Characterizing anisotropic scale-specific variations in soil salinity from a reclaimed marshland in China

In the recent past, much of the marshland in China has been reclaimed for agriculture and to create urban buffer zones. Information about the variability of soil properties (e.g. salinity) in these reclaimed areas is mostly limited, but is crucial for improving agricultural production as well as sustaining their environmental benefits. Characterization of variations in soil salinity and their correlation with other soil properties at different scales can improve the use of these reclaimed areas, as well as assess the efficacy of the reclamation process. In this study, we examined variations in soil salinity at different directions using anisotropic analysis (AA) and separated the variations at different scales using empirical mode decomposition (EMD). The AA identified two pairs of major and minor axes along which variations in soil salinity were dominant. The EMD separated the variations in soil salinity along these axes into six intrinsic mode functions according to the dominant scales of variations. The importance of these scale components were assessed based on their percent variance contribution towards the overall variance. Large variations in soil salinity were observed mainly at 200-500 m and 3500-4500 m scales. The scale-specific variations in soil salinity and their correlation with other soil properties were direction-specific. A strong correlation between the large scale components of soil salinity (>5 km) and soil total nitrogen and clay content were identified along the directions from inland to the ocean. This may be attributed to the distance from the ocean as well as the length of time for which reclaimed land has been cultivated. Information about this variability at different scales and directions acquired by combining AA and EMD provided a unique way to characterize reclaimed lands for agricultural or environmental use. This information will be helpful in developing future sampling strategies and preparing detailed digital soil maps for improved management decisions. (C) 2015 Elsevier B.V. All rights reserved.