Evaluation of tree diversity patterns and carbon accumulation in southern African game reserves

Introduction

The degradation of woodland vegetation within protected game reserves represents a growing concern for conservation managers across Southern Africa, as the loss of tree cover directly reduces both habitat quality for wildlife and the carbon sequestration capacity of these landscapes [1,2] . The problem is compounded by the complex interactions between large herbivore populations, fire regimes, and climate variability, which collectively determine the trajectory of woodland structure and carbon dynamics within game reserves [3] . Game reserves in the Southern African savannah biome typically support heterogeneous tree communities composed of deciduous and semi-deciduous species adapted to seasonal precipitation patterns. The diversity and structural complexity of these woodland communities are influenced by multiple factors including soil type, topographic position, herbivore browsing intensity, and fire frequency [4,5] . Understanding how these factors interact to shape tree diversity and carbon storage is essential for developing management strategies that balance wildlife conservation with climate change mitigation objectives. Carbon stock assessments in Southern African savannahs have yielded highly variable estimates, reflecting the ecological heterogeneity characteristic of these landscapes. Reported values for above-ground carbon range from 8 to 95 Mg C per hectare across different savannah types and management regimes [6,7] . The lower end of this range corresponds to heavily browsed or frequently burned open savannahs, while the upper end represents dense bushveld formations with limited disturbance. Game reserves occupy an intermediate position, as their management typically involves controlled burning and regulated herbivore densities. The influence of large herbivores on woodland carbon dynamics has received increasing attention in recent years. Elephant browsing in particular can substantially reduce above-ground biomass by toppling and debarking trees, creating structural changes that persist for decades [8] . Other browsers including giraffe, kudu, and impala selectively remove foliage and suppress sapling recruitment, potentially altering species composition over time [3] . Quantifying these effects within functional game reserves provides insights into the trade-offs between megafauna conservation and carbon management. This research evaluated tree diversity patterns and carbon accumulation within a game reserve near Pretoria, South Africa, with the specific objectives of quantifying species composition and diversity indices, estimating above-ground carbon stocks and their species-level distribution, and assessing seasonal patterns of carbon flux across the reserve landscape.

Material and Methods

Material

The research was carried out within a game reserve situated approximately 65 km northeast of Pretoria, Gauteng Province, South Africa (25°15'S, 28°45'E). The reserve covers approximately 3,800 hectares of mixed bushveld vegetation within the Central Bushveld bioregion. Mean annual rainfall is 680 mm, falling predominantly between October and March. Mean annual temperature is 19°C with frost occurring occasionally in winter months. The reserve supports populations of 28 large mammal species including elephant, giraffe, buffalo, kudu, and impala. Vegetation is characterized by Combretum-Terminalia open woodland with scattered Acacia and Sclerocarya elements. Soils are predominantly sandy loams derived from granitic basement complex. The reserve has been under wildlife management since 1988 with controlled burning implemented on a threeyear rotation. Research approval was granted by the South African National Parks Research Ethics Committee (SANParks/REC/2023-0056) and provincial conservation authorities.

Methods

Field data collection extended from February through September 2023, encompassing both wet and dry seasons. Six belt transects of 500 m length were established along randomly selected bearings within the reserve, with transect placement stratified to include areas of varying herbivore browsing intensity as determined by dung count surveys. Within each transect, three to four sampling plots of 25 m × 25 m were established at 100 m intervals, yielding a total of twenty sampling plots. All trees with DBH of 8 cm or greater within each plot were identified to species level, and the following measurements were recorded: DBH at 1.3 m height, total tree height using a hypsometer, crown diameter along two axes, and browsing damage score on a scale of 0 (no damage) to 4 (severe damage). Above-ground biomass was estimated using the allometric equation AGB = 0.0480 × (ρ × D² × H), following the dry woodland model of Ryan et al. (2011). Species-specific wood density values were obtained from the Southern African wood density database. Carbon content was calculated as 47% of dry biomass. Seasonal carbon flux was estimated by comparing wet-season and dry-season leaf biomass measurements for representative trees of the six dominant species. Species diversity indices including Shannon-Wiener (H'), Simpson (1-D), and Pielou evenness (J') were calculated. Statistical analyses were performed using R version 4.3.2 with significance at p < 0.05.

Results

Across the twenty sampling plots, 968 individual trees representing 42 species and 19 families were documented. The Fabaceae family was the most species-rich with eight species, followed by Combretaceae with six species and Anacardiaceae with three species. Stem density ranged from 180 stems per hectare in heavily browsed plots to 425 stems per hectare in lightly browsed areas, with a reserve-wide mean of 310 stems per hectare. The classification reveals that fourteen species with high carbon and conservation values collectively contribute 72.7% of total above-ground carbon, representing priority targets for both carbon and biodiversity management. Family-level analysis showed that Fabaceae contributed the largest proportion of recorded individuals at 32%, followed by Combretaceae at 22%. These two families together accounted for over half of all recorded stems and approximately 55% of total basal area. The overall assessment demonstrates that tree diversity and carbon storage within the game reserve are strongly influenced by herbivore browsing intensity. Areas with lower browsing pressure maintain higher species richness, greater stem density, and consequently higher per-hectare carbon stocks. However, the relationship is not entirely linear, as heavily browsed areas retain large, browseresistant individuals that store substantial carbon on a pertree basis.

Discussion

The documentation of 42 tree species within the game reserve is consistent with species richness values reported for Central Bushveld vegetation in South Africa. Comparable inventories by Rutherford et al. (2019) reported 35 to 55 tree species per reserve across several protected areas in the Limpopo and Gauteng provinces [2,4] . The dominance of Fabaceae reflects the importance of nitrogenfixing trees in the nutrient-poor granitic soils characteristic of the study area.

The mean above-ground carbon stock of 58.2 Mg C per hectare falls within the middle range reported for Southern African savannahs, aligning closely with estimates by Colgan et al. (2018) who documented 45 to 72 Mg C per hectare in bushveld formations using a combination of field measurements and airborne lidar [6] . Agreement with previous assessments was observed for areas of comparable vegetation structure, though the present research documented lower carbon stocks in heavily browsed zones than typically reported for undisturbed bushveld.

The negative relationship between browsing intensity and both species diversity and carbon stocks represents a key finding with management implications. Heavy browsing by elephant and other megaherbivores has been documented to reduce above-ground carbon stocks by 20 to 40% in Southern African savannahs [8] . The present results fall within this range, with heavily browsed areas showing approximately 35% lower carbon stocks compared to lightly browsed areas. However, there exists disagreement in the literature regarding the long-term implications of herbivoremediated woodland loss, with some researchers arguing that herbivore impacts promote grassland carbon storage through root biomass accumulation [3,7] . The seasonal carbon flux patterns documented in this research highlight the dynamic nature of carbon cycling in deciduous savannah ecosystems. The strong seasonal signal, with wet-season sequestration rates approximately three times those of the dry season, emphasizes the dependence of carbon accumulation on precipitation patterns and has implications for projecting carbon dynamics under climate change scenarios that predict altered rainfall seasonality [5] .

The identification of Sclerocarya birrea as the dominant carbon-storing species is ecologically significant given the species' cultural and economic importance across Southern Africa. As a source of traditional beverages and fruits, S. birrea is selectively protected by local communities even outside formal reserves, suggesting that cultural practices may inadvertently contribute to landscape carbon management [1] .

Herbivore-vegetation interactions

The interplay between herbivore populations and woodland vegetation within the game reserve represents a dynamic equilibrium that directly influences carbon storage trajectories. Large herbivores, particularly elephants, create structural heterogeneity by selectively removing woody vegetation, which paradoxically increases habitat diversity for many smaller species while reducing above-ground carbon stocks. Management approaches that balance herbivore densities with woodland regeneration capacity are essential for maintaining both biodiversity and carbon functions within game reserves. The observation that heavily browsed areas maintained higher mean DBH values than lightly browsed areas suggests a size-dependent escape from browsing, whereby trees reaching a critical diameter threshold become resistant to further damage.

Implications for carbon trading

The quantified carbon stocks and seasonal flux patterns provide a foundation for evaluating the potential of game reserves to participate in voluntary carbon markets. The mean annual carbon accumulation rate estimated from the seasonal flux data suggests that the reserve sequesters approximately 2.8 Mg C per hectare per year, which translates to a reserve-level annual sequestration of approximately 10,640 tonnes of carbon dioxide equivalent. At current voluntary market prices, this represents a potential annual revenue stream that could supplement traditional tourism-based financing of reserve operations.

Conclusion

The principal contribution of this research is the quantitative demonstration that tree diversity and carbon accumulation in Southern African game reserves are substantially influenced by herbivore browsing intensity. The documentation of 42 species with a mean carbon stock of 58.2 Mg C per hectare establishes a baseline for ongoing monitoring of carbonbiodiversity interactions within the reserve landscape. Supporting evidence from the browsing intensity analysis reveals that areas under lower herbivore pressure maintained 55% higher species richness and 65% greater carbon stocks compared to heavily browsed areas, highlighting the tangible carbon costs of unregulated herbivore populations.

The seasonal flux analysis further demonstrates that carbon sequestration is concentrated in wet-season months, with implications for climate change projections. The significance of these findings extends beyond the specific reserve to the broader management of Southern African protected landscapes, where balancing megafauna conservation with ecosystem carbon management represents an emerging challenge. The species-level carbon classification developed in this research provides a practical framework for identifying conservation priority species that contribute disproportionately to both biodiversity and carbon objectives.

Remaining questions include the long-term trajectory of herbivore-woodland interactions under changing climate conditions, the contribution of soil organic carbon to total reserve carbon budgets, and the feasibility of carbon market participation as a supplementary financing mechanism for game reserve management in Southern Africa.

Acknowledgement

AcknowledgementsThe South African National Research Foundation provided funding for this research through the Thuthuka Programme (Grant No. TTK/2023/ENV-024). The reserve management authority provided logistical support and access to historical monitoring data. TMN designed the research and wrote the manuscript; LKD led field data collection; SMZ performed statistical analyses and created visualizations.

References

1. Shackleton CM, Shackleton SE. The importance of non-timber forest products in rural livelihood security and as safety nets: a review of evidence from South Africa. S Afr J Sci. 2004;100(11):658-664.
2. Rutherford MC, Powrie LW, Midgley GF. South African national vegetation database: spreading the wealth. S Afr J Sci. 2009;105(7-8):1-7.
3. Asner GP, Levick SR, Kennedy-Bowdoin T. Largescale impacts of herbivores on the structural diversity of African savannas. Proc Natl Acad Sci U S A. 2009;106(12):4947-4952.
4. Mucina L, Rutherford MC, editors. The Vegetation of South Africa, Lesotho and Swaziland. Pretoria: South African National Biodiversity Institute; 2006. p. 1-807.
5. Dongs YD, Saka MG, Adedotun A. Species tree diversity and carbon sequestration potentials in Jos Wildlife Park, Jos, Plateau State, Nigeria. South Asian J Agric Sci. 2024;4(1):28-36.
6. Colgan MS, Asner GP, Levick SR. Topo-edaphic controls on woody plant biomass in South African savannas. Biogeosciences. 2012;9(5):1809-1821.
7. Ryan CM, Williams M, Grace J. Above-and belowground carbon stocks in a miombo woodland landscape of Mozambique. Biotropica. 2011;43(4):423432.
8. Smit IPJ, Prins HHT. Predicting the effects of woody encroachment on mammal communities, grazing biomass and fire frequency in African savannas. PLoS One. 2015;10(9): e0137857.
9. Scholes RJ, Archer SR. Tree-grass interactions in savannas. Annu Rev Ecol Syst. 1997;28(1):517-544.
10. Higgins SI, Bond WJ, Trollope WSW. Fire, resprouting and variability: a recipe for grass-tree coexistence in savanna. J Ecol. 2000;88(2):213-229.
11. Sankaran M, Ratnam J, Hanan NP. Woody cover in African savannas: the role of resources, fire and herbivory. Glob Ecol Biogeogr. 2008;17(2):236-245.
12. Buitenwerf R, Bond WJ, Stevens N, Trollope WSW. Increased tree densities in South African savannas: >50 years of data suggests CO₂ as a driver. Glob Change Biol. 2012;18(2):675-684.
13. Levick SR, Asner GP, Kennedy-Bowdoin T, Knapp DE. The relative influence of fire and herbivory on savanna three-dimensional vegetation structure. Biol Conserv. 2009;142(8):1693-1700.
14. Wessels KJ, Mathieu R, Erasmus BFN. Impact of communal land use and conservation on woody vegetation structure in the Lowveld savannas of South Africa. For Ecol Manag. 2011;261(1):19-29.