A quick google search located this extract from FAO’s Water Report on “Climate Change Water and Food Security” quoting CSIRO’s report of the previous year.
There are a number of important aspects to the changes in runoff: where yields
are expected to decline, we can cautiously assume a reduction in groundwater
recharge, but this may not always be the case. An expected increase in the
frequency of larger rainfall events is likely to cause increases in peak runoff rate
and probable maximum flood. This has implications for storage management
in that the proportion of currently available storage will decrease unless peak
flows can be captured and stored. Where runoff declines and the proportion of
large events increases, we can expect lower median annual storage volumes and
supply security. At the same time, spillway sizes will have to be increased to
pass larger probable maximum floods, especially if more dams are designed or
modified to harvest peak flows and carry storage from year to year. Thus the
costs of surface water storage can be expected to increase, especially in terms
of unit costs of median annual volume stored. In Australia, there has been
a revision of estimated Probable Maximum floods (Australian Rainfall and
Runoff, 1999) and a revision of spillway capacity, overseen by the Australian
National Committee on Large Dams (ANCOLD)(CSIRO, 2007). If this logic
is correct, then there will be considerable interest in enhancing groundwater
recharge as an alternative and possibly cheaper means of storage.