Testing alternative stable state theory at Afromontane and Milkwood forest-fynbos ecotones

Thesis / Dissertation


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Alternative stable state (ASS) and ‘sharpening switch' theories provide a conceptual framework within which to understand abrupt ecotones between two (or more) contrasting ecosystem configurations. ASS theory is more suited for explaining ecosystems that exist within the same abiotic conditions. In the Greater Cape Region, forest-fynbos ecotones are often juxtaposed in landscapes where fire is the most important disturbance that maintains open canopy fire-prone fynbos and closed- canopy forests in fire refugia. Forest and fynbos ecotones are, however, additionally maintained by biotic feedbacks that result in edaphic differences between the contrasting ecosystems. Anthropogenic modifications to disturbance regimes alter the stability of forests and fynbos and may shift forest and fynbos into a degraded species-poor state. Conservation practitioners are concerned over perceived loss of forest habitat, and tree planting initiatives are common in parts of the Western Cape of South Africa. I hypothesised that diverse (e.g., Afromontane-and Milkwood-forests) forest and fynbos (e.g., different fynbos types) assemblages are ASS on the same nutrient-poor geology, and that biotic feedbacks with the soil result in edaphic discontinuities across the forest-fynbos ecotone. A sub- hypothesis is that anthropogenically altered disturbance regimes (e.g., timber harvesting, herbivory, fire management) have the potential to keep forest and fynbos in a transient state, so that a regime shift may be more likely. Soil samples (n=5 per vegetation type) were taken, and vegetation plot surveys conducted (n=3 per vegetation type) at seven study sites in Grootbos Private Nature Reserve (GPNR) and Platbos Forest Reserve (PFR) in South Africa. These data were collected in areas identified as forests, transitional vegetation, reforested areas (from conservation management), and the surrounding fynbos matrix. Forests were floristically distinct from fynbos at all study sites (sharing max. = 5% spp. at one site). Most forests had similar soil texture to fynbos however, one site associated with an Afromontane Forest had a much higher silt: sand ratio (ca. 1: 0.9) than the fynbos matrix (ca. 1: 1.6), indicating differences in pedogenic processes in the forest compared to the surrounding fynbos. Forest soils had higher soil nutrient content (N, P, K, Mg, Ca, Fe, and loss-on-ignition) than fynbos soils. The edaphic nutrient differences across ecotones were not due to geological variability, indicating that edaphic differences are largely due to differences in nutrient cycling and vegetation-soil feedback mechanisms. The soil nutrient status of transitional vegetation was variable and not always intermediate between forest and fynbos. Historical imagery revealed that a complex history of disturbance has negatively affected all of the Southern Coastal/Milkwood Forests in the past, but that the forests have recovered in the absence of continued disturbances. Tree cover for most Southern Coastal/Milkwood forests has increased since 1938 (ca. 65% tree cover increase at one of the forests). Afromontane Forests have, however, not changed in extent or tree density since 1938. These findings add to the evidence that soil nutrient differences are sufficient to prevent a regime shift between forest and fynbos. The Southern Coastal/Milkwood Forests can be considered a fire-derived ASS, where edaphic differences reinforce the stability of forest and fynbos boundaries. Afromontane Forest- fynbos ecotones are perhaps better explained by a ‘sharpening switch' model than ASS. Most existing transitional vegetation should not be considered a degraded vegetation state since transitional vegetation is a natural spatial and temporal gradation between fynbos and forest and also a natural buffer for forests against fire. Keywords: alternative stable state, regime shift, bistability, multi-stability, hysteresis, tipping point, resilience, vulnerability, edaphic properties