整合生物学期刊网

林业研究(英文版) ›› 2023, Vol. 34 ›› Issue (5): 1195-1206.DOI: 10.1007/s11676-023-01599-1

• •    下一篇

Bountouraby Balde1,a, Cristina Vega-Garcia1, Pere Joan Gelabert1, Aitor Ameztegui1, Marcos Rodrigues2   

  • 收稿日期:2022-06-14 接受日期:2022-11-24 出版日期:2024-10-16 发布日期:2024-10-16
  • 通讯作者: Bountouraby Balde

The relationship between fire severity and burning efficiency for estimating wildfire emissions in Mediterranean forests

Bountouraby Balde1,a, Cristina Vega-Garcia1, Pere Joan Gelabert1, Aitor Ameztegui1, Marcos Rodrigues2   

  1. 1 Department of Agriculture and Forest Engineering, University of Lleida, , Av. Alcalde Rovira Roure, 191, 25198, Lleida, Spain
    2 Department of Geography and Land Management, University of Zaragoza, Pedro Cerbuna 12, 50009, Saragossa, Spain
  • Received:2022-06-14 Accepted:2022-11-24 Online:2024-10-16 Published:2024-10-16
  • Contact: Bountouraby Balde
  • Supported by:
    Universitat de Lleida

Abstract:

Forests are exposed to changing climatic conditions reflected by increasing drought and heat waves that increase the risk of wildfire ignition and spread. Climatic variables such as rain and wind as well as vegetation structure, land configuration and forest management practices are all factors that determine the burning potential of wildfires. The assessment of emissions released by vegetation combustion is essential for determining greenhouse gases and air pollutants. The estimation of wildfire-related emissions depends on factors such as the type and fraction of fuel (i.e., live biomass, ground litter, dead wood) consumed by the fire in a given area, termed the burning efficiency. Most approaches estimate live burning efficiency from optical remote sensing data. This study used a data-driven method to estimate live burning efficiency in a Mediterranean area. Burning severity estimations from Landsat imagery (dNBR), which relate to fuel consumption, and quantitative field data from three national forest inventory data were combined to establish the relationship between burning severity and live burning efficiency. Several proxies explored these relationships based on dNBR interval classes, as well as regression models. The correlation results between live burning efficiency and dNBR for conifers (R = 0.63) and broad-leaved vegetation (R = 0.95) indicated ways for improving emissions estimations. Median estimations by severity class (low, moderate-low, moderate-high, and high) are provided for conifers (0 .44 − 0.81) and broad-leaves (0.64 − 0.86), and regression models for the live fraction of the tree canopy susceptible to burning (< 2 cm, 2 − 7 cm, > 7 branches, and leaves). The live burning efficiency values by severity class were higher than previous studies.

Key words: Forest wildfires, Emissions, Greenhouse gases, Satellite images