Climate change on water resources is not a unique concept that can be measured using any single software, rather you may use climate change models for water resources to study environmental consequences. For instance, UK PRECIS may be best suit for your study.

I agree with the previous answer. The effects of climate change on water resources are manifold with a great variety of dimensions: physical, biological, economic, social.

But there are a number of databases that have data you might be able to work into programs you have.  Take a look at Yoffe, Wolf, Giordano, Conflict and Cooperation Over International Freshwater Resources: Indicators of Basins at Risk, J of the American Water Resources Association, Oct. 2003, 1109-1126.

Once you have the necessary climate inputs (see Md Rahman's answer), you need to look at how these will impact on the water resources. This could be done using existing studies (Mary's answer), but this requires your catchment to have similar behaviour. You could use a model (I think this is what you are asking), in which case there are many - the problem is determining how the parameter values will be affected by the change in climate. Examples of the factors to consider here include the vegetation response, the change in the catchment moisture status, as well as the impact of a change in the rainfall intensity. You should also consider the impact of demographic changes, as well as the possibility of economic and social drivers (Gidean's answer). There is a symposium at the next IUGG meeting focusing on this, so you could check the red book that will come out towards the middle of the year for a sample of the recent work.

In terms of software, you could go a simpler, more parsimonious model - has the advantage of generally needing less data and being easier to calibrate (due to fewer parameters), but the disadvantage is that the parameter values will tend to represent more than 1 process. This can make it more difficult to determine how the parameter values might be affected by the change in climate. An example is the hydromad package (http://hydromad.catchment.org) which runs in R.

You could try using a more "physically-based" model - these tend to have a significantly larger number of parameters, trading off predictive uncertainty for accuracy in representing the processes operating in the catchment. These often need more data (e.g. soil, veg cover, ...), and can be more difficult to calibrate. While the parameters have a stronger physical basis, the information in the streamflow data used to calibrate the model is diluted across more parameters, and there is also greater risk of correlation between parameters. This leads to increased uncertainty, as well as potentially more difficult in determining the impact of future change on the parameter values.

Constructively answered by Dr. Croke. Nice.

I would like to replace 'Catchment' with 'Climate Envelope' for a climate change model/ software study.

Dr. Ahemd need to consider uncertainties in prediction because it may change expected outputs in a climate model/ software.

Sid