I really don't think much can be done about trying to predict the future, uncertainty is sure and has to be accounted for in the model outcome. That's why most models are based on the probability of a hydrological event occurrence rather than an absolute value. So it may be worth communicating the uncertainty in your results and presenting the outcome as a probability.

Hi Asong. explore this web link too..Best.

https://www.data.gov/climate/water/

Hydrologic models can be calibrated using historical time series to capture the internal dynamics. The assumption is that these dynamics will continue into the future and future climate prediction impacts these dynamics. Predicting exactly the future hydrologic process is impossible to predict. One way we approach is to use scenarios to capture the uncertainty or parameterize for potential changes in parameters. This the process using historical data to calibrate hydrologic simulations and using GCM/ESM predictions for climate series is standard procedure with an explicit discussion of assumptions made. 

GCMs provide scenarios for the possible future development of climate, but unfortunately their spatial resolution is coarse and therefore can not be used directly for the assessment of regional consequences of climate change. You can use regional climate models (RCMs) data which are in finer resolution or for the best result you can implement downscaling approaches for modeling the climate change in future to reduce the possible inherited bias from GCMs.

Thanks a lot for your various insightful answers. It is absolutely true that we cannot predict with certainty the future evolution of the global climate system due to many obvious reasons which the climate modelling community continue to grapple with. It is also true that we tend to communicate our modelling results in probabilistic terms and then communicate the associated uncertainty and outstanding assumptions made during the modelling process. As you might be aware, communicating such uncertainties to the end-user is itself a daunting task. I came in touch with many climate skeptics at the Fourth Santa Fe climate conference on regional/global climate modelling. They constitute a very multi-disciplinary end-user group although punctuated with a few hydrologic modellers of renowned potentials. The basic question was: why do we calibrate hydrologic models using historically measured inputs, then, for climate change impact analyses, we force these models with ESM/AOGCM/RCM/LAM outputs for the future? You will bear with me that every watershed modeller using this approach to account for future climate change on water resources understands this level of uncertainty which is even more difficult to communicate to any end-user. As pointed out by @Timothy, "the assumption is that these dynamics will continue into the future and future climate prediction impacts these dynamics".  My take is that we are kind of glued in the conventional approach of using hydrologic models for climate change studies. I am in search of alternative approaches to handle this issue and will welcome all kinds of brainstorming on this issue from both the climate,  hydrologic modelling, and end-user community.