We present a simulation model developed to communicate a potential investment trap associated with using man-made river engineering to protect built infrastructure. A small system dynamics model in STELLA TM was constructed following a collaborative model-building process to increase understanding among stakeholders of the role natural capital plays in wealth creation. We set out to explore the dynamic relationship between investing tax revenue in natural capital (specifically forested headwaters and low land wetlands) rather than built capital (specifically stopbanks) for flood protection in the Manawatū watershed, New Zealand. Significant investment is currently required to maintain and enhance river engineering infrastructure and keep pace with changes in the river's geomorphology. Viewed from a systems perspective, we suggest diversion of a proportion of existing funding into restoration of forested headwaters on steep slopes and restoration of functioning wetlands on floodplains could in the longer term provide an effective approach to flood protection. Co-benefits of increased natural capital include the ecosystem services nutrient cycling, sediment capture, water purification, biodiversity, pollination, and cultural and recreational values. Overcoming an investment trap requires a longer term perspective. This simple model consisting of two feedback loops and two delays aims to contribute to an ongoing stakeholder dialogue concerning the Manawatū River watershed in New Zealand.(KEY TERMS: flood protection; natural capital; built capital; reforestation; wetland restoration; investment trap; ecosystem services.) van den Belt, Marjan, Thomas Bowen, Kimberley Slee, and Vicky Forgie, 2013. Flood Protection: Highlighting an Investment Trap Between Built and Natural Capital.