Abstract This study describes the architecture and capabilities of an open source agent-based Java platform that permits the systematic study of interactions among hydrology, climate, and strategic human decision-making in a watershed over time. To demonstrate the platform's use and capabilities, an application is presented in accordance with ODD protocol requirements that captures, in simplified form, the structural attributes of the Squaw Creek watershed in central Iowa. Illustrative findings are reported for the sensitivity of farmer and city social welfare outcomes to changes in three key treatment factors: farmer land-allocation decision method, farmer targeted savings, and levee quality effectiveness for the mitigation of city flood damage.