Water resources operation is facing difficulties due to climate change-induced unprecedented
rainfall patterns and aging water infrastructures that are vulnerable to flood. In order
to control flooding, dams must open their gates and discharge reservoir water. However, since the
water level of the downstream river also rises, more scientific dam operation is required. ‘Digital
twin’ technology, which implements the real world into the digital world, can support better decision-
making through intuitive monitoring and various simulations. This paper introduces a digital
twin watershed management platform that is pilot-built to respond to floods. The platform
name is K-Twin SJ 2.0. A GIS-based geospatial digital twin is built for the entire Sumjin dam and
river water system in Korea (watershed area 4,913 km2, river extension 173 km). Real-time data
such as rainfall, dam and river water level, flow rate, and CCTV are intuitively synchronized.
High-precision topography and facility information for dams and rivers are twinned visibly. Three
types of simulation models (rainfall-runoff, river water level prediction, and flooding) are loaded
for efficient dam operation considering the river conditions. It can also predict the river water
level and suggest the optimal dam discharge scenario with the aid of AI technology. A safety
evaluation module for the river levees is implemented. It is also equipped with an advanced drone
monitoring system for dams and rivers and an AI-intelligent CCTV video surveillance function.
The digital twin-based platform supports efficient decision-making for the smart flood response
in a wide range of water systems. In addition, the comprehensive configuration of dam and river
can be considered together on the same line, contributing to flooding damage reduction and optimal
operation through better smart water management.