Fieldwork Report

‘Dung beetles (Scarabaeinae): Diel flight activity, response to bait size, bait type, and habitat structure in central Zambia'

The fieldwork carried out has resulted in approximately 35kg of specimens, which are now stored in the Hope Entomological Collections (OUMNH). The material is currently being cleaned and sorted in readiness for mounting and identification. All required permits were obtained, as well as data for nearly all of the study sites for vegetation type, soil type, local climate (during trapping period), farming regime, and chemical usage. It is therefore hoped that data analysis can be constructed from several different perspectives, formulating reports on diverse, topical issues, and not only those mentioned in the original proposal.

There were a number of factors that contributed to slight changes to the original plan as laid out in the proposal. The rains, which usually begin during September, did not arrive until the end of November, thus delaying the emergence of much insect life and making the sampling periods closer together than planned. A country-wide fuel crisis and subsequent limited transport altered plans slightly, and made it impossible to get data on flight activity.

Below, the separate experiments are outlined, with any modifications to the initial proposal noted.

Habitat Comparison Experiment

Ten sites of variable farming intensity were chosen within a 300km² area in central Zambia (with Lusaka at the centre.) Ten cattle-dung baited pitfall traps were set at each site: 5 in intensive grazing land, and five in the surrounding non-intensively grazed area. All traps were set at distances of 50m apart within the same habitat, and of between 50m to 1km of those in the other habitat. Although it was originally planned for sites to be at least 5km apart, this was not possible due to the fuel crisis. Thus the distance was thus reduced to 2km. The traps were left for 3 days and the bait was changed, and specimens collected in every 24 hours.

The first repeat was carried out at the end of October/beginning of November rather than September, in an attempt to wait for the arrival of the rains. However, trapping had to begin in order to fit in all repeats in the time available. Thus, there was no rain during the first repeat. This may have affected the quantity of specimens collected, but was consistent within the 10 sites, as trapping at all sites was carried out in a 2 week-long period before the arrival of rain. The second repeat was carried out during December.

As an initial observation, this experiment has provided a reasonable sized amount of material. However, the distinction between intensive grazing and non-intensive grazing land was not obvious. Almost all land in Zambia is within a fenced area belonging to a farmer, with animals allowed to roam freely. The only truly ‘unfarmed' area is within National Parks, where there was no available transport, due to lack of fuel. Thus two, rather than three habitat types were sampled: intensive grazing land and non-intensive grazing land. Farmers often moved animals without prior notification, in some instances the ‘non-intensive grazing' area was found to contain higher numbers of cattle within the sampling period. Farming methods within intensive/non-intensive grazing sites was changed between trapping repeats, and in some cases, the area was burnt/deforested. Thus the site had to be moved (less than 1km) between repeats. There were also problems with cattle destroying traps within intensive grazing land. One repeat provided no data, thus was used as a pilot for the actual experiment, and traps were not laid in highly intensive grazing land.

Optimum Bait Size Experiment

Four size categories of cattle dung and chicken carrion bait (and 1 bait-free control) were investigated: 15g, 30g, 60g, and 120g. The experiment was carried out on a single site of non-intensive grazed land. Traps were randomly laid at distances of 50m apart, and emptied once every 24 hours over a 3-day trapping period. The first repeat was carried out in November, and the second in December: thus all fieldwork was done during the rains when insect life is at its most abundant.

There were slight problems with baits disappearing, but data was eventually collected. This experiment has provided a very large data set with observational differences in abundance of scarabs attracted to different bait sizes.

Bait Type Experiment

Bait types were modified from the initial proposal depending on what was available in sufficient quantity. 6 categories (and one bait-free control) of 30g baits were used: cattle dung, chicken carrion, chicken dung, horse dung, fish and fruit. The experiment was carried out at a single site of non-intensive grazed land. Traps were randomly laid at distances of 50m apart, and emptied once every 24 hours over a 3-day trapping period. The first repeat was carried out in November, and the second in December. During the first repeat, the site suffered through an accidental bush fire, and the traps were destroyed! Subsequent trapping was therefore carried out at a different site (within 1km). However, traps were left in place in the burnt area for an additional unexpected experiment looking at the effect of burning.

Effect of Burning Experiment

30g baited chicken carrion, cattle dung, and standard traps were emptied every 24 hours on the burnt area from 26/11/05 until 20/12/05. This, combined with the 2 days of pre-burning data will enable the investigation of recovery rates after burning. As burning is a procedure that is widely used to clear land both in and outside of National Parks, this should provide a fascinating and highly relevant data set.

Extra Trapping

Extra trapping was carried out at Lusaka, Luangwa, Mkushi and Chisamba, using techniques such as flight intercept trapping, light trapping, fruit baited pitfalls and hand collection. This extra trapping has provided a large number of specimens for identification.

Current Situation

I am currently sorting specimens to Order, and then will be working with Darren J. Mann (Hope Entomological Collections, OUMNH) pinning out the Scarabaeinae and sorting them to species. I am working on one experiment at a time, enabling each experiment to be written up as the specimens are identified. This should allow reports to be processed at a quicker pace than if specimens from all experiments were identified before any writing-up was carried out.


Back to Zambia