Green infrastructure for source water protection in the form of forest protection and afforestation is gaining interest worldwide. It is considered more sustainable in the long-term than traditional engineering-based approaches. This paper presents a theoretical model to support investment decisions in green and grey infrastructure to deliver safe drinking water. We first develop a static optimal control model accounting for the uncertainties surrounding green infrastructure. This model is then extended to factor in key characteristics surrounding investment decisions aimed at optimizing the stock of green and grey infrastructure. We first include dynamic forest growth, followed by the risk of wildfires and finally the potential offsetting effect of carbon sequestration on long-term climate change and the reduced risk of wildfires. We provide a numerical example to analyze the performance of the different model specifications, interpret their outcomes and draw conclusions to guide future investment decisions in green and grey infrastructure.