Using Land Surface Temperature to Quantify Fuel Moisture in Complex Terrain
Abstract
The moisture of dead and down vegetation, referred to as dead fuel moisture (FM), is important for understanding fire risk and behavior and is critical for performing lower intensity prescribed fires as part of fuels management. FM is estimated using air temperature, precipitation, and/or vapor pressure deficit (VPD) data from meteorological stations. However, fire behavior is inherently dependent on spatial processes that can vary substantially in regions with complex topography. Spatially distributed measurements of land surface temperature (LST) are becoming increasingly accessible through unmanned aerial system technology and may provide an opportunity to account for this complexity and improve our estimates of FM. LST reflects energy and water states and fluxes, specifically longwave radiation and turbulent sensible and latent energy. Drier fuel partitions more incoming energy to heating the fuel rather than evaporating the moisture. Thermal infrared (TIR) observations can be used to detect LST from different platforms but have seen limited application for FM detection where they could aid in constraining models in complex terrain. We collected a unique LST and FM dataset across a hydrological gradient to ask two questions: 1. Can LST be used to determine FM? 2. What does LST variability over larger scales indicate about the utility of TIR for monitoring FM dynamics? At Sagehen Experimental Forest in the Sierra Nevada, we continuously monitor the LST in ground locations that we relate to ground-based FM. We derive LST using our TIR datasets at ~10 cm resolution over an area of ~0.5 hectare, with an accuracy of ~1.5 °C. As expected, UAS-derived LST measurements have high variability across our study hillslope of ~250 m. There is 10-20 °C difference in the fuel temperature between the cooler meadow to the warmer south-facing hillslope, which corresponds to VPD differences of 25-100%. We then compare UAS-derived LST and VPD maps to ground-based observations of VPD and FM. Finally, we use ECOSTRESS observations coincident with the UAS flights to determine the potential utility of LST collected from space for estimating FM Our LST-based results show promise for reducing uncertainty in existing empirical FM models and providing a powerful tool for observing and modeling FM in complex terrain.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.A23J2951S
- Keywords:
-
- 3379 Turbulence;
- ATMOSPHERIC PROCESSES;
- 3390 Wildland fire model;
- ATMOSPHERIC PROCESSES;
- 4301 Atmospheric;
- NATURAL HAZARDS;
- 4313 Extreme events;
- NATURAL HAZARDS