This paper presents a physics-based digital twin solution for floating offshore wind turbines, validated using measurement data from the full-scale TetraSpar prototype. The digital twin integrates a Kalman filter to estimate structural states, an aerodynamic estimator, and a physics-based virtual sensing procedure to obtain loads along the tower. The Kalman filter uses a linear model of the structure and measurements from the turbine, while the aerodynamic estimator and virtual sensing procedure are designed to estimate aerodynamic loads and section loads, respectively. The digital twin relies on standard measurements such as power, pitch, rotor speed, tower acceleration, inclinometers, and GPS sensors. The authors explore two approaches to obtain physics-based models: dedicated Python tools and OpenFAST linearization. The final version of the digital twin combines components from both approaches. Numerical experiments and measurements from the TetraSpar prototype are used to verify the accuracy of the digital twin, with estimated damage equivalent loads of the tower fore-aft bending moment achieving an accuracy of approximately 5% to 10%. The overall accuracy of the digital twin is promising, demonstrating its potential for estimating fatigue loads on floating offshore wind turbines. The paper also discusses the future potential of integrating monitoring and diagnostics aspects to inform operation and maintenance decisions.This paper presents a physics-based digital twin solution for floating offshore wind turbines, validated using measurement data from the full-scale TetraSpar prototype. The digital twin integrates a Kalman filter to estimate structural states, an aerodynamic estimator, and a physics-based virtual sensing procedure to obtain loads along the tower. The Kalman filter uses a linear model of the structure and measurements from the turbine, while the aerodynamic estimator and virtual sensing procedure are designed to estimate aerodynamic loads and section loads, respectively. The digital twin relies on standard measurements such as power, pitch, rotor speed, tower acceleration, inclinometers, and GPS sensors. The authors explore two approaches to obtain physics-based models: dedicated Python tools and OpenFAST linearization. The final version of the digital twin combines components from both approaches. Numerical experiments and measurements from the TetraSpar prototype are used to verify the accuracy of the digital twin, with estimated damage equivalent loads of the tower fore-aft bending moment achieving an accuracy of approximately 5% to 10%. The overall accuracy of the digital twin is promising, demonstrating its potential for estimating fatigue loads on floating offshore wind turbines. The paper also discusses the future potential of integrating monitoring and diagnostics aspects to inform operation and maintenance decisions.