The occasional flooding of the two rivers and underground flow of water in the upstream river beds are inadequate and too slow to replenish what is being taken out. In other words, demand is outstripping supply. What few people are aware of is that there is a partial collapse of the pore space in sand aquifers when the water table falls. This results in lower porosity, lower permeability and ultimately to a lower yield when the aquifer is filled up again. Examples of this are Mexico City and the town of Lagunillas on the east coast of Lake Maracaibo in Venezuela where both have subsided 6 m due to the respective abstraction of groundwater and oil from sandstone reservoirs deep beneath them. Replenishment of water in river beds only takes place when rivers flood. The aquifers in the coastal deltas depend on rains inland heavy enough for floods to reach the coast. However, floods flow over the top of the aquifer allowing little time for infiltration. Long-term records in the delta areas have shown that most of the flood water stays on or near the surface and evaporates (Kuiseb) or reaches the sea (Omaruru) and there is very little replenishment of the underground water. Because of this, the Omdel Dam was built near the top end of the Omaruru delta to catch the flood waters. After allowing the contained silt to settle, the silt-free water is fed into infiltration ponds in the coarse sand of the river bed below the dam wall. This has been partially successful in replenishing the delta aquifer but would be more efficient using more infiltration ponds and allowing each to drain completely before refilling it. Flooding in the Kuiseb and Swakop Rivers has also been dramatically reduced by farm dams and the Von Bach and Swakoppoort Dams. Thus, even less water reaches the coast. Add the slow flow rate of water in sand and the vagaries of our climate to the above and we are compelled to the conclusion that the problem of providing fresh water is going to get ever bigger, but most particularly at the coast. Groundwater in sandy beds flows downstream at a rate of 1-5 m per year or more depending on coarseness of the river alluvium. At Koichab Pan, which supplies water to Lüderitz, the rate is between 5 and 13 m per year, i.e. it takes between 8 000 and 20 000 years for water from the catchment in the escarpment area between Aus and Helmeringhausen to travel the 100 km underground to Koichab. Above-average rainfall is essential to ensure significant flooding of rivers. During the 20-year period from 1979 to 1999, there was not a single year with above-average rainfall in the Helmeringhausen area and there was no large-scale flooding of the major rivers. Many boreholes dried up and even deep wells dug next to rivers prior and just subsequent to the First World War dried up for the first time. The Windhoek records show the same pattern. It was also in this period that the large Ana trees (Faidherbia albida) in the Gobabeb area were dying as a result of reduced flooding and a falling water table. Riparian vegetation also consumes large volumes of groundwater. In response to the looming coastal water shortage, Areva was the fastest out of the blocks with their desalination plant. Limping a long way behind is Namwater. Yet Namwater, being responsible for the supply of water in Namibia, should have been up front leading. Nevertheless, the reality of the situation is that demand for water from mining alone is increasing with the present expansion at Rössing and Langer Heinrich and will increase dramatically as Etango and Rössing South come on stream. We are already using more water than we have. We cannot endanger our coastal aquifers for short-term benefit. The coastal potable water has to be reserved for our coastal towns. Mining has to go the way of desalination. The Department of Water Affairs should not issue any further licenses for the abstraction of water from our coastal rivers for mining. I therefore find it an affront to Namibia and Namibians for Langer Heinrich to even consider the abstraction of water from the Swakop River bed for its expansion. This also leaves one with the uncomfortable notion that Langer Heinrich has scant regard for the environment in its pursuit of profit. The responsibility for mitigating the environmental impacts and for rehabilitation as mining progresses rests with top management. It has become long overdue for Langer Heinrich to issue an invitation to the public to view its progress with environmental management and rehabilitation. Other mines in Namibia, such as Navachab, have done this openly and have been able to implement positive suggestions made by the public. In addition, Rössing has made huge progress in recycling water. Effective environmental management and progressive rehabilitation are possible with forethought and are being implemented on some of the mines in Namibia. We would like to see it transparently applied on all mines and the mines owe it to the future sustainable development of their host, Namibia. Roy Miller Via e-mail