Search

Despite the importance of South Africa's Afromontane grasslands for ecosystem services (water supply and biodiversity), soil organic carbon (SOC) research remains limited. These grasslands evolved with fire, and fire exclusion leads to native plant afforestation. This study investigated SOC fractions and origin to understand the impact of fire‐exclusion‐driven afforestation and aspect on SOC storage in Afromontane grasslands. This study in Cathedral Peak Research Catchments, initiated in the 1940s, compared an afforested fire‐excluded site (AF) to a periodically burnt (accidental fires, 2–5 years interval) grassland (PB) within the same catchment (Catchment‐IX). Additionally, it compared a south‐facing periodically burnt grassland (Catchment‐IX) to a north‐facing biennially burnt grassland (Catchment‐VI). Soil samples collected at soil‐depth increments (0–5, 5–10, 10–15, 15–20, 20–30, 30–60 and 60–100 cm) revealed that, within Catchment IX, PB had more topsoil SOC stocks and microbial activity than AF but similar active carbon (C) concentrations. As expected, δ13C values revealed that SOC in PB originates from C4 grasses, whilst it mostly originates from C3 plants in AF. The south‐facing slope (Catchment‐IX) had more SOC stocks, microbial activity and active C compared to the north‐facing slope (Catchment‐VI). Fire‐exclusion‐driven afforestation changed SOC input from roots to litter, thus reducing SOC storage. Cooler south‐facing slopes are better C reservoirs. Afromontane grasslands show greater potential for C sequestration than afforested systems. [ABSTRACT FROM AUTHOR]

Climatic and land management factors, such as water availability and grazing intensity, play an important role in seasonal and annual variability of the ecosystem-atmosphere exchange of CO2 in semi-arid ecosystems. However, the semi-arid South African ecosystems have been poorly studied. Four years of measurements (November 2015-October 2019) were collected and analysed from two eddy covariance towers near Middelburg in the Karoo, Eastern Cape, South Africa. We studied the impact of grazing intensity on the CO2 exchange by comparing seasonal and interannual CO2 fluxes for two sites with almost identical climatic conditions but different intensity of current and historical livestock grazing. The first site represents lenient grazing (LG) and the vegetation comprises a diverse balance of dwarf shrubs and grasses, while the second site has been degraded through heavy grazing (HG) in the past but then rested for the past 10 years and mainly consists of unpalatable grasses and ephemeral species. Over the observation period, we found that the LG site was a considerable carbon source (82.11 g C m-2), while the HG site was a slight carbon sink (-36.43 g C m-2 ). The annual carbon budgets ranged from -90 ± 51 g C m -2 yr-1 to 84 ± 43 g C m-2 yr-1 for the LG site and from -92 ± 66 g C m -2 yr-1 to 59 ± 46 g C m -2 yr-1 for the heavily grazed site over the four years of eddy covariance measurements. The significant variation in carbon sequestration rates between the last two years of measurement was explained by water availability (25 % of the precipitation deficit in 2019 compared to the long-term mean precipitation). This indicates that studied ecosystems can quickly switch from a considerable carbon sink to a considerable carbon source ecosystem. Our study shows that the CO2 dynamics in the Karoo are largely driven by water availability and the current and historical effects of livestock grazing intensity on aboveground biomass (AGB). The higher carbon uptake at the HG site indicates that resting period after overgrazing, together with the transition to unpalatable drought-tolerant grass species, creates conditions that are favourable for carbon sequestration in the Karoo ecosystems, but unproductive as Dorper sheep pasture. Furthermore, we observed a slight decrease in carbon uptake peaks at the HG site in response to resuming continuous grazing (July 2017). [ABSTRACT FROM AUTHOR]

Reliable estimates of daily, monthly and seasonal soil moisture are useful in a variety of disciplines. The availability of continuous in situ soil moisture observation records in Southern Africa barely exists. In this regard, process based simulation model outputs turns out to be a valuable source of climate information, which is needed for guiding farming practises and policy interventions at various spatio-temporal scales. Despite their ability to yield historic and future projections of climatic conditions, simulation model outputs often reflect a certain degree of systematic uncertainty hence it is very important to evaluate their representativeness of spatial and temporal patterns against observations. To this effect, this study presents an evaluation of soil moisture outputs from a simulation and satellite data based soil moisture products. The simulation model consists of a global circulation model known as the conformal-cubic atmospheric model (CCAM), coupled to the CSIRO Atmosphere Biosphere Land Exchange model (CABLE). The satellite based soil moisture products include; satellite observations from the European space agency (ESA) and satellite observation based model estimates from the Global Land Evaporation Amsterdam model (GLEAM). The evaluation is done for both the surface (0–10cm) and root zone (10–100cm) using in situ soil moisture measurements collected from two savanna sites, located in the Kruger National Park, South Africa. For the two chosen sites with different soil types and vegetation cover, the evaluation considers soil moisture time series aggregated to a monthly time scale from all the data sources. In order to reflect the inter-comparability of CCAM-CABLE simulation output, and GLEAM model estimates, a qualitative analysis of phase agreement, using wavelet analysis is presented. The onset and offset of the wet period, for the two specific sites, is calculated for each of the models and the soil moisture time series covariance between CCAM-CABLE and the GLEAM is discussed. Our results indicate that both the simulation and satellite observation based model outputs are generally consistent with the in situ soil moisture observations at the two study sites, especially at the surface. CCAM-CABLE and GLEAM inter-comparison also shows that the models are generally in phase, however with a time lag of about 12 and 20 days on average, for the surface and root zone respectively. In general the simulation compare well with the GLEAM model estimates, hence indicating that the key physical processes that drive soil moisture in CCAM-CABLE and GLEAM, at the surface and root zone, lead to an appreciable degree of mutual information. This is reinforced by a predominantly positive measure of covariance between the respective two soil moisture outputs. [ABSTRACT FROM AUTHOR]