Flood Forecasting on a Shoestring Budget - a new Open Source Project

Operational hydrometeorological forecast centers whose principal focus is flood and hydrologic forecasting do not exist in all countries. In many other countries the forecasting methods used are remarkably antiquated and only provide the most rudimentary flood alert and warning services to the public and other significant stakeholders. Many international development programs aimed at addressing the deficiencies in low- and middle-income countries have been undertaken, largely through financing from international development sources, generally end with suboptimal outcomes. Two major problems can be found with these programs (1) once external funding ends there is insufficient in-country funding for continued operation and maintenance of alert and warning systems and (2) the needed human capacity building to sustain the modernized alert and warning systems is inadequate and diminishes over time.

In response to these shortcomings many global and continental scale flood forecasting systems have been developed. But these small-scale forecast systems lack the needed spatial and temporal scale detail to meet village level needs, that is, site specificity for impact based forecasts. Large-scale flood forecasting is achievable, but generally at significant cost. What is needed is a low cost alternative that is financially sustainable which, at the same time, grows country technical capacity. The approach suggested in Flood Forecasting on a Shoestring Budget attempts to close this gap with free, open source software tools and models. The architecture of the system is presented with a prototype example.