Analysis of surface soil moisture patterns in agricultural landscapes using Empirical Orthogonal Functions

摘要

Soil moisture is one of the fundamental variables in hydrology, meteorologyand agriculture. Nevertheless, its spatio-temporal patterns inagriculturally used landscapes that are affected by multiple natural(rainfall, soil, topography etc.) and agronomic (fertilisation, soilmanagement etc.) factors are often not well known. The aim of this study isto determine the dominant factors governing the spatio-temporal patterns ofsurface soil moisture in a grassland and an arable test site that arelocated within the Rur catchment in Western Germany. Surface soil moisture(0–6 cm) was measured in an approx. 50×50 m grid during 14 and 17measurement campaigns (May 2007 to November 2008) in both test sites. Toanalyse the spatio-temporal patterns of surface soil moisture, an EmpiricalOrthogonal Function (EOF) analysis was applied and the results werecorrelated with parameters derived from topography, soil, vegetation andland management to link the patterns to related factors and processes. Forthe grassland test site, the analysis resulted in one significant spatialstructure (first EOF), which explained 57.5% of the spatial variabilityconnected to soil properties and topography. The statistical weight of thefirst spatial EOF is stronger on wet days. The highest temporal variabilitycan be found in locations with a high percentage of soil organic carbon(SOC). For the arable test site, the analysis resulted in two significantspatial structures, the first EOF, which explained 38.4% of the spatialvariability, and showed a highly significant correlation to soil properties,namely soil texture and soil stone content. The second EOF, which explained28.3% of the spatial variability, is linked to differences in landmanagement. The soil moisture in the arable test site varied more stronglyduring dry and wet periods at locations with low porosity. The methodapplied is capable of identifying the dominant parameters controllingspatio-temporal patterns of surface soil moisture without being affected bysingle random processes, even in intensively managed agricultural areas.