Soil moisture memory and soil properties: An analysis with the stored precipitation fraction

It is of primary interest to determine the extent that natural processes are linked to understand natural system dynamics. There is growing interest in investigating the coupling and coevolution occurring between natural processes and the interaction between natural systems. Soil moisture is a variable that is at the core of the interaction between multiple physical and biological processes; therefore, it is very important to understand the influence of soil moisture dynamics on that coevolution. Soil moisture memory (SMM) shows soil moisture dynamics and the nature of their interaction with other related processes.Several studies have recognized that SMM depends on soil properties and some other meteorological and biophysical variables; however, there has not been a study specifically devoted to studying the relationship between SMM and the main soil properties. The present study analyzes the relationship between several relevant soil properties (texture, bulk density and organic matter content) involved in soil water dynamics and the stored precipitation fraction (Fₚ) as a suitable expression of the SMM. The relationship between SMM and precipitation and soil water content was also analyzed. These relationships were analyzed both at the surface layer and at several soil layer depths and with four soil moisture sampling frequencies (3 and 12 h, and 1 and 3 days). The study was conducted with data collected over a period of 12 years, at two soil moisture monitoring networks (REMEDHUS and SCAN) located in Spain and the USA, respectively, covering a wide range of soil characteristics and environmental conditions.The results showed that soil texture plays the main role in understanding SMM dynamics and that organic matter content has a good relationship with Fₚ. The sand fraction is the variable more correlated with Fₚ, with a clear inverse relationship (correlation coefficient, from −0.62 to −0.92 in REMEDHUS, and from −0.65 to −0.76 in SCAN). The variations in Fₚ observed with the different soil layer depths are also mainly correlated with texture and organic matter content, but only for sampling intervals of 1 and 3 days. The temporal variability of Fₚ in the different soil layers is more correlated with the soil water content than with annual precipitation, especially in the surface layer. At 0–5 depth, R ranges between 0.42 and 0.82. For sampling intervals of 1 and 3 days, the R values between Fₚ and mean soil moisture are always significant (p < 0.01) and higher than 0.66 in all soil layers. Therefore, in view of the cases studied and the results obtained, the most critical aspect of understanding SMM dynamics is precise characterization of the water storage conditions, especially in terms of texture and organic matter content.