FEATURE
runoff and trap seed and sediment ( Figure 1b-d ). Together the treatments covered an area of approximately 0.5 ha . On completion of the erosion control structures , the whole area was sown with 4 kg of locally collected seed comprising equal weights of Osteospermum sinuatum , Salsola aphylla , Augea capensis and Stipagrostis ciliata . The vegetation surrounding the eroded area was dominated by Augea capensis , Pteronia pallens , Salsola spp . and Stipagrostis spp .
Site 2 : Skeletal soil on mudstone
Figure 2 . Rehabilitation structures on skeletal soil when installed in 2017 and four years later in 2021 .
In September 2017 , a group of NMMU 25 students under the supervision set out rehabilitation structures on bare ground where Ecca Mudstone bedrock was covered with 10-40 mm of skeletal silty soil . The structures included furrows , brush fencing , mulched hollows , brush-packs of reeds and T . ramossissima and V . karroo branches ( Figure 2 ). Restoration infrastructure and controls covered approximately 0.25 ha . On completion of the rehabilitation works ,
the whole area was sown with 4 kg of locally-collected seed comprising a mixture of Cenchrus ciliaris , Fingerhuthia africana , Osteospermum sinuatum , Stipagrostis ciliata , Stipagrostis obtusa and Augea capensis . The vegetation surrounding the bare area was dominated by Ruschia spinosa , Enneapogon scaber , Kleinia longiflora and Rhigozum obovatum .
The sites were monitored in August 2018 , February 2020 and September 2021 by counting and identifying all live plants in an 80 cm diameter hoop ( 0.5 sq m ) placed alternatively in or adjacent to rehabilitation structures ( hollows , fences , brush-packs ). Average species richness and plant densities were calculated for each site and treatment and compared visually using bar graphs .
Results in 2021
Figure 3 . Plant establishment in various rehabilitation interventions at Sites 1 and 2 .
spent three days installing rehabilitation infrastructure on wind-eroded and water-eroded bare patches on deep silty sand . The rehabilitation structures comprised 50 hand-dug small ( 0.5 m wide and 0.2 m deep ), and 25 large ( 1.0 m wide and 0.3 m deep ) minicatchments . The excavated soil was used to form an elevated rim ( 150 mm high ) on the down-slope side of each catchment . The catchments were partly filled with a mulch of chipped Vachelia karroo . Branches were placed over the mini-catchments to attract perching birds in the hope that they would bring seeds of fleshy-fruited plant species to these microsites ( Figure 1a ). The microcatchments were interspersed with the erosion control fences on the watereroded area and small gullies were brush-packed with V . karroo and Tamarix ramossissima branches to slow down
Despite the extreme drought plants established on both sites . Mulched hollows , with or without brush-packing , supported more plants and greater species richness than did untreated soil surfaces ( controls ), loosened soil surfaces or brush-packed surfaces without hollows on both deep and shallow soils ( Figure 3 ). Annual ( or ephemeral ) plants , shrubs and grasses were all more abundant in hollows than on other treatments . As the drought continued into 2021 , plants on untreated and brushpacked sites tended to die and disappear , whereas many of those in hollows survived .
At Site 1 , 12 % of all plants and 17 % of all species recorded in 2020 were species included in the seed mix . This increased to 24 % of plants and 26 % of
Grassroots Vol 22 No 1 March 2022 12