I contend that a science response to these large issues should be to tackle the overall system, not the problems one by one. We well know that tackling one problem on its own might exacerbate another. For example, widespread planting of trees might (if placed in the right areas) help combat salinity, but will also lead to reductions in water supply and hence reduce our opportunity to reintroduce environmental flows. The systems science required will certainly engage in monitoring and modelling, and in future will increasingly couple them (rather than, as at present, data being merely input into or tests of models).