Machine Learning Enables Real-Time Proactive Quality Control: A Proof-Of-Concept Study

To improve the forecast accuracy of numerical weather prediction, it is essential to obtain better initial conditions by combining simulations and available observations via data assimilation. It has been known that a part of observations degrade the forecast accuracy. Detecting and discarding such detrimental observations via proactive quality control (PQC) could improve the forecast accuracy. However, conventional methods for diagnosing observation impacts require future observations as a reference state and PQC cannot be real-time in general. This study proposes using machine learning (ML) trained by a time series of analyses to obtain a reference state without future observations and enable real-time ML-based PQC. This study presents proof-of-concept using a low-dimensional dynamical system. The results indicate that ML-based and model-based estimates of observation impacts are generally consistent. Furthermore, ML-based real-time PQC successfully improves the forecast accuracy compared to a baseline experiment without PQC. Simulation-based weather prediction needs observations to obtain accurate initial conditions and to improve the forecast accuracy. However, it has been reported that a part of observations degrade the forecast accuracy. Detecting such detrimental observations using future observations and proactively denying them could result in skillful predictions. This approach is referred to as proactive quality control (PQC) and cannot be real-time in general because PQC uses future information. To enable real-time PQC, this study proposes using machine learning (ML) predictions instead of future observations in PQC. This study demonstrates for the first time that ML-based real-time PQC successfully improves the forecast accuracy in a low-dimensional dynamical system. Future research could apply ML-based PQC for more complicated weather prediction systems. Proactive quality control (PQC) uses future information for detecting detrimental observations and cannot be real-time in general We propose using machine learning (ML) predictions instead of future information and enabling real-time PQC ML-based real-time PQC successfully improves the forecast accuracy in a low-dimensional dynamical system