Bird surveys at windfarms in the Eastern Cape, South Africa

Bird surveys were conducted at four windfarm sites and four control sites in the Eastern Cape, South Africa. Numbers of the Blue Crane Grus paradisea decreased at three out of four windfarm sites after commencement of windfarm activities and did not decrease at any of the control sites. Numbers for several bird species of the fynbos and grassland tended to increase at windfarm sites after commencement of windfarm activities, and did not increase at the control sites.


Introduction
The introduction of windfarms as a renewable energy source has been accompanied by concerns about the possible negative effects on bird populations in the vicinity of the windfarms. The St Francis Bay bird club undertook a long term monitoring exercise in order to detect and report on changes in the abundance of bird species at and around the windfarms.

Methods
Since 2011, the St Francis Bay Bird Club has carried out regular bird surveys at four windfarm sites (Tsitsikamma, Jeffreys Bay, Gibson Bay and Kouga) as well at four control sites (Banna-ba-Phifu, Oyster Bay, Tsitsikamma West and Ubuntu) (Figure 1 and 2) in the coastal region of the Eastern Cape, South Africa, between Jeffreys Bay in the east and Plettenberg Bay in the west. The control sites are proposed windfarm sites where no windfarm activity occurred during the study period.
Surveys were conducted by groups of two or more observers by vehicle on roads around the perimeter of the sites, with observers dismounting from time to time to scan the surroundings. Except in the case of Jeffreys Bay, observers were not permitted to enter the windfarm sites. At Jeffreys Bay, initial surveys were done around the perimeter, and later surveys were done on roads traversing the site (after permission to enter the site was granted). The surveys were conducted once every six weeks at each of the sites. At least 70 surveys were conducted at each site. Data on bird mortalities due to collision with the wind turbines are kept by the windfarm managements. Those data were not available to this study. The data accumulated during the period 2011 to 2020 were analysed. For windfarm sites, the average numbers of birds counted per visit for each species after commencement of windfarm activities were compared to the average number counted before commencement. For control sites, the average numbers counted for the second half of the study period were compared to the average numbers for the first half.

Results and Conclusions
In general, no differences between the outcomes for windfarm sites compared to control sites that were consistent across all sites emerged. However some trends emerge which if not conclusive, at least indicate a need for further investigation. In Table 1, the outcomes for species of conservation concern (Red Data species listed in Taylor et al. (2015)) are summarised. Table 1 indicates for which sites the average numbers for each species showed an increase or decrease over the study period, or was essentially stable. The numbers of Blue Cranes Grus paradisea decreased at three out of four windfarm sites after commencement of windfarm activities and did not decrease at any of the control sites. For the other species in this group, trends in abundance were not consistently different for windfarm sites compared to control sites. Biodiversity in grassland and fynbos habitats is dependent on a suitable regime of grazing by livestock or game animals and periodic burning. A change in the observed numbers of grassland and fynbos bird species which is not observed at control sites in the same region might be expected to reflect a change in land management, particularly changes in grazing pressure or burning frequency. That does not seem to be the case here.
The owners of the land on which windfarms were developed retained ownership of the land while wind turbines were constructed and came into operation, and continue to use the land surrounding the turbines as before, predominantly for grazing livestock. It is therefore unknown if and why there was any change in land use management across the windfarm properties subsequent to the commencement of windfarm operations, and there is no ready explanation for why the average numbers counted for some fynbos and grassland species increased at the windfarm sites and not at the control sites. For those species that were observed only occasionally at one or more of the sites, the

Banna-ba-Phifu Oyster Bay Tsitsikamma West Ubuntu Tsitsikamma Jeffreys Bay Gibson Bay Kouga
African Marsh Harrier + + --n/a n/a --Black Harrier n/a n/a n/a -n/a + -n/a Black-winged Lapwing + + -n/a + n/a -- White-bellied Bustard + O n/a -n/a n/a n/a +

Key n/a
The species was not regularly reported at the site O There was little change in the average count (no more than 10% difference) + An increase of more than 10% in average count -A decrease of more than 10% in average count Table 1: Changes in the average number counted per visit for species of conservation concern (Red Data species).

Key n/a
The species was not regularly reported at the site O There was little change in the average count (no more than 10% difference) + An increase of more than 10% in average count -A decrease of more than 10% in average count Table 2: Species showing differences in trends of average numbers counted for windfarm sites compared to control sites.

Key n/a
The species was not regularly reported at the site O There was little change in the average count (no more than 10% difference) + An increase of more than 10% in average count -A decrease of more than 10% in average count Table 3: Changes in average numbers counted per visit across windfarm and control sites for selected species.

Banna-ba-Phifu Oyster Bay Tsitsikamma West Ubuntu Tsitsikamma Jeffreys Bay Gibson Bay Kouga
African Hoopoe n/a n/a -n/a n/a n/a n/a O Cape Batis n/a n/a -n/a O n/a -n/a Cape Bulbul -

Key n/a
The species was not regularly reported at the site O There was little change in the average count (no more than 10% difference) + An increase of more than 10% in average count -A decrease of more than 10% in average count Table 3: Changes in average numbers counted per visit across windfarm and control sites for selected species (continued).

Banna-ba-Phifu Oyster Bay Tsitsikamma West Ubuntu Tsitsikamma Jeffries Bay Gibson Bay Kouga
Cape Canary - n/a n/a -n/a n/a n/a n/a + Cape TurtleDove + -- n/a n/a n/a - n/a n/a -n/a n/a n/a -+ Little Swift -n/a n/a n/a n/a n/a n/a - n/a O n/a n/a n/a + n/a Pintailed Whydah ---n/a n/a n/a -+ Speckled Mousebird -n/a --n/a n/a + -Speckled Pigeon + + O --+ -+ Terrestrial Brownbul + n/a + n/a n/a n/a -n/a White stork -+ -+ + + O + Whitebreasted Cormorant + n/a + n/a n/a n/a + n/a White-necked Raven -+ -+ + + + + Yellow Bishop O + --+ + + + data are insufficient for analysis. Table 3 summarises the results for species which were observed regularly at least one of the windfarm and one of the control sites. The most striking result here is that numbers counted of Cape Starling Lamprotornis nitens declined across three out of four control sites as well as the two windfarm sites where they were present. The detailed data summaries for each site can be obtained from the author.