Journal
GEODERMA
Volume 230, Issue -, Pages 79-94Publisher
ELSEVIER
DOI: 10.1016/j.geoderma.2014.04.008
Keywords
Boreal forest; Chronosequence; Machine learning; Soil carbon; Organic layer thickness; Remote sensing; Succession; Tundra; Wetlands
Categories
Funding
- U.S. Geological Survey (USGS) [G08PC91508, G10PC00044]
- Climate Effects Network Yukon River show Basin program
Ask authors/readers for more resources
Understanding of the organic layer thickness COLT) and organic layer carbon (OLC) stocks in subarctic ecosystems is critical due to their importance in the global carbon cycle. Moreover, post-fire OLT provides an indicator of long-term successional trajectories and permafrost susceptibility to thaw. To these ends, we 1) mapped OLT and associated uncertainty at 30 m resolution in the Yukon River Basin (YRB), Alaska, employing decision tree models linking remotely sensed imagery with field and ancillary data, 2) converted OLT to OLC using a nonlinear regression, 3) evaluate landscape controls on OLT and OLC, and 4) quantified the post-fire recovery of OLT and OLC. Areas of shallow (<10 cm), moderate (>= 10 cm and <20 cm), moderately thick (>= 20 cm and <30 cm), and thick (>= 30 cm) OLT, composed 34, 20, 14, and 18% of the YRB, respectively; the average OLT was 19.4 cm. Total OLC was estimated to be 338 Pg. A regional chronosequence analysis over 30 years revealed that OLT and OLC increased with stand age COLT: R-2 = 0.68; OLC: R-2 = 0.66), where an average of 16 cm OLT and 5.3 kg/m(2) OLC were consumed by fires. Strong predictors of OLT included climate, topography, near-surface permafrost distributions, soil wetness, and spectral information. Our modeling approach enabled us to produce regional maps of OLT and OLC, which will be useful in understanding risks and feedbacks associated with fires and climate feedbacks.(C) 2014 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available