Skip to main content
NCBI home page
Heliyon logoLink to Heliyon
. 2020 Sep 3;6(9):e04806. doi: 10.1016/j.heliyon.2020.e04806

Impact of Pteridium aquilinum on vegetation in Nyungwe Forest, Rwanda

JMV Senyanzobe a,∗, Josephine M Mulei b, Elias Bizuru a, Concorde Nsengimuremyi c
PMCID: PMC7481561  PMID: 32944668

Abstract

Pteridium aquilinum acts as an important ecological filters in dominated communities. A study to investigate the effects of its dominance in the vegetation of Nyungwe was conducted. Sampling was done in Mubuga and Uwajerome mountains. A total of 53 alternate plots measuring 10 m × 10 m were sampled along a transect at regular interval of 10 m. In each plot, the species were identified and the cover abundance measured subjectively. Plant strategies, succession, biological forms, distribution and conservation status of each species were also determined. Data on species composition and cover abundance were analyzed using MVSP software and Shannon -Weiner index was used to determine diversity of communities. Descriptive statistics were used to assess the characteristics of the species.

A total of 141 species belonging to 100 genera and 54 families and distributed in four plant communities were identified. Pteridium aquilinum, Macaranga kilimandischarica, Lycopodium clavatum and Microglossa parvifolia were dominant in communities I, II and III, with average cover of 31%, 6% and 4% respectively. The primary forest was dominated by Pavetta rwandensis and Allophyllus chaunostachys in community IV, with 21% and 10% coverage respectively. Shannon- Weiner and evenness indices were 1.538, 2.925, 3.251 and 2.940 and 0.436, 0.716, 0.791 and 0.768 in communities I, II, III and IV respectively. Species richness were 34, 36, 61 and 46 in communities I, II, III and IV respectively.

Ruderal, chamaephytes, secondary, Africa tropical and least concerns plant species predominated in Pteridium vegetation areas with 76% 48%, 69%, 43% and 90% of total species respectively. Competitive, phanerophytes, primary, Afromontane and least concerns plant species dominated in non-dominated area with 54%, 52%, 58%, 40% and 88% of total species respectively.

Pteridium aquilinum restricted the growth of trees as exhibited by the presence of few phanerophytes and enhanced the growth of ruderal species, both of which are indicators of disturbed forest. The tree species observed in P. aquilinum cut-areas was Macaranga kilimandischarica.

Keywords: Ecology, Environmental science, Evolutionary biology, Plant biology, Species diversity, Nyungwe forest, Plant communities, Pteridium aquilinum, Ecological characteristics

Ecology; Environmental science; Evolutionary biology; Plant biology; Species diversity; Nyungwe forest; Plant communities; Pteridium aquilinum; Ecological characteristics.

1. Introduction

Fires, deforestations, human clearing, cattle grazing, agricultural activities etc. are ecological disturbances that fundamentally affect an ecosystem (Cochrane and Schulze, 1999; Cochrane, 2001, Silva Matos et al., 2003).

In the case of fire, a portion of the nutrients previously held in plant biomass is returned to the soil as biomass burns (Pringle, 1979) and vegetation with the potential for rapid growth takes advantage of the lack of competition and quickly fill the gaps created by fire. Ferns species have been reported as having the ability to dominate such nutrients-deficit soils (Akomolade & Rahmad, 2018). Every fire occurred in the forest, kills trees either by burning or because the resulting heat and a few days after that fire, Pteridium species is sprouting vigorously on both burnt and heat-killed areas, while it is absent in the intact primary forest (Hartig and Beck, 2003).

After deforestation by fires, Pteridium species fast colonizes and dominates the vegetation for long time periods (Silva Gallegos et al., 2014). In addition, these species inhibit the establishment of other associated plants (Pakeman and Marrs, 1992; Calvert, 1998; Silva Matos et al., 2003).

Grass species like Panicum maximum and Pteridium species are widely known to colonize rapidly such burned areas developing large accumulation of litter (Watt, 1940; Humphrey and Swaine, 1997) and the growth of Pteridium rhizome system is always promoted by fire (Hartig; Beck, 2003). In tropical habitats, the widespread use of fire always favors an ever-increasing spread of bracken ferns (Gliessman, 1978).

Pteridium species, the most dominant species worldwide (Taylor, 1990; Marrs and Watt, 2006) establishes itself on disturbed areas by fires, deforestations and agricultural activities (Page, 1986; Schneider, 2006; Silva and Silva, 2006; and Miatto et al., 2011). The main strategies for the increase of this bracken fern are: the high resistance to diseases and pests (Cooper-Driver, 1990; Den Ouden, 2000), vegetative reproduction (Page, 1986), high density of the frond canopy and tolerance to abroad range of climatic, edaphic conditions and the presence of allopathic chemicals (Gliessman and Muller, 1976), the resistance of the rhizome to fire and adverse weather conditions allowing the colony to spread vegetatively (Fletcher and Kirkwood, 1979). Its rhizomes run deeply into the soil and is the main factor responsible for colony expansion. Its spores are dispersed all through the year and once the individuals are established, deep rhizomes allow their local persistence (Schneider, 2004).

Due to its clonal growth, longevity and persistence, P. aquilinum builds a dense understorey canopy of fronds that dominates the forest floor for decades and centuries (Page, 1986; Schneider, 2004). The impacts of bracken fern dominance are many and most of them are known as negative effects:(i) dense bracken competes for water and nutrients and reduces the growth of the canopy trees (Richardson, 1993; Den Ouden, 2000), (ii) the long term failure of tree regeneration results to the disappearance of tree canopy in bracken vegetation (Den Ouden, 2000; Brandt; Black, 2001), such treeless situation occurred in Mexico (Schneider, 2004), Italy (Vos & Stortelder, 1992), Great Britain (Taylor, 1990), Dutch forest ecosystems (Den Ouden, 2000), (iv) Pteridium aquilinum reduces biodiversity as it tends to replace species in existing habitats and there is evidence that decrease in species richness over time is due to its competitive exclusion (Pakeman et al., 2000; Lameire et al., 2000), (v) bracken fern contributes to forest fires due to the large amount of leaflet litters they produce (Adie et al., 2011) and (vi) it affects forest seed banks after invasion success thereby preventing restoration of the environment (Ghorbani et al., 2006; Silva and Silva, 2006; Miatto et al., 2011; Akomolade and Rahmad, 2018).

Nyungwe Forest in Rwanda has also experienced destructive fires in dry EI nino years of 1997 and 1998 and vegetation in the affected areas dried out as most of the species are not adapted to wildfires disturbance. Majority of the species have slow growth rate. As a result, the forest canopy was opened up in various areas totaling approximately 12% of the total forest area. Light tolerant species such as P. aquilinum rapidly colonized the burned areas in the forest floor and had negatively impacted seedlings and saplings leading to lack of forest regeneration (Masozera and Mulindahabi, 2007).

The aim of this study was to assess the impacts of Pteridium aquilinum dominance on the vegetation of Nyungwe forest. The species composition and ecological characteristics of species were assessed in paired sites, with and without Pteridium dominance. The aim was also to evaluate the species diversity in Pteridium-dominated sites and intact primary forest. We aimed to answer the following questions: (i) Are there differences in species composition between Pteridium-dominated site and Pteridium-free area site? has species diversity of Nyungwe forest reduced in areas dominated by Pteridium? (iii) has vegetation structure changed in areas dominated by Pteridium?

2. Materials and methods

2.1. Description of study area

We carried out the study in Nyungwe Forest which is situated in Western Rwanda between latitude 2◦15–2◦55 S, and longitude 29◦00–29◦30 E, and at an altitude between 1600 m and 2950 m above sea level and about 1000 square kilometers in size. Nyungwe forest experiences a cool, humid climate that is highly influenced by its topography and proximity to Lake Kivu. It receives mean annual rainfall of approximately 1800 mm and an average minimum and maximum temperature of 10.9 °C in April and 19.6 °C in August respectively. Nyungwe is the largest protected area in Rwanda and is the most biologically important montane rainforest in Central Africa (Bizuru et al., 2014).

Floristically, Nyungwe was regarded as the richest forest remaining in Rwanda with more than 240 plant species included dominant species like Syzygium parvifolium, Macaranga kilimandischarica, Hagenia abyssinica, Carapa grandiflora, Newtonia buchananii, Neoboutonia macrocalyx Prunus africana, Symphonia globulifera, Cyathea manniana, Polyscias fulva, Parinaria excelsa, Podocarpus latifolius, Erica johnstonii, Entandophragma excelsium and Maesa lanceolata (Plumptre et al., 2002). Fires occurred in Nyungwe during dry EI nino years of 1997 and 1998 allowed the forest canopy to open up in various areas and light tolerant species such Pteridium aquilinum var centrali-africanum a native of Central Africa (Thomson et al., 2005), quickly colonized the burnt areas especially in its central east part. For this reason, fire was considered as a major threat to conservation of plant biodiversity (Masozera and Mulindahabi, 2007).

Specifically, the study was carried out in that affected part and two sites were selected namely Mubuga and Uwajerome mountains (Figure 1). Mubuga site was chosen because of the great natural regeneration of bracken fern after fires, both in its highland and lowland. Uwajerome site, 50 m away to the East of Mubuga, has both secondary and primary forests. The secondary forest is due to the natural regeneration after bracken fern clearing by Wildlife Conservation Society (WSC, 2013) and the primary forest being exempt of 1997–1998 fires was taken as control site.

Figure 1.

Figure 1

Open in a new tab

Location of Mubuga and Uwajerome Mountains in Nyungwe forest.

2.2. Methods

2.2.1. Vegetation sampling

The vegetation sampling was done systematically according Brawn-Blanquet (1932). Plant species were inventoried in plots of 10 m × 10 m which were regularly and alternatively spaced at interval of 10 m. Geographic coordinates of each plot were recorded using a GPS Garmin 12xl (Figure 2).

Figure 2.

Figure 2

Open in a new tab

Areas sampled in four sites.

A total of 53 plots were sampled with 41 plots in fern bracken vegetation and 12 plots in primary forest. In the field, the sampling in fern bracken vegetation was done in the following sites: Mubuga highland (18 plots), Mubuga lowland (13 plots) and Uwajerome secondary forest after the mechanical removal of the fern occurred after fire (10 plots). The succession of the later was started with Macaranga kilimandischarica but also the presence of bracken fern was still significant. Uwajerome primary forest (12 plots) was sampled as control site. Because of the heterogeneity in terms of vegetation in the highland of Mubuga site, the area was stratified into 3 subzones: top, middle and lower zone (Figure 2).

2.2.2. Vegetation data recording

In each plot the following data were recorded:

(i) Cover-abundance values.

This was based on an analysis of vegetation plots using Braun-Blanquet'coefficents which use presence/absence and cover-abundance scores. Each species in the plots was identified and cover class recorded (Table 1). Cover-abundance values of each species observed in the field were recorded.

Table 1.

Braun-Blanquet'coefficients used.

Classes Range of cover Mean cover percent values
5
4
3
2
1
+		 75–100
50–75
25–50
5–25
1–5
<1		 87.5
62.5
37.5
15
2.5
0.5
Open in a new tab

(ii) Mapping data.

GPS data were recorded by using a GPS Garmin 12xl. Areas sampled were mapped according to obtained geographic coordinates for each sampling plot (Figure 2).

(iii) Plant ecological strategies.

In the field, ecological strategies to community species were assessed according three primary ecological strategies proposed by Grime (1977):

  • (i)

    competitive strategy (C) which allocates species with low disturbance and stress,

  • (ii)

    stress-tolerant strategy (S) which allocates species with low disturbance and high stress,

  • (iii)

    and ruderal strategy (R) with species adapted to low stress and high disturbance habitats.

(iv) Plant succession.

The succession of each plant species in the communities where assessed according Gibson et al. (1983) where the following were observed:

  • (a)

    primary species (Pf) were species which started from a point with no biological material present,

  • (b)

    secondary species (Sf) were species after disturbance of fire,

  • (c)

    pioneer species (Pn) were plant species that dominate a community in early stage.

(v) Biological forms.

Biological forms were measured according Raunkiaer methodology (1934). In the field, different forms were found as follows:

Phanerophytes (Pha): were mostly woody plants with perennating buds above the ground surface, fully exposed to the atmosphere and at a height of more than 40 cm.

Chamaephytes (Cha): were woody or semi-woody perennials with buds above the ground surface but below 40 cm.

Therophytes (The): were annual plants surviving only as seeds in unfavorable season.

Hemicryptophytes (Hcry): were perennial herbs perennating tissues at the soil surface, leaf litter or dead plants remains give the buds some protection.

Geophytes (Ge): were plants with subterranean organs such as bulbs, rhizomes and tubers from which shoots emerge in the next growing season.

Epiphytes (Ep): were plants that harmlessly grow upon other plants.

(vi) Plant phytogeographic distribution.

Community species were classified into their phytogeographic distribution (White, 1978) and (Fischer and Killmann, 2008). Cosmopolitans (Cos) were species found over the world, Pantropicals (Pan) found in all tropics, Tropicals (Trop) were species of the tropic, Afrotropicals (Af-trp) were species found in Africa tropics, Afromontane (AME) were species of Africa Mountain and Alberine Rift Endemics (ARE) were species endemic to Albertine Rift.

(vii) Conservation status.

The status of plant species was assessed in order to establish a list of threatened species following IUCN criteria (IUCN, 2012). Species were assessed in categories of threats:

  • (a)

    Critically Endangered species (CRE): were plant species with extreme high risk of extinction,

  • (b)

    Endangered species (EN): were plant species with high risk of extinction,

  • (c)

    Vulnerable species (VU): were plant species with high risk of endangerment,

  • (d)

    Least concern species (LC): were plant species with lowest risk, widespread and abundant.

(viii) Plant species identification.

Plant species were identified using taxonomic keys in Combe (1977), Troupin, 1971, Troupin, 1978, Troupin, 1982, Troupin, 1983, Troupin, 1985, Troupin, 1988, Bleosch (2009), and Fischer and Killmann (2008) and comparison with already identified herbarium held at the National Herbarium of Rwanda, based in Huye District. For the checklist, data from International Plant Names List (IPN) were used. The study done by Thomson et al. (2005) was used to identify Pteridium aquilinum var centrali-africanum (African fern) from Pteridium aquilinum var aquilinum (European fern).

2.2.3. Data analysis

Data on vegetation were analyzed using MVSP (Multivariate Statistical Package) software where variables are plots and samples (cases) are species. Correspondence Analysis of that software was used to delineate communities in all areas sampled and plant diversity of communities was calculated using Shannon- Weiner index which took into account both species and presence/absence values.

3. Results

3.1. Species composition, diversity and communities

During sampling, 141 species belonging to 100 genera and 54 families were recorded in 53 plots. Cover-abundance values of the species were analyzed and delineated four communities dominated by Pteridium aquilinum and Microglossa parvifolia (16 plots); Pteridium aquilinum and Lycopodium clavatum (13 plots); Macaranga kilimandischarica and Pteridium aquilinum (12 plots) and Pavetta rwandensis and Allophylus chaunostachys (12 plots) respectively (Figure 3). Among the recorded species, Pteridium aquilinum was dominant in three communities with average coverage of 31% (Tables 1, 2, and 3) and Pavetta rwandensis and Allophyllus chaunostachys were dominant in community IV with 21% and 10 % of coverage respectively (Table 4). The species diversity in communities calculated by Shannon-Weiner formula were presented in Table 5. Shannon's indices were between 1.5 and 3.5, values accepted in most ecological studies during data analysis (Mahmuda et al., 2018). Community III whose plots representing samples in secondary forest after removal of ferns was the more diverse with average values of 61 species with diversity and evenness indices of 3.251and 0.791 respectively (Table 2) followed by the community IV of 46 species with diversity and evenness indices of 2.940 and 0.768 respectively (Table 2). The lowest richness was observed in communities of ferns vegetation after fires represented by communities I&II of 34 and 36 species, diversity indices of 1.538 and 2.923 and evenness of 0.436 and 0.716 respectively (Figures 4, 5, and 6).

Figure 3.

Figure 3

Open in a new tab

Individualization of communities. I: Community dominated by Pteridium aquilinum and Microglossa parvifolia. II: Community dominated by Pteridium aquilinum and Lycopodium clavatum. III: Community dominated by Macaranga kilimandischarica and Pteridium aquilinum. IV: Community dominated by Pavetta rwandensis and Allophylus chaunostachys.

Table 2.

Species diversity in communities.

Community Shannon-Weiner index
Diversity Evenness Number of species
I 1.54 0.44 34
II 2.93 0.73 36
III 3.25 0.79 61
IV 2.94 0.77 46
Open in a new tab

Table 3.

Percentage of floristic data in community I.

Ecological strategies % Biological forms % Plant succession % Plant distribution % Conservation status %
R 37 71 Pha 20 38 Pf 12 23 Cos 1 2 Lc 44 84
C 15 29 Cha 18 36 Sf 35 67 Pan 4 8 VU 2 4
The 1 2 Pn 3 6 Trop 1 2 EN 6 12
Ge 4 8 Epf 2 4 Af-trp 24 46 CRE 0 0
Hcry 5 10 AME 16 30
Ep 3 6 ARE 6 12
Tot. of species 52 100 52 100 52 100 52 100 52 100
Open in a new tab

Table 4.

Percentage of floristic data in community II.

Ecological strategies % Biological forms % Plant succession % Plant distribution % Conservation status %
R 37 88 Pha 7 17 Pf 2 5 Cos 2 5 Lc 40 96
C 5 12 Cha 22 53 Sf 36 85 Pan 4 10 VU 0 0
The 1 2 Pn 2 5 Trop 3 7 EN 1 2
Ge 2 5 Epf 2 5 Af-trp 20 48 CRE 1 2
Hcry 9 21 AME 7 16
Ep 1 2 ARE 6 14
Tot. of species 42 42 100 42 100 42 100 42 100
Open in a new tab

Table 5.

Percentage of floristic data in community III.

Ecological strategies % Biological forms % Plant succession % Plant distribution % Conservation status %
R 61 73 Pha 30 35 Pf 6 7 Cos 2 2 Lc 76 91
C 23 27 Cha 45 55 Sf 73 87 Pan 7 8 VU 2 2
The 1 1 Pn 5 6 Trop 2 2 EN 6 7
Ge 2 2 Epf 0 0 Af-trp 30 35 CRE 0 0
Hcry 4 5 AME 30 35
Ep 2 2 ARE 15 18
Tot. of species 84 100 84 100 84 100 84 100 84 100
Open in a new tab

Figure 4.

Figure 4

Open in a new tab

Shannon index in communities.

Figure 5.

Figure 5

Open in a new tab

Species evenness in communities.

Figure 6.

Figure 6

Open in a new tab

Species richness in communities.

3.2. Species characteristics

Species characteristics such as ecological strategies, biological forms, plant succession, and plant distribution and conservation status in the four communities were determined and each community was shown the percentage of species recorded as vegetation characteristics (Tables 3, 4, 5, and 6).

Table 6.

Percentage of floristic data in community IV.

Ecological strategies % Biological forms % Plant succession % Plant distribution % Conservation status %
R 24 46 Pha 27 52 Pf 30 58 Cos 1 2 Lc 46 88
C 28 54 Cha 16 30 Sf 22 42 Pan 3 6 VU 2 4
The 1 2 Pn 0 0 Trop 0 0 EN 4 8
Ge 1 2 Epf 0 0 Af-trp 14 27 CRE 0 0
Hcry 4 8 AME 21 40
Ep 3 6 ARE 13 25
Tot. of species 52 100 52 100 52 100 52 100 52 100
Open in a new tab

R: Ruderal; C: Competitive; Pf: Primary forest; Sf: Secondary forest; Pn: Pioneer; Epf: Earlier primary forest; Pha: Phanerophyte; The: Therophyte; Cha: Chamaephyte; Ep: Epiphyte; Hcry: Hemicryptophyte; Ge: Geophyte; COS: Cosmopolitan; PAN: Pantropical; ARE: Albertine Rift Endemic; Af-trp: Afrotropical; Trop: Tropical; AME: Afromontane Element; CRE: Critical Endangered; EN: Endangered; VU: Vulnerable; LC: Least Concern.

4. Discussion

The species in plots varied considerably between the sites of sampling areas. The vegetation was dominated by one species of fern (Pteridium aquilinum) in plots characterized as burned zones with a coverage of 61.8% and 17.4 respectively in highland and lowland areas of Mubuga site (Tables 1 and 2).

In the Uwajerome primary forest which was taken as control site, Pteridium aquilinum was rarely represented with 0.1% of the coverage (Table 4). However, in the Uwajerome secondary forest where the bracken fern was removed mechanically, Macaranga kilimandischarica was the most dominant species with coverage of 16.3% as compared to 13.4 % of P. aquilinum (Table 3). The species of Macaranga kilimandischarica was dominant because of its fast colonization in clear-cut areas and its requirements of high light and therefore tends to outgrow other trees (Fiala et al., 2011).

A biodiversity surveys done in 13 sites of Nyungwe Forest by Plumptre et al. (2002) recorded an herbaceous layer dominated by one species of Pteridium aquilinum which corroborates with the current study. A study curried out by Matos et al. (2002) recorded that fire events during dry seasons from 1991- 2000 in the National park of Tijuca (Brazil) has enabled dominant vegetation including Pteridium aquilinum. Fire was observed in 1993, 1995, 1999 years with 76 events/year including large and small events. The author stated that even if fire events are short, their frequency produce enormous changes in plant community. As pointed out by Cochrane and Schulze (1999), frequent fires may create forested areas with less tree density and reduce regeneration and tropical trees are vulnerable to fire (Uhl and Kauffman, 1990).

Our results suggest hence that, the Pteridium dominance has reduced the densities of wood species. The densities of phanerophytes (most shrubs and trees) were lower in dominated areas compared to the non-invaded ones. However, the removal of ferns has promoted Macaranga species.

Scali et al. (2019) experimentally examined seed germination, seedling survival and seedling growth of 14 woody species including Macaranga capensis in Bwindi National Park (Uganda). Experimental plots were under six field conditions of bracken fern stands including removing ferns by cutting fronds and leaving the litter intact. Authors recorded that Macaranga capensis germinated more frequently in that treatment than in the five other treatments. This corroborates with our results in Uwajerome plots where Macaranga kilimandischarica was dominant because the ferns were removed but litter remained. This confirms that Macaranga species cannot grow under bracken shade but when its fronds are removed, its opens up the light to Macaranga which tends to outgrow other trees. Another study done by Miatto et al. (2011) in Brazilian Cerrado showed that the bracken dominance has changed the species composition and structure of the wood vegetation. In Cerrado sites, areas invaded by neotropical bracken had fewer woody plant species than the non-invaded areas. This corroborates with our findings that bracken communities had restricted the growth of trees with the presence of few phanerophytes compared to the non-bracken community.

Species diversity was shown that Pteridium aquilinum and Microglossa parvifolia and Pteridium aquilinum and Lycopodium clavatum communities in both Mubuga highland and lowland were less floristically diversified: diversity index equals to 1.538 and 2.925 respectively than the community without ferns: diversity index equals to 2.940 (Figure 4). The most diversified of all was the Macaranga kilimandischarica and Pteridium aquilinum community which constitutes ferns -cut area: diversity index equals to 3.251 with 61 species (Figure 4 and Figure 6). Miatto et al. (2011) have demonstrated Pteridium invasion can impact and effect the vegetation structure, but also is reducing floristic diversity. Pakeman and Marrs (1992) also reported that the cutting of bracken stands can reduce frond density and break up the litter layer and those can accelerate colonization by other species. There is evidence that if Pteridium cover is reduced, especially where the litter layer has been disturbed example by animal tracks, there can be rapid tree invasion and also experiments done for reducing Pteridium, mortality of seedlings is reduced and their performance increases (Marrs and Watt, 2006). Our results also found that bracken-cut area was the more diverse site than bracken vegetation. As bracken was cut in Uwajerome site (WCS, 2013), a natural regeneration was proceeded owning more species than before. The increase of species in bracken – cut area (61 species) compared to other communities is due to the advantages of cutting mechanism cited above.

Phanerophytes (shrubs and trees) were less represented in fern vegetation contributing 38%, 17% and 35% of the total species in communities I, II and III respectively and most abundant in community IV with 52 % of total species (Figure 7). This agrees with the theory that bracken fern, once established, inhibits further colonization by other species including trees due to its large accumulation of litter which prevent other species from colonizing (Pakeman and Marrs, 1992) and the suppression of tree colonization makes treeless situation in bracken vegetation (Marrs and Watt, 2006).

Figure 7.

Figure 7

Open in a new tab

Species characteristics between communities.

According Grime (1979) ruderals are species adapted to disturbed areas. This correlated with our study where ruderal species were abundant in bracken vegetation represented by the communities I, II and III with 71%, 88% and 73% of the species respectively (Tables 3, 4, and 5 and Figure 7) while competitive species were most abundant in community IV contributing 54% of the total species (Table 6 and Figure 7). Bracken fern dominance seen as a perturbation, similar to secondary forests, leads to a change in plant distributions patterns (Schneider, 2006). Nyungwe Forest belonging to the afromontane forest and the Albertine Rift region (Plumptre et al., 2002, 2006; Ayebare et al., 2018), the afromontane (AME) and Albertine Rift (ARE) species were supposed to dominate in areas sampled but in this study, those species were decreased gradually from the primary forest to the natural regenerations after removal of fern and after fires. In the primary forest, 40 % and 25 % of total species belonged to AME and ARE respectively while in the secondary forest of lowland of Mubuga site, they were 16% and 14% of total species respectively (Table 6 and Figure 7). The reduction of characteristic species distribution in Pteridium - dominated area is an indicator of extinction of some species. According the Rwanda Environment Management Authority (REMA, 2015), the threats that affected the Nyungwe forest were also affecting its biodiversity. Among them, wildfires caused a substantial loss of the forest. The burned area was immediately colonized with a fern species (Pteridium aquilinum) which formed a dense layer of 1–2 m in height, blocking out light and reducing or preventing the natural regeneration of tree seedlings. In this study, the same species was in general dominant in secondary forests represented by communities I, II and III. In the report on national list of threatened species established by REMA, 2015, the following species were recorded in Nyungwe forest as threatened species: Lobelia milibraedii (CRE), Bersama abyssinica (EN), Caesaeria runssonica (EN), Chassalia subochreata (EN), Erica johnstonii (EN), Harungana montana (EN), Ocotea usambarensis (EN), Prinus africana (EN) and Lobelia petiolata (VU). In this study, the same species were also found as threatened species in remaining forest community. It means tree species recorded as threatened species were more concerned at any disturbance that why fire events cleared up the dense forest and promoted Pteridium aquilinum which is wide tolerant and high competitive.

5. Conclusion

A study was conducted in Nyungwe forest with the objective to investigate the effects of Pteridium aquilinum on vegetation after 1997–1998 years of wildfires. Ecological data were taken in Mubuga and Uwajerome sites. The data analysis was shown the dominance of Pteridium aquilinum, Microglossa parvifolia, Lycopodium clavatum, in Mubuga whereas Pavetta rwandensis and Allophylus chaunostachys were dominant Uwajerome site. However, the removal of P. aquilinum was promoted the dominance of Macaranga kilimandischarica and also increases species diversity. In forest community, phanerophytes and primary species remained dominant. Disturbed communities were dominated by ruderal species and most of them were chamaephytes and secondary species and afro-tropical species were also dominant. Majority of the encountered species were not threatened. In conclusion, the presence of Pteridium aquilinum changed the vegetation pattern of Nyungwe forest by suppressing phanerophytes communities and promoting chamaephytes and ruderal species, characteristic of disturbed areas but also of less trees.

5.1. Recommendation

Fire events always enable dominant species including Pteridium aquilinum which changes the vegetation structure. From this study, the following recommendations are made:

  • 1

    Control fires since it promotes bracken ferns in the forest

  • 2

    Device a mechanism of controlling further spread of the bracken fern

  • 3

    Device a mechanism of regenerating wood species in bracken community

Declarations

Author contribution statement

Jean Marie Vianney Senyanzobe: Contributed reagents, materials, analysis tools or data; Wrote the paper.

Josephine Mumbe Mulei: Conceived and designed the experiments.

Elias Bizuru: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data.

Concorde Nsengumuremyi: Analyzed and interpreted the data.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Acknowledgements

I acknowledge the Government of Rwanda through the University of Rwanda for providing training expenses, the Government of Kenya through the University of Eldoret for providing tuition waiver and accommodation, the Reginal Universities Forum for Capacity Building in Agriculture for funding my research project, The RDB for accepting to conduct research in Nyungwe and lastly the supervisors for a lot of contributions to my research.

References

  1. Adie H., Richert S., Kirkman K.P., Lawes M.J. The heat is on: frequent high intensity fire in bracken (Pteridium aquilinum) drives mortality of the sprouting tree Protea caffra in temperate grasslands. Plant Ecol. 2011;212(12):2013–2022. Springer. [Google Scholar]
  2. Akomolade G.F., Rahmad Z.R. A review of global ferns invasions: mechanisms, management and control. J. Res. Forest. Wildlife Environ. 2018;10(3):14. Lafia, Nigeria. [Google Scholar]
  3. Ayebare S., Plumptre A.J., Kujirakwinja D., Segan D. Conservation of the endemic species of the Albertine Rift under future climate change. Biol. Conserv. 2018;220:67–75. [Google Scholar]
  4. Bizuru E., Niyigaba P., Mujawamariya M. Phytosociology study of Nyungwe montane savannas. J. Nat. Sci. Res. 2014;4 no 9. United Kingdom. [Google Scholar]
  5. Bleosch U., Troupin G., Derungs N. Shaker Verlag; Aachen, Germany: 2009. Les plantes ligneuses du Rwanda. Flore, ecologie et usages. [Google Scholar]
  6. Brandt L.A., Black D.W. Impacts of the introduced fern, Lycopodium microphyllum, on the native vegetation of tree islands in the Arthur R. Marshall Loxahatchee National Wildlife Refuge. Fla. Sci. 2001;64(3):191–196. 7p. [Google Scholar]
  7. Brawn-Blanquet J. George D. Fuller and Henry S. Conard; 1932. Plant Sociology. The Study of Plant Communities. Translated, revised and edited by. [Google Scholar]
  8. Calvert G. Conference of the Society for Growing Australian Plants ’Queensland Region - 28 June- 5 July, 1998. Townsville, Australia. 1998. Weeds – the silent invaders. [Google Scholar]
  9. Cochrane M.A., Schulze M.D. Fire as a recurrent event in tropical forests of eastern Amazon: effect on forest structure, biomass, and species composition. Biotropica. 1999;31(1):2–16. 15. [Google Scholar]
  10. Cochrane M.A. Synergistic interactions between habit fragmentation and fire in evergreen tropical forests. Conserv. Biol. 2001;15(16):1515–1521. 7p. [Google Scholar]
  11. Combe J. C.T. Suisse. Kibuye.; Rwanda: 1977. Guide des principales essences de la forêt de montagne du Rwanda. Projet Pilote Forestier; p. 241. [Google Scholar]
  12. Cooper-Driver G. Defense strategies in bracken (Pteridium aquilinum (L.) Kuhn) Ann. Mo. Bot. Gard. 1990;77:281–286. 6 p. [Google Scholar]
  13. Den Ouden J. Wageningen University; The Netherlands: 2000. The Role of Bracken (Pteridium Aquilinum) in forest Dynamics; p. 218. PhD thesis. [Google Scholar]
  14. Fiala B., Meyer U., Hashim R., Maschwitz U. Pollination systems in pioneer trees of the genus Macaranga (Euphorbiaceae) in Malaysian rainforests. Biol. J. Linn. Soc. 2011;103(3):935–953. London. [Google Scholar]
  15. Fischer E., Killmann D. first ed. Vol. 1. Series Biogeographical Monograghs; Germany: 2008. Illustrated Field Guide to the Plants of Nyungwe National Park, Rwanda. Koblenz Geographical Colloquia. [Google Scholar]
  16. Fletcher W.W., Kirkwood R.C. The bracken fern (Pteridium aquilinum) L. (Kuhn): its biology and control. In: Dyer A.F., editor. The Experimental Biology of Ferns. Academic Press; London: 1979. pp. 607–635. 29p. [Google Scholar]
  17. Ghorbani J., Le Duc M.G., McAllister H.A., Pakeman R.J., Marrs R.H. Effects of the litter layer of Pteridium aquilinum on seed banks under experimental restoration. Appl. Veg. Sci. 2006;9(1):127–136. 10 p. Opulus Press, Uppsala, Sweden. [Google Scholar]
  18. Gibson C.W.D., Guifold T.C., Humbler C., Sterling P.H. Transition matrix models and succession after release from grazing on Aldabra Atoll. Vegetario. 1983;52:151–159. [Google Scholar]
  19. Gliessman S.R., Muller C.H. Allelopathy in a broad spectrum of environments as illustrated by bracken. Bot. J. Linn. Soc. 1976;73:95–104. 10. [Google Scholar]
  20. Gliessman S.R. The establishment of bracken following fire in tropical habitats. Am. Fern J. 1978;68:41–44. [Google Scholar]
  21. Grime J.P. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. Am. Nat. 1977;111(982):1169–1194. [Google Scholar]
  22. Grime J.P. 1979. Plant Strategies and Vegetation Process; p. 349. Sheffield, UK. [Google Scholar]
  23. Hartig K., Beck E. The bracken fern (Pteridium arachnoideum (kaulf.) maxon) dilemma in the andes of southern Ecuador. Ecotropica. 2003;9:3–13. 2003. Society for Tropical Ecology. [Google Scholar]
  24. Humphrey J.W., Swaine M.P. Factors affecting the natural regeneration of Quercus in Scottish oak woods. Competition from Pteridium aquilinum. J. Appl. Ecol. 1997;34:577–584. [Google Scholar]
  25. IUCN . List of Critical Endangered, Endangered and Vulnerable Plant Species. 2012. The IUCN red list of threatened species. [Google Scholar]
  26. Lameire S., Henry M., Honnay O. Two decades of Change in the ground vegetation of a mixed deciduous forest in an agricultural landscape. J. Veg. Sci. 2000;11:695–704. [Google Scholar]
  27. Mahmuda A., Md Animul I., Md Abdul A. Orthopteran diversity at keraniganj, Dhaka. Bangladesh J. Zool. 2018 [Google Scholar]
  28. Masozera A.B., Mulindahabi F. Wildlife Conservation Society; 2007. Post- fire regeneration in Nyungwe national park, Rwanda; p. 28. [Google Scholar]
  29. Marrs R.H., Watt A.S. Biological flora of the British isles: Pteridium aquilinum (L.) kuhn. J. Ecol. 2006;94:1272–1321. 50. [Google Scholar]
  30. Matos S.D.M., Santos C.J.F., Chevalier D.D.E.R. Fire and restoration of the largest urban forest of the world in Rio de Janeiro City, Brazil. Urban Ecosys. 2002;6:151–161. 12 p. Netherlands. [Google Scholar]
  31. Miatto R.C., Silva I.A., Silva Matos D.M., Marrs R.H. Wood vegetation structure of Brazilian Cerrado invaded by Pteridium arachnoideum (kaulf) maxon (Dennstaedtiaceae) Flora Morpho. Distri. Func. Eco. Plants. 2011;206(8):757–762. 6 p. [Google Scholar]
  32. Page C.N. The strategies of bracken as permanent ecological opportunist. In: Smith R.T., Taylor J.A., editors. Bracken: Ecology, Land Use and Control Technology. Parthenon Press; Carnforth, England: 1986. pp. 173–180. 8. [Google Scholar]
  33. Pakeman R.J., Marrs R.H. The conservation value of bracken Pteridium aquilinum (L.) Kuhn dominated communities in the UK, and an assessment of the ecological impact of bracken expansion or its removal. Biol. Conserv. 1992;62:101–114. 14. [Google Scholar]
  34. Pakeman R.J., Duc M.G., Marrs R.H. Bracken distribution of Great Britain: strategies for its control and sustainable management of marginal land. Ann. Bot. 2000;85(Sub B):37–46. [Google Scholar]
  35. Plumptre A.J., Masozera M., Fashing P.J., MCNeilage A., Ewango C., Kaplin B.K., Liongola I. WCS working paper series; 2002. Biodivesrity Surveys of the Nyungwe forest reserve in S.W. Rwanda; p. 96. [Google Scholar]
  36. Plumptre A.J., Davenport T.R.B., Behangans M., Kityo, Eilu G., Ssepawa P., Ewango C., Meirte D., Kalinda C., Harremans M., Peterhans J.K., Pilgrim J.D., Wilson, Languy M., Moyer D. The biodiversity of the Albertine Rift. Biological. Conserv. J. 2006;134:178–194. (2007) [Google Scholar]
  37. Pringle Laurence L. Natural fire : Its Ecology in Forests. William Morrow and Company; New York: 1979. pp. 22–25. [Google Scholar]
  38. Raunkiaer C. Claredon; Oxford, UK: 1934. The Life Forms of Plants and Statistical Geography. [Google Scholar]
  39. REMA . Study to establish a national list of threatened ecosystems and species in need of protection in Rwanda. Final report. REMA; Rwanda: 2015. Terrestrial threatened plant species; pp. 117–141. [Google Scholar]
  40. Richardson B. Vegetation management practices in plantation forests of Australia and New Zealand. Can. J. For. Res. 1993;23:1989–2005. 16p. [Google Scholar]
  41. Schneider L.C. Bracken fern invasion in Southern Yucatan: a case for land- change Science. Geogr. Rev. 2004;94(No. 2):229–241. People, Place, & Invasive Species. 13 p. [Google Scholar]
  42. Schneider L.C. Invasive species and Land use: the effect of land management practices on bracken fern invasion in the region of Calakmul, Mexico. J. Lat. Am. Geogr. 2006;5(2):91–107. [Google Scholar]
  43. Silva U.S.R., Silva D.M. The invasion of Pteridium aquilinum and the impoverishment of the seed bank in fire prone areas of Brazilian Atlantic Forest. Biodivers. Conserv. 2006;15:3035–3043. 9. [Google Scholar]
  44. Silva Gallegos C., Hassen I., Saavedra F., Schleuning M. Bracken fern facilitates tree seedlings recruitment in tropical fire-degraded habitats. For. Ecol. Manag. 2014;337:135–143. (2015) [Google Scholar]
  45. Silva Matos, D.M., Giovana Fonseca, D.F.M., Silva- LIma Differences on post-fires of the pionneer trees: Cecropia glazioui and Trema micrantha in a lowland Brazilian Atlantic Forest. Revista de Biologia Tropical. 2003;53 n. 1-2 San José, ISSN 0034-7744. [PubMed] [Google Scholar]
  46. Scali F., Moe S.R., Sheil D. The differential effects of Bracken (Pteridium aquilinum (L.) Kuhn) on germination and seedling performance of tree species in the African tropics. Plant Ecol. 2019;220:41–55. [Google Scholar]
  47. Taylor J.A. Bracken Biology and Management. Austr Inst of Agric Sci; Sydney: 1990. The Bracken problem: a global perspective; pp. 3–9. 10. [Google Scholar]
  48. Thomson J.A., Chikuni A.C., McMaster C.S. The taxonomic status and relationships of bracken ferns (Pteridium: Dennstaedtiaceae) from sub-Saharan Africa. Bot. J. Linn. Soc. 2005;148:311–321. [Google Scholar]
  49. Troupin G. Musée Royal de l’Afrique Centrale; 1971. Syllabus de la flore du Rwanda, Spermatophytes. Tervuren, Belgium; p. 143. [Google Scholar]
  50. Troupin G. Musée Royal de l’Afrique Centrale, Sciences Economiques, Série in-8o, no 9, Tervuren. Vol. 1. 1978. Flore du Rwanda, Spermatophytes; p. 413. [Google Scholar]
  51. Troupin G. Musée Royal de l’Afrique Centrale, Sciences Economiques, Série in-8o, no 12, Tervuren. 1982. Flore des plantes ligneuses du Rwanda; p. 747. [Google Scholar]
  52. Troupin G. Agence de Coopération Culturelle et Technique. Vol. 2. Institut National de Recherche Scientifique, Publication; Butare, Rwanda: 1983. Flore du Rwanda, Spermatophytes; p. 604. no22. [Google Scholar]
  53. Troupin G. Agence de Coopération Culturelle et Technique. Vol. 3. Institut National de Recherche Scientifique, Publication; Butare, Rwanda: 1985. Flore du Rwanda, Spermatophytes; p. 729. no30. [Google Scholar]
  54. Troupin G. Agence de Coopération Culturelle et Technique. Vol. 4. Institut National de Recherche Scientifique, Publication; Butare, Rwanda: 1988. Flore du Rwanda, Spermatophytes; p. 651. no41. [Google Scholar]
  55. Uhl C.A., Kauffman J.B. Deforestation, fire susceptibility, and potential tree responses to fire in the eastern Amazon. Ecology. 1990;71:437–449. [Google Scholar]
  56. Vos W., Stortelder A.F.H. Pudoc, press; Wageningen, The Netherlands: 1992. Vanishing Tuscan Landscapes. Landscape Ecology of a Submediterranean-Montane Area (Solano Basin, Tuscany, Italy) [Google Scholar]
  57. Watt A.S. Contributions to the ecology of bracken (Pteridium aquilinum). I. The rhizome. New Phytol. 1940;39(4):22. Willey, UK. [Google Scholar]
  58. WCS . 2013. Factsheet WCS- Biodiversity Research and Monitoring in Nyungwe forest; p. 4. 2009. Rwanda. [Google Scholar]
  59. White F. Biogeography and Ecology of Southern Africa. Junk publishers; The Hague: 1978. The Afromontane region; p. 1439. [Google Scholar]

Articles from Heliyon are provided here courtesy of Elsevier

RESOURCES