Lake Mboandong (4˚45'N and 9˚26'E, 120 m altitude) is an elliptical crater lake of about 25 hectares, with an average depth of 5 m, lying in the rain shadow of Mt. Cameroon, with very close proximity to Lake Barombi Kotto ( Figure 1 ). Its bottom is relatively flat at 5 m with a characteristically thick layer of organically-rich mud and silty clay ( Richards, 2021 ). It has a tropical climate, with an average annual temperature of 26.6˚C and an annual precipitation of 3216 mm. Precipitation is highest in July and August, while January and February are known to be the driest months ( Campbell et al., 2017 ). The vegetation is made up of secondary forests where cocoa and oil palms are cultivated together with staples like cassava, plantains and cocoyams ( Awo et al., 2020 ; Richards, 2021 ).

Field sampling was done in two seasons (dry and rainy), with samples collected at the 4 cardinal points, from Stations 1 to 4 ( Table 1 ). Water samples were collected within 1 m depth using a Van Don Water sampler and transferred to 1 litre plastic bottles. This was repeated twice to collect 3 replicates per sampling point. These bottles were immediately put in coolers containing ice blocks after three drops of Lugol’s iodine were added to each of these plastic bottles to preserve the algae.

Physicochemical parameters were determined very early in the morning between 6 and 7 a.m. before sunrise to avoid the influence of the sunlight on water temperature. Water temperature, total dissolved solids, Hydrogen potential (pH), and electrical conductivity, were determined in situ using a Hanna HI 98,494 multi-parameter probe. A different set of water samples were labelled and put in a cooler containing ice blocks, parcelled and sent to the Hydrobiology and Environmental Laboratory of the University of Yaoundé 1 within 24 hours for the analysis of seven nutrients and anions and cations (Phosphates, magnesium, nitrates, nitrites, potassium, sodium, and ammonium ions) following standard methods ( APHA, 2005 ).

Samples were preserved in a 5% formalin solution ( Suthers et al., 2019 ), and taken to the Plant Science Laboratory of the University of Buea for identification and counting of the phytoplankton cells, using a Sedgwick-Rafter counting chamber under a compound microscope equipped with a micrometre. The identification was done using comparative morphology based on phytoplankton identification keys ( Bellinger & Sigee, 2010 , 2015 ; Suthers et al., 2019 ; van Vuuren et al., 2006 ; Verlecar & Desai, 2004 ). Algal abundance was expressed in cells/L.

Data was keyed into Microsoft Excel Version 2021 and analysed with the R package for statistical computing version 4.2 ( R Core Team, 2024 ). Various ecological indices (species occurrence, abundance, richness, evenness and Shannon-Weiner diversity index) were used to elucidate species composition, structure and diversity across sites and seasons. The dissimilarity between study sites was assessed using Jaccard’s dissimilarity index. Pearson’s correlation analysis was used to shed light on the relationship between phytoplankton diversity and water physicochemical parameters. Principal component analysis was used to simplify the complexity of the dataset, uncover patterns, and identify relationships between variables and the key variables driving the dynamics of the system.

Shannon-Weiner diversity index ( $H^{\prime }$): $H^{\prime }=-{\displaystyle {\sum }_{i=1}^s{P}_i\ln P_i}$, with P_i = n_i/N; n_i was the number of individuals belonging to a species i and N corresponded to the total number of species.

$H^{\prime }$ takes into account species abundance and it is sensitive to species with low frequencies. Typical values usually range between 1.5 and 3.5 bits in most ecological studies, and rarely more than 4 in most ecological studies. It increases as both the richness and the evenness of the community increase ( Fai et al., 2023 ; Inyang & Wang, 2020 ).

Pielou Evenness Index (J) was computed according to Pielou (1975) . The value of J varies between 0 (one species dominates) and 1 (all the species tend to have the same abundance). J = H'/log₂S, where: S is the total number of species or taxonomic richness.

A total of 77 phytoplankton species were identified from seven algal divisions ( Table 2 ). These included Bacillariophyta (18 species), Charophyta (7 species), Chlorophyta (23 species), Chysophyta (5 species), Cyanophyta (11 species), Dinophyta (6 species), and Euglenophyta (7 species). Amongst these, Chlorophyta exhibited the highest species diversity, encompassing 23 different species. There were differences in the phytoplankton abundance across seasons, with higher abundances recorded in the dry season than during the rainy season. Concerning the distribution of algal groups, Cyanophyta was the most dominant in both seasons (56.31% and 28.13% in the dry and rainy seasons respectively). The composition of Cyanophyta was higher in both seasons relative to the other algal divisions. Dry season algal composition was in the following order: Cyanophyta > Bacillariophyta > Chlorophyta > Euglenophyta > Dinophyta > Chrysophyta and Charophyta while rainy season composition was Cyanophyta > Chlorophyta > Bacillariophyta > Euglenophyta > Dinophyta > Chrysophyta > Charophyta.

Higher phytoplankton was recorded during the rainy season than in the dry season ( Table 3 ). Species diversity ranged from 1.69 to 2.48 bits, with S3 and S4 having the lowest and highest algal diversities respectively. During the rainy season, diversity ranged from 2.35 in S1 to 2.83 bits in S4. Differences in diversity across sites were equally recorded in both the dry and rainy seasons. Site 4 (S4) had the highest phytoplankton diversity during the dry season (2.48 bits) while the lowest was in S3 (1.63 bits).

Generally, according to the classification by Shekhar et al. (2008) , which associated Shannon’s diversity values with water quality, it could be inferred that all sites, except Site 3 in the dry season are mildly polluted since (3 ≤ H ≤ 4).

D = dry season, R = rainy season, A = abundance, S = species richness, J = evenness index, H' = Shannon-Weiner diversity index.

The similarity between study sites permitted to grouping of them according to the common species ( Figure 2 ). Group 1 brought together sites 1 and 2 during rainy and dry seasons. Group 2 brought together sites 3 and 4 during rainy and dry seasons. This grouping revealed the proximity of sites 1 and 2, and sites 3 and 4, hence similar algal species.

Generally, these parameters did not display remarkable differences between sampling sites, indicating the homogeneity of water quality in most portions of the lake. That notwithstanding, seasonal differences were recorded for most of the parameters measured.

Water temperature showcased fluctuations with higher mean water temperature in the dry season of 19.04˚C ± 0.44˚C than in the rainy season of 18.38˚C ± 0.52˚C (p = 0.016). The recorded low temperatures could be attributed to the shielding effect of the large canopies of trees which dominated the study site.

Contrary to adjacent Lake Barombi Kotto, a greater portion of this lake was shaded by tree canopies, thus accounting for the lower temperatures. pH was neutral to slightly basic, 7.49 ± 0.28 and 8.66 ± 0.07 in the rainy and dry seasons respectively. Higher pH recorded during the dry season could be because the high phytoplankton biomass carries out photosynthesis, thus reducing inorganic carbon from the water which shifts the pH towards high values by loss of alkaline reserve. This mechanism is thought to drive higher pH values during warmer periods such as the dry season ( Staehr & Sand‐Jensen, 2006 ). The pH was within the range of that recorded in Lake Barombi Kotto ( Awo et al., 2020 ; Campbell et al., 2017 ). The lake’s electrical conductivity was in the range of 236-243 but with significant seasonal differences. Higher turbidity occurred during the rainy season at 10.93 ± 3.82 NTU than in the dry season at 6.56 ± 0.44 NTU as a result of the dead organic matter and debris brought into the lake by runoffs during the rainy season ( Table 4 ). Nitrates were significantly higher during the rainy season of 1.92 ± 0.99 mg/l as opposed to 0.52 ± 0.27 mg/l in the dry season (p = 0.002). A similar trend was observed for nitrites and ammonium levels. Apart from allochthonous nutrient sources, cyanobacteria are known to increase nitrates in water via processes of nitrogen fixation by the nitrogen-fixing genera. In freshwater lake ecosystems, NH₃-N, NO₃-N and NO₂-N are the three main forms of soluble nitrogen which are very important to phytoplankton and aquatic plants ( Liu et al., 2019 ). NH₃-N is generated by heterotrophic bacteria during the decomposition of organic material and remains a very vital source of nitrogen to phytoplankton and this is readily converted via nitrification by bacteria ( Datta, 2012 ). Very low dissolved oxygen levels were recorded in the lake, although significantly higher during the rainy than in the dry season at 3.78 ± 0.43 mg/l and 2.35 ± 0.14 mg/l respectively. Low dissolved oxygen levels suggest high organic pollution possibly from nearby cocoa farms and high inputs of eroded materials from these soils. This is evidenced by the thick mud layer that overlays the lake’s bottom. Similar observations were made by Bessa et al. (2025) . With the very low dissolved oxygen levels recorded during the study, more ammonium nitrogen could be released from the sediments, leading to an increase in its concentration.

All macronutrients tested were present in Lake Mboandong. Calcium, and Mg, did not show seasonal variations in concentrations but the concentrations of K and Na were higher during the dry season. This could be associated with high rates of evaporation in the dry season. Similarly, orthophosphate concentration was higher during the dry season (3.08 ± 1.87 mg/l) as opposed to 1.53 ± 0.57 mg/l recorded in the rainy season.

Total hardness was higher in the dry season at 32.36 ± 4.84 mg/l than in the rainy season at 14.40 ± 2.49 mg/l.

¹Two samples t-test, ^nsNot significant at the 5% level of significance, *Significant at the 5% level of significance, **Significant at the 1% level of significance, ***Significant at the 0.1% level of significance, Temp = temperature, Cond = conductivity, Turbid = turbidity, NH₄ = ammonium, NO₃ = nitrates, NO₂ = nitrites, Bicarb = bicarbonate, Ca = calcium, Mg = magnesium, K = potassium, Na = sodium.

Principal component analyses with Dim1 × Dim2 axes grouped 16 variables in the dry season ( Figure 3 ). Dim1 (51.4%) was positively correlated to Bicarbonate, potassium, H’ (Shannon diversity), total H, water pH, magnesium, conductivity, ammonium, and sodium. It was negatively correlated to nitrites, nitrates, DO, calcium, temperature and turbidity. Dim2 (31.9%) was positively correlated to nitrates, calcium, temperature, turbidity, bicarbonate, potassium, H’, water pH, and magnesium. It was negatively correlated to DO, nitrites, sodium, total H, ammonium, and conductivity. During dry season, Sites 1 and 2 were similar and constituted a group, and were different from Site 3, likewise Site 4. Site 4, which was closer to the small outflowing stream received more pollutants than other sites.

Principal component analyses with Dim1 × Dim 2 axes grouped 16 variables in the rainy season ( Figure 4 ). Dim1 (50.7% of inertia) was positively correlated to H, nitrites, water pH, nitrates, Total H, temperature, and sodium. It was negatively correlated to ammonium, bicarbonate, potassium, conductivity, magnesium, DO, orthophosphates, turbidity and calcium. Dim2 (35.2% of inertia) was positively correlated to magnesium, conductivity, potassium, bicarbonate, ammonium, H, nitrites and water pH. It was negatively correlated to total H, nitrates, temperature, sodium, calcium, orthophosphates, turbidity, and DO. During the rainy season, Sites 1 and 3 were more similar and different to Site 2, and also different to Site 4. Site 4 received more pollutants than other sites ( Figure 4 ).

Physicochemical parameters were found higher during the dry season as opposed to the rainy season. NO₃, NO₂, DO, NH₄, Turb, Bicarb and Mg were higher during the rainy season as opposed to the others high during the dry season. Species related to these parameters during rainy season included Achnanthidium minutissimum,

Cymbella sp., Pinnularia sp., Pinnularia viridis, Nitzschia palea, Urosolenia eriensis, Asterionella formosa, Tabellaria flocculosa, Closterium sp.2, Cosmarium minitum, Micrasterias sp., Spirogyra sp., Mougoetia sp., Chlorella sp., Oedegonium sp.2, Tetraedron caudatum, Pediastrum duplex, Ankistrodesmus sp., Ankyra sp., Chlamydomonas sp., Haematococcus sp., Cladophora fracta, Dinobryon divergens, Synura sp., Rhodomonas sp., Anabaena circinalis, Gloeotrichia sp., Microcystis flos-aquae, Arthrospria sp., Merismopedia sp., Aphanocapsa sp., Ceratium hirundinella, Peridinium umbonatum, Euglena viridis, Trachelomonas abrupta, Trachelomonas sp., and Phacus sp.

In most phytoplankton studies, nutrients have been implicated as one of the key variables controlling their community structure and biomass. In this study, there were some species highly related to some physicochemical parameters during the dry season. These included Cocconeis pediculus, Aulacoseira granulata, Cymbella cistula, Navicula amphora, Navicula sp., Gyrosigma acuminatum, Urosolenia gracilis, Melosira varians, Meridion circulare, Synedra ulna, Closterium sp.1, Cosmarium contractum, Cateria sp., Oedogonium sp., Scenedesmus quadricauda, Scenedesmus obliquus, Lagerhemia sp., Tetraedron caudatum, Tetraedron sp., Selenastrum sp., Chlorogonium sp., Haematococcus pluvialis, Pandorina sp., Dictyosphaerium sp., Chorella vulgaris, Oocystis lacustris, Dinobryon cylindricum, Mallomonas sp., Aphanizomenon flos-aquae, Anabaena sp., Microcystis aeruginosa, Oscillatoria rubescens, Synechococcus sp., Ceratium sp., Gymnodinium rotundatum, Gymnodinium sp., Peridinium sp., Euglena gracilis, Trachelomonas africana, and Trachelomonas volvocina ( Figure 5 ). Nitrogen and phosphorus are two nutrients having dramatic influences on algal composition as demonstrated in bioassay experiments ( Mebane et al., 2021 ). The appearance of more species in the dry season could be explained by the nutrient limitation theory. Most of these species have been identified in mild to heavily polluted waters in different ecosystems ( Fai et al., 2023 ; Tabot et al., 2016 ). Typical Cyanophyte genera such as Microcystis, Oscillatoria, and Synechococcus were recorded in adjacent Lake Barombi Kotto ( Awo et al., 2020 ). Higher abundances of these species during the dry season could be attributed to higher temperatures and nutrients associated with this season that favoured the growth and proliferation of these species. These findings aligned with the outcomes of several studies of phytoplankton community structure, underscoring a high level of correlation between key water quality variables. Qu and Zhou (2024) recorded similar observations in Xuanwu Lake, China. The variation of phytoplankton composition between the rainy and dry seasons could be significant for their use as indicators of water quality. The concentration of phosphates in the rainy season is attributed to the fact that the lake is surrounded by agricultural farms which make use of higher doses of agrochemicals during this season to prevent the blackening and rot of cocoa pods. These chemicals are easily leached during the high rains into the lake. Other distinctive properties of the lake, different from surrounding land uses could explain variability noticed in terms of phytoplankton species occurrence in different sites. The lake’s pH falls within the 6.5 - 8.5 range, suitable for aquatic biodiversity ( WHO, 2017 ). This study does not address its suitability for drinking since its murky appearance and high organic (muddy benthic layer) deter people from using it for this purpose.

(Achnanthidium minutissimum = Achmi, Cocconeis pediculus = Cocpe, Aulacoseira granulata = Aulgr, Cymbella cistula = Cymci, Cymbella sp. = Cymsp, Navicula amphora = Navam, Navicula sp. = Navsp, Pinnularia sp. = Pinsp, Pinnularia viridis = Pinvi, Gyrosigma acuminatum = Gyrac, Nitzschia palea = Nitpa, Urosolenia gracilis = Urogr, Urosolenia eriensis = Uroer, Melosira varians = Melva, Meridion circulare = Merci, Synedra ulna = Synul, Asterionella Formosa = Astfo, Tabellaria flocculosa = Tabfl, Closterium sp.1 = Closp1, Closterium sp.2 = Closp2, Cosmarium contractum = Cosco, Cosmarium minitum = Cosmi, Micrasterias sp. = Micsp, Spirogyra sp. = Spisp, Mougoetia sp. = Mousp, Cateria sp. = Catsp, Chlorella sp. = Chlsp, Oedogonium sp. = Oedsp1, Oedegonium sp. = Oedsp2, Scenedesmus quadricauda = Scequ, Scenedesmus obliquus = Sceob, Tetraedron caudatum = Tetca, Pediastrum duplex = Peddu, Lagerhemia sp = Lagsp, Tetraedron caudatum = Tetca, Tetraedron sp. = Tetsp, Ankistrodesmus sp. = Anksp, Selenastrum sp. = Selsp, Ankyra sp. = Anksp, Chlamydomonas sp. = Chlsp, Chlorogonium sp. = Chlosp, Haematococcus sp. = Haesp, Haematococcus pluvialis = Haepl, Pandorina sp. = Pansp, Dictyosphaerium sp. = Dicsp, Cladophora fracta = Clafr, Chorella vulgaris = Chovu, Oocystis lacustris = Oocla, Dinobryon cylindricum = Dincy, Dinobryon divergens = Dindi, Mallomonas sp. = Malsp, Synura sp. = Synsp, Rhodomonas sp. = Rhosp, Aphanizomenon flos-aquae = Aphfl, Anabaena circinalis = Anaci, Anabaena sp. = Anasp, Gloeotrichia sp. = Glosp, Microcystis aeruginosa = Micae, Microcystis flos-aquae = Micfl, Oscillatoria rubescens = Oscru, Arthrospria sp. = Artsp, Synechococcus sp. = Synsp, Merismopedia sp. = Mersp, Aphanocapsa sp. = Aphsp, Ceratium hirundinella = Cerhi, Ceratium sp. = Cersp, Gymnodinium rotundatum = Gymro, Gymnodinium sp. = Gymsp, Peridinium sp. = Persp, Peridinium umbonatum = Perum, Euglena gracilis = Euggr, Euglena viridis = Eugvi, Trachelomonas abrupta = Traab, Trachelomonas africana = Traaf, Trachelomonas volvocina = Travo, Trachelomonas sp. = Trasp, Phacus sp. = Phasp, Temp = temperature, pH = hydrogen potential, Cond. = conductivity, Turbid = turbidity, NH₄ = ammonium, NO₃ = nitrates, NO₂ = nitrites, Bicarb = bicarbonate, Ca = calcium, Mg = magnesium, K = potassium, Na = sodium, OP = orthophosphates, DO = dissolved oxygen, Total_H = total hardness).

Based on findings from this study and previous studies on this lake and adjacent Lake Barombi Kotto, agriculture has profound effects on the lake ecosystem through runoff, sedimentation and chemical pollution. The following recommendations could mitigate the effects of agriculture on Lake Mboandong: The village Council should establish vegetated buffer strips along the Lake’s shoreline to filter runoff and reduce the influx of sediments and nutrients from the surrounding agricultural farmlands. This riparian vegetation should comprise native trees already existing in the area, which can effectively absorb nutrients without altering the habitat characteristics. Regional councils should offer training programs to communities on tree selection/planting, silvicultural best practices and riparian buffer zone establishment and maintenance. Climate-Smart agricultural practices that maintain soil cover could be implemented by the community to reduce soil disturbance and erosion, thus preventing siltation in and around this lake.