Topological data analysis (TDA) applied to reveal pedogenetic principles of European topsoil system

In short period of time, topological data analysis (TDA) allowed deeper insight and understanding of various phenomena, from proper choice of sport teams to cancer treatments. For the first time, this state-of-the-art technique was applied in soil science. As the evaluation set of data, the Land Use/Land Cover Area Frame Survey (LUCAS) was used. Database consisted of approximately 20000 samples, while 12 physical and chemical parameters were measured for each sample. Without any pre-judgment for the grouping of data TDA was applied. The obtained results were cross-checked against known grouping parameters including 5 types of land cover, 9 types of clime and organic carbon content of soil. Results showed that TDA method is reliable for exploring the soil system and understanding relationships of topsoil properties. Some of the grouping characteristics achieved using standard approaches were confirmed by TDA (e.g. organic carbon content) but some novel, subtle relationships (e.g. magnitude of anthropogenic effect in soil formation), were discovered as well. In further analysis, after the reliability of the method was proven, TDA was applied on specific types of topsoil according to carbon content leading to diverse conclusions about both biotic and abiotic factors that affect soil formation and soil composition.

SAVIC Alexander;  TOTH Gergely;  DUPONCHEL Ludaovic; 

2017-03-15

NATURE PUBLISHING GROUP

JRC101753

2045-2322,   

http://www.sciencedirect.com/science/article/pii/S0048969717303431,    https://publications.jrc.ec.europa.eu/repository/handle/JRC101753,   

10.1016/j.scitotenv.2017.02.095,