Evaluating Remote Sensing Approaches to Describe Forest Structure for Ladder Fuel Estimation and to Predict Wildfire Burn Severity
Wildfires are becoming larger and more severe due to climate change and historical fire suppression. Technology to quantify ladder fuels, which bridge the gap between the surface and the canopy, can help manage forest structure to reduce fire hazard. In this study we evaluated several remote sensing techniques and field measurements to quantify ladder fuels and relate these metrics to burn severity. Ladder fuel data at 1-m strata from 1-8 m were collected using a 4 x 0.5-m photo banner, a terrestrial laser scanner (TLS), a handheld-mobile laser scanner (HMLS), an unoccupied aerial system with multispectral camera and Structure from Motion processing (UAS-SfM), and airborne laser scanner (ALS) data in 35 plots in oak woodlands and mixed-conifer forests in Sonoma County, California, USA, before wildfires occurred. Canopy base height (CBH) was also measured and post-wildfire burn severity was calculated using relativized delta normalized burn ratio (RdNBR). The linear relationships among ladder fuel metrics at each strata were compared and RdNBR prediction was evaluated with and without CBH as an interaction term. All approaches quantified ladder fuels across plots but were not consistently related to each other, unless CBH height class was included as a means of categorizing structural differences among plots. The UAS-SfM could not measure relative differences across plots due to lack of penetration to the ground. Ladder fuels between 1-2m and 2-3m best predicted RdNBR across most methods, where HMLS had the strongest correlation (R2 = +0.72). By accounting for interactions between ladder fuels from 1-3 m, CBH, and burn severity, diverse remote sensing approaches can be used to estimate and validate ladder fuels. Importantly, forest structure has important implications for estimating ladder fuels and may be crucial to consider if ladder fuels are extrapolated across larger spatial scales.
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021