Mapping global depth to bedrock based on borehole logs and soil profile data using machine learning
Journal of Geology & Geophysics

Journal of Geology & Geophysics
Open Access

ISSN: 2381-8719

+44 20 3868 9735

Mapping global depth to bedrock based on borehole logs and soil profile data using machine learning

International Conference on Geosciences and Geophysics

October 06-07, 2016 Orlando, USA

Wei Shangguan, Tomislav Hengl, Jorge Mendes de Jesus, Hua Yuan and Yongjiu Dai

Beijing Normal University, China
International Soil Reference and Information Centre-World Soil Information, Netherlands
Sun Yat Sen University, China

Posters & Accepted Abstracts: J Geol Geophys

Abstract :

Understanding the global pattern of underground boundaries such as bedrock occurrence is of continuous interest to Earth and geosciences. This work presents a framework for global estimation of depth to bedrock (DTB). Observations were extracted from a global compilation of soil profile data (ca. 130,000 locations) and borehole data (ca. 1.6 million locations). Additional pseudoobservations generated by expert knowledge were added to fill in large sampling gaps. The model training points were then overlaid on a stack of 155 covariates including DEM-based hydrological and morphological derivatives, lithologic units, MODIS surface reflectance bands and vegetation indices derived from the MODIS land products. Global spatial prediction models were developed using random forests and gradient boosting tree algorithms. The final predictions were generated at the spatial resolution of 250 m as an ensemble prediction of the two independently fitted models. The 10ΓΆΒ?Β?fold cross-validation shows that the models explain 59% for absolute DTB and 34% for censored DTB (depths deep than 200 cm are predicted as 200 cm). The model for occurrence of R horizon (bedrock) within 200 cm does a good job. Visual comparisons of predictions in the study areas, where more detailed maps of depth to bedrock exist, show that there is a general match with spatial patterns from similar local studies. Limitation of the data set and extrapolation in data spare areas should not be ignored in applications. To improve accuracy of spatial prediction, more borehole drilling logs will need to be added to supplement the existing training points in under-represented areas.

Biography :

Email: [email protected]