Mapping Fuel Loads and Dynamics in Rangelands Using Multi-Sensor Data in the Great Basin, USA
Abstract
Fuel conditions in rangelands are influenced by disturbances such as wildfires, and is also strongly controlled by weather and climate. These factors impact the availability of fuel loads, which is the key component to stimulate burned area and severity. In this paper, we developed an approach for mapping live fuel loads (biomass density) and their dynamics using field collection, Landsat 8, and MODIS data sets at a spatial resolution of 30 m from the growing season. Using the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) modelling process, we generated monthly shrub and grassland greenness levels for 2015. The spatial resolution of Landsat and the temporal resolution of MODIS complimented each other to allow us to produce monthly products. Understanding the dynamics of these greenness patterns helps the fire management community to recognize areas that have high likelihood of burning in the future, thus enabling them to anticipate and plan accordingly. We obtained field biomass information from selected shrub and grass sites located throughout the Great Basin. This information was used to calibrate fire models and generate remotely-sensed data sets. We then used Landsat 8 NDVI dates representing the phenological profile, regression tree models, and product validation. The calculated fuel loads were further examined and validated using high resolution images (World View 2/3), field measurements, and Google Earth. Once we have the requisite image data converted to biomass, we anticipate fire conditions and behavior using various models developed by the fire community. One key element is to use information from this study to improve and inform the Rangeland Vegetation Simulator. Finally, we analyzed the correlations of fire occurrence (frequency) and burn severity with live fuel loads and climate conditions. Our results show modeled fuel loads and their dynamics in rangelands capture the spatiotemporal heterogeneity of non-forest live fuel types and the variations in both wildfire disturbances and climate/weather conditions. This suggests the developed approach to map fuel loads is robust and can improve the existing LANDFIRE fuel data in rangelands. It can also be used to monitor the changes in fuel conditions in response to management activities and climate change.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFMGC51E1219L
- Keywords:
-
- 0315 Biosphere/atmosphere interactions;
- ATMOSPHERIC COMPOSITION AND STRUCTUREDE: 3390 Wildland fire model;
- ATMOSPHERIC PROCESSESDE: 1615 Biogeochemical cycles;
- processes;
- and modeling;
- GLOBAL CHANGEDE: 1640 Remote sensing;
- GLOBAL CHANGE