2.0 METHODOLOGY

Malawi is subdivided into northern, central and southern regions. This study was conducted in the southern region. Studies on small water bodies in Malawi commenced in 1985 when an inventory of small water bodies was made and about 800 small water bodies were identified (Brooks 1993). In 1991, a study by EEC/Government of Malawi project looked at species composition, fish production of a sample of 12 small water bodies in the northern region (Brooks 1993).

In Malawi, small water bodies are placed in groups on the basis of ownership into communal, private and public (government owned). To select small water bodies for the study which is reported here a set of criteria was used: the small water body located either in the central or southern regions should be between 5–20 ha in size (to allow proper mixing of the stock for tag-recapture studies), but not privately owned e.g. by an estate (it was felt that the small water body should be utilised by the community so that results could be used to manage small water bodies which would benefit a lot of people); should be stable (i.e. not new and not prone to drying) and have the spillway and dam in good condition; should have species composition and abundance to allow effective tagging and lastly, it should be accessible in terms of transportation.

Initially, test fishing was carried out in various small water bodies to determine the species composition: Mpasankhuli, Kuchimimbe, Misale, Nkhande and Bunda in the central region, Makoka, Mpalanganga, Chipale, Toleza, Chileka, Bvumbwe in the southern region. For most of these small water bodies the species were composed of Oreochromis shiranus and Barbus paludinosus and in some cases, Clarias gariepinus with relative abundance decreasing in that order.

Due to size of most small water bodies, it was not possible to select small water bodies of size between 5–20 ha. At first, Mpasankhuli and Mchoka in the central, Mbvoniha and Bvumbwe in the southern region were selected. Problems of transportation and logistics to effectively supervise the study outweighed the criteria outlined above for selecting the small water bodies. Small water bodies from the southern region were selected. Of the ones that were selected, Bvumbwe had obstacles which prevented random sampling using seine net and was hence dropped out. It was replaced by Mikolongwe.

Mikolongwe is approximately 2.5 ha while Mbvoniha is 3 ha. The former is owned by the public and represented small water bodies which are owned by the government. The latter is in rural area and therefore represented small water bodies available to the rural community.

Since Oreochromis shiranus was found in higher abundance than other species, and that it is the species which was most commonly farmed Malawi, it was chosen for the study. Sampling was carried out using two sampling gears, multimesh nylon monofilament benthic gill nets (supplied by Lundgrens, Fiskeredskapsfabrik, Stockholm) and a 100 metres long, 1 inch (2.54 cm) stretched mesh seine net. The gillnets consisted of 14 randomly placed mesh panels in a geometric series (Table 1). Each panel was 3 m long and 1.5 deep. The hanging ratio (ratio of length of rope to stretched webbing) was 0.50

Each sampling took approximately 3 days. On the first day 8 gillnets were set randomly over night in all areas of the small water body at dusk and were lifted at dawn the following morning. On the second day fish removed from different nets/meshes were counted and measured to the nearest mm (total length, T.L., measured from the anterior-most extremity to the tip of the median rays of the tail in its natural position), and weighed in grams to one decimal.

On the third day the seine net was hauled several times to get enough fish for tagging. Fish above 10 centimetres were weighed, sexed, and measured for total length as stated above and then tagged using floy tags. The number on the tag was recorded against each fish.

From gill nets and seine net catches, recaptures were weighed, measured for total length and recorded accordingly. A watchman was deployed at each small water body to record recaptures from other gears by local fishermen.

Table 1. Characteristics of the multi-mesh nylon monofilament benthic gillnets

Limnological data collected were dissolved oxygen, temperature, conductivity, transparency, total phosphorus, and total alkalinity. Dissolved oxygen (DO) and temperature were measured using YSI model DO/temperature meter. A depth profile of the two parameters was taken at 0.5 metre interval. Conductivity was measured using a conductivity meter. A secchi disk was used to measure transparency. Total phosphorus and alkalinity were measured using the Hach kit.

Sampling and measurement of limnological parameters were made from February 1993 to March 1994 at 1 to 2 months interval. The sampling schedules for the two small water bodies are presented in Table 2.

Table 2. Sampling schedule for Mikolongwe and Mbvoniha