Spatial and regression-based missing precipitation data imputation: Western Black Sea region

Abstract

The study of natural phenomena in the environment influences the shaping of human geography. Investigating the occurring physical events is achieved by measuring the magnitudes in nature. These measurements are then structured within certain models, and the resulting outputs are used in engineering applications. However, measurements taken from nature or a system may not provide continuous data due to human and sensor-related errors or inadequacies, resulting in gaps or discontinuities in data acquisition. The success of the method in the missing data completion problem is still an important research topic, as it is influenced by various factors such as the characteristics of the data and the type of missing data. Particularly, the lack of precipitation observation data due to climate change poses serious risks in the planning of water structures. In this study, spatial-based inverse weighted distance (IDW), regression, and statistical methods such as mean and median values are used to fill in and complete missing precipitation data obtained from meteorological stations in the Western Black Sea Region. The results of the study conducted at 10 stations showed that the spatial-based method, IDW, produced more successful results.