Coert Johannes Geldenhuys

This study sought to develop an allometric model for estimating the stem volume and aboveground biomass of the karee (Searsia lancea) tree, which is characterised by its multi-stemmed structure. St...

Persistent forest fire refugia are areas within fire-prone landscapes that remain fire-free over long periods of time and are crucial for ecosystem resilience. Modelling to develop maps of these refugia is key to informing fire and land use management. We predict persistent forest fire refugia using variables linked to the fire triangle (aspect, slope, elevation, topographic wetness, convergence and roughness, solar irradiation, temperature, surface wind direction, and speed) in machine learning algorithms (Random Forest, XGBoost; two ensemble models) and K-Nearest Neighbour. All models were run with and without ADASYN over-sampling and grid search hyperparameterisation. Six iterations were run per algorithm to assess the impact of omitting variables. Aspect is twice as influential as any other variable across all models. Solar radiation and surface wind direction are also highlighted, although the order of importance differs between algorithms. The predominant importance of aspect relates to solar radiation received by sun-facing slopes and resultant heat and moisture balances and, in this study area, the predominant fire wind direction. Ensemble models consistently produced the most accurate results. The findings highlight the importance of topographic and microclimatic variables in persistent forest fire refugia prediction, with ensemble machine learning providing reliable forecasting frameworks.

Aims: This study aims to: i) differentiate the plant associations in the Biosphere Reserve of Pendjari (BRP), ii) determine the ecological characteristics of their habitats and iii) present distribution maps on different soil types. Study area: The BRP, located in the Sudanian Zone of Benin. Methods: 202 phytosociological relevés were sampled according to the Braun-Blanquet method within the BRP. Ordination was performed using Detrended Correspondence Analysis to evaluate vegetation patterns. Soil parameters were used to characterize the vegetation types. Results: The numerical analysis of 202 plots and 249 plant species showed two major floristic groups that correlated with a moisture gradient: drylands versus wetlands. The dryland group was a mixture of woodland and shrub savanna, the dominant ecosystems of the study area. The wetland group encompassed species primarily from riparian forest, tree savanna and grass savanna on floodplains. Syntaxonomical analysis of the dryland group showed rocky and gravelly soil associations (Burkeo africanae-Detarietum microcarpi) and soils associated with or without fine gravels (Andropogono gayani-Terminalietum avicennioidis, Andropogono gayani-Senegalietum dudgeonii and Terminalietum leiocarpae). Syntaxonomical analysis of the wetland group showed riparian forest associations on sandy-clay soil (Coletum laurifoliae, Borassetum aethiopi and Hyparrhenio glabriusculae-Mitragynetum inermis) and floodplain associations on silt-clay soil (Terminalio macropterae-Mitragynetum inermis, Brachiario jubatae-Terminalietum macropterae, Sorghastro bipennati-Vachellietum hockii). Conclusions: Eleven new associations were identified in this study. If the distribution of plant associations was determined by different soil properties, the soil humidity would be one of the main ecological factors determining the establishment of plant species and thus plant association development. Taxonomic reference: Akoègninou et al. (2006), Angiosperm Phylogeny Group classification for the orders and families of flowering plants (APG IV, 2016). Abbreviations: BRP = Biosphere Reserve of Pendjari; CBD = Convention on Biological Diversity; CCA = Constrained Correspondence Analysis; DCA = Detrended Correspondence Analysis; GPS UTM = Global Positioning System Universal Transverse Mercator.

The general perception is that forest resources exploitation can cause the degradation and loss of resources and biodiversity, promoted by landscape fragmentation and pressure over habitats. But such assumption might not be adjusted to situations where long term exploitation by traditional communities is based on management strategies adapted to recovery, adjusting pressure to kept resources available. In this work, we assess the recovery of the Mopane woodlands considering different land use practices and stand development stages on areas explored by local communities. In detail, we want to explore: i) changes on stem size across plant comunities dominated by Mopane (Kirk ex Benth ex J. Leonard); ii) relationships between the variation in stem size across communities and the associated land use practices and vegetation stand development stage. To answer those questions, an inventory was carried out to collect data from 50 temporary circular plots of different sizes spatially nested, covering four stand development stages. Each nested circular plot was composed of a large plot (0.02 ha), an intermediate plot (0.04 ha), and an inner plot (0.01 ha) for counting stems of different DBH. Data was analysed using Excel, Two-Way Indicator Species Analysis (TWINSPAN) and CANOCO, and several parametric and non-parametric tests. TWINSPAN analysis, using stem counts, grouped the 1746 stems from 29 species (with Mopane as one species) into 5 communities and 10 sub-communities, based on indicator species and eigenvalues (level of stability). This study was carried out in the Limpopo National Park (LNP), in Mozambique, a conservation area established in 2001 and part of the Great Limpopo Transfrontier Park (GLTP) since 2002. Acording to the results, population structure (size class distribution of trees) of Mopane communities showed variation related to stand development stage, as expected. The initial stages show good regeneration, indicating that recovery is not hampered by the exploitation regime in use, a critical aspect once ecosystem dynamics of Mopane woodland has a deep influence on the way that local communities manage harvesting of Mopane products for different uses, and traditional uses do also have influence on recovery dynamics. So, management of such resource is critical to ensure sustainable resource use and guarantee provision for future generations. For that, a zonation of Mopane woodlands within the Park, considering different land uses, might enhance a quick system recovery on specific areas and contribute to woodland productivity, good quality products and reduce used areas, promoting habitat conservation.

The persistence niche (studied through different modes of regeneration) of three key species (Baikiaea plurijuga (Harms), Guibourtia coleosperma (Benth.) J. Leonard, and Pterocarpus angolensis) (DC) under five different disturbance regimes (eight years after cessation of disturbance) was studied in the Baikiaea–Guibourtia–Pterocarpus woodlands of north-western Zimbabwe. The disturbance regimes were: protected area (no disturbance); timber concession area (timber harvesting); pole and firewood harvesting sites; abandoned crop fields (after clearing for agriculture); and burnt sites (fire disturbance). Influence of stump diameter and height on regeneration factors was investigated. For each disturbance regime, six main plots (100 m × 100 m) each were systematically laid out (20 m apart) to assess regeneration. In each main plot, 10 m wide transects (5 m apart) were surveyed for seedlings and saplings of the target species (at least 100 plants of each species), resulting in 7 transects per main plot, 42 transects in each study site and 210 transects in total. Number of seedlings with or without a rootstock (i.e., just grown from seed) was recorded. Diameter of stumps was measured at the cut surface. Shoot height was measured from their origin. Each target species was associated with mode of regeneration (χ² = 27 642, p < 0.001). Most plants regenerated from root suckers. Regeneration mode is dependent on disturbance regime (B. plurijuga χ² = 225.66, p < 0.001; G. coleosperma χ² = 158.62, p < 0.001; and P. angolensis χ² = 144.01, p < 0.001). Stump diameter negatively influenced number of sprouts for B. plurijuga and P. angolensis, but positively influenced sprout height for all species. Stump height positively influenced sprout height for G. coleosperma and P. angolensis. The study therefore concludes that disturbances are necessary in facilitating sprouting and growth of suppressed shoots. Stump height and diameter influence the ability of stumps to regrow vegetatively through sprouting. Weak relationships between stump diameter/height and coppice density and shoot height suggest that there are other factors that affect vegetative regrowth through coppicing.

Worldwide scientific publications on the remarkably diverse Orchidaceae family are comprehensive, but Africa's craze for the taxon was only felt a few decades ago. This largest and threatened family of flowering plants is widely distributed in tropical areas and recognized by several authors for its environmental, ornamental, and medicinal importance. This review aims to identify research patterns on African orchids using co-citation analysis and bibliometric information visualization. 386 publications concerning African orchids listed in the Scopus database from 1968 to May 2020 were recorded and reviewed. The description of new orchid taxa, the adaptation of orchids to various pollinators and phylogeny emerged as the most discussed aspects. In total, 889 authors have invested in research on orchids in Africa. Johnson Steven D. was the most impactful author with h-index 37. Interestingly, several institutions were involved in these studies with a large dominance of the University of KwaZulu-Natal, a South African University. Future research efforts should target the identification of new orchid species and emphasize on orchid distribution, ecology, and species’ threats. They should also focus on developing efficient management and domestication strategies for orchid species to reduce threats to this family. The restoration and conservation planning of degraded orchid habitats are additional avenues to be explored by future research.