Assisted natural regeneration of selected tree species through silviculture interventions as a tool for restoring degraded natural forests in northwestern Ethiopia

Abstract Natural forest ecosystem assets and services are increasingly deteriorating due to forest degradation and deforestation in Northern parts of Ethiopia. Enhancing the natural regeneration status of native tree species through different silviculture practices is highly Forest regeneration method recommended. Thus, this study aims to investigate the role of different silvicultural interventions in enhancing the regeneration condition of selected ecological and economical important tree species. Three natural forests stands from different agro-ecologies were chosen and three silvicultural treatments including Fencing, Fence + hoeing, Fence + gap creation, and control (without any intervention) with 400 m2 plot areas established with three replications and randomized complete block design (RCBD) with a total of 36 plots. In each plot, seedling, and saplings under different mod of regeneration were recorded while height and diameter at breast height were measured and analyzed using descriptive and inferential statistics using R version 4.1. There was a significant difference (p < .05) in seedling density and regeneration status among forest stands, methods of regeneration, silvicultural methods, and season of regeneration. The Camboo forest stand recorded the highest regeneration (971 seedlings ha−1); seedlings regenerated from seed recorded the highest regeneration (368 seedlings ha−1), fencing and gap creation silvicultural intervention recorded the highest seedlings (961 seedlings ha−1), the highest regeneration density was recorded at mid-rainy season (570 seedlings ha−1); and Prunus africana and Albizia gummifera have the best regeneration capacity (245 seedlings ha−1) in the study forest stand. The regeneration status of trees under fencing and gap creation plots changed from J-shape to an inverted J-shape before and after silviculture intervention, respectively. The regeneration and population structure of many studied tree species showed particularly good condition in fenced plots than control plots which shows future management interventions for the restoration of degraded natural forest ecosystems in Ethiopia’s tropical forest ecosystems. Some native tree species, such as Hagenia abyssinica, Apodytes dimidata, and Ekebrgia capensis, have a weak response to silvicultural treatments that require advanced regeneration techniques such as clonal nurseries and invitro propagation.


Introduction
Deforestation and forest degradation have a significant impact on biophysical and climatic change. These changes have an impact on the ecology, society, and economy, which can harm populations whose livelihoods rely on the forest (Tigabu 2016). Such deviations result in a decrease in the ecosystem's use of the forest's commodities and services. In other words, the main barrier to ensuring the sustainability of forest resources and preserving terrestrial biodiversity is the increase in demand for forests and forest-related products caused by the world's growing population (Nolet et al. 2018). In the last ten years, as an encounter measure, a forest management strategy concerned with forest ecosystem management, reduction of forest degradation, and deforestation has been launched and implemented. One of the degraded forest management restoration strategies is assisted natural regeneration (Shono et al. 2007;Lohbeck et al. 2021). However, it is still underutilized due to a lack of awareness and research results showing its effectiveness (Shono et al. 2007).
Assisted natural regeneration is an adaptive forest ecosystem management techniques for the restoration and enhancement of the natural forest particularly valuable and desired species (Abella et al. 2020). Assisted natural regeneration is an easy implemented and cost-effective forest ecosystem restoration method for the successful conversion of degraded forest tree species and lead to a more productive forest ecosystem. The technique targeted on accelerating regeneration through silvicultural treatments as an applied ecology, removing or reducing degradation agents and extreme disturbances. Assisted natural regeneration reduces the cost of seedling production, planting, and management as compared to conventional reforestation methods. It is the most effective and applicable approach for landscape-level forest restoration initiatives (Shono et al. 2007;Uriarte and Chazdon 2016;Abella et al. 2020;Lohbeck et al. 2021). It is also a flexible mechanism that allows the integration of timber production, biodiversity recovery, and cultivation of crops, fruit trees, and non-timber forest products in the restored forest (Shono et al. 2007).
Different assisted natural regeneration techniques have been synthesized and applied (Shono et al. 2007;Abella et al. 2020;Lohbeck et al. 2021) in various parts of the world liberation and tending of degraded forest, reduce and remove extreme disturbances, maintenance, and enrichment planting. The most common and widely used method of assisting natural regeneration is silvicultural treatment as applied forest ecology. Although regeneration after total loss may support economic production, it does not help the original ecological productivity (Shono et al. 2007). This is the most common method of management against deforestation and forest degradation.
Assisted natural regeneration techniques particularly silvicultural practices is the common forest ecosystem restoration mechanism for degraded tropical forest ecosystems and forest tree species (Khaine et al. 2018). It is a successful mechanism based on the desirable species methods of regeneration (seed, sucker, and coppice) with the combination of soil seed bank (Simonsen 2013). However, how effectively silviculture practices can accelerate natural regeneration is not well investigated associated with the method regeneration is not well evaluated in degraded forest ecosystems of Ethiopia's humid and dry tropical forests (Strassburg et al. 2016). However, there is a lack of application of assisted natural regeneration particularly silviculture treatments to restore Ethiopian native forest tree species due to lack of knowledge and experience. Because of this knowledge gap, most economically and ecologically important and valuable tree species are at risk of overexploitation, which could lead to extinction. Because of the scarcity of data-based silvicultural knowledge and skill on native forest tree species, communities, and institutions are forced to focus on fastgrowing exotic forest tree species with well-known and organized silvicultural knowledge. As a result of this scenario, preferable and desirable native tree species with multipurpose benefits receive insufficient attention. Furthermore, it is critical to generate scientific information on the method of regeneration and to recommend and prove the role of silvicultural application in accelerating regeneration for sustainability and sustainable use of the forest and forest products (Kennard et al. 2002;Yan et al. 2013).
As a result, the primary goals of this study are to: (a) figure out the size and method of natural regeneration; (b) assess the status of regeneration; and (c) find the best silvicultural practice for assisting natural regeneration of valuable native tree species in degraded forest ecosystems. This is critical for the long-term management and use of native forest ecosystems for wood and non-wood forest products, as well as for preserving biodiversity and human well-being (Yan et al. 2013). Furthermore, this research is critical for short and long-term adaptive forest ecosystem management planning to achieve and maximize the intended economic and ecological goals of forest resources as one of the United Nations Sustainable Development Goals (SDGs) (Tom-dery et al. 2014;Strassburg et al. 2016).

Description of the study area
The research was conducted in Banja and Guangua districts of Awi administrative zone, Amhara region, northwestern Ethiopia which is situated between 11 to 10 85 0 N latitude and 36 30 0 to 36 57 00 E longitude ( Figure 1).
The elevation of Banja district ranges between 1800 and 2953 meters above sea level, with an annual rainfall of 2300 mm and an average temperature of 15-25 C. The Guangua district is traditionally classified as part of the Woina-Daga climatic zone. It has a long unimodal rainfall pattern that lasts from April to the end of October with an annual rainfall of 400 mm and an average annual temperature of 20 C respectively (Nigatu et al. 2021) (Figure 2).
The forests that an experiment applied were highly degraded due to selective illegal logging and extensive over grazing. This makes the forest have tree cover and lower floristic diversity (Table 1). Based on forest degradation and assessment the two forests face sever degradation and due to participatory forest management Dikuma forest have better tree cover and species diversity (Ahmed et al. 2023).

Research design and data analysis
Prior to study plot establishment and actual data collection, a reconnaissance survey was conducted in different forest areas. Based on the information from reconnaissance, three forest stands were selected from various agro-ecological areas including Lewi Forest (1870-2570 m. ab.sl) from the high-altitude agro-ecology, Dukima (1800 À 2200 m. ab.sl) and Cambo (1200 À 1900 m. ab.sl) forests from mid-altitude agroecology. Tree crown cover, species diversity, and forest degradation level were evaluated based on the method of vegetation description and degradation assessment methods. Tree crown cover can be measured using different methods. One way is to use the crown sectional area of all trees in a stand and dividing by the plot area (and expressed as a percentage). The level of forest degradation were measured based on growing stock, biomass, biodiversity (Elliott 2016;Kent 2012;Korom et al. 2022). Then each forest stands were stratified based on elevation as a criterion for blocking replication. Then, three silvicultural treatments including Fence only, Fence þ hoeing, Fence þ gap creation, and control (open without a fence and without any treatment) were applied in three replications.
Gap creation treatments is mainly for light penetration to initiate light demander species applied by reducing the 60% of tree crown cover and remove   lianas with pruning (Patterson et al. 2022). Hoeing treatment is aimed to remove weeds, undesirable species, and soil losing for the initiation of dormant seeds from the soil seed bank. Then each treatment was randomized and replicate at each elevation strata. The three silvicultural treatments and one control plot were assigned at lower, medium and higher elevation of the mountain. The silvicultural treatments were applied once before the rainy season was started. Thus, a total of forest stands 12 circular plots (400 m 2 each) were established at each forest stands based on RCBD experimental design. The plots in each strata were established across the elevation with a distance of 250 m The fence were built by thorn steel wire in a row and zigzag pattern with the support of wooden poles having 1.8 m in a circular shape to exclude small ruminants and large mammals for prevention of trampling and browsing effects.
In each sample plot, the regeneration status of selected tree species, as the method of regeneration (seed, sucker, and/or coppice), were recorded during three seasons (beginning of rainy season, mid of rainy season, and mid of dry season). The tree species targeted for this study were chosen based on their ecological and economic value which includes; Albizia gummifera, Apodytes dimidata, Cordia africana, Croton macrostachus, Dombiya torrida, Hagenia abyssinica, and Prunus africana. The experiment was conducted for two years in 2021 and 2022.
A number of seedlings (DBH <2.5 cm and height <1 m), saplings (DBH < 2.5 cm, height 2 m), and trees (DBH >2.5 cm and height >2.5 m) were recorded based on (Teketay and Granstrom 1997) for each method of regeneration in each sample plots while mature trees for sources of regeneration as a mother tree height and diameter at breast height (DBH) were measured using hypsometer and diameter tape before the experiment. Descriptive statistics (mean and standard deviation) and inferential statistics (ANOVA and MANOVA) were calculated using R ver.4.1 statistical software for independent variables (method of regeneration, silvicultural practices, and method of regeneration) versus dependent variables (density of seedling).
In comparing regeneration methods by species, Albizia gummifera was the highest seedlings in suckeroriginated regeneration while Prunus africana and Croton macrostachus were the highest in seed sourced and coppicing regeneration, respectively (Figure 4).

Effects of silvicultural application on regeneration
The regeneration status of tree species varied significantly (p < .01, N ¼ 36) within and between forest stands (or sites). The use of silvicultural interventions as intermediate disturbances has a significant (p < .01) effect on natural regeneration of native tree species. Camboo forest stand has the highest seedling regeneration with 971 ± 132.02 seedlings ha À1 , followed by Lewi Forest 556 ± 263.3 seedlings ha À1 , and Dikuma forest 196 ± 24.8seedlings ha À1 . In comparing the silvicultural interventions, the combined intervention of fencing and gap creation resulted in the highest regeneration (961 ± 261.5seedlings ha À1 ), and the least regeneration (140 ± 47.5 seedlings ha À1 ) was recorded in control/open plots ( Figure 5).

Tree regeneration and seasonal variation
Seasonal variation had no statically significant effect on tree species regeneration (p > .5, N ¼ 36) but the highest seedling density was recorded in the mid-wet season (570 ± 258.4 seedlings ha À1 ), followed by the dry season (471 ± 158.5 seedlings ha À1 ) and the beginning of the rainy season (389 ± 122.95 seedlings ha À1 ) ( Figure 6).

Population status of targeted tree species
Tree species in the study forest have different regeneration, growth performance, and population status (p < .05). As a result, Prunus africana (312 ± 32.6 seedlings ha À1 ) and Albizia gummifera (177 ± 10.01seedlings ha À1 ) with medium growth performance (23-30 cm DBH and 47-65 m height), while Hagenia abyssinica (28 ± 3.28 seedlings ha À1 ) has extremely low regeneration with the presence of gigantic tree sizes (>1 m DBH and 47.7 m height) in the forest ( Table 2).
The highest number of regenerated seedlings were recorded under silvicultural intervention of fencing and gap creation as the silviculture promotes natural regeneration and protect seedlings and saplings from grazing and browsing disturbances. After application, the structure pattern of this intervention shows an inverted J-shape (a high number of seedlings were recorded followed by sapling and mature trees respectively). In the Camboo forest stand, a healthy and remarkable population structure was seen and recorded ( Figure 7).

Method of regeneration
Seed-originated regeneration method supported by fencing combined gap creation silvicultural interventions became the dominant type regeneration over all targeted forest stands. Additionally, suckering and coppicing method regeneration were also promising methods for regenerating degraded forest stand if they are supported with fencing combined with gap creation and hoeing treatments. The findings of this research is consistent with reports of (Jannat et al. 2020) which states the dominancy of seed coppicing as the method of regeneration for most of tree species. On the other hand, there is great variation between methods of regenerations over species. Accordingly, suckering and sprouting were dominant and potential methods for regenerating Albizia gummifera. The findings of this research implies that the identification of promising and potential methods of regenerations for desired tree species plays a crucial role for the successful implementation of forest resources restoration and conservation initiatives (Busby et al. 2010). Figure 7 depicts the vertical structural of tree species. It follows an inverted J-shape structure over forest locations and silvicultural treatments. This means that the forest stands were subjected for a critical risk of degradation and lack optimum mature trees able to be used as mother trees as a source of viable seeds. This scenario (an inverted J-shape structure) is a key indicator for the need of improving forest structure so as to secure suitability through the application of various environmentally suitable and economically sound interventions. Thus, the findings of this research contributes for the regeneration of degraded forest resources through the identification and suggestion of promising, potential and applicable as well as economically feasible methods of regenerations. Additionally the finding outlines the possibility of converting the vertical structure of a given forest stands into a healthy ecosystem via the use of silvicultural interventions (Vieira and Scariot 2006). This means that silvicultural management promotes regeneration and can change the structure of the forest from J-shape to inverted J-shape.

Effects of silvicultural application on regeneration
The highest variation in tree regeneration status observed among silivicultural interventions in this study, particularly the combined intervention of fencing and gap creation supports, Alemayehu's (2007) statement that grazing disturbances and insufficient sunlight are major factors that hinder natural regeneration. This study found that plots managed using a mix of fence and gap creation had a high rate of seedling regeneration. This implies that minimizing human and grazing disturbances could aid to ensure the regeneration of many native tree species in forest ecosystems. This idea is claimed by Hishe et al. (2021), that silvicultural practices such as gap creation and deliberate fire enhance natural regeneration with the exclusion of anthropogenic disturbances through fencing. A study conducted by Ruo (2018) in northern Ethiopia also found that fencing was effective at increasing seedling regeneration in a degraded forest. They observed that plots that were fenced had higher seedling densities than unfenced plots, this suggests that fencing can be an effective approach for excluding grazing animals and reducing human disturbance, while gap creation can provide suitable microsites and light conditions for seedling establishment and  growth. This indicates that grazing animals were a major impediment to natural regeneration.
Promoting natural regeneration through silvicultural treatment plays a significant role in accelerating the restoration and productivity of a given degraded forest ecosystem (Chazdon and Guariguata 2016;Chazdon and Uriarte 2016;Uriarte and Chazdon 2016). Most natural forests in Ethiopia are being degraded and deforested because of anthropogenic factors, and there is a tradeoff between economic benefit and ecological values. As a result, practices like assisted natural regeneration, which entails reducing human and grazing disturbances and executing particular silvicultural treatments, can aid in the restoration of these forests (Shoo and Catterall 2013;Elliott 2016).

Regeneration variation across seasons
According to the findings of this study, silvicultural treatments and seasonality can have a positive impact on tree regeneration and the highest regeneration was recorded during the mid-rainy season under fencing and gap creation silvicultural treatments. The result is consistent with the statement of Brose and Van Lear (1998), that season and silvicultural treatments aid accelerate the regeneration of native tree species in tropical forest ecosystems. They also stated that the application of silvicultural treatments before the growing season can accelerate the regeneration of native tree species in tropical forest ecosystem. Also, the study conducted by Noul ekoun et al. (2017) in a semi-arid region of Ethiopia demonstrated the importance of seasonality in promoting tree regeneration, it found that higher rainfall during the growing season increased seedling establishment. As a result, silvicultural treatment should be applied before the growing season begins.
Additionally, Bekele-Tesemma (2007) reported that most native tree species in Ethiopia regenerate between the beginning of June and the middle of August when there is enough moisture. This timeframe is crucial for successful restoration and planting activities since seedling recruitment of woody species is typically highest during the mid-wet season and lowest during the dry season, as demonstrated by a study carried out in a semi-arid region of Ethiopia (Yirdaw 2001). The pattern of seedling density observed in this study is also consistent with other research such as a study conducted in a lowland Amazonian forest, which found that the highest seedling density occurred during the peak of the wet season (Condit et al. 1998). Thus, managing and transplanting this abundant regeneration during the optimal regeneration period can enhance restoration and planting activities in Ethiopia.

Population status of targeted tree species
The findings from our study indicate that tree species in the study forest exhibit different regeneration, growth performance, and population status, as statistically significant differences were observed (p < .05). The result in (Table 1) shows that these species do not have a natural regeneration problem. The result showed that Prunus africana and Albizia gummifera exhibited higher regeneration rates, seedlings, and medium growth performance (Kebrom et al. 2016). Even though the species are extremely vulnerable to animal browsing and human disturbances (Bekele-Tesemma 2007). In contrast, Hagenia abyssinica showed extremely low regeneration rates, despite the presence of gigantic trees within the forest (Table 2) (Aynekulu et al. 2016). These results suggest that the regeneration potentials of the various native species should be taken into account when planning conservation and management strategies for the forest.
Our study found that the highest number of regenerated seedlings was observed under the silvicultural interventions of fencing and gap creation. These silvicultural treatments promote natural regeneration and protect seedlings and saplings from disturbances such as grazing and browsing. As a result, the structure pattern of this intervention shows an inverted J-shape, where a high number of seedlings were recorded followed by sampling and mature trees respectively (Fichtler et al. 2003). In the Camboo forest stand, a healthy and remarkable population structure was observed and recorded as a result of these interventions (Figure 7) (Ribeiro et al. 2015). These results highlight the effectiveness of fencing and gap-creation silvicultural treatments as an adaptive management strategy for maintaining the long-term conservation of forest resources and promoting sustainable use of forest products.

Conclusions
The synergistic deployment of seed-borne regeneration with encroachment prevention and canopy control has resulted in an excellent regeneration response. Identifying successful methods of regeneration for each native species is crucial for future conservation efforts. Gap-creation silvicultural treatments improve the regeneration of valuable native tree species within gated plots and maintain healthy and integrative forest ecosystems. Silvicultural practices operate as an intermediate disturbance and are extremely successful in enhancing natural regeneration. Fencing and gap creation silvicultural treatments are an effective adaptive management approach for long-term conservation and increasing the sustainable use of forest products.
According to the result of the silivicultural treatment, Prunus africana and Albizia gummifera species have no natural regeneration problems but are severely managed by anthropogenic disturbances. However, some species, particularly Hagenia abyssinica, have severe regeneration constraint and require advanced regeneration tools such as clonal nursery and invitro propagation techniques. To enhance the restoration of degraded natural forest ecosystems, intermediate silvicultural disturbances such as temporary fencing and gap creation can be effective methods of protecting forests from artificial disturbances. Properly implementing these silvicultural treatments can help facilitate natural regeneration. Identifying and documenting valuable regeneration sources for desired species is crucial for ensuring resource sustainability, successful conservation protocols, and promoting the use of native species.

Disclosure statement
No potential conflict of interest was reported by the author(s).