7-8c32-3a1724ed0dca","added_by":"auto","created_at":"2026-04-08 06:43:09","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":120982,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmaps of predicted seedling density across fallow age and Chromolaena odorata cover levels for each species based on ZINB model predictions.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/a80e77cfeca056c68374cdc0.png"},{"id":106388387,"identity":"ff6f311f-c053-480d-a53c-00329b618be9","added_by":"auto","created_at":"2026-04-08 06:43:18","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":112072,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted relationship between seedling density and distance from seed tree across fallow age classes and Chromolaena odorata cover levels based on ZINB models with spline functions (mean ± 95% confidence intervals).\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/23465b8873d4c1d2d17a7b98.png"},{"id":106388345,"identity":"0f8e69ba-6571-496b-ad9d-0d5119e48f15","added_by":"auto","created_at":"2026-04-08 06:43:14","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":70771,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted seedling density as a function of distance from seed tree and Chromolaena odorata cover across fallow age classes and species. Panels represent individual species (A: Erythrophleum suaveolens, B: Petersianthus macrocarpus, C: Ricinodendron heudelotii, D: Piptadeniastrum africanum, E: Pycnanthus angolensis).\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/922a17dfd5c55d98c0a3ef1a.png"},{"id":106388381,"identity":"1686d608-1d03-4f8b-8eb1-8318b7538a8b","added_by":"auto","created_at":"2026-04-08 06:43:15","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":41505,"visible":true,"origin":"","legend":"\u003cp\u003eRelative composition of seedling height classes across fallow age and Chromolaena odorata cover levels, showing dominance of early-stage seedlings and reduced proportion of taller individuals under high invasion.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/186947c8f3335b276e39f717.png"},{"id":106388426,"identity":"59c426f6-2ed7-4cd4-ac98-657877d0537f","added_by":"auto","created_at":"2026-04-08 06:43:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1889878,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/aeeaa67f-4ca5-4fac-9343-40d8a9abad16.pdf"},{"id":106388338,"identity":"4579c2b1-56ef-4916-a258-1a75366be0a6","added_by":"auto","created_at":"2026-04-08 06:43:08","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":927952,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarymaterialAzengeetal23032026.docx","url":"https://assets-eu.researchsquare.com/files/rs-9203304/v1/a42717f3dbb3392ad7cb0db1.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Interactive effects of fallow age and Chromolaena odorata invasion on tree regeneration in Central African agricultural landscapes","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAcross tropical Africa, agricultural expansion and shifting cultivation have profoundly transformed forest landscapes, resulting in extensive mosaics of croplands, fallows, and remnant tree cover (Curtis et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Mpanda et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In these human-modified systems, trees outside forests (TOF) play a crucial ecological and socio-economic role by maintaining biodiversity, supporting ecosystem services, and contributing to forest recovery processes (FAO \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Zomer et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). In particular, natural regeneration within agricultural fallows represents a major pathway for the persistence and renewal of woody species in landscapes where access to intact forests is increasingly limited (Curtis et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Loubota Panzou et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNatural regeneration in fallow systems is shaped by multiple interacting drivers, including land-use history, successional stage, seed availability, and biotic interactions (Chazdon et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Mwampamba and Schwartz \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Among these factors, fallow age is widely recognised as a key determinant of regeneration dynamics, as it influences light availability, soil conditions, and vegetation structure (Chazdon \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Poorter et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Early successional stages may favour seedling establishment due to higher light availability, whereas later stages may enhance survival through improved microclimatic buffering and soil development. However, the relationship between fallow age and regeneration is often non-linear and species-dependent, reflecting differences in ecological strategies among tree species (Poorter et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Rozendaal et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn tropical agricultural landscapes, these successional processes are increasingly altered by invasive plant species (Pyšek et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Among them, \u003cem\u003eChromolaena odorata\u003c/em\u003e (L.) R.M. King \u0026amp; H. Rob. is one of the most aggressive invasive shrubs in tropical Africa and is now widespread in post-cultivation fallows (Rai and Singh \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Its rapid growth, dense canopy, and potential allelopathic effects can reduce light availability, limit seedling establishment, and modify competitive interactions (Kato-Noguchi and Kato \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Juru et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). While \u003cem\u003eC. odorata\u003c/em\u003e is often associated with reduced tree regeneration, its effects are not uniformly negative. In some contexts, dense shrub cover may alter competitive dynamics or microenvironmental conditions in ways that influence seedling recruitment differently across species and successional stages (Gb\u0026egrave;toho et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Hardy et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite increasing attention to the role of TOF in landscape resilience, the combined effects of fallow age and invasive shrub cover on tree regeneration remain insufficiently understood, particularly in Central African agricultural systems (Reed et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Chazdon et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Most studies have examined these drivers independently, overlooking potential interactions between successional stage and invasion intensity. In addition, species-specific responses to these interacting factors remain poorly documented, limiting our ability to predict regeneration trajectories and design context-adapted management strategies.\u003c/p\u003e \u003cp\u003eIn the Democratic Republic of Congo (DRC), and particularly in Mongala Province, shifting cultivation dominates rural land use and generates extensive networks of agricultural fallows where TOF constitute the main source of natural forest regeneration (FAO \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Loubota Panzou et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These landscapes provide an appropriate context for analysing how regeneration patterns vary across successional stages and invasion gradients. In such systems, understanding both the magnitude and variability of regeneration responses is essential for informing sustainable land management and restoration-oriented practices.\u003c/p\u003e \u003cp\u003eAgainst this background, this study investigates the natural regeneration of five dominant TOF species (\u003cem\u003ePetersianthus macrocarpus\u003c/em\u003e (P. Beauv.) Liben, \u003cem\u003ePycnanthus angolensis\u003c/em\u003e (Welw.) Warb., \u003cem\u003eRicinodendron heudelotii\u003c/em\u003e (Baill.) Heckel, \u003cem\u003eErythrophleum suaveolens\u003c/em\u003e (Guill. \u0026amp; Perr.) Brenan, and \u003cem\u003ePiptadeniastrum africanum\u003c/em\u003e (Hook.f.) Brenan) across agricultural fallows differing in age and \u003cem\u003eC. odorata\u003c/em\u003e cover in Mongala Province, Democratic Republic of the Congo. Specifically, we aimed to: (i) quantify the effects of fallow age and invasion level on seedling density; (ii) assess whether the effect of \u003cem\u003eC. odorata\u003c/em\u003e varies across successional stages; (iii) analyse species-specific responses to these interacting drivers; and (iv) evaluate how these factors influence the spatial and structural patterns of regeneration.\u003c/p\u003e \u003cp\u003eWe hypothesised that: (1) seedling density decreases with increasing \u003cem\u003eC. odorata\u003c/em\u003e cover;\u003c/p\u003e \u003cp\u003e(2) this negative effect is strongest in early fallows and weakens with increasing fallow age; (3) species exhibit contrasting responses to invasion and succession; and (4) both spatial patterns of recruitment and seedling size structure are jointly influenced by distance from seed trees, fallow age, and invasion level.\u003c/p\u003e \u003cp\u003eBy explicitly testing these hypotheses, this study provides a quantitative assessment of how successional dynamics and biological invasion interact to shape tree regeneration in fallow-based tropical landscapes, and contributes to a better understanding of forest recovery processes outside forests.\u003c/p\u003e"},{"header":"2. Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n\u003ch2\u003e2.1. Study area\u003c/h2\u003e\n\u003cp\u003eThe study was conducted in Mongala Province, located in the north-western part of the Democratic Republic of the Congo within the central Congo Basin (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). The province is characterised by lowland tropical landscapes dominated by moist evergreen and semi-evergreen forests interspersed with agricultural fields and fallows resulting from shifting cultivation.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe climate is humid tropical, with relatively stable high temperatures throughout the year and mean annual rainfall generally exceeding 1,500 mm (Azenge et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). Rainfall is distributed across a long rainy season and a shorter dry season, creating favourable conditions for continuous plant growth and rapid vegetation recovery following agricultural abandonment.\u003c/p\u003e\n\u003cp\u003eVegetation is largely composed of dense tropical rainforest formations, although extensive areas have been converted into agricultural mosaics. Shifting cultivation is the dominant land-use system, with fields typically cultivated for one to two years before being abandoned to fallow. Fallow duration varies widely, resulting in a heterogeneous patchwork of fallows of different ages that constitute key sites for the regeneration of trees outside forests.\u003c/p\u003e\n\u003cp\u003eSoils are predominantly highly weathered tropical soils with low to moderate fertility. Repeated cultivation can further reduce nutrient availability, making fallow periods essential for restoring soil structure and fertility. However, vegetation recovery during fallow stages is often constrained by competition with herbaceous and shrub species, particularly invasive plants (Diyarzola and Bernard \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eC. odorata\u003c/em\u003e is among the most widespread invasive species in agricultural fallows of Mongala Province, where it forms dense shrub layers in recently abandoned fields.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\n\u003ch2\u003e2.2. Data collection\u003c/h2\u003e\n\u003cp\u003eNatural regeneration of TOF was assessed through a field-based seedling inventory conducted in agricultural fallows across Mongala Province, (DRC). The study focused on five dominant TOF species (\u003cem\u003eP. macrocarpus\u003c/em\u003e, \u003cem\u003eP. angolensis\u003c/em\u003e, \u003cem\u003eR. heudelotii\u003c/em\u003e, \u003cem\u003eE. suaveolens\u003c/em\u003e and \u003cem\u003eP. africanum\u003c/em\u003e) and aimed to quantify how regeneration varies across gradients of fallow age, \u003cem\u003eC. odorata\u003c/em\u003e cover, and distance from seed trees.\u003c/p\u003e\n\u003cp\u003eThe sampling design was structured to capture variability in both successional stage and invasion intensity while ensuring spatial representativeness. Fifteen villages were selected across the three administrative territories of Mongala Province to reflect the range of agroecological and land-use conditions in the study area. Within each village, agricultural fallows were identified in collaboration with local farmers to determine fallow age and recent land-use history.\u003c/p\u003e\n\u003cp\u003eFallow age was classified into three categories (0\u0026ndash;3 years, 3\u0026ndash;6 years and \u0026gt;\u0026thinsp;6 years), representing early, intermediate and advanced stages of post-cultivation succession. The cover of \u003cem\u003eC. odorata\u003c/em\u003e was visually estimated around each seed tree and assigned to four ordinal classes (C1: 0\u0026ndash;25%, C2: 25\u0026ndash;50%, C3: 50\u0026ndash;75% and C4: \u0026gt;75%), capturing increasing levels of invasion.\u003c/p\u003e\n\u003cp\u003eFor each of the five species, 60 mature seed trees were selected as focal sampling units, resulting in a total of 300 seed trees. Seed trees were distributed across fallow age classes and invasion levels to ensure that all combinations of successional stage and \u003cem\u003eC. odorata\u003c/em\u003e cover were represented at the landscape scale. While minor imbalances occurred at the village level due to local availability constraints, the overall design remained well balanced across treatments.\u003c/p\u003e\n\u003cp\u003eAround each seed tree, seedling abundance was assessed using four rectangular transects oriented along the cardinal directions. Each transect was 40 m long and 3 m wide, corresponding to the average radius of cultivated fields, and was subdivided into eight segments of 5 m. This design allowed the spatial distribution of seedlings to be captured along a gradient of distance from the seed tree while standardising sampling effort across sites.\u003c/p\u003e\n\u003cp\u003eAll seedlings encountered within transects were counted and assigned to one of three height classes (0\u0026ndash;30 cm, 30\u0026ndash;50 cm and \u0026gt;\u0026thinsp;50 cm), providing information on both recruitment and size structure. Seedling counts were subsequently aggregated at the seed-tree level prior to statistical analysis, so that each seed tree represented an independent sampling unit associated with a specific combination of fallow age, invasion level and local environmental conditions. This aggregation reduced the risk of pseudoreplication associated with multiple transects and ensured consistency with the analytical framework used in the study.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n\u003ch2\u003e2.3. Data processing and analysis\u003c/h2\u003e\n\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.1. Data preparation\u003c/h2\u003e\n\u003cp\u003eAll statistical analyses were conducted using the R statistical environment (R version 4.5.3; R Core Team, 2025). The dataset comprised seedling counts recorded around individual seed trees across combinations of fallow age, \u003cem\u003eC. odorata\u003c/em\u003e cover, species identity, and distance from the seed tree.\u003c/p\u003e\n\u003cp\u003eSeedling density was expressed as the number of individuals per hectare (ind. ha⁻\u0026sup1;). Total seedling density was calculated as the sum of individuals across all height classes. Additional variables described seedling structure by three height classes (0\u0026ndash;30 cm, 30\u0026ndash;50 cm, and \u0026gt;\u0026thinsp;50 cm).\u003c/p\u003e\n\u003cp\u003eExplanatory variables were defined as follows:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eFallow age\u003c/strong\u003e: three ordered classes (0\u0026ndash;3 years, 3\u0026ndash;6 years, \u0026gt;\u0026thinsp;6 years);\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eC. odorata\u003c/strong\u003e \u003cstrong\u003ecover\u003c/strong\u003e: four ordered levels (C1: 0\u0026ndash;25%, C2: 25\u0026ndash;50%, C3: 50\u0026ndash;75%, C4: \u0026gt;75%);\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eSpecies\u003c/strong\u003e: five focal tree species (\u003cem\u003eE. suaveolens\u003c/em\u003e, \u003cem\u003eP. macrocarpus\u003c/em\u003e, \u003cem\u003eR. heudelotii\u003c/em\u003e, \u003cem\u003eP. angolensis\u003c/em\u003e, and \u003cem\u003eP. africanum\u003c/em\u003e).\u003c/p\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eDistance from the seed tree was treated as a continuous variable.\u003c/p\u003e\n\u003cp\u003ePrior to analysis, data were checked for consistency, missing values, and outliers. Distributions of response variables were examined using histograms and boxplots (see Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e). Seedling density exhibited strong right-skewness, high variance, and a large proportion of zero counts, indicating overdispersion and zero inflation typical of ecological count data.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.2. Modelling of seedling density\u003c/h2\u003e\n\u003cp\u003eTo analyse the effects of fallow age, \u003cem\u003eC. odorata\u003c/em\u003e cover, species identity, and distance on seedling density, we fitted zero-inflated negative binomial (ZINB) models using the \u003cem\u003eglmmTMB\u003c/em\u003e package. This modelling framework was selected to simultaneously account for: (i) overdispersion in count data, and (ii) excess zeros arising from ecological processes such as recruitment failure.\u003c/p\u003e\n\u003cp\u003eThe final model structure was:\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\( \\text{Seedling density}\\sim \\text{species}\\times \\text{fallow age}\\times \\text{cover}+\\text{ns(distance, df = 3)}\\)\u003c/span\u003e \u003c/span\u003e\u003c/p\u003e\u003cp\u003ewith a zero-inflation component specified as:\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\( \\text{Zero inflation}\\sim \\text{fallow age}+\\text{cover}\\)\u003c/span\u003e \u003c/span\u003e\u003c/p\u003e\u003cp\u003eDistance from the seed tree was modelled using a natural spline (df\u0026thinsp;=\u0026thinsp;3) to capture non-linear spatial patterns of seed dispersal and establishment.\u003c/p\u003e\n\u003cp\u003eModel coefficients were interpreted on the log scale and exponentiated to obtain rate ratios. Estimated marginal means (EMMs) and associated confidence intervals were computed using the \u003cem\u003eemmeans\u003c/em\u003e package and used to generate predicted response curves and interaction plots.\u003c/p\u003e\n\u003cp\u003ePairwise comparisons among \u003cem\u003eC. odorata\u003c/em\u003e cover levels within each fallow age class and species were performed using Tukey-adjusted contrasts to control for multiple testing. Results were presented as:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003ea simplified summary table in the main text (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e), and\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003ea full set of contrasts in the Supplementary Material (Table S2).\u003c/p\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.3. Spatial patterns of seedling recruitment\u003c/h2\u003e\n\u003cp\u003eSpatial variation in seedling density relative to distance from the seed tree was analysed within the ZINB modelling framework described above. The inclusion of a spline function of distance allowed flexible modelling of non-linear dispersal and establishment patterns.\u003c/p\u003e\n\u003cp\u003ePredicted seedling densities were generated across continuous gradients of distance and invasion levels for each fallow age class and species. These predictions were visualised using heatmaps, facilitating the interpretation of interactions between spatial processes, successional stage, and invasion intensity.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.4. Analysis of seedling size structure\u003c/h2\u003e\n\u003cp\u003eTo characterise regeneration dynamics beyond total density, seedling populations were analysed by height class.\u003c/p\u003e\n\u003cp\u003eRelative proportions of each size class were calculated as:\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\( {p}_{i}=\\frac{\\text{density of size class }i}{\\text{total seedling density}}\\)\u003c/span\u003e \u003c/span\u003e\u003c/p\u003e\u003cp\u003eVariation in size-class composition across fallow age and invasion gradients was analysed descriptively and visualised using stacked bar charts. In addition, ternary plots were used to represent the relative contribution of each size class across environmental gradients, providing an integrated view of regeneration structure (Fig. S5).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.5. Model diagnostics and validation\u003c/h2\u003e\n\u003cp\u003eModel adequacy was assessed using simulation-based diagnostics implemented in the \u003cem\u003eDHARMa\u003c/em\u003e package. Diagnostic procedures included tests of residual uniformity, dispersion, zero inflation, and outliers.\u003c/p\u003e\n\u003cp\u003eResults indicated that the zero-inflated negative binomial model adequately captured the excess of zero observations (zero-inflation test: non-significant) and did not exhibit problematic outliers. Although dispersion and uniformity tests indicated statistically significant deviations, visual inspection of residual plots suggested only moderate departures from model assumptions, which are common in large ecological count datasets.\u003c/p\u003e\n\u003cp\u003eDiagnostic results supported the adequacy of the model despite minor deviations typical of large ecological datasets (Fig. S2, Fig. S3; Text S2).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\n\u003ch2\u003e2.3.6. Data visualisation\u003c/h2\u003e\n\u003cp\u003eAll figures were produced using the \u003cem\u003eggplot2\u003c/em\u003e package, with additional extensions (\u003cem\u003eggdist\u003c/em\u003e, \u003cem\u003epatchwork\u003c/em\u003e, and \u003cem\u003eggtern\u003c/em\u003e) for advanced visualisation. Figures were designed to maximise interpretability and facilitate comparison across species, fallow age classes, invasion levels, and spatial gradients.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Dataset structure and distribution of observations\u003c/h2\u003e \u003cp\u003eThe dataset followed a fully balanced factorial design combining fallow age (0\u0026ndash;3, 3\u0026ndash;6, \u0026gt;\u0026thinsp;6 years), Chromolaena odorata cover (C1-C4), and five tree species, resulting in an equal number of observations per treatment combination (n\u0026thinsp;=\u0026thinsp;160). This balanced structure ensured robust estimation of main and interaction effects while minimising potential biases associated with unequal sampling effort (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Effects of fallow age and invasion level on total seedling density\u003c/h2\u003e \u003cp\u003eSeedling density varied markedly across both fallow age and Chromolaena odorata cover gradients (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The zero-inflated negative binomial (ZINB) model revealed significant effects of fallow age, invasion level, and their interaction (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating that the impact of invasion depends strongly on successional stage.\u003c/p\u003e \u003cp\u003ePredicted seedling densities declined sharply with increasing invasion intensity, particularly in early fallows (0\u0026ndash;3 years). Model-based estimates showed a decrease from 1051 ind. ha⁻\u0026sup1; under low invasion (C1) to 270 ind. ha⁻\u0026sup1; under high invasion (C4), corresponding to a reduction of approximately 74%. This strong decline was consistent across descriptive statistics, which showed a reduction in mean density from 1343 to 203 ind. ha⁻\u0026sup1; across the same gradient.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive statistics of seedling density (individuals ha⁻\u0026sup1;) across fallow age and \u003cem\u003eChromolaena odorata\u003c/em\u003e cover gradients.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFallow age\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eC. odorata\u003c/em\u003e cover\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003en\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMedian\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eQ1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eQ3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMin\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eMax\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1343\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1185\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1270\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e197\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e2235\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e5264\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e771\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e673\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e646\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3394\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e324\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e240\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e304\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e488\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1565\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e203\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e208\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e291\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1279\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e932\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e772\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e168\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1524\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e6141\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e803\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e774\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e586\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e201\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1210\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3749\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e621\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e772\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e324\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e757\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3542\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e668\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e832\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e242\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1060\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3481\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e704\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e496\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e194\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1222\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3226\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e651\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e590\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e266\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e2811\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e706\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e659\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e539\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e134\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1070\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3077\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e615\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1244\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e2779\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn intermediate fallows (3\u0026ndash;6 years), invasion effects remained significant but were less pronounced, with predicted densities declining from 784 ind. ha⁻\u0026sup1; (C1) to 454 ind. ha⁻\u0026sup1; (C4), representing a reduction of approximately 42%. In contrast, in older fallows (\u0026gt;\u0026thinsp;6 years), predicted densities were relatively stable across invasion levels (698\u0026ndash;760 ind. ha⁻\u0026sup1;), indicating a marked attenuation of invasion effects.\u003c/p\u003e \u003cp\u003eThe zero-inflation component of the model further indicated that the probability of structural zeros increased under high invasion levels and in early fallows, suggesting that invasion not only reduces seedling density but also increases the likelihood of recruitment failure.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Species-specific responses to fallow age and invasion gradients\u003c/h2\u003e \u003cp\u003eSpecies responses to invasion and succession varied significantly, as indicated by a strong three-way interaction between species, fallow age, and invasion level (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Predicted marginal means revealed contrasting response patterns among species (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Table S4).\u003c/p\u003e \u003cp\u003e \u003cem\u003eE. suaveolens\u003c/em\u003e exhibited the strongest negative response to invasion, with predicted densities declining from 1454 ind. ha⁻\u0026sup1; under low invasion (C1) to 149 ind. ha⁻\u0026sup1; under high invasion (C4) in early fallows, corresponding to a reduction of approximately 90%. Similarly, Pycnanthus angolensis showed pronounced declines across invasion gradients, particularly in early and intermediate fallows.\u003c/p\u003e \u003cp\u003e \u003cem\u003eP. africanum\u003c/em\u003e showed a moderate response, with significant declines under high invasion in early and intermediate fallows, but weaker and non-significant responses in older fallows.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn contrast, \u003cem\u003eP. macrocarpus\u003c/em\u003e exhibited a positive response to invasion, with predicted densities increasing from 203 ind. ha⁻\u0026sup1; (C1) to 396 ind. ha⁻\u0026sup1; (C4) in early fallows, and up to 1090 ind. ha⁻\u0026sup1; under high invasion in older fallows, representing more than a fourfold increase.\u003c/p\u003e \u003cp\u003e \u003cem\u003eR. heudelotii\u003c/em\u003e showed relatively stable responses across invasion gradients, particularly in intermediate and older fallows, where differences among invasion levels were small and non-significant.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThese results indicate that \u003cem\u003eC. odorata\u003c/em\u003e acts as a strong ecological filter, suppressing regeneration in some species while favouring others, thereby generating substantial interspecific variability in regeneration dynamics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.4. Spatial patterns of seedling recruitment\u003c/h2\u003e \u003cp\u003eSeedling density decreased non-linearly with increasing distance from seed trees, as captured by the spline term included in the ZINB model (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). This decline was steepest in early fallows, indicating strong dispersal limitation during initial successional stages.\u003c/p\u003e \u003cp\u003ePredicted densities were highest near seed trees and declined rapidly within the first tens of metres, after which the decline became more gradual. In intermediate and older fallows, spatial gradients were less pronounced, suggesting more diffuse recruitment patterns.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eInvasion intensity significantly modified these spatial patterns. High \u003cem\u003eC. odorata\u003c/em\u003e cover reduced seedling density across all distances and flattened spatial gradients, resulting in more homogeneous distributions (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). This indicates that invasion weakens distance-dependent recruitment processes and reduces spatial heterogeneity in seedling establishment.\u003c/p\u003e \u003cp\u003eOverall, these results demonstrate that regeneration patterns are jointly structured by dispersal limitation and environmental filtering, with invasion affecting both the magnitude and spatial configuration of recruitment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.5. Changes in regeneration structure across invasion gradients\u003c/h2\u003e \u003cp\u003eSeedling size structure varied systematically across invasion gradients and fallow age classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Across all conditions, the smallest size class (0\u0026ndash;30 cm) dominated seedling populations.\u003c/p\u003e \u003cp\u003eHowever, the proportion of larger seedlings (\u0026gt;\u0026thinsp;50 cm) declined consistently with increasing invasion intensity. Under high invasion levels (C3-C4), the relative contribution of larger size classes was substantially reduced compared to low invasion conditions, indicating a shift towards younger developmental stages.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThis pattern suggests that \u003cem\u003eC. odorata\u003c/em\u003e not only limits seedling recruitment but also constrains growth and progression to later developmental stages, potentially affecting long-term regeneration trajectories.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.6. Pairwise comparisons of invasion effects\u003c/h2\u003e \u003cp\u003ePairwise comparisons based on estimated marginal means confirmed significant differences in seedling density across invasion levels for most species (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Table S2). High invasion levels (C3 and C4) were generally associated with significantly lower seedling densities compared to low invasion (C1), particularly in early fallows.\u003c/p\u003e \u003cp\u003eFor example, in Erythrophleum suaveolens, rate ratios decreased from 0.53 (C2 vs C1) to 0.06 (C4 vs C1) in early fallows, indicating a progressive and substantial decline in seedling density with increasing invasion intensity.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePairwise comparisons of seedling density across invasion levels within fallow age classes (rate ratios from negative binomial models).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpecies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFallow age\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC2 vs C1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC3 vs C1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC4 vs C1\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eE. suaveolens\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.53***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.22***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.06***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.60***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.30***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.12***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.70***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.39***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.15***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eP. africanum\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.73*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.28***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.19***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.81 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.21***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.17***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.91 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.61***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.63***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eP. angolensis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.50***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.16***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.13***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.76 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.26***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.16***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.92 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.97 ns\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eP. macrocarpus\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.44 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.36***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.60***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.51*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.72***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7.32***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.76***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.15***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.51***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eR. heudelotii\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026ndash;3 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.52***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.13***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.08***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026ndash;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.00 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.04 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.07 ns\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;6 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.07 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00 ns\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.01 ns\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cb\u003eNote\u003c/b\u003e: \u003cem\u003eValues represent rate ratios relative to low invasion (C1). Values\u0026thinsp;\u0026lt;\u0026thinsp;1 indicate a decrease in seedling density, whereas values\u0026thinsp;\u0026gt;\u0026thinsp;1 indicate an increase. Significance levels: *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; ns\u0026thinsp;=\u0026thinsp;not significant\u003c/em\u003e.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn contrast, \u003cem\u003eP. macrocarpus\u003c/em\u003e showed the opposite pattern, with rate ratios exceeding 1 under moderate to high invasion levels and reaching up to 8.51 in older fallows, indicating a strong positive response to invasion.\u003c/p\u003e \u003cp\u003eOther species, such as \u003cem\u003eR. heudelotii\u003c/em\u003e, showed no significant differences among invasion levels in intermediate and older fallows, confirming species-specific variability in sensitivity to invasion.\u003c/p\u003e \u003cp\u003eOverall, these pairwise comparisons reinforce the results of the main model and highlight the contrasting responses of species to invasion across successional stages.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Successional stage as a major driver of regeneration dynamics\u003c/h2\u003e \u003cp\u003eThis study demonstrates that fallow age is a major determinant of tree regeneration dynamics in Central African agricultural landscapes, but its effect is strongly contingent on invasion intensity. While seedling density was highest in early fallows under low invasion (\u0026asymp;\u0026thinsp;1050 ind. ha⁻\u0026sup1;), this advantage was rapidly eroded under high \u003cem\u003eC. odorata\u003c/em\u003e cover, with densities declining to \u0026asymp;\u0026thinsp;270 ind. ha⁻\u0026sup1;, representing a reduction of more than 70%.\u003c/p\u003e \u003cp\u003eThis result refines the classical view that early successional stages favour regeneration due to higher light availability and reduced structural competition (Poorter et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Our findings show that this advantage is conditional, and can be overridden by strong competitive exclusion from invasive species.\u003c/p\u003e \u003cp\u003eIn contrast, older fallows (\u0026gt;\u0026thinsp;6 years) maintained relatively stable regeneration levels across invasion gradients (\u0026asymp;\u0026thinsp;700\u0026ndash;760 ind. ha⁻\u0026sup1;), suggesting that successional development buffers the negative effects of invasion. This pattern indicates that regeneration dynamics are not simply driven by successional stage, but by interactions between succession and biotic filtering processes.\u003c/p\u003e \u003cp\u003eThese findings support a more nuanced view of succession, where early stages represent both a window of opportunity for recruitment and a period of high vulnerability to invasion, while later stages provide greater ecological resistance to competitive exclusion, consistent with increasing biotic resistance and structural complexity along successional gradients (Chazdon et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Poorter et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Strong but non-linear effects of \u003cem\u003eChromolaena odorata\u003c/em\u003e on regeneration\u003c/h2\u003e \u003cp\u003eThe results provide strong evidence that increasing \u003cem\u003eC. odorata\u003c/em\u003e cover reduces seedling density, but this effect is clearly non-linear and threshold-dependent. Across species, major declines occurred primarily between low (C1) and high invasion levels (C3-C4), with reductions exceeding 70\u0026ndash;90% in early fallows for several species.\u003c/p\u003e \u003cp\u003eThis pattern suggests the existence of ecological thresholds, beyond which invasion shifts from a moderate influence to a dominant limiting factor for regeneration. Below these thresholds, regeneration may still occur, albeit at reduced levels; above them, dense shrub cover likely imposes severe constraints through light limitation, space pre-emption, and possibly allelopathic effects (Kato-Noguchi and Kato \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eImportantly, the zero-inflation component of the model showed that invasion increased the probability of structural zeros, particularly in early fallows. This indicates that invasion does not merely reduce seedling abundance but also increases the likelihood of complete recruitment failure, a critical but often overlooked mechanism in regeneration studies.\u003c/p\u003e \u003cp\u003eThese results extend previous findings by demonstrating that invasion impacts are not gradual but may involve non-linear transitions in regeneration success, reinforcing the need to explicitly account for invasion intensity rather than presence alone, as ecological impacts of invasive species are often strongly dependent on their abundance and may exhibit threshold responses (Gb\u0026egrave;toho et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Pyšek et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Interactive effects of succession and invasion: a context-dependent process\u003c/h2\u003e \u003cp\u003eA central contribution of this study is the demonstration that the impact of \u003cem\u003eC. odorata\u003c/em\u003e is strongly modulated by successional stage. The negative effects of invasion were most pronounced in early fallows, where density reductions exceeded 70%, but became negligible in older fallows.\u003c/p\u003e \u003cp\u003eThis interaction highlights that early successional stages constitute a critical ecological bottleneck, where recruitment is both highly active and highly sensitive to environmental filtering. In such conditions, invasive shrubs can effectively suppress regeneration and potentially alter successional trajectories.\u003c/p\u003e \u003cp\u003eIn contrast, the reduced impact of invasion in older fallows suggests that increasing vegetation complexity, canopy development, and microclimatic buffering enhance ecosystem resistance to invasion. This aligns with the concept of biotic resistance, whereby more developed plant communities limit the establishment and impact of invasive species (Chazdon et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOverall, these findings demonstrate that invasion effects cannot be generalised across successional gradients, but instead emerge from dynamic interactions between ecosystem development and competitive processes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Species-specific responses and ecological filtering\u003c/h2\u003e \u003cp\u003eThe strong interspecific variation observed in this study confirms that \u003cem\u003eC. odorata\u003c/em\u003e acts as a selective ecological filter shaping regeneration patterns. Species such as \u003cem\u003eE. suaveolens\u003c/em\u003e experienced drastic declines (\u0026gt;\u0026thinsp;90%) under high invasion in early fallows, whereas \u003cem\u003eP. macrocarpus\u003c/em\u003e showed a marked increase in density, with values exceeding four times those observed under low invasion.\u003c/p\u003e \u003cp\u003eThese contrasting responses suggest that invasion favours species with traits adapted to shaded and competitive environments, while excluding more light-demanding or competition-sensitive species. This pattern is consistent with trait-based community assembly theory, where environmental filters select species according to functional traits (Funk et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Kraft et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe relatively stable response of \u003cem\u003eR. heudelotii\u003c/em\u003e across invasion gradients further illustrates that some species may be tolerant to a wide range of environmental conditions, contributing to their persistence in disturbed landscapes.\u003c/p\u003e \u003cp\u003eSuch species-specific responses have important implications for long-term forest composition. By selectively favouring certain species, invasion may lead to functional and compositional shifts, potentially reducing diversity and altering ecosystem functioning even when overall seedling abundance remains high.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e4.5. Spatial constraints on regeneration and their interaction with invasion\u003c/h2\u003e \u003cp\u003eThe observed decline in seedling density with increasing distance from seed trees confirms the central role of dispersal limitation in structuring regeneration patterns. This effect was particularly strong in early fallows, where recruitment was highly concentrated near seed sources, indicating limited dispersal distances and strong spatial clustering.\u003c/p\u003e \u003cp\u003eHowever, invasion significantly modified these spatial dynamics. High \u003cem\u003eC. odorata\u003c/em\u003e cover reduced seedling density across all distances and flattened spatial gradients, leading to more homogeneous distributions. This suggests that invasion weakens distance-dependent recruitment processes, likely by imposing strong environmental constraints that override dispersal-driven patterns.\u003c/p\u003e \u003cp\u003eThese findings highlight the joint role of dispersal limitation and environmental filtering in shaping regeneration, and demonstrate that invasive species can alter not only the magnitude but also the spatial structure of recruitment.\u003c/p\u003e \u003cp\u003eSuch alterations in spatial patterns may have cascading effects on community assembly and spatial heterogeneity, which are key components of ecosystem resilience and biodiversity maintenance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e4.6. Implications for regeneration dynamics and management\u003c/h2\u003e \u003cp\u003eThe results of this study have important implications for the management of agricultural fallows and forest restoration in Central Africa.\u003c/p\u003e \u003cp\u003eFirst, the high regeneration potential observed in early fallows under low invasion confirms their importance as key regeneration niches. However, the strong decline in seedling density under high invasion (up to 70\u0026ndash;90%) indicates that these systems are highly vulnerable to competitive exclusion during early successional stages.\u003c/p\u003e \u003cp\u003eSecond, the identification of strong invasion effects in early fallows suggests that management interventions should prioritise early-stage control of \u003cem\u003eC. odorata\u003c/em\u003e, before dense stands become established. Preventing invasion during this critical window is likely to be more effective than attempting to restore regeneration after invasion thresholds have been exceeded.\u003c/p\u003e \u003cp\u003eThird, the marked species-specific responses observed highlight the need for context-dependent management strategies. While some species are strongly suppressed by invasion, others may tolerate or even benefit from altered environmental conditions. Management approaches should therefore consider species traits and ecological strategies rather than applying uniform interventions.\u003c/p\u003e \u003cp\u003eFinally, the observed effects of invasion on both spatial patterns and size structure indicate that invasion can influence not only recruitment but also subsequent growth and community development. This underscores the importance of integrating multiple dimensions of regeneration (density, structure, spatial patterns) in restoration planning.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study provides a robust quantitative assessment of how successional dynamics and biological invasion jointly shape tree regeneration in agricultural fallows of Central Africa. By integrating field data with zero-inflated modelling approaches, we demonstrate that fallow age and \u003cem\u003eChromolaena odorata\u003c/em\u003e cover interact strongly to determine not only seedling density, but also the spatial structure and developmental trajectory of regeneration.\u003c/p\u003e \u003cp\u003eOur results show that early fallows constitute a critical phase for regeneration, characterised by high recruitment potential under low invasion but also high vulnerability to competitive exclusion. Under high invasion levels, seedling density declined by more than 70%, and the probability of recruitment failure increased substantially, highlighting the sensitivity of early successional stages to biotic constraints. In contrast, older fallows exhibited more stable regeneration across invasion gradients, suggesting increasing ecological resistance with successional development.\u003c/p\u003e \u003cp\u003eImportantly, the effects of \u003cem\u003eC. odorata\u003c/em\u003e were not gradual but strongly non-linear, with marked declines in regeneration occurring beyond intermediate levels of invasion. This threshold-like behaviour indicates that invasion intensity is a key determinant of ecological impact and should be explicitly considered when assessing regeneration processes and designing management interventions.\u003c/p\u003e \u003cp\u003eThe study further reveals pronounced species-specific responses to invasion. While several species experienced strong declines in seedling density under high invasion, others showed neutral or even positive responses, indicating that invasion acts as a selective ecological filter shaping regeneration trajectories and potentially altering long-term community composition.\u003c/p\u003e \u003cp\u003eBeyond overall density patterns, invasion was also found to modify spatial recruitment processes and seedling size structure, reducing distance-dependent gradients and limiting the progression of seedlings to larger size classes. These findings demonstrate that invasion affects multiple dimensions of regeneration, with potential cascading effects on forest structure and recovery dynamics.\u003c/p\u003e \u003cp\u003eFrom a management perspective, our results emphasise the importance of early intervention. Preventing the establishment of dense Chromolaena odorata stands during early successional stages appears critical to maintaining regeneration potential. At the same time, the strong interspecific variability observed suggests that management strategies should be adapted to ecological context and species-specific responses rather than relying on uniform control approaches.\u003c/p\u003e \u003cp\u003eOverall, this study highlights that regeneration in human-modified tropical landscapes is governed by complex and context-dependent interactions between succession, invasion intensity, dispersal processes, and species traits. By demonstrating the importance of non-linear invasion effects and successional context, it contributes to a more mechanistic understanding of forest recovery processes outside forests and provides a foundation for more targeted and effective restoration strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eJean Pierre Azenge conceptualised the study, designed the methodology, coordinated all research data collection and analysis interpretation, and drafted the manuscript. Paxie W. Chirwa and Justin N\u0026rsquo;Dja Kassi provided crucial academic supervision throughout the study, offering substantial intellectual inputs and critical revisions to the manuscript. All authors critically reviewed and approved the final version of the manuscript submitted for publication.\u003c/p\u003e\u003ch2\u003e6. Acknowledgements\u003c/h2\u003e \u003cp\u003eWe thank the Regional Scholarship and Innovation Fund (RSIF) for the scholarship that made this research possible.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAzenge JP, Wassila IS, Kassi JN, Chirwa PW (2025) Diversity and ethnobotanical use-value of trees outside forests on the agricultural landscape of the Mongala Province, Democratic Republic of Congo. Agrofor Syst 99:1\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10457-025-01332-3\u003c/span\u003e\u003cspan address=\"10.1007/s10457-025-01332-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL (2014) Second Growth The Promise of Tropical Forest Regeneration in an Age of Deforestation, Th e Unive. Th e University of Chicago, Chicago and London\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL, Harvey CA, Komar O et al (2009) Beyond reserves: A research agenda for conserving biodiversity in human-modified tropical landscapes. Biotropica 41:142\u0026ndash;153. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1744-7429.2008.00471.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1744-7429.2008.00471.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL, Lindenmayer D, Guariguata MR et al (2020) Fostering natural forest regeneration on former agricultural land through economic and policy interventions. Environ Res Lett 15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/ab79e6\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/ab79e6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCurtis PG, Slay CM, Harris NL et al (2018) Classifying drivers of global forest loss. Sci (80-) 361:1108\u0026ndash;1111\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiyarzola J, Bernard C (2024) Diagnostic agraire de la province de la Mongala en RDC. Enabel-RDC, Kinshasa, R\u0026eacute;publique D\u0026eacute;mocratique du Congo\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO (2016) Trees, forests and land use in drylands: the first global assessment. Rome\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFunk JL, Cleland EE, Suding KN, Zavaleta ES (2008) Restoration through reassembly: plant traits and invasion resistance. Trends Ecol Evol 23:695\u0026ndash;703. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tree.2008.07.013\u003c/span\u003e\u003cspan address=\"10.1016/j.tree.2008.07.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGb\u0026egrave;toho AJ, Kingbo A, Gnangu\u0026egrave;non-gu\u0026eacute;ss\u0026egrave; D et al (2018) Impacts of Chromolaena odorata on native trees\u0026rsquo; regeneration in the Lama secondary forests in Benin, West Africa. Bois Forets des Trop 338:5\u0026ndash;14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHardy NG, Kuebbing SE, Duguid MC et al (2025) Non-native invasive plants in tropical dry forests: a global review of presence, impacts, and management. Restor Ecol 33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/rec.14288\u003c/span\u003e\u003cspan address=\"10.1111/rec.14288\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJuru VN, Ndam LM, Tatah BN, Fonge BA (2024) Rhizospheric soil chemical properties and microbial response to a gradient of Chromolaena odorata(L) invasion in the Mount Cameroon Region. PLoS ONE 19:1\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0312199\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0312199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKato-Noguchi H, Kato M (2023) Evolution of the Secondary Metabolites in Invasive Plant Species Chromolaena odorata for the Defense and Allelopathic Functions. Plants 12:6\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/plants12030521\u003c/span\u003e\u003cspan address=\"10.3390/plants12030521\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKraft NJB, Adler PB, Godoy O et al (2015) Community assembly, coexistence and the environmental filtering metaphor. Funct Ecol 29:592\u0026ndash;599. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2435.12345\u003c/span\u003e\u003cspan address=\"10.1111/1365-2435.12345\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoubota Panzou GJ, Mankessi F, Tsiba Ngambou FC et al (2024) Natural forest regeneration over a fallow age chronosequence in central African moist forests. Afr J Ecol 62:1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/aje.13255\u003c/span\u003e\u003cspan address=\"10.1111/aje.13255\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMpanda M, Kashindye A, Aynekulu E et al (2021) Forests, farms, and fallows: The dynamics of tree cover transition in the southern part of the uluguru mountains, tanzania. Land 10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land10060571\u003c/span\u003e\u003cspan address=\"10.3390/land10060571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMwampamba TH, Schwartz MW (2011) The effects of cultivation history on forest recovery in fallows in the Eastern Arc Mountain, Tanzania. Ecol Manage 261:1042\u0026ndash;1052. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.foreco.2010.12.026\u003c/span\u003e\u003cspan address=\"10.1016/j.foreco.2010.12.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoorter L, Craven D, Jakovac CC et al (2021) Multidimensional tropical forest recovery. Sci (80-) 374:1370\u0026ndash;1376\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoorter L, Rozendaal D, Bongers F et al (2019) Wet and dry tropical forests show opposite successional pathways in wood density but converge over time. Nat Ecol Evol 3:928\u0026ndash;934. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41559-019-0882-6\u003c/span\u003e\u003cspan address=\"10.1038/s41559-019-0882-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePyšek P, Hulme PE, Simberloff D et al (2020) Scientists\u0026rsquo; warning on invasive alien species. Biol Rev 95:1511\u0026ndash;1534. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/brv.12627\u003c/span\u003e\u003cspan address=\"10.1111/brv.12627\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRai PK, Singh JS (2024) Ecological insights and environmental threats of invasive alien plant Chromolaena odorata: Prospects for sustainable management. Weed Biol Manag 24:15\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/wbm.12286\u003c/span\u003e\u003cspan address=\"10.1111/wbm.12286\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReed J, Van Vianen J, Deakin EL et al (2016) Integrated landscape approaches to managing social and environmental issues in the tropics: learning from the past to guide the future. Glob Chang Biol 22:2540\u0026ndash;2554. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.13284\u003c/span\u003e\u003cspan address=\"10.1111/gcb.13284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozendaal DMA, Bongers F, Aide TM et al (2019) Biodiversity recovery of Neotropical secondary forests. Sci Adv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.aau3114\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.aau3114\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. 5:\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZomer RJ, Neufeldt H, Xu J et al (2016) Global Tree Cover and Biomass Carbon on Agricultural Land: The contribution of agroforestry to global and national carbon budgets. Sci Rep 6:1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep29987\u003c/span\u003e\u003cspan address=\"10.1038/srep29987\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAzenge JP, Wassila IS, Kassi JN, Chirwa PW (2025) Diversity and ethnobotanical use-value of trees outside forests on the agricultural landscape of the Mongala Province, Democratic Republic of Congo. Agrofor Syst 99:1\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10457-025-01332-3\u003c/span\u003e\u003cspan address=\"10.1007/s10457-025-01332-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL (2014) Second Growth The Promise of Tropical Forest Regeneration in an Age of Deforestation, Th e Unive. Th e University of Chicago, Chicago and London\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL, Harvey CA, Komar O et al (2009) Beyond reserves: A research agenda for conserving biodiversity in human-modified tropical landscapes. Biotropica 41:142\u0026ndash;153. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1744-7429.2008.00471.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1744-7429.2008.00471.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChazdon RL, Lindenmayer D, Guariguata MR et al (2020) Fostering natural forest regeneration on former agricultural land through economic and policy interventions. Environ Res Lett 15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/ab79e6\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/ab79e6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCurtis PG, Slay CM, Harris NL et al (2018) Classifying drivers of global forest loss. Sci (80-) 361:1108\u0026ndash;1111\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiyarzola J, Bernard C (2024) Diagnostic agraire de la province de la Mongala en RDC. Enabel-RDC, Kinshasa, R\u0026eacute;publique D\u0026eacute;mocratique du Congo\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO (2016) Trees, forests and land use in drylands: the first global assessment. Rome\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFunk JL, Cleland EE, Suding KN, Zavaleta ES (2008) Restoration through reassembly: plant traits and invasion resistance. Trends Ecol Evol 23:695\u0026ndash;703. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tree.2008.07.013\u003c/span\u003e\u003cspan address=\"10.1016/j.tree.2008.07.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGb\u0026egrave;toho AJ, Kingbo A, Gnangu\u0026egrave;non-gu\u0026eacute;ss\u0026egrave; D et al (2018) Impacts of Chromolaena odorata on native trees\u0026rsquo; regeneration in the Lama secondary forests in Benin, West Africa. Bois Forets des Trop 338:5\u0026ndash;14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHardy NG, Kuebbing SE, Duguid MC et al (2025) Non-native invasive plants in tropical dry forests: a global review of presence, impacts, and management. Restor Ecol 33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/rec.14288\u003c/span\u003e\u003cspan address=\"10.1111/rec.14288\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJuru VN, Ndam LM, Tatah BN, Fonge BA (2024) Rhizospheric soil chemical properties and microbial response to a gradient of Chromolaena odorata(L) invasion in the Mount Cameroon Region. PLoS ONE 19:1\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0312199\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0312199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKato-Noguchi H, Kato M (2023) Evolution of the Secondary Metabolites in Invasive Plant Species Chromolaena odorata for the Defense and Allelopathic Functions. Plants 12:6\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/plants12030521\u003c/span\u003e\u003cspan address=\"10.3390/plants12030521\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKraft NJB, Adler PB, Godoy O et al (2015) Community assembly, coexistence and the environmental filtering metaphor. Funct Ecol 29:592\u0026ndash;599. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2435.12345\u003c/span\u003e\u003cspan address=\"10.1111/1365-2435.12345\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoubota Panzou GJ, Mankessi F, Tsiba Ngambou FC et al (2024) Natural forest regeneration over a fallow age chronosequence in central African moist forests. Afr J Ecol 62:1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/aje.13255\u003c/span\u003e\u003cspan address=\"10.1111/aje.13255\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMpanda M, Kashindye A, Aynekulu E et al (2021) Forests, farms, and fallows: The dynamics of tree cover transition in the southern part of the uluguru mountains, tanzania. Land 10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land10060571\u003c/span\u003e\u003cspan address=\"10.3390/land10060571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMwampamba TH, Schwartz MW (2011) The effects of cultivation history on forest recovery in fallows in the Eastern Arc Mountain, Tanzania. Ecol Manage 261:1042\u0026ndash;1052. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.foreco.2010.12.026\u003c/span\u003e\u003cspan address=\"10.1016/j.foreco.2010.12.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoorter L, Craven D, Jakovac CC et al (2021) Multidimensional tropical forest recovery. Sci (80-) 374:1370\u0026ndash;1376\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoorter L, Rozendaal D, Bongers F et al (2019) Wet and dry tropical forests show opposite successional pathways in wood density but converge over time. Nat Ecol Evol 3:928\u0026ndash;934. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41559-019-0882-6\u003c/span\u003e\u003cspan address=\"10.1038/s41559-019-0882-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePyšek P, Hulme PE, Simberloff D et al (2020) Scientists\u0026rsquo; warning on invasive alien species. Biol Rev 95:1511\u0026ndash;1534. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/brv.12627\u003c/span\u003e\u003cspan address=\"10.1111/brv.12627\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRai PK, Singh JS (2024) Ecological insights and environmental threats of invasive alien plant Chromolaena odorata: Prospects for sustainable management. Weed Biol Manag 24:15\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/wbm.12286\u003c/span\u003e\u003cspan address=\"10.1111/wbm.12286\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReed J, Van Vianen J, Deakin EL et al (2016) Integrated landscape approaches to managing social and environmental issues in the tropics: learning from the past to guide the future. Glob Chang Biol 22:2540\u0026ndash;2554. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.13284\u003c/span\u003e\u003cspan address=\"10.1111/gcb.13284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozendaal DMA, Bongers F, Aide TM et al (2019) Biodiversity recovery of Neotropical secondary forests. Sci Adv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.aau3114\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.aau3114\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. 5:\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZomer RJ, Neufeldt H, Xu J et al (2016) Global Tree Cover and Biomass Carbon on Agricultural Land: The contribution of agroforestry to global and national carbon budgets. Sci Rep 6:1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep29987\u003c/span\u003e\u003cspan address=\"10.1038/srep29987\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Tree regeneration, Chromolaena odorata, Fallow age, Invasion intensity, Non-linear effects, Ecological filtering, Tropical agricultural landscapes","lastPublishedDoi":"10.21203/rs.3.rs-9203304/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9203304/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNatural regeneration of trees outside forests (TOF) is a key process sustaining biodiversity and ecosystem functions in tropical agricultural landscapes, yet it is increasingly shaped by interactions between succession and biological invasion. This study quantified the combined effects of fallow age and \u003cem\u003eChromolaena odorata\u003c/em\u003e cover on tree regeneration in agricultural fallows of Mongala Province (Democratic Republic of the Congo). Seedling inventories were conducted around 300 seed trees across gradients of fallow age (0\u0026ndash;3, 3\u0026ndash;6, \u0026gt;\u0026thinsp;6 years) and invasion intensity (0\u0026ndash;25% to \u0026gt;\u0026thinsp;75% cover). Seedling density, spatial patterns, and size structure were analysed using zero-inflated negative binomial models. Seedling density declined significantly with increasing \u003cem\u003eC. odorata\u003c/em\u003e cover, particularly in early fallows, where predicted densities decreased from 1051 to 270 ind. ha⁻\u0026sup1; (\u0026asymp;\u0026thinsp;74% reduction). This effect weakened along the successional gradient and became negligible in older fallows, indicating a strong interaction between succession and invasion. Invasion effects were non-linear, with sharp declines occurring beyond intermediate levels of cover, suggesting threshold responses and increased probability of recruitment failure. Species responses varied markedly: \u003cem\u003eErythrophleum suaveolens\u003c/em\u003e and \u003cem\u003ePycnanthus angolensis\u003c/em\u003e showed strong declines, whereas \u003cem\u003ePetersianthus macrocarpus\u003c/em\u003e increased under high invasion, indicating species-specific ecological filtering. Invasion also reduced spatial heterogeneity and limited progression to larger seedling size classes. These findings highlight that regeneration is governed by context-dependent interactions between succession and invasion intensity, with early fallows representing both a window of opportunity and a phase of high vulnerability. Accounting for non-linear invasion effects is critical for understanding forest recovery and designing targeted management strategies in tropical agricultural landscapes.\u003c/p\u003e","manuscriptTitle":"Interactive effects of fallow age and Chromolaena odorata invasion on tree regeneration in Central African agricultural landscapes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-08 06:41:39","doi":"10.21203/rs.3.rs-9203304/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8c249f89-f69b-4ff4-81b8-b4085b674d1b","owner":[],"postedDate":"April 8th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-08T06:41:39+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-08 06:41:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9203304","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9203304","identity":"rs-9203304","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}