Water source for the Matrimandir lake

Will the Matrimandir lake require desalinated water to support its water level against evaporation and seepage losses, or can the lake be maintained with harvested rainwater?

Desalinated water is the purest drinking water one can imagine. It is like distilled water, with no bacteria inside, no minerals, no pollution, just pure water. Rainwater is already quite polluted once it reaches the earth, at least in our Auroville area. Once it then comes in contact with the surface earth, it will absorb in addition a lot of phosphates and other organic and inorganic substances. To convert harvested rainwater into drinking water would be a difficult process, which is costly and energy-consuming.
Therefore it seems to be natural that one should try to save the desalinated water for drinking water use and utilize harvested rainwater to maintain the water level of the future Matrimandir lake.
We could actually collect during the monsoon enough surface water of the Matrimandir oval and the immediate surroundings of the Matrimandir to compensate for the annual water loss of the future lake, no matter how big or small the lake may be. The only problem is, we cannot store it in the Matrimandir lake itself, unless we allow the water level to vary drastically.
The presently available figures are telling us that the difference between annual rainfall and annual evaporation in our area is only 40 cms of water-column. This of course has to be first verified with our test-pond experiment, but if this is true then the “trans-evaporation losses” for a surface of 160,000 square metres would amount to 64,000 cubic metres of water per year. The Matrimandir oval itself will deliver around 40,000 cubic metres of collectable rainwater every year. Along with the harvested rainwater from the eastern and southern shores of the lake, which are considerably higher than the lake itself, we would definitely have enough water to compensate for evaporation from the lake. We have no figures yet for seepage losses. These also will have to be quantified by our test-pond experiment. But we presume that they will be less than the evaporation losses. All-in-all, we are confident that the lake area would be able to provide a self-sustainable water body, if we can store the collected surface water from the monsoon time till summer time.
In our present scheme of linking the Matrimandir lake with a second lake on top of a hill in the Green Belt, we are thinking of providing such a water reservoir for the Matrimandir lake. Behind the proposed hill there is ample land available for sale. If we could manage to purchase some additional land there and could create an open water body there, it could provide storage capacity for the future Matrimandir lake.
A link with pipelines between the lake and the hill will be there in any case, so nothing much extra would have to be built to provide a facility for transporting the water from the MM lake to the storage water-body and back to the MM lake. Since the water-body behind the proposed hill does not have to fulfill tight aesthetic parameters like the Matrimandir lake, its water level could vary without limit.
If we presume that the trans-evaporation losses plus the seepage losses would amount to about 100,000 cubic metres per annum for the Matrimandir lake, then an open water-body of 15,000 square metres surface area and a depth of 10 metres would provide enough storage volume to absorb the collected rainwater of the Matrimandir area, and would thereby allow the Matrimandir lake to function in the long run with harvested rainwater only.
We are at present in the process of looking at the option of purchasing land, not only for the proposed hill but also for such an additional (so-to-say “third”) lake. It would mean that our integrated system would have the Matrimandir lake, an uphill-lake and a small rainwater storage lake. Desalinated water then would be required only in the first period for the initial filling of the Matrimandir lake, not for its ongoing maintenance.