Wednesday, 31 October 2012

The Impact of FIRE on Fynbos Birds

Most of the blog entries for this year have somehow focused on the fieldwork aspects of the research I have been conducting focused on the impact of climate change on Fynbos endemic birds, starting with the adventure by bicycle across the biome.

At the beginning of October I finished my winter survey across Protea, Erica and Restio dominated habitats of the Fynbos biome. I spent most of October entering data, and starting some preliminary analysis for a talk presented by my supervisor Phoebe Barnard at the Pan-African Ornithological Conference just finished in Arusha, Tanzania. Annelise Vlok then invited me to give a talk at the Gouritz Cluster Biosphere Reserve quarterly meeting, and knowing that she is concerned about the fate of certain Protea species with frequently returning fire intervals in the Boland area, I decided to push on with analysis of the impact of fire on Fynbos birds. After all, climate change means not only an increase in temperature, but also for the winter rainfall areas we have already recorded a decrease in annual rainfall over the last 15 years (although I'd be hard pressed to convince anyone of that this year with record high rainfall! - but such the issue between weather and climate). A drier climate means an environment more prone to fire.

In fact, Tineke Kraaij, who did her PhD at Port Elizabeth has recently published 2 papers on fire and weather on the eastern Fynbos. There are her findings in a nutshell:

1. Fynbos is fire prone and fire adapted, with the frequency, season and intensity of fires being important determinants of vegetation structure and composition
2. She cites a paper that says fire return intervals are from 8 to 40 years
3. In western sections, best fynbos recovery occurs after summer or autumn fire, while in the eastern Fynbos fire patterns are aseasonal. However, existing guidelines for the management of fire in fynbos are largely based on research carried out in the west.
4. Lightning causes 59% of fires, and she used data to calculate trends in a Fire Danger Index (FDI).
5. Mean annual FDI has increased since 1939
6. Large fires have increased in the south - eastern CFK
8. Frequent recurrence of very large fires and the virtual absence of vegetation in older postfire age classes are potential causes for concern in achieving fynbos conservation objectives

So my research focuses on Fynbos Endemic birds. A poster describing these, as well as their status as far as I am concerned, can be downloaded here:

Basically, we have 2 nectarivores: Cape Sugarbird and Orange-breasted Sunbird; 2 seed eaters: Cape Siskin and Protea Seedeater; and 2 insectivores: Victorin's Warbler and Cape Rock-jumper. All of these are listed by the IUCN as being species of Least Concern i.e. none are endangered. However, a review of the South African Bird Atlas Project (SABAP2) data shows that 2 species meet IUCN criteria for vulnerable – Cape Rock-jumper and Protea Seedeater due to a decrease in reported range >30% and this range is close to 20 000 square kilometers, another criteria for vulnerable. The thing is, these are mountain birds, so there is some concern that people who submit lists may not be getting into mountains. However, since I am looking at similar protocols, why were they reported for these sites previously? An article reviewing the data can be read here:

So in order to really understand what was going on, what was needed was a survey through the mountains by one observer to avoid observer bias, with a repeatable protocol. So I jumped on my bike and travelled 2500km to check out the situation. The following is a map of the points I conducted, highlighting the occurrence of Cape Sugarbird along the way. I did over 800 points. The pink area are selected bioregions of the Fynbos that includes mountain Fynbos, and excludes renosterveld habitats. This area is around 56 000 square kilometers.

For the analysis on fire, I use around 500 points that I had data for at the time I started to analyse the impact of fire, so it is not the complete data set. However, it does include points conducted in both winter and summer surveys in some cases. I did this to improve the number of encounters for calculating bird density in the Distance 6.0 program, which needs lots of encounters to figure out how many birds there are.

I should stress these data are preliminary – they have not yet been peer-reviewed and a final analysis will include all the data points, not just the data that was ready at the time the first analysis were needed. However, I feel the trends have been identified.

So the first chart is the distribution of Fynbos age, as determined using growth rings on Protea species. For simplicity, I've grouped the data into 4 age categories, and a mixed category for points where different fire histories were clearly present at a site. This chart shows that there are fewer areas with old Fynbos – so few for 20 years or more I had to lump them into the category of 15 years or more in order to run the density analysis.

This chart shows the average number of species I recorded per point. This was lowest for the youngest age class, around 2, and higher for points were age was greater than 10.

To show that these few species are not dominating the landscape, the following chart shows bird density for all the birds that were recorded during the survey. Again, bird density was lowest for the youngest age class, and greatest for the oldest age class.

Of all the over 3000 bird group encounters analysed, 12 species made up 62 percent of these. So the Fynbos is dominated by a handfull of species. Of these, four were considered Fynbos endemics. The following series of charts shows bird density by species for these four – where there was enough data to analyse. There were not enough group encounters to do this analysis for Cape Rock-jumper and Protea Seedeater. The strongest relationship is shown for Cape Sugarbird – where density is strongly correlated to Fynbos age. Cape Siskins on the other hand don't seem to care. Victorin's Warbler are happy once age since fire is greater than 5 years.

Orange-breasted Sunbird density in relation to fire age categories

Cape Sugarbird density in relation to fire age categories

Cape Siskin
Victorin's Warbler
Lastly, this chart shows relative capture rates of birds in mist nets from Blue Hill Nature Reserve grouped according to dietary guild. All bird groups were captured less frequently, with nectarivores the most impacted.

Relative capture rates of Fynbos bird species in burnt and unburnt Fynbos (standardised by 1000meters of net hours).

1 comment:

  1. "So I jumped on my bike and travelled 2500km to check out the situation." You are a legend!
    Really nice work, and a great website. I look forward to seeing what comes next.


Related Posts Plugin for WordPress, Blogger...