McNicol-2019-NPCTR-SOC-map

• Main Authors: McNicol, Gavin, Bulmer, Chuck, D'Amore, David D., Sanborn, Paul, Saunders, Sari, Giesbrecht, Ian, Gonzalez Arriola, Santiago, Bidlack, Allison, Butman, David, Buma, Brian
• Format: Dataset
• Language: unknown
• Published: 2018
• Subjects: soil carbon; random forest; temperate rainforest; pedology; machine learning; Canada; Pacific; British Columbia; glaciers; Alaska
• Online Access: http://hdl.handle.net/10255/dryad.197516; https://doi.org/10.5061/dryad.5jf6j1r/1

This raster [.tif] is the predicted soil organic carbon for the North Pacific coastal temperate rainforest. Content is displayed in megagrams of carbon per hectare (Mg ha-1) to 1 m in mineral soil, plus overlying organic horizons. Map values are the output of a random forest machine learning algorithm trained on pedon data from within British Columbia and southeast Alaska only, therefore confidence is low for predictions south of the US-Canada border and predictions in that region have not been validated. Lakes, glaciers and ice-fields have also not been masked from the map. More information on the map can be found in the associated manuscript.