Generally groundwater has been regarded as a better source of water in terms its low biological contamination and so have been favoured for this reason. However, there may be other contaminates that can make it less desirable.
I have just been looking into the issue of arsenic contamination in groundwater in Bangladesh which has led to a return to the use of surface water sources. Other contaminates from groundwater might be iron, fluoride, manganese, salts and nitrates.
Practical Action has a technical brief water-treatment-systems http://practicalaction.org/water-treatment-systems which gives a little bit of information about this.
For a general introduction to these issues the book Environmental Health Engineering in The Tropics: An Introductory Text by Sany Cairncross and Richard Feachem, published by Wiley is a good place to look.
For information on rainwater harvesting you can look at the book Rainwater Catchment Systems for Domestic Supply by John Gould and Erik Nissan Practical Action Publishing http://developmentbookshop.com/rainwater-catchment-systems-for-domestic-supply-pb
Adding to Neil's answer above, rainwater harvesting has advantages if:
- there is a suitable pattern of rainfall for collection
- the system is affordable to build and maintain
- the community want to have a rainwater harvesting system and are willing to maintain it
Typically it is used where other sources of water are insufficiently available (perhaps seasonal) and unreliable, inconveniently located (such as too far away), or polluted (as in the case of arsenic contamination in parts of Asia and SEAsia).
The disadvantages are:
- Installation of a collection and storage system can be surprisingly costly, depending on the location, and is likely to be the highest cost water supply available
- The system does need some maintenance, particularly in terms of cleaning
- Clean, stored water can become polluted by leaf debris, birds and small animals (as in any water storage system). Particular care should be taken to avoid creating breeding sites for mosquitos.
If you do a search you will find a lot of references for information on rainwater harvesting systems. A good source is Chapter 7 in the book by Smet and van Wijk (2002).
Harriette's first point of "a suitable pattern of rainfall for collection" is the most common reason for rainwater harvesting to be deemed economically infeasible in projects I have worked on.
The issue is that you have a base-load of water demand for potable purposes. This is relatively constant, hence you need a solution that can meet this demand all year round. If rainfall is seasonal, then you either need a back-up source when there is no rain, or massive storage to see you through the dry spells. If you are storing, then you also need to keep the water from going stagnant. If you have a back-up source then it is often more economical to use that all year rather than providing the treatment plant to make rainwater potable.
It is easier to use rainfall for irrigation than treating for potable use, but again, in areas of the world with highly seasonal rainfall, you end up with irrigation water at the time when fields are waterlogged from the rains anyway.
Conversely, if you have regular rainfall, interspersed with shorter dry periods, rainwater harvesting for irrigation can be a highly effective solution. If you have extremely regular, year-round rain, then treatment of rainwater for potable use can become economically feasible too.
This thread is public, all members of KnowledgePoint can read this page.