Hancox, G.T. 1977 Preliminary geological investigation and reservoir permeability studies for the proposed Vaipu and Afulilo hydro-electric schemes, Western Samoa. Lower Hutt: New Zealand Geological Survey. New Zealand Geological Survey report EG 288 75 p.
Abstract: The proposed Vaipu and Afulilo hydropower schemes at the head of the Salani River in eastern Upolu, incorporate the only significant hydro storage possible in Western Samoa. With a combined possible installed capacity of 31 GWh per annum, they could together carry a major part of the base load during the next 10 to 15 years. Over-year storage is desirable to overcome the dry season fall off in hydro generating capacity. The schemes involve the construction of two dams across the Salani River at the outlets to the Vaipu and Afulilo catchments, two high level swamp basins some 250m and 310m above sea level. One powerhouse would be sited in Fagaloa Bay and the other one would be at the base of the Afulilo Falls. Engineering geological investigations were carried out during 1977 to investigate the geology of the project area and assess the water-tightness of the proposed reservoirs. The investigation included geological mapping, diamond drilling, permeability testing, soils testing and ground water studies. The rocks in the project area are intercalated basaltic lavas and breccias of varying ages. To the north of the basins, Fagaloa basalts form the steep impounding ridges above Fagaloa Bay, while the southern side of the basins are bordered by several volcanic cones of younger Salani and Mulifanua basalt. Superficial deposits of clayey-silt soils and unconsolidated volcanic sediments mantle the area; in places there are deposits up to 14m thick. The basalts are moderately permeable, with K values ranging from 10-3 to 10-5 cm/s. The permeabilities of the soils and superficial deposits are more variable with K values ranging from 10-7 to 10-3 cm/s, the average value being about 10-5 cm/s. The water table in the basins is generally shallow, ranging from 3 to 8m deep in most drill holes, while in the swamps the water table is generally less than 1.5m deep. On the flanks of the proposed Vaipu reservoir, at the south Vaipu Saddle, the water is some 5-6m above the proposed top reservoir level. The hydrological studies between 1971 and 1977 suggest the possibility of high seepage losses from the catchments. At Vaipu, the losses attributed to seepage are within the error limits of the data, especially if interception losses are added to the evapotranspiration losses. The losses from the Afulilo catchment however, are too high to be accounted for by errors alone. Geological differences between the two basins, outside the area of the proposed reservoir, are thought to account for the disparity. It is concluded from the engineering geological investigations that the Vaipu and Afulilo basins are likely to be suitable for over-year water storage. The soil and rock permeabilities are sufficiently low, and under the existing groundwater conditions the leakage paths are thought to be long enough to limit reservoir leakage to acceptable levels. With effective grouting and cut-off measures at the dam sites, and at the north Vaipu Saddle, leakage i n these areas should be minimal. With the provisional maximum storage levels that have been proposed, there is no evidence to show that leakage losses will be greater for Afulilo than they will be for Vaipu. Truly watertight catchments are rare and some leakages are Inevitable. It is doubtful whether the magnitude of the loss can be proven conclusively by standard investigation methods. Some form of trial run-of-river scheme with minor storage would probably provide more effective data for the assessment of the leakage problem if this is desired. (auth)