Temporal Interpolation Methods

Temporal interpolation (TI) or univariate interpolation aims to fill missing values within a time series of observations. TI methods are generally referred to as the within-site (station) methods that use only the observations at the base site (i.e., a site with missing data) for imputation. TI uses available data before and after missing values (World Meteorological Organization (WMO) in Guide to climatological practices. WMO-No-100, Geneva, Switzerland 2011). In contrast, spatial interpolation (SI) is referred to as a between-site (station) method, which uses observations at several regional locations to fill values at a site with missing data. The main feature of SI is that it relies on the quantifiable strength of association between observations at any two sites. TI uses the information available from a univariate series of hydroclimatic or other variables at a site by utilizing time-dependences in a time series.