A Contribution to the Study of the Flora and Vegetation of Mnemba Island, Zanzibar

Abstract

Swahili coastal forests, spanning the Kenyan and Tanzanian coastlines and the Zanzibar Archipelago, are integral to the biodiverse Eastern Arc and Coastal Forest region. These forest ecosystems face considerable anthropogenic threats. This study contributes to the available knowledge on the floral composition and plant community assemblages of Mnemba Island, a small sandy cay in the Zanzibar Archipelago, which helps to establish baseline data for conservation purposes. Two main coastal forest plant communities were identified: (1) the Casuarina cunninghamiana–Suriana maritima open to closed sandy dry coastal forest, and (2) the Eugenia capensis–Mimusops obtusifolia coastal forest. A total of 91 different plant species belonging to 54 plant families were identified for the island. Community 2, an indigenous forest, supports diverse bird breeding colonies and is a crucial habitat for the threatened Aders’ duiker. Understory development is limited due to the historical Suni antelope overpopulation. Environmental factors like salt spray, allelochemicals, herbivore browsing, and climate fluctuations influence the vegetation abundance and composition. The study underscores differences in species diversity and composition between the planted Casuarina community and the natural atoll vegetation. The natural vegetation shows affinities with the Zanzibar–Inhambane edaphic coral-rag scrub forest and the transitional rainforest. Several species characteristic of Indian Ocean atolls were identified, providing insights into invasion ecology and conservation strategies. The study contributes conceptually to our understanding of vegetation dynamics in island ecosystems by highlighting the interplay between plant communities, environmental processes, and human activities. The fragile yet resilient nature of Mnemba’s unique ecosystem is emphasised, offering insights for conservation management, long-term monitoring, and adaptive approaches tailored to island environments.

1. Introduction

The coastal forest of the Swahili region, encompassing the Kenyan and Tanzanian coastlines, including the islands of Unguja, Pemba, and other smaller islands collectively known as Zanzibar, form a vital part of the Eastern Arc and Coastal Forest region. These forest ecosystems, despite their isolation, face significant anthropogenic threats. A comprehensive report on biodiversity, climate change, and the energy crisis [1] highlights the vulnerability of Zanzibar’s forests to increased human activity, emphasising the pressing need for improved conservation efforts.

Within the Zanzibar Archipelago, Mnemba Island, a small forested sandy cay mounted on a sub-tidal atoll reef [2,3,4,5], possesses distinct morphological characteristics and unique ecological features. This island serves as a crucial habitat for the vulnerable Aders’ duiker and supports various bird breeding colonies. Research on such small sandy cays is pivotal for understanding species distributions, recognising isolated extinction events [6,7], and understanding the dynamics of island ecosystems, including the impact of human activities, the associated potential for biological invasions, and their ecological effects [8,9].

The company AndBeyond has created a luxury eco-friendly lodge on Mnemba Island that is known for its exclusive and intimate setting. The lodge features pristine beaches, crystal-clear waters, and a tranquil atmosphere. With its limited beachfront bandas, guests can enjoy personalised experiences through activities like snorkelling, diving, and nature walks.

Seed-eating animals like birds and bats play an important role in dispersing seeds, fruits, and parts of plants among and on the islands [9]. Rainfall has also been identified as an important factor that influences the vegetation distribution, species composition, plant phenology, and diversity on the islands [10,11]. Low-lying atolls like Mnemba are characterised by low endemism, fluctuating colonisation rates, and species extinction [12].

According to [13], information about floral composition and vegetation assemblages is important for making scientifically defensible decisions regarding environmental conservation and management. Despite the importance of Mnemba Island, comprehensive diversity and ecological baseline vegetation data are lacking. Apart from two short-term studies on the Aders’ duiker and Suni antelope on Mnemba and a brief terrestrial survey conducted in 1990 [14] that provided limited information on the vegetation, no other studies have been identified for the island. This study aims to classify, describe, and ecologically interpret the flora and vegetation of Mnemba Island. Detailed descriptions of the plant assemblages, including their structural breakdown, spatial distribution, floral composition, and environmental characteristics, are provided.

This investigation was part of a PhD study on the ecology of the Aders’ duiker (Cephalophus adersi) on Mnemba Island, Zanzibar [2], and contributes conceptually to a better understanding of the vegetation composition and dynamics in island ecosystems by highlighting the intricate relationship between plant communities, environmental factors, and human activities. Key findings include the divergence in species diversity and composition between the planted Casuarina cunninghamiana section and the natural atoll vegetation, the ecological significance of the island’s natural forest in supporting diverse bird breeding and providing a habitat for the vulnerable Aders’ duiker [15], and the observed lack of understory development due to the historical Suni antelope overpopulation. No research exists on bat species specific to Mnemba. Although the island hosts various geckos, no snakes have been found on the island. Crustaceans include three species of ghost crabs, the coconut or robber crab, and other land crab species. Additionally, various insects, arachnids, millipedes, and worms contribute to the island’s biodiversity.

This study offers valuable insights for conservation management, emphasising the fragile yet resilient nature of Mnemba Atoll’s unique ecosystem. It highlights the affinity of the island’s flora with the regional ecosystems, the interplay between natural processes and human activities, and the complex relationships among plant species and communities. These findings lay a solid foundation for future research endeavours and long-term monitoring and conservation initiatives, aligning with the island ecology and the unique challenges faced by island ecosystems.

2. Methods

2.1. Study Area

Mnemba Island (5°49.219′ S, 39°22.959′ E) is situated approximately 2.7 km off the northeastern coast of Unguja, the main island of Zanzibar, being separated from it by a 75 m deep channel (Figure 1) [16]. The island is a sandy cay located within the Mnemba Atoll, which forms part of a coral reef complex characterised by a reef flat or lagoon [4,5]. Positioned on the western edge of the atoll reef, Mnemba Island has a triangular shape, roughly situated at the 8 o’clock position relative to north (with north being 12 o’clock). The entire atoll area spans approximately 150 ha, with a distance of 7 km between the northern and southern reef edges along a very shallow reef flat [16,17]. Mnemba Island itself covers an area of only 18 ha, of which 11 ha are vegetated. Its sandy cay morphology contributes towards its unique ecological features [4,5].

Mnemba Island is surrounded by a fringing coral reef complex, boasting an approximate circumference of 3 km that is subject to a tidal range of 4 m between low and high tides. At low tide, the reef flat is fully exposed and drained [17]. Aerial imagery, as depicted in Figure 2, offers a clear perspective from the west to the east, showcasing the shallow reef flat and the semi-submerged reef crest along the atoll’s eastern edge.

The reef surrounding Mnemba Island predominantly consists of hard corals, creating a diverse array of reef habitats. This diversity arises from the varying depths at which the different corals occur. When exploring the underwater reef face, one observes distinct structures such as terraces, shelves, and inclines. Notably, the reef boasts a high diversity of coral species that contribute to Mnemba Atoll’s reputation as having the highest coral diversity among the reefs in Northern Unguja. An investigation of the reef around the island, conducted by the Department of Fisheries and Marine Resources, recorded 36 genera of hard coral [19].

On the northern side of the cay is a smaller reef known as the “house reef”. This reef covers an area of 0.25 km² and reaches a depth of only 7 m [20]. The eastern and southern shores of the island are bordered by a reef flat consisting of sediments of fine carbonates [17]. This shallow reef flat experiences filling during high tide and serves as a natural wave attenuation system that protects the fine sandy beaches from excessive wave impacts [17,20]. Additionally, the seabed on the southern side of the cay supports a sea-grass marine ecosystem [20].

Mnemba Island hosts two distinct vegetation communities: an outer band of planted Casuarina cunninghamiana and an inner core of natural atoll vegetation. The natural vegetation follows a concentric atoll succession pattern [21] featuring a diverse mix of shrubs, trees, lianas, creepers, and Pandanus spp. The Casuarina cunninghamiana was introduced before the 1970s as windbreaks.

The fauna found on the island includes approximately 50 bird species [22], introduced mice, and intentionally reintroduced Suni and Aders’ duikers.

2.2. Climate

Mnemba Island’s unique size and location relative to Unguja, coupled with the influence of ocean currents and trade winds, give rise to distinctive climatic conditions that differ from those experienced on Unguja. To capture the island’s microclimate data, an impromptu Ambient Weather WS-1000-WiFi OBSERVER Solar Powered Wireless Wi-Fi Weather Station was procured from AWR Smith Process Instrumentation, Randburg, South Africa. The weather station was erected on the atoll for the duration of the study to record the first baseline data for the island’s microclimate, enabling the assessment of weather effects on its vegetation.

Temperature and rainfall trends on Mnemba Island for 2016 are depicted in Figure 3. The recorded temperatures exhibited distinct patterns throughout the year. From January to September, the temperatures ranged from an average of 25 °C to 31 °C. From October to December, the average temperature sharply increased from 31 °C to 38 °C. The highest daily average temperatures were observed in October (31 °C), November (34 °C), and December (39 °C), while the lowest average temperatures occurred in June (26 °C), July (26 °C), and August (27 °C). Several temperature spikes exceeding 40 °C were recorded in July, November, and December 2016.

Mnemba Island experienced 490 mm of rainfall from March to May 2016, whereas during the period from October to December, the rainfall was drastically reduced to only 70 mm. This decline in rainfall during the short rainy season may be linked to the notable increase in temperature during the latter part of the year. Throughout January to November, the island’s humidity ranged from 79% to 89%, peaking at 89% in April. Subsequently, the humidity gradually decreased, reaching 74% in December.

2.3. Vegetation Sampling

The study was conducted from January 2016 to December 2017, during which an onsite weather station (WS-1001-WiFi Observer) was installed and maintained to record the local weather conditions on Mnemba Island.

To survey the island’s vegetation, a total of 24 sampling plots, each covering an area of 400 m² [2,22], were placed out in a randomly stratified manner within the two identified physiognomic–physiographic plant communities. Eight plots were assigned to the planted area, and 16 to the natural forest. Plots were placed sufficiently far away from guest accommodation to prevent disturbance during the data collection. All plant names are according to the Plants of the World (POWO) database.

A Garmin eTrex GPS was used to record the coordinates of each sample plot for easy relocation during follow-up visits, while digital photographs were taken in eight directions from the centre of each relevé (north; northeast; east; southeast; south; southwest; west; northwest).

All plant species found in the sample plots were identified, and their cover abundance estimated using the modified Braun-Blanquet Cover-Abundance scale (

Table 1. Modified Braun-Blanquet cover-abundance scale [23] and the predicted mean (M) value for each class, as calculated by [24].

Scale	Description	Predicted Mean (PM) [24]
r	One or a few individuals with less than 1% cover of the total sample relevés area.	0.01
+	Occasional and less than 1% cover of the total sample relevés area.	0.49
1	Abundant with low cover or less abundant but with higher cover; 1–5% cover of the total sample relevés area.	0.74
2a	6–12.5% cover.	8.95
2b	12.6–25% cover.	17.5
3	26–50% cover of the total sample relevés area, irrespective of the number of individuals.	38.77
4	51–75% cover of the total sample relevés area, irrespective of the number of individuals.	62.43
5	76–100% cover of the total sample relevés area, irrespective of the number of individuals.	81.24

) [13]. To determine the woody species density, the number of individuals of each woody species present within every sample plot was counted.

The advanced stick method was employed for the tree height determination, which is based on the methodology of [25]:

$$TH={\displaystyle \frac{(LS\times DT)}{DE}}$$

where

• TH = Tree height
• LS = Length of the stick
• DT = Distance to the tree
• DE = Distance to the eye.

In cases where dense vegetation impeded the application of the advanced stick method, the tree height was estimated using a nearby tree of similar canopy height for which the advanced stick method was applicable.

Following data collection, trees with a breast-height diameter (BHD) greater than 6 cm were categorised, based on their heights, into low-range (4–9 m), mid-range (>9–14 m), and high-range (>14 m) (adapted from [26]). Additionally, the heights of the shortest and tallest individuals for each species were recorded.

Additional environmental data recorded in each sample plot included the percentage rockiness, litter type and depth, and the broad soil type.

2.4. Data Analysis

Data were recorded in an Excel spreadsheet and subsequently imported into the vegetation classification software package JUICE 7.1. The modified TWINSPAN algorithm was then used, which clusters species based on their internal heterogeneity, to delineate the distinct plant communities [27]. A phytosociological table was generated, and plant community names were assigned following the guidelines outlined in [13]. To facilitate analysis, Braun-Blanquet cover-abundance classes (Table 1) were converted into continuous nominal values. This conversion was carried out using the predicted mean (PM) values for each ordinal class as documented by [24]. In cases where the PM table did not include certain classes, appropriate values were assigned to ensure accuracy and consistency for the purpose of this study. Such values were selected based on their logical relationship to the existing values in the table and their alignment with the relative abundance of vegetation cover observed in the study area. For example, if a specific cover-abundance class fell between two existing classes in the PM table, an appropriate value might be determined by interpolation between the neighbouring classes. Alternatively, if the class represented a significant deviation from existing classes, the value might be determined based on expert judgment or previous research findings. The goal was to ensure that the assigned values accurately reflected the relative abundance of vegetation cover for the specific class in the study area.

The characterisation of each plant community involved identifying characteristic species based on their constancy of occurrence (the number of sample plots within which each species occurred), which ranged from 11% to 100%, rendering them indicative of that specific community or habitat [28,29].

Species diversity within each plant community was calculated using the Shannon–Weiner Index, a metric that considers both evenness and abundance, producing values between 0 and 5, with higher values indicating greater species diversity. The Shannon–Weiner Index was calculated using the formula:

$$H=-\sum \left[\left(pi\right)\ast \mathit{ln}\left(pi\right)\right]$$

where

• $H$ = Shannon–Weiner Index;
• ∑ = summation;
• $pi$ = proportion of individuals belonging to the i-th species in the community; and
• $ln$ = natural logarithm.

To determine evenness, the Shannon–Weiner Index was divided by the maximum possible value of the index (H max), resulting in an evenness value (E) that provides insights into the homogeneity of each community. The formula for calculating evenness is

$$E={\displaystyle \frac{H}{H max}}$$

where

• $E$ = evenness;
• $H$ = Shannon–Weiner Index;
• $H max$ = maximum possible value of the Shannon–Weiner Index.

The diversity index and evenness values were compared to assess differences in diversity among the plant communities.

The woody species density was determined per woody species using the following formula:

$$D={\displaystyle \frac{N}{A}}$$

where

• D = density, expressed as individuals per unit area (ind/ha);
• N = number of individuals counted in all sample plots of a plant community × 10,000 (i.e., p/ha);
• A = area size of plots within which the individuals were counted.

3. Results

3.1. Vegetation Classification and Description

The results of the modified TWINSPAN classification revealed the presence of four plant communities that were grouped into two main plant communities (Figure 4 and

Table 2. Phytosociological table with modified Braun-Blanquet cover-abundance scale values for all species within the plant communities found on Mnemba Island.

Community Number	1								2
	1.1				1.2				2.1									2.2
																									% Constancy
Relevé numbers	1	1					1		1	2	2	2	2	1	1	1	2			1	1		1		Main
	3	4	7	6	3	2	5	1	7	0	2	4	1	6	8	9	3	9	5	2	1	4	0	8	Community
Species group A
Casuarina cunninghamiana (*)	5	4	1	2b	1	5	4	5					r				r								100%
Suriana maritima (***)	+	+	+	r	2b	3	2a	2a																	100%
Enteropogon macrostachyus (***)	+	2a	1	+	+	+	+								r					r					88%
Evernia prunastri (***)	+	+	+		+		+												r						63%
Brachiaria leersioides (***)			+				+																		25%
Ricinus communis (***)	r	r	+	1	+		r	+										+			+	+		+	46%
Rhynchosia sublobata (***)	+		+	+	+	r		r												+	+	+			38%
Aneilema indehiscence (***)	+	+				+		r		+								r		+		+			33%
Dactyloctenium aegyptium (***)	+	+	+								r							+		+		+		r	33%
Polyporus tubaeformis (***)	+	+	+	+				+	+	+															63%
Species group B
Melanthera biflora (***)			1	5																					50%
Panicum repens (***)	+	+		+																					75%
Ipomoea pes-caprae (***)	r	r			r																				50%
Scaevola sericea (***)	+																								25%
Kyllinga erecta (***)	+																								25%
Sesbania bispinosa (***)			2a																						25%
Gymnosphoria heterophylla (***)	r		+																		r				50%
Species group C
Bougainvillea spectabilis (**)					r	r		r																	75%
Chrysothrix sp. (***)						1	+														+				50%
Fomitopsis pinicola (***)							+																		25%
Bidens pilosa (**)					r																				25%
Panicum maximum (***)								+																r	25%
Species group D
Eugenia capensis (***)								+	2a	3	2b	2a	2a	1	3	2b	3	2a	4	4	2b	2b	2a	2a	100%
Mimusops obtusifolia (***)	+				+				3	3	2a	1	+	+	+	2b	2b	+	1	2a	2b	1	1	1	100%
Pandanus kirkii (***)	r	2a	2a						2b	2b	1	+	2a	+	2b	1	2a	1	1	+	+	+	+	1	100%
Secamone puntulata (***)					r				+	1	2a	+	2a	+	1	1	1	+	+	1	1	+	+	1	100%
Cassytha filiformis (***)										+	1	1	+	1	+	1	1	+	2a		+		+	+	81%
Grewia glandulosa (***)	r	r			r			r	+	+		r	+		+	+	+	2a	1	2a		1	1	1	81%
Maerua triphylla (***)	r								+	+	+	+	+	+	+		1			r		r	+	+	75%
Synaptolepis kirkii (***)	r	+							+		+	r	+		+	+	1	+	+		+	+		1	75%
Pisonia grandis (*****)		r	r								+	3		+			2a			+		r	+		44%
Ecbolium ligustrinum (***)		1			+						+	+	+	1	r	r	+				+		2b	1	63%
Suregada zanibariensis (***)	r	+		+					r		+	+	+				+	+		+	+	+		+	63%
Ficus scassellati (***)									r								+					+			19%
Afroligusticum linderi (***)														+								r			13%
Hypoxidia sp. (***)															r					r					13%
Adenia gummifera (***)															r									r	13%
Species group E
Scutia myrtina (***)											2b	+		+			+				+				44%
Pavetta stenosepala (***)											+			+	+		+					+			44%
Bourreria petiolaris (***)											+	+					+	r							33%
Ficus lutea (***)											r		+		r	r			r						44%
Psychotria psychotrioides (***)										r	+	+									r				33%
Ficus polita (***)	+									r	+														22%
Conyza newii (***)									r																11%
Justicia capensis (***)														1											11%
Auricularia auricular-judae (***)																	+								11%
Xanthoria sp. (***)																+									11%
Ludia mauritiana (***)														+											11%
Phaeotrametes decipiens (***)																	+								11%
Daldinia concentrica (***)																r									11%
Solanum viarum (*)																r									11%
Species group F
Clerodendrum glabrum (***)													r					1	+	+	+	+	+	1	100%
Daedaleopsis confragosa (***)											r			+				+	+	r	+	+	+	+	100%
Pavonia sp. (***)	r																	+	1		+	+			57%
Phyllanthus amarus (**)																				+	+	r		+	57%
Polysphaeria parvifolia (***)										+		+						+	+	+				+	57%
Abutilon sp. (***)										+								r				+	+	+	57%
Ipomoea sp. (***)																		+				r			29%
Morinda citrifolia (****)	r			+	r		+												+		r	r			29%
Kyllinga platyphylla (***)				+			+													+	+				17%
Acacia auriculiformis (**)			r		r														r			r			17%
Ehretia amoena (***)																							+		14%
Parmelia sulcata (***)																			+						14%
Pycnoporus sp. (***)																			+						14%
Trametes hirsuta (***)																						+			14%
Boerhavia sp. (***)																						+			14%
Morinda citrifolia (****)																				+					14%
Coccinia grandis (***)																				r					14%
Pupalia lappacea (***)		r				+														+			r		17%
Species group G
Sideroxylon inerme (***)	+	2a	1			+	+	r	+	2b	+	1	+	+	2a	1	+	1	r	+	+	r	+	+	92%
Capparis viminea var. orthacantha (***)	+		1	+	+	+	+	+		+	+	+	r		+	+	+	+	r	+	r	+	+	+	88%
Cyphostemma adenocaule (***)	+	1	1	1	+	+	+	+	r		r		r			+	+	+	+	+	+	+	1	1	83%
Eragrostis ciliaris (***)	+												r						+	r	r				21%
Terminalia catappa (****)			r		r					r			r									r			21%
Dalechampia scandens (***)	+	+	1	2a	2a	1	1	+		+	+		+	+	+	+	+	+	1	+	+	+		+	88%

Species status descriptions (* = Alien, ** = Introduced, *** = Native to the region, **** = Naturalized, ***** = Unclassified).

):

• The Casuarina cunninghamiana–Suriana maritima open to closed sandy dry coastal forest:
• Casuarina cunninghamiana–Suriana maritima–Enteropogon macrostachyus sandy dry coastal forest community;
• Casuarina cunninghamiana–Suriana maritima–Chrysothrix sp. open sandy dry coastal forest.
• The Eugenia capensis–Mimusops obtusifolia coastal forest:
• Eugenia capensis–Mimusops obtusifolia–Scutia myrtina coastal forest;
• Eugenia capensis–Mimusops obtusifolia–Clerodendrum glabrum coastal scrub.

The phytosociological table (Table 2) delineates two main communities and four sub-communities. Dominant species within each plant community are organised based on relevés (represented by columns) and species (represented by rows).

The island’s forested terrain exhibits a consistent elevation, with an overall altitudinal variation of just 1 m across the entire vegetated area. The island’s vegetation is situated at an elevation ranging from approximately 4 to 5 m above mean sea level (MSL). Apart from a 40 m stretch of natural vegetation along the southern edge of the island, plant community 2 is surrounded by plant community 1. This forms a protective barrier between the frontal dune zone and the island’s inner forested area (Figure 5).

3.1.1. The Casuarina cunninghamiana–Suriana maritime Open to Closed Sandy Dry Coastal Forest

Plant community 1 is the dominant coastal forest community on the island, accounting for 53% of the total vegetated area. This community is primarily situated along the island’s outer periphery, with the Casuarina cunninghamiana belt extending into the island’s interior on the eastern side. The community is characterised by species from species group A (Table 2), including the Australian she-oak Casuarina cunninghamiana, the evergreen shrub Suriana maritima, the castor oil plant Ricinus communis, the perennial grass Enteropogon macrostachyus, the pioneering grass Dactyloctenium aegyptium, the climbing perennial herb Rhynchosia sublobata, and the creeping herb Aneilema indehiscens. Fungi and lichen species such as Evernia prunastri, Brachiaria leersioides, and Polyporus tubaeformis are also present in this community.

The upper tree layer is dominated by Casuarina cunninghamiana (species group A), with an estimated 350 trees per hectare. Suriana maritima (species group A), Pandanus kirkii (species group D), and Sideroxylon inerme (species group G) are also prominent in the woody vegetation. The herbaceous layer is dominated by Melanthera biflora (species group B), while Dalechampia scandens (species group G) is conspicuous in some areas. The grass layer is dominated by the perennial grass Enteropogon macrostachyus (species group A). Dactyloctenium aegyptium (species group A) and Panicum repens (species group B) also occur in the grass layer. A “typical” litter layer composed of dropped Casuarina cunninghamiana branchlets is present, with an average depth of 2.5 to 3 cm. Beneath this layer is a fermentation stratum measuring between 6 and 7 cm in depth. Well-drained sandy soils predominate in this community, with the rock cover being less than 0.1%.

The community is subject to disturbance primarily from infrastructure construction projects and the uprooting of Casuarina cunninghamiana trees driven by high-tide interactions along the eastern shore. This community housed 60% of the island’s buildings at the time of the survey.

Two sub-communities were classified for Community 1.

The Casuarina cunninghamiana–Suriana maritima–Enteropogon macrostachyus Sandy Dry Coastal Forest Community

Situated between sub-communities 2.1 and 1.2, this coastal forest sub-community covers 31.4% of Community 1 and 17% of the island’s total vegetated area (Figure 4). This sub-community is dominated by Casuarina cunninghamiana and is characterised by a high litter layer, composed mainly of dropped branchlets from this species, seeds, small twigs, and branches. The litter layer covers 87% of the sub-community and has an average depth of 2.5 cm. The fermentation layer below the litter layer has an average depth of 6 cm. There are 11 fallen trees per hectare, and an average of two deceased trees per hectare that remain standing.

Sandy soils with good drainage prevail throughout this sub-community, and the rock cover is less than 0.01%. Non-vegetated areas, including footpaths and a water pipe trench concealed beneath the litter, make up around 11% of this sub-community. Disturbance within this community is attributed to it housing 17% of the island’s hotel support infrastructure, including operational facilities such as a desalination plant and a power generation facility, which encompassing ~1150 m².

Characteristic species for sub-community 1.1 are represented by species group B in Table 2, and include the herbaceous scrub Melanthera biflora, the perennial grass species Panicum repens, the evergreen shrub Gymnosporia heterophylla, the pioneer dune creeper Ipomoea pes-caprae, the beach-stabilising pioneer shrubs Scaevola sericea and S. taccada, the pike sedge Kyllinga erecta, and the leguminous shrub Sesbania bispinosa.

The woody layer is dominated by Casuarina cunninghamiana (species group A), while the screw pine Pandanus kirkii (species group D) and the broadleaf evergreen coastal species Sideroxylon inerme (species group G) in its shrub growth form are dominant in localised areas. The erect shrub Suregada zanzibariensis (species group D) is prominent, and the catch-bird tree Pisonia grandis (species group D) has a cover-abundance of less than 1%. In the understory, scamper climbers like Cyphostemma dysocarpum and Dalechampia scandens (species group G) are prominent. The scrambling shrub Melanthera biflora (species group B) is locally dominant in the shrub layer. Enteropogon macrostachyus (species group A) is dominant in the grass layer, with Dactyloctenium aegyptium (species group A) and Panicum repens (species group B) being locally prominent. The herbaceous layer is represented by Boerhavia diffusa (species group F). A locally prominent lichen, Evernia prunastri (species group A), and two fungi species, Polyporus tubaeformis (species group A) and Phaeotrametes decipiens (species group E), are also present.

The overall vegetation cover for this community is 79%, with an average of 21 species per 400 m² of relevé. A total of 37 plant species were recorded for this sub-community, representing 49% of all the species identified during the relevé sampling process. The tree cover ranged from 54% to 92% across the sample relevés, with an average of 74%. Trees with a BHD above 6 cm at a low level in the canopy constitute 29%, while those at higher canopy levels account for 71%. There is no mid-level canopy in this sub-community. The average minimum tree height is 12.1 m, and the average maximum tree height is 16.7 m. The average tree BHD is 27.4 cm, with the shortest tree being 3 m high and the tallest being 28 m. The number of Pisonia grandis trees is 100 individuals per hectare, with an average BHD of 23.5 cm and a height of 12.0 m. The number of Casuarina cunninghamiana trees is 419 per hectare, with an average BHD of 29.4 cm and height of 22.5 m.

The shrub cover in this community ranges from 25% to 75%, with an average of 46%. The shrub height ranges from 0.4 m to 1.37 m, with an average height of 1.0 m. The average grass/sedge cover is 10%, the average climber/creeper height is 1.7 m, and the average herb/forb cover is 24%.

The Casuarina cunninghamiana–Suriana maritime–Chrysothrix Open Sandy Dry Coastal Forest

This sub-community comprises the salt-spray zone and frontal dune vegetation, which is situated along the periphery of the island. It covers 36% of the island’s total vegetated area and 68.6% of Community 1 (Figure 4). This sub-community is characterised by sandy soils with good drainage and a predominant litter layer covering 96% of the area, consisting mainly of Casuarina cunninghamiana dropped branchlets, seeds, small twigs, and branches, with an average depth of 3.0 cm. The fermentation layer below the litter layer extends to approximately 7 cm deep. Fallen trees or sections thereof amount to 53/ha, with an average of 61/ha of deceased trees still standing.

Sandy soils with good drainage prevail throughout sub-community 1.2, and rock cover is absent. Approximately 8% of the sampled areas consist of non-vegetated areas, which include small footpaths and beach extensions that encroach into the vegetation’s edge. This sub-community houses 43% of the island’s hotel support infrastructure, including laundry services, staff houses, and guest facilities, displacing approximately 2828 m2 of the island’s vegetation. The uprooting of C. cunninghamiana trees due to high-tide wave interaction on the eastern side of the island also contributes to disturbance in this sub-community.

Characteristic species for sub-community 1.2 are represented by species group C in Table 2, including the exotic woody scrambling shrub Bougainvillea spectabilis, the gold-dust lichen Chrysothrix sp., the stem-decaying bracket fungus Fomitopsis pinicola, the perennial herb Bidens pilosa, and the perennial tufted grass Panicum maximum.

The woody layer is dominated by Casuarina cunninghamiana and Suriana maritima (species group A), while broadleaf evergreen coastal species like Sideroxylon inerme and the evergreen shrub Capparis viminea (species group G) are locally prominent. Dalechampia scandens (species group G) prevails as a climber in the understory, and Cyphostemma dysocarpum (species group G) is similarly prominent. The grass layer is dominated by Enteropogon macrostachyus (species group A), and Panicum maximum (species group C) is locally prominent. The herb and forb layer are represented by the creeping herb Aneilema indehiscence, the climbing perennial herb Rhynchosia sublobata (species group A), and the annuals Bidens pilosa (species group C) and Boerhavia diffusa (species group F). The locally prominent lichen Chrysothrix sp. and the fungus Fomitopsis pinicola (species group C) were also identified in this sub-community. The castor oil shrub Ricinus communis (species group A) is locally prominent in the undergrowth.

The overall vegetation cover for this community is 73%, with an average of 15 plants per 400 m2 of relevé. A total of 29 plant species were recorded for this sub-community, representing 39% of all the species identified during the relevé sampling process. The tree cover ranges from 50% to 83%, with an average of 70%. The average percentage of trees with a BHD above 6 cm in the lower canopy averages 38%; in the medium canopy, 15%; and in the upper canopy, 30%. The minimum tree height is approximately 4.0 m high, the maximum is 28 m, and the average tree BHD is approximately 14.8 cm. Within this context, a total of 538 Casuarina cunninghamiana trees were counted per hectare.

The shrub cover in this sub-community ranges from 5% to 35%, with an average of 22%. The minimum shrub height is 0.2 m, the maximum is 10.8 m, and the average is 0.64 m. The average grass/sedge cover is 0.3%, and the average herb cover is 3%. The average climber/creeper height is 1.4 m, with the tallest climber reaching 8 m.

3.1.2. The Eugenia capensis–Mimusops obtusifolia Coastal Forest

The second main community on the island is a triangular-shaped natural coastal forest community that covers 47% of the island’s vegetated area and is located within the interior of the island. This community is surrounded almost entirely by the first main plant community, and a portion of it extends onto the southern beach of the island, covering a 40 m stretch (Figure 5). The dominant species in this community is the evergreen dune myrtle shrub Eugenia capensis.

Characteristic species for this community include the evergreen milkwood tree Mimusops obtusifolia, the screw pine Pandanus kirkii, the woody vine Secamone punctulate, the parasitic climber Cassytha filiformis, the evergreen coastal shrub Grewia glandulosa, the scrambling shrub Maerua triphylla, the climbing shrub Synaptolepis kirkii, the broadleaf evergreen “catch-bird” tree Pisonia grandis, the small and semi-evergreen woody shrub Ecbolium ligustrinum, the evergreen coastal shrub Suregada zanzibariensis, the strangling fig Ficus Scassellati, the hardy perennial herb Afroligusticum linderi, seedlings from the palm species Cocos nucifera, and the perennial liana Adenia gummifera (all from species group D) (Table 2).

The woody layer in this community is dominated by the multi-stemmed shrub Eugenia capensis (species group D), the trees Mimusops obtusifolia, Pandanus kirkii, and Pisonia grandis (species group D) and Sideroxylon inerme (species group G), and the lianas Secamone punctulate (species group D) and Capparis viminea var. orthacantha (species group G). The strangling fig Ficus Scassellati, the exotic palm Cocos nucifera, the perennial liana Adenia gummifera (species group D), and the shrubs Suregada zanzibariensis, Grewia glandulosa, Ecbolium ligustrinum, and Maerua triphylla (species group D) and Polysphaeria parvifolia (group F) are locally prominent. Dalechampia scandens (species group G) dominates the climber/creeper layer in the understory, while the parasitic vine Cassytha filiformis (species group D) and the climbing Cyphostemma adenocaula (species group G) are prominent in the canopy layer.

In the herb layer, the perennial herb Afroligusticum linderi (species group D) is locally prominent, and in the grass layer, Dactyloctenium aegyptium (species group A) is locally prominent. The forest floor of this community is covered in a diverse litter layer, making up 82% of the surface, and it includes dry leaves, twigs, seeds, fruits, flowers and coconut husks. The average litter depth ranges from 2.1 to 5.8 cm, and the fermentation layer underneath measures between 4.1 and 4.7 cm. Sandy loam soils with good drainage predominate throughout this community, with the rock cover constituting 1.0%. Approximately 8% of this community comprises non-vegetated areas, which include sandy clearings, paths, and antelope middens. Infrastructure, making up 40% of all buildings on the island, displaces 2611 m² of vegetation. Additional disturbances arise from guest activities and operational activities, which include waste recycling.

Two sub-communities were distinguished for Community 2.

The Eugenia capensis–Mimusops obtusifolia–Scutia myrtina Coastal Forest

This sub-community covers 41% of the island’s vegetated area and is predominantly situated within the island’s interior, extending to the beach and the frontal dune zone in the south (Figure 4). It is characterised by a diverse litter layer covering 86% of the forest floor, with an average depth of 5.8 cm, including leaves, twigs, branches, seeds, fruits, coconut husks, bark, and dropped Casuarina cunninghamiana branchlets. Below this layer, the fermentation layer is approximately 4.1 cm thick. The sub-community has an average of 19 fallen trees and 5 standing dead trees or shrubs per hectare.

Well-drained sandy loam soils are prevalent, with an average rock cover of 1.7%. Non-vegetated areas, including sandy footpaths, animal lairs, and clearings, make up 6% of this sub-community. Infrastructure, such as operational buildings and staff accommodation, covers 35% of sub-community 2.1. Infrastructure covers 35% of this sub-community and includes all lodge operational buildings such as the kitchen, dining areas for guests and senior staff, and staff accommodation. Additional disturbances to the vegetation in this sub-community are the various operational activities, such as waste recycling and guest engagements.

Characteristic vegetation in sub-community 2.1 is represented by species group E (Table 2), consisting of two fig tree species (Ficus lutea and F. polita), seven shrub species (Pavetta stenosepala, Bourreria petiolaris, Psychotria psychotrioides, Justicia capensis, Ludia mauritiana, Conyza newii, and Solanum viarum), along with one vine species (Scutia myrtina). Characteristic fungal species found are the jelly ear fungus Auricularia auricular-judae, the lichenised fungus Xanthoria sp., the polypore fungus Phaeotrametes decipiens, and the inedible fungus Daldinia concentrica.

The woody layer is dominated by four tree species (Mimusops obtusifolia, Pandanus kirkii, Pisonia grandis (species group D), and Sideroxylon inerme (species group G)), the multi-stemmed shrub Eugenia capensis (species group D), and the lianas Secamone punctulate (species group D) and Scutia myrtina (species group E). Ficus lutea (species group E) is locally prominent in this layer. In the shrub layer, Ecbolium ligustrinum, Suregada zanzibariensis, Maerua triphylla, and Grewia glandulosa (species group D), Pavetta stenosepala (species group E), and Capparis viminea var. orthacantha (species group G) are prominent. Dalechampia scandens and Cyphostemma adenocaule (species group G) are prominent climbers/creepers in the understory, while the parasitic vine Cassytha filiformis (species group D) is prominent in the canopy layer. In the grass layer, the monotypic fungus Phaeotrametes decipiens (species group E) is locally prominent. The herb layer is absent in sub-community 2.1.

A total of 44 plant species (59% of all the species identified in this study) were recorded in this sub-community, which equates to an average 18 species per 400 m². The tree cover ranges from 13% to 69%, with an average of 36%. The average percentage of trees in the lower canopy is 16%, 56% in the medium canopy, and 28% in the upper canopy. The minimum tree height is 1.5 m, and the maximum height is 38 m, with an average of 7.5 m. On average, the trees have a BHD of 22.9 cm, with 244/ha having a BHD larger than 6 cm.

The shrub cover ranges from 30% to 75%, with an average of 52%. The minimum shrub height is 2.0 m, and the maximum height is 12 m, with an average of 2.6 m. No grasses, sedges, or herbs are present in sub-community 2.1. The average climber/creeper height is 0.8 m, with the tallest climber measuring 18.0 m.

The Eugenia capensis–Mimusops obtusifolia–Clerodendrum glabrum Coastal Scrub

This sub-community constitutes a smaller portion of Community 2 and covers 6% of the island’s vegetated area. It is divided into three separate segments, staggered from north to south, and is characterised by a diverse litter layer that covers 82% of the forest floor in the sub-community, with an average depth of 2.1 cm. This litter layer includes various materials such as leaves, twigs/branches, seeds, pods, remnants of birds’ nests, guano, and antelope dung. The fermentation layer beneath this litter layer is approximately 4.7 cm thick. There is an average of 34 fallen trees and 8 standing dead trees/shrubs per hectare.

Sandy loam soils, known for their good drainage properties, serve as the substrate for this sub-community. Coral rock, scattered across the sandy substrate, accounts for 0.3% of the rock cover in sub-community 2.2. Non-vegetated areas, consisting of sandy footpaths, clearings, middens, and lairs, make up 9% of this sub-community. Infrastructure, in the form of staff accommodation, constitutes 304 m² of this sub-community. Additional disturbances in this sub-community include an old well and a water trench.

Characteristic vegetation in this sub-community is represented by species group F (Table 2) and includes the shrub species Morinda citrifolia, Pavonia sp., Polysphaeria parvifolia, Acacia auriculiformis, Ehretia amoena, and Clerodendrum glabrum. Herbaceous species found in this sub-community are Abutilon sp., Ipomoea sp., Boerhavia sp., Pupalia lappacea, Phyllanthus amarus, and Boerhavia diffusa. The sedge Kyllinga platyphylla and the perennial vine Coccinia grandis are also characteristic. Characteristic fungi include the polypores Daedaleopsis confragosa and Pycnoporus sp., the foliose lichen Parmelia sulcate, and the hairy bracket Trametes hirsuta.

The woody vegetation in sub-community 2.2. is dominated by the multi-stemmed Eugenia capensis, Grewia glandulosa, and Ecbolium ligustrinum (species group D) in the shrub layer, as well as by Mimusops obtusifolia (species group D) in the tree layer. Prominent species include the trees Pandanus kirkii (species group D) and Sideroxylon inerme (species group G), the shrubs Aneilema indehiscens and Ricinus communis (species group A), Suregada zanzibariensis, Ecbolium ligustrinum, Maerua triphylla and Synaptolepis kirkii (species group D), and Pavonia sp., Abutilon sp., Polysphaeria parvifolia, and Phyllanthus amarus (species group F). Prominent climber/creeper species are Rhynchosia sublobata (species group A), Pisonia grandis, Secamone punctulate, and Cassytha filiformis (species group D), Clerodendrum glabrum and Ipomoea sp. (species group F), and Cyphostemma adenocaule, Capparis viminea var. orthacantha, and Dalechampia scandens (species group G). The grass layer is represented by the locally prominent species Dactyloctenium aegyptium (species group A) and Eragrostis ciliaris (species group F). Daedaleopsis confragosa (species group F) is a prominent fungus species. In the herb layer, Boerhavia diffusa (group F) is prominent.

A total of 53 plant species, approximately 24 plants/400 m², were recorded for this plant community, which constitutes 71% of all the species identified during the study. The tree cover in sub-community 2.2 ranges from 8% to 71%, with an average of 31%. The average percentage of trees in the lower canopy is 52%, 42% in the medium canopy, and there are no trees in the upper canopy. The minimum tree height is 1.5 m, and the maximum height is 12 m, with an average of 5.9 m. The average tree BHD is 14.4 cm, and the number of trees with a BHD greater than 6 cm is 40/ ha.

The shrub cover in sub-community 2.2 ranges from 25% to 86%, with an average of 54%. The minimum shrub height is 0.8 m, and maximum height is 9.0 m, with an average of 1.8 m. No grasses or sedges are present in this sub-community. The average climber/creeper height is 2.2 m, with the tallest climber being at a height of 10.0 m.

3.2. Plant Species Not Recorded in Sample Relevés

The subshrub Aneilema indehiscens was observed growing within a mature, fenced-off enclosure, which effectively prevented browsing by Suni and Aders’ antelopes, both of which are present on the island. Due to the size and configuration of the plant communities on Mnemba Island, particularly in the frontal dune beach zone, some plot placements inadvertently excluded the peripheral areas of these communities. Species identified in these excluded areas along the shoreline included the salt-tolerant shrub Pemphis acidula, coastal grasses such as Sporobolus virginicus, Halopyrum mucronatum, and Paspalum virgatum, the perennial beach shrub Guettarda speciose, and the coastal shrub Cordia subcordata. Notable species encountered within the interior of the island but not recorded within any of the surveyed relevés included the pioneer species Trema orientalis, the perennial vines Parquetina nigrescens and Searsia natalensis from the sumac family, the coastal shrub Euclea natalensis, the arid forest shrub Flueggea virosa, the liana Caesalpinia bonduc, and the strangling fig Ficus ingens. Additionally, the naturalised shrub Tacca leontopetaloides was observed.

3.3. Floristic Characteristics

The vegetation survey on Mnemba Island identified 91 species belonging to 54 families. A floristic analysis was conducted to compare the composition of the two distinct plant communities present.

Community 1 comprised 45 species from 43 genera, constituting 51% of the total species inventory (

Table 3. Families and number of genera and species for plant community 1 (includes fungi and lichens).

Family	Genera	Species	Family	Genera	Species
Acanthaceae	1	1	Fomitopsidaceae	1	1
Amaranthaceae	1	1	Goodeniaceae	1	1
Apocynaceae	1	1	Malvaceae	1	1
Asclepiadaceae	1	1	Moraceae	1	1
Asteraceae	2	2	Myrtaceae	1	1
Capparaceae	2	2	Nyctaginaceae	3	3
Casuarinaceae	1	1	Pandanaceae	1	1
Celastraceae	1	1	Parmeliaceae	1	1
Chrysothricaceae	1	1	Poaceae	5	6
Combretaceae	1	1	Polyporaceae	1	1
Commelinaceae	1	1	Sapotaceae	2	2
Convolvulaceae	1	1	Surianaceae	1	1
Cyperaceae	1	2	Thymelaeaceae	1	1
Euphorbiaceae	3	3	Tiliaceae	1	1
Fabaceae	3	3	Vitaceae	1	1

). These species originated from 31 plant families, representing 61% of all families identified. The most well-represented families in Community 1 were Euphorbiaceae, Fabaceae, Nyctaginaceae, and Poaceae. Collectively, these families, along with Asteraceae, Capparaceae, Sapotaceae and Cyperaceae, accounted for 49% of the species richness in this community.

Community 2 consisted of 64 species from 64 genera, accounting for 72% of the total species on the island (

Table 4. Families and number of genera and species for plant community 2 (includes fungi and lichens).

Family	Genera	Species	Family	Genera	Species
Acanthaceae	2	2	Myrtaceae	1	1
Amaranthaceae	1	1	Nyctaginaceae	3	3
Apiaceae	1	1	Palmae	1	1
Apocynaceae	1	1	Pandanaceae	1	1
Araucariaceae	1	1	Parmeliaceae	2	2
Asteraceae	1	1	Passifloraceae	1	1
Boraginaceae	2	2	Phyllanthaceae	1	1
Capparaceae	2	2	Poaceae	4	4
Casuarinaceae	1	1	Polyporaceae	5	5
Celastraceae	1	1	Rhamnaceae	1	1
Chrysothricaceae	1	1	Rubiaceae	4	4
Combretaceae	1	1	Salicaceae	1	1
Commelinaceae	1	1	Sapotaceae	2	2
Convolvulaceae	1	1	Solanaceae	1	1
Curbitaceae	1	1	Teloschistaceae	1	1
Cyperaceae	1	1	Thymelaeaceae	1	1
Euphorbiaceae	3	3	Tiliaceae	1	1
Fabaceae	2	2	Verbenaceae	1	1
Lauraceae	1	1	Vitaceae	1	1
Malvaceae	2	2	Xylariaceae	1	1
Moraceae	3	3

). These species belonged to 42 families, constituting 82% of the identified families. Parmeliaceae, Polyporaceae, Acanthaceae, Boraginaceae, Capparaceae, Euphorbiaceae, Fabaceae, Malvaceae, Moraceae, Nyctaginaceae, Rubiaceae, Sapotaceae, and Poaceae were the most well-represented families, collectively contributing to 56% of the species richness in Community 2.

The two communities shared 28 plant families. Community 1 was distinguished by the presence of Fomitopsidaceae, Goodeniaceae, and Surianaceae, which were absent in Community 2. Conversely, Community 2 featured 14 families that were not present in Community 1, including Apiaceae, Verbenaceae, Phyllanthaceae, Lauraceae, Boraginaceae, Salicaceae, Palmae, Passifloraceae, Rhamnaceae, Rubiaceae, Solanaceae, Araucariaceae, Xylariaceae, and Teloschistaceae.

In Community 1, Casuarinaceae and Asteraceae accounted for 41% and 21% of the composition, respectively (

Table 5. Cover-abundance percentages for plant families in plant communities 1 and 2.

Community 1		Community 2
Family	Cover (%)	Family	Cover (%)
Casuarinaceae	41%	Myrtaceae	23%
Asteraceae	21%	Sapotaceae	12%
Surianaceae	10%	Pandanaceae	5%
Euphorbiaceae	3%	Nyctaginaceae	3%
Pandanaceae	2%	Apocynaceae	8%
Fabaceae	2%	Acanthaceae	1%
		Rhamnaceae	1%

). In Community 2, Myrtaceae and Sapotaceae contributed 23% and 12% of the composition, respectively (Table 5).

Shannon–Weiner Index values and evenness for the main and sub-communities are depicted in

Table 6. Mnemba plant community species diversity (Shannon–Weiner Index) and evenness.

Plant Community	Shannon–Weiner Index (H)	Evenness (E)
1	2.24	0.97
1.1	1.84	0.95
1.2	1.06	0.96
2	2.54	0.94
2.1	2.53	0.96
2.2	2.53	0.88

.

Higher Shannon–Weiner Index values indicate increased richness and evenness, while higher evenness values suggest greater homogeneity within the plant communities. Community 2 and its associated sub-communities exhibited the highest diversity, while Community 1 displayed a higher degree of homogeneity. Among the sub-communities, sub-community 2.2 had the lowest evenness value of 0.88, indicating the lowest level of homogeneity.

4. Discussion

The island’s vegetation comprises two main communities: Community 1, the Casuarina cunninghamiana–Suriana maritima open to closed sandy dry coastal forest; and Community 2, the Eugenia capensis–Mimusops obtusifolia coastal forest. Each main community has two sub-communities. The indigenous forest, designated as Community 2, plays a vital role in the island’s ecology by supporting bird breeding colonies and providing a habitat for the threatened Aders’ duiker. However, the scarcity of understory development in this community indicates the historical impact of the Suni antelope on shrub and tree regeneration, necessitating strategic antelope population management to facilitate ecosystem recovery [30]. The dominance of Pandanus kirkii and the presence of ecologically important species like Terminalia catappa and Pisonia grandis in this plant community underscore the island’s unique ecological profile. Notably, Pisonia grandis’s adaptability and potential as a pioneer species hold implications for island afforestation efforts in similar challenging environments [31]. Eugenia capensis is the most abundant species in the indigenous forest (Community 2), which exhibits regional affinity with the coral-rag forest from the Tondooni peninsula of Ngezi Forest in Pemba [32], where Eugenia spp., Parvifolia spp. and various Ficus spp. are listed as abundant.

Human settlement has notably influenced Mnemba Island’s vegetation, as evidenced by the established Casuarina cunninghamiana forest (Community 1). This fast-growing pioneer species facilitated the initial expansion of the island’s landmass. However, the long-term impact of this species on vegetation diversity and understory development warrants consideration [33]. The observed dominance of Casuarina stands encircling the indigenous forest reflects the protective role of this species against salt spray and wind. Nevertheless, their presence also raises concerns regarding suppressed understory growth [34,35] and a reduction in overall species diversity, highlighting the trade-off between ecological stability and biodiversity [33].

Species diversity variations among plant communities provide insights into the island’s historical ecological development. Community 2 demonstrates a higher species diversity than Community 1 and contains fragments of natural vegetation interspersed with alien plants, revealing a more developed understory structure [33]. Due to the island’s small size and close proximity to Zanzibar, species richness was anticipated to fall between the high and low ranges of Unguja’s diversity, possibly at the lower end of the range for evenness. Smaller habitats often reflect lower evenness, while richness and diversity can still be relatively high. Low evenness serves as an indicator of the fragmentation or isolation of forest patches [30]. Additionally, the local incidence of ruderals could be attributed to various types of disturbances, including the impact of tourism and related activities. The high species diversity and presence of non-pioneer plants in sub-community 2.2 suggest an older, more established ecosystem, while the lack of understory development, due to the previous overpopulation of the Suni antelope, suggests the recent degradation of this sub-community. In all sub-communities where herbs were identified, many species were only present as seedlings, which is typical for small islands in the Indian Ocean as a result of low survival rates [6,7]. These findings corroborate the island’s evolutionary trajectory, reflecting the influence of time and ecological forces [3].

Our study highlights the contrasting development of the shrub and herb layers across different plant communities. While the indigenous forest (Community 2) exhibits a more developed understory structure, the Casuarina-dominated outer edge (Community 1) demonstrates lower diversity. Factors such as salt spray, allelochemicals, and herbivore browsing influence the herb layer’s abundance and composition [34,36]. Furthermore, the influence of recent drought conditions underscores the susceptibility of Mnemba’s vegetation to climatic fluctuations [37].

The dynamic interplay between vegetation succession and environmental processes on Mnemba Island aligns with concepts in restoration ecology. The indigenous forest’s characterisation by numerous species showcases the island’s mature natural state, reflecting the atoll’s natural succession process [38,39]. This emphasises the potential for ecological restoration by understanding and mimicking natural succession processes to restore degraded ecosystems. Additionally, the emphasis on environmental processes reflects the broader concept of ecological succession, highlighting the continuous and dynamic nature of ecosystem development. Our findings highlight the vulnerability of the island’s vegetation to continuous oceanic changes, resulting in a delicate balance between erosion and succession. Ongoing changes in ocean currents, combined with continuous erosion and sedimentation, perpetuate a dynamic environment that is conducive to atoll vegetation adaptation. The cyclical effects of the East African Coastal Current (EACC) drive the deposition and erosion of sand, shaping the island’s topography and vegetation distribution [37,40,41].

The indigenous flora of Mnemba comprises species like Grewia glandulosa, Ludia mauritiana, Euclea racemosa, Flueggea virosa, Mimusops obtusifolia, Pandanus kirkii, and Sideroxylon inerme, which align with the Zanzibar–Inhambane edaphic coral-rag scrub forest and undifferentiated forest types. Species typical of scrub forest in the absence of coral karst include Adenia spp., Suregada zanzibariensis, and Terminalia catappa. Zanzibar–Inhambane transitional rainforest species such as various Ficus spp., Morinda citrifolia, Cordia subcordata, Gymnosporia heterophylla, and Trema orientalis were also identified at the study site. This affinity with the East African coastal mosaic affirms the island’s ecological position within the region [37,42,43,44].

Our study identified several plant species typical of Indian ocean atolls [45,46,47], including Casuarina cunninghamiana (an introduced alien invasive species), Trema orientalis, Fluggea virosa, Searsia natalensis, Caesalpinia bonduc, Melanthera biflora, Boerhavia diffusa, Bidens pilosa (a cosmopolitan weed), Aneilema indehiscence, Dalechampia scandens, Rhynchosia sublobata, Enteropogon macrostachyus, Panicum repens, Dactyloctenium aegyptium, Kyllinga erecta, Ipomoea pes-caprae, Scaevola sericea, and S. taccada. Understanding the composition and distribution of these species aligns with invasion ecology, providing insights into the potential for biological invasions and their ecological impacts on the atoll ecosystem. This is crucial for conservation and management strategies to prevent and control invasive species that may threaten the native biodiversity.

5. Conclusions

The investigation of Mnemba Atoll’s flora and vegetation contributes conceptually to a better understanding of vegetation structure and dynamics in island ecosystems by highlighting the intricate relationship between plant species composition, environmental factors, and human activities. This study provides valuable insights for conservation and management by emphasising the fragile yet resilient nature of Mnemba Atoll’s unique ecosystem. It offers insights into the island’s vegetation ecology and its affinities with analogous regions, indicating the interplay between natural processes and human influence.

Our examination of Mnemba Island’s flora and vegetation, covering an area of 11 ha, revealed distinct vegetation patterns. The island’s established indigenous plant communities and the frontal beach zone showed affinities with Indo-pacific sandy cays that have been studied by numerous botanists [31,48,49,50]. The predominant vegetation type, the Sideroxylon inerme–Grewia glandulosa coastal scrub forest, exhibits strong affinities with the Zanzibar–Inhambane edaphic coral-rag scrub forest, aligning with the larger coastal mosaic of East Africa [42,43,44].

A notable divergence in species diversity and composition between the “planted” Casuarina cunninghamiana section and the natural atoll vegetation emerged as a key finding. Contrary to expectations, the natural forest (Community 2) demonstrated remarkable species richness and an even distribution of species, indicative of ecological connectivity with the main island of Zanzibar, which is particularly facilitated by seed dispersal mechanisms involving migratory birds and bats [30].

Future mitigation of erosion and the enhancement of island stability should be supported through the gradual replacement of Casuarina cunninghamiana with indigenous beach-stabilising plant species. Suitable species include Ipomoea pes-caprae (a spreading vine that has the ability to stabilise dunes and sandy soils with its extensive root system), Scaevola taccada (a salt-tolerant shrub that helps bind sand with its spreading roots and branches), Suriana maritima (an evergreen salt-pruned shrub or small tree that stabilises sandy soils), Terminalia catappa (a tree with an extensive root system that helps stabilise coastal areas), and Pandanus kirkii (a palm with fibrous roots that binds sand effectively). This strategic transition, coupled with active management of the island’s vegetation, has the potential to halt further erosion and to stabilise both the soil and the island’s ecosystem. The transformation observed since [14]’s assessment highlights the rapid establishment of monostands by Casuarina cunninghamiana, prompting the consideration of controlled dune formation and sedimentation processes as mechanisms to counter erosion.

Findings from this study expand the vegetation inventory for the island, which now comprises 91 species across 54 families, enhancing our understanding of the island’s botanical diversity. The observed similarities with regional ecosystems, the interplay between natural processes and human activities, and the complex relationships among species and communities lay a solid foundation for future research endeavours. Long-term monitoring and conservation initiatives should consider the delicate balance between erosion and ecological succession, the implications of dominant species, and the potential role of adaptable pioneer species like Pisonia grandis in shaping island ecosystems. Given that the Casuarina cunninghamiana community surrounds most of Community 2, there is an increased risk of future invasion into Community 2. Monitoring is essential in this context, and removing alien plant species whenever they are encountered is recommended. This practice should also be extended to other non-native species, especially since the higher diversity in Community 2 appears to be partly due to the increased presence of alien plants. The study’s findings can be applied beyond the study region by serving as a reference for understanding the ecological dynamics, regional affinities, and conservation prospects of similar island ecosystems in other parts of the world.