The map illustrates depth to the redox interface in meters below ground. The interface marks the transition from oxidised to reduced conditions in the subsurface. Based on approx. 13.000 observations of colour changes in sediment cores, a model has been developed describing relations between explanatory variables and depth to the redox interface. The machine learning method random forest has been used to develop the model. The map shows the best estimate of the location of the interface in a 100 m grid resolution. The applied method additional provides an estimate of the uncertainty. The method is described further in:

Koch J, Stisen S, Refsgaard JC, Ernstsen V, Jakobsen PR, Højberg AL (2019) Modelling depth of the redox interface at high resolution at national scale using Random Forest and residual Gaussian simulation, Water Resour. Res., https://doi.org/10.1029/2018WR023939

Koch J, Stisen S, Refsgaard JC, Ernstsen V, Jakobsen PR, Højberg AL (2019) Nyt nationalt redoxkort i høj opløsning. In Tema: TReNDS. Vand & Jord, 26 (1), 26-29.