Multi-scale digital soil mapping with deep learning. Issue 1 (December 2018)

• Record Type: Journal Article
• Authors: Behrens, Thorsten; Schmidt, Karsten; MacMillan, Robert; Viscarra Rossel, Raphael
• Is Part Of: Scientific reports. Volume 8:Issue 1(2018)
• Page Start: 1
• Page End: 9
• Publication Date: 2018-12
• Subjects: Natural history -- Research -- Periodicals; Biology -- Research -- Periodicals; Physical sciences -- Research -- Periodicals; Earth sciences -- Research -- Periodicals; Environmental sciences -- Research -- Periodicals; 502.85
• DOI: 10.1038/s41598-018-33516-6
• Languages: English
• ISSNs: 2045-2322

Abstract

Abstract We compared different methods of multi-scale terrain feature construction and their relative effectiveness for digital soil mapping with a Deep Learning algorithm. The most common approach for multi-scale feature construction in DSM is to filter terrain attributes based on different neighborhood sizes, however results can be difficult to interpret because the approach is affected by outliers. Alternatively, one can derive the terrain attributes on decomposed elevation data, but the resulting maps can have artefacts rendering the approach undesirable. Here, we introduce 'mixed scaling' a new method that overcomes these issues and preserves the landscape features that are identifiable at different scales. The new method also extends the Gaussian pyramid by introducing additional intermediate scales. This minimizes the risk that the scales that are important for soil formation are not available in the model. In our extended implementation of the Gaussian pyramid, we tested four intermediate scales between any two consecutive octaves of the Gaussian pyramid and modelled the data with Deep Learning and Random Forests. We performed the experiments using three different datasets and show that mixed scaling with the extended Gaussian pyramid produced the best performing set of covariates and that modelling with Deep Learning produced the most accurate predictions, which on average were 4–7% more accurate compared to modelling with Random Forests.