FBIP: The effect of land-use on the butterfly diversity of ecological refugia within an agricultural landscape

Study Extent

Free State (Eastern Free State)

Sampling

At each inselberg, butterflies were sampled at each transect using a modified 'Pollard Walk method' [1]. This entailed walking the transect between 10:00 and 15:00 and recording all the butterflies observed along the transect, as well as estimating the distance of the specimen from the observer. The species identity was recorded for each specimen as was the linear distance from the centre transect line to the nearest 50 cm to quantify species-specific detection curves used to calculate the butterfly abundance [2]. This was be repeated for each survey transect, to cover the north- and south facing slopes as well as matrix surrounding each inselberg. References 1 - Pollard & Yates (1993) Monitoring butterflies for ecology and conservation. Chapman and Hall, London. 2 - Isaac et al. (2011) Distance sampling and the challenge of monitoring butterfly populations. Methods Ecol. Evol., 2, 285-594.

Method steps

1. This study requires five work-packages: 1. A GIS-based desktop study and survey design, 2. A field feasibility and pilot survey, 3. Aseasonal survey of the butterfly assemblage, 4. The management of data and estimation of population abundances and 5. The publication of field data. Work-package 1 is already in progress and will be completed before the FBIP grants are awarded. It is described here to provide context for the rest of the study. Similarly, Work-package 5 will only occur after the end of the FBIP grant-cycle because time-lines are variable during the scientific peer-review process. Nevertheless, the data from this study will be publicly accessible at the end of Work-package 4 by 15 December 2017. Below is an outline for each of the five work-packages. Work-package 1 - GIS desktop study and survey design Timeline: Currently ongoing - 03 October 2016 Purpose: To establish a scientifically-robust and representative survey design for inselbergs in the agricultural landscape of the eastern Free State. Activities: Characterising the inselbergs of the eastern Free State based on key biophysical characteristics (e.g. elevation, area, isolation, slope, habitat type) using a 30 m digital elevation model as well as national and provincial land-cover data. Selecting a representative subset of inselbergs from a 40,000 hectare portion to the south-west of Bethlehem. The aim is to establish 50 independent survey transects (300 - 700 m depending on habitat uniformity). Of these, 20 transects will be from 10 different inselbergs varying in their biophysical characteristics (2 transects per inselberg) and 20 transects will be in a 2km buffer around each inselberg (2 transects per inselberg). Ten random transects will be in the surrounding matrix. Where possible, the selection of transects will be done using randomised sampling tools in a GIS (e.g. Research Tools plugin for QGIS). Deliverables: A map with inselbergs, land-cover and geographic location of sampling transects. A report on the biophysical characteristics of each inselberg and the habitat characteristics of each transect Work-package 2 - Field feasibility and pilot survey Timeline: (3 weeks) 21 November - 09 December 2016 Purpose: Validate the accessibility of survey transects and compile a species reference data set. Activities: A field survey to ground-truth the selected sampling transects to confirm access to properties of local landowners and confirm physical accessibility to each transect. Each transect will be characterised in terms of length, habitat type and average vegetation height. During this visit, a photo database will be set up for as many butterflies species as possible to set up a reference dataset for subsequent surveys refine the sampling protocol. Deliverable: An updated map of survey transects that incorporates land-owner permission and accessibility. A photo database of butterfly species. Work-package 3 - Seasonal survey of the butterfly assemblage Timeline: (cumulative 8 weeks) 06 - 17 February 2017; 02 - 12 May 2017; 07 - 18 August 2017; 06 - 17 November 2017. Purpose: To gather a site-by-species matrix of butterflies observed in each transect. Activities: Sample butterflies at each transect using the standardised 'Pollard Walk method' [1]. This entails walking the transect between 10:00 and 15:00 and recording all the butterflies observed in a three-dimensional forward-facing 5m box. The species identity will be recorded for each specimen as will be the linear distance from the centre transect line to the nearest 10 cm to quantify species-specific detection curves used to calculate the butterfly abundance [2]. This will be repeated for each survey transect during four seasons: February, May, August and November. Accompanying the main observer is a second observer tasked with capturing any observed, but unidentified, specimens with a butterfly net to add to the photo-database. The second observer will not make observations themselves (to avoid observer bias), but will only support the main observer. Deliverables: Four site-by-species matrices that include records for each observed specimen. Work-package 4 - Management of data and estimation of population abundances Timeline: (3 weeks) 20 November - 08 December 2017 Purpose: To determine butterfly abundances and habitat affinities Activities: Calculating the abundances of butterflies in each transect after correcting for species-specific detectability [2]. Once these abundances are calculated, it is possible to determine habitat affinities for each species. Deliverables: A dataset with abundances and habitat affinities for each species. Also, a summary report on how the butterfly assemblage varies with inselberg characteristics and surrounding land-use.

Butterflies

1. Lepidoptera
   common name: Butterfly rank: order

Free State (Eastern Free State)