| Summary: | This study evaluates the effect of considering data intermittency and log-normality in applications of simple Kriging. Several sets of synthetic data, both intermittent and log-normal, were prepared for this purpose, and then four different Kriging applications were repeated with these synthetic data under different assumptions of data intermittency and log-normality. The effects of these assumptions on the simple Kriging applications were evaluated and compared with each other. As a result, it was found that the derived correlation length of a variogram becomes longer when considering both data intermittency and log-normality, and the sill height becomes smaller when data intermittency is high. The data field generated by simple Kriging was also closer to the original data when considering both data intermittency and log-normality. In the application to rain rate data, the effect of considering data intermittency was confirmed. However, the effect of considering data log-normality was found to be vague. The general assumption of log-normality in relation to the rain rate data seems not to be so valid, at least not for the rain rate data considered in this study. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.
|
|---|
