Details

Autor(en) / Beteiligte

• Seymour, Colleen L.
• Joseph, Grant S.
• Makumbe, Milton
• Cumming, Graeme S.
• Mahlangu, Zacheus
• Cumming, David H.M.
• Prinzing, Andreas

Titel

Woody species composition in an African savanna: determined by centuries of termite activity but modulated by 50 years of ungulate herbivory

Ist Teil von

• Journal of vegetation science, 2016-07, Vol.27 (4), p.824-833

Ort / Verlag

Blackwell Publishing Ltd

Erscheinungsjahr

2016

Quelle

Wiley Online Library

Beschreibungen/Notizen

• Questions: Ungulate herbivory and formation of large termite mounds occur over different time scales, but both can affect plant community composition in savannas. Human-managed savanna systems are increasingly dominated by domesticated grazing herbivores. These have replaced a mix of indigenous browsers and grazers, leading to changes in plant communities such as increased bush encroachment. We compared the relative impacts of five decades of different types of ungulate herbivory (cattle grazing, browsing and grazing by wild herbivores, and exclusion of all larger herbivores) on woody plant assemblages in two habitats, namely, large termite mounds, which may be thousands of years old, and the surrounding matrix. Location: Miombo woodland savanna, Zimbabwe. Methods: To determine the influence of termite mounds and herbivory on spatial distribution and composition of woody assemblages, we compared 40 paired mound-matrix plots from the three herbivory treatments using PERMANOVA. We assessed whether mound plots were more similar to each other than matrix plots were to each other, and whether herbivory influenced similarity. We used SIMPER analysis to identify changes in abundance of indicator species for each habitat (i.e. mound or matrix). Results: Species composition differed significantly between mound and matrix and also between herbivory treatments. Woody plant assemblages on mounds were more similar to each other than woody plant assemblages in matrix plots were to each other, regardless of type of herbivory, but herbivory reduced the mound/matrix contrast. Sample location (i.e. on mound or in matrix) explained 23% of the variation in woody species composition, compared to 12% explained by herbivory. Woody plant abundance was lowest where there were both grazers and browsers. Conclusions: The influence of large termite mounds on plant assemblages was roughly double that of herbivory, which has occurred over five decades. Type of herbivory emerged as relatively influential on woody plant species composition, given that the time periods involved were short by comparison to the slow formation of termite mounds. Type of herbivory also influenced species composition and vegetation structure, with a diversity of herbivores necessary to reduce encroachment by woody species.