Infilling missing precipitation records using variants of spatial interpolation and data‐driven methods: use of optimal weighting parameters and nearest neighbour‐based corrections

Abstract

Variants of spatial interpolation and data-driven methods to fill gaps in daily precipitation records are developed and evaluated in this study. The evaluated methods include variations of inverse distance and correlation weighting procedures, linear weight optimization and artificial neural networks. An already existing method, support vector logistic regression-based copula, is also assessed. Optimal weights are estimated using inverse distance and correlation-based weighting methods, post-corrections of spatially interpolated estimates for rain or no rain classifications using support vector machine (SVM), and variations of a single best classifier (SBC) are used. The optimal number of gauges for use in spatial interpolation methods and for artificial neural network-based method are selected. Three benchmark methods provide a basis against which all the methods are compared: single best estimator (SBE), and spatial and climatological mean estimators (SME and CME). All of the methods are tested for estimating varying amounts of missing precipitation data at 53 rain gauges located in South Florida, USA. Results show that the linear weight optimization method with an SBE provides the best estimates of daily precipitation values based on several performance metrics. Results from evaluation of different methods and their variants indicate use of optimized exponents in distance and correlation-based weighting methods, classifiers for rain or no rain conditions, and an optimal number of neighbours in spatial interpolation improve estimates of missing data. Corrections to missing data estimates using nearest neighbours can help in improving the accuracy of rain and no rain state determinations with a possibility of introducing bias in estimates.

An assessment of global satellite‐based precipitation datasets in capturing precipitation extremes: A comparison with observed precipitation dataset in India

“…Precipitation datasets often contain discrepancies and large‐gaps (Kim et al ., ; Singh and Xiaosheng, , ), which makes it hard to produce accurate information by utilizing such unreliable or incomplete datasets. Several studies were performed to rectify such distortions and filling data gaps in gauged time series precipitation datasets (Teegavarapu et al ., ; Singh and Xiaosheng, , ). Also, the magnitude of the precipitation varies in the spatio‐temporal domains.…”

“…In India, the efficiency evaluation of satellite‐based precipitation products (e.g., SM2RAIN‐ASCAT, CHIRPS, etc.) is limited and a comprehensive assessment of these satellite‐based precipitation datasets to check their ability to simulate the extreme rain events has not been explored, which may produce uncertain outcomes in case of hydrological extreme events such as floods and droughts (Teegavarapu et al ., ; Singh and Xiaosheng, ; ).…”

Rainfall variability in Malay Peninsula region of Southeast Asia using gridded data

“…This study utilizes standard mathematical methods for filling the rainfall data gaps in SA-OBS after a careful review of literatures [6,12]. Initially, the minor data gaps (<2%) in the SA-OBS data is filled by using the average of the nearest neighbouring rainfall grids.…”

“…The missing values in rainfall time series is common but long data gaps could be critical since these gaps in rainfall time series can produce biased results [4,5]. To reduce the uncertainty in the filled rainfall time series, the main prominence should be maintaining rainfall frequency, extremity and their patterns [6]. The objective of this study is to analyse rainfall variabilities over the Malaya Peninsula region in a relatively long-term (1981-2007) duration using gridded rainfall dataset at a resolution of 0.50°×0.50°.…”

A General Overview of the Risk-Reduction Strategies for Floods and Droughts

“…The common approach is to adaptively correct remote sensing-based data with ground-based observations [108,109]. Recent studies have used data-driven or data-mining techniques to identify possible data bias or extremes and further correct them simultaneously to meet the needs of disaster response [110][111][112]. Finally, because of rapid progress in the internet of things (IoT), sensors and associated components are becoming cheaper than ever before.…”