Conservation of flower-arthropod associations in remnant African grassland corridors in an afforested pine mosaic
-
Published source details
Bullock W.L. & Samways M.J. (2005) Conservation of flower-arthropod associations in remnant African grassland corridors in an afforested pine mosaic. Biodiversity and Conservation, 14, 3093-3103.
Published source details Bullock W.L. & Samways M.J. (2005) Conservation of flower-arthropod associations in remnant African grassland corridors in an afforested pine mosaic. Biodiversity and Conservation, 14, 3093-3103.
Summary
Parts of South Africa in recent years have undergone intensive afforestation with exotic tree species such as pines, for timber production. Much of this planting has been undertaken on former grasslands, but with some grassland areas persisting between the afforested patches. These unplanted areas managed in some cases to maintain the grassland, have been shown to have considerable conservation value for certain native invertebrate assemblages. This study focused on indigenous plant and arthropod associations within these grassland corridors.
Study area: The study area was in and around the Mondi Goodhope Estate (1873 ha) in Kwa-Zulu Natal, eastern South Africa. About two thirds of the area had been planted with Pinus patula (c. 30 m tall at the time of study), the remainder was remnant, indigenous grassland. Twelve grassland sites were selected: one was a natural grassland outside the estate (a 'full control'); one was marginal (a second control); 10 sites were grassland corridors that were narrow (20–100 m wide) (4 sites), medium (101–200 m wide) (2 sites) or wide (>200 m wide) (4 sites).
Study plants: The main study plant was Berkheya speciosa a common, indigenous herbaceous plant of (<1 m tall) open grassland chosen as it occurred in all the corridors. Three less common indigenous herbs Watsonia densiflora and Kniphofia linearifolia provided additional data.
Sampling: Sampling was undertaken between 24 February and 30 April 2001. At each site, six samples of five B.speciosa flower heads were collected. In the laboratory arthropods in these heads were assigned to 'morphospecies' (or at least 'morphotypes'), size class and functional taxonomic groups.
At each site 50 seed heads were also collected and leaf samples taken: seeds were counted, weighed, and arthropods on or in the seeds identified to morphospecies; 50 fresh leaves per site were dipped into 70% ethyl alcohol and the arthropods collected; leaves were analysed to determine percentage leaf loss to arthropod herbivory. W.densiflora and K.linearifolia were sampled in two corridors and a control grassland.
A selection of environmental variables thought most likely to affect the flower-arthropod association were recorded: minimum width of corridor, distance to the nearest natural grassland area beyond the plantation, number of sides of the site that was bordered by pine trees, altitude, and level of cattle impact (low, medium or high).
A total of 5,379 individual arthropods of 107 ‘morphospecies’ were collected. There was a positive correlation between corridor width and number of species, but the relationship was not significant (p = 0.056). and there were no clear correlations with any other of the environmental variables measured.
The arthropod-plant (flowers, seed heads and leaves) associations were mostly similar throughout the network of corridors studied; narrow corridors (20–100 m wide) still had high biodiversity value. The most threatening factor was anthropogenic disturbance, particularly cattle grazing.
Conclusions: Corridors of remnant indigenous grassland such as those studied may have a significant role to play in maintaining arthropod/flower interactions, and flora and fauna in general, through a network of interconnected habitat patches. Whilst the corridors mitigate the effects of afforestation to an extent, not all associations, particularly those involving animals with large home ranges, will be conserved.
Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at:
Output references
|