Quantifying soil carbon loss and uncertainty from a peatland wildfire using multi-temporal LiDAR

• Published in: Remote sensing of environment 2015-12, Vol.170, p.306-316
• Main Authors: Reddy, A.D., Hawbaker, T.J., Wurster, F., Zhu, Z., Ward, S., Newcomb, D., Murray, R.
• Format: Article
• Language: English
• Subjects: Carbon; Elevation; Errors; Estimates; Fire; Fires; LiDAR; Peatlands; Soil (material); Soil carbon; Uncertainty; Vertical accuracy
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2015.09.017

Peatlands are a major reservoir of global soil carbon, yet account for just 3% of global land cover. Human impacts like draining can hinder the ability of peatlands to sequester carbon and expose their soils to fire under dry conditions. Estimating soil carbon loss from peat fires can be challenging due to uncertainty about pre-fire surface elevations. This study uses multi-temporal LiDAR to obtain pre- and post-fire elevations and estimate soil carbon loss caused by the 2011 Lateral West fire in the Great Dismal Swamp National Wildlife Refuge, VA, USA. We also determine how LiDAR elevation error affects uncertainty in our carbon loss estimate by randomly perturbing the LiDAR point elevations and recalculating elevation change and carbon loss, iterating this process 1000 times. We calculated a total loss using LiDAR of 1.10TgC across the 25km2 burned area. The fire burned an average of 47cm deep, equivalent to 44kgC/m2, a value larger than the 1997 Indonesian peat fires (29kgC/m2). Carbon loss via the First-Order Fire Effects Model (FOFEM) was estimated to be 0.06TgC. Propagating the LiDAR elevation error to the carbon loss estimates, we calculated a standard deviation of 0.00009TgC, equivalent to 0.008% of total carbon loss. We conclude that LiDAR elevation error is not a significant contributor to uncertainty in soil carbon loss under severe fire conditions with substantial peat consumption. However, uncertainties may be more substantial when soil elevation loss is of a similar or smaller magnitude than the reported LiDAR error. •Multi-temporal LiDAR can provide estimates of soil carbon loss from fire.•1.1 Tg (44kg/m2) soil carbon were burned in a 2011 Virginia, USA peat fire.•Soil loss exceeded 1m in some parts of the drained peatland.•LiDAR elevation error was not a major contributor to uncertainty in carbon loss.