Small earth dams are most valuable in arid and semi-arid areas where they are used for both domestic and agricultural purposes. These dams however, continue to fail. The causes of such failures are interconnected in the sense that one can trigger the other. Most research into earth dams nevertheless, uses reductionist approaches. Such approaches do not consider the complex interactions between these modes and/or causes. This research used fuzzy cognitive mapping to identify the prominent modes and causes of small earth dam failure in Swaziland and to capture their interactions. A sample of seven earth dam construction experts was purposively selected from five institutions for individual interviews. An individual map was developed from each interview. An aggregated map was thereafter developed by combining seven individual maps. The results indicated that overtopping, piping and sliding were the common modes of earth dam failure. Overtopping was mainly due to siltation whilst animal barrows and tree roots were largely responsible for piping. Sliding was mostly associated construction defects and sudden drawdown. It was concluded that most of the failures were linked to poor management of catchments and that of the dams. It is recommended that future designs and management should increase the level of community participation in order to limit some of the causes associated with land use practices.

Chemical industries in Onitsha urban area of southeastern Nigeria have been discharging large quantities of effluents into surface streams. These streams are the primary sources of water used by poor households for domestic purposes. This study examines the effects of effluents on the physicochemical and microbiological characteristics of the recipient streams. This objective was achieved by collecting eight effluents and twenty-two water samples from control points, discharge lo-cations and exit chutes of the effluents for analysis. The results of the study characterised the effluents and their effects on the recipient streams. The effluents cause gross pollution of the streams as most of the parameters including pH, total dis-solved solids (TDS), turbidity, biological oxygen demand (BOD), chemical oxygen demand (COD), Mg, NO3, Fe, Cu, Pb, Cr, total heterotrophic count (THC) and total coliform group (TCG) returned high values that exceeded the World Health Organisation’s (WHO) benchmark from 2011 for drinking water quality. Only dissolved oxygen (DO), Na, Zn, Ca, and Na returned values lower than the WHO guideline. E. coli was found in all the samples; TCG was also high. This paper, there-fore, recommends that the effluent generating industries should treat their effluents before disposal.