Assessing the time stability of soil moisture patterns using statistical and geostatistical approaches

Understanding the variation of soil moisture patterns over time can support field investigative efforts specific to managing irrigation input, environmental risk assessment, or water resource planning. The objective of this study was to apply an integrated suite of traditional (e.g. Vachaud’s time stability analysis) and less common techniques (e.g. confusion matrix, Castrignano’s average of the differences, cross correlogram analysis, and polygon kriging) to gain deeper insight into the temporal persistence of soil moisture patterns, especially for site-specific water management purposes. This study used soil moisture estimates generated in previous work that was carried out in a field in Central Kentucky for three dates ranging from permanent wilting point up to field capacity. The results obtained from this study provide richer evidence of time stable soil moisture patterns and lend greater insight into the controlling factors of spatial and temporal variation, including soil moisture status, soil physicochemical properties, and landscape position, that otherwise would not be attainable using a single metric alone.