Imputation Methods for Missing Hydrometeorological Data Estimation

En savoir plus sur le livre

Missing data is a common issue in hydrometeorological datasets, necessitating effective spatial and temporal imputation methods to achieve error-free and continuous data. This comprehensive guide serves as a reference for both basic and advanced interpolation techniques for imputing missing data. It explores various methods, including spatial and temporal interpolation, universal function approximation, and data mining-assisted approaches. The book introduces innovative spatial deterministic and stochastic methods, emphasizing the objective selection of control points and optimal spatial interpolation. Additionally, it delves into emerging machine learning techniques applicable to interpolation schemes, along with error and performance measures for assessing and validating imputed data. Practical applications of these methods are demonstrated using real-world hydrometeorological data. The content is tailored for a diverse audience, including graduate students and researchers in climatology, hydrology, and earth sciences, as well as water engineering professionals from government and private sectors engaged in hydrometeorological and climatological data processing. Key topics include an introduction to missing data, various imputation methods, temporal and spatial interpolation, data-driven models, multiple imputation methods, and evaluation techniques for imputation methods and data.