Alamillo et al. (2025) Post-Fire Vegetation Recovery Response: A Case Study of the 2020 Bobcat Fire in Los Angeles, California

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-12-13
• Authors: Andrew Alamillo, Jingjing Li, Alireza Farahmand, Madeleine Pascolini‐Campbell, Christine Lee
• DOI: 10.3390/rs17244023

Research Groups

• Department of Geography, Geology, and Environment, California State University, Los Angeles, USA
• NASA Jet Propulsion Laboratory, California Institute of Technology, USA

Short Summary

This study assessed post-fire vegetation recovery following the 2020 Bobcat Fire in Los Angeles, California, by analyzing changes in evapotranspiration (ET) and Normalized Difference Vegetation Index (NDVI) across different burn severity levels and vegetation classes. It found that while ET and NDVI generally increased, reflecting partial functional recovery, most burned forest and shrub areas have shifted to grassland, indicating a potential ecosystem shift.

Objective

• To assess changes in vegetation and understand how hydrological components in wildfire-affected areas contribute to potential vegetation shifts after the 2020 Bobcat Fire.
• To understand the variation in vegetation growth between wet and dry seasons within each burn severity class and National Land Cover Database (NLCD) vegetation class.
• To expand the study period to 2016–2024, including pre-fire years for comparison and additional post-fire recovery years.

Study Configuration

• Spatial Scale: The 2020 Bobcat Fire boundary in the San Gabriel Mountains, Los Angeles, California, covering approximately 468.61 square kilometers. Data products were analyzed at a 30 meter by 30 meter spatial resolution.
• Temporal Scale: Monthly and seasonal analysis from January 2016 to December 2024, encompassing pre-fire (2016–2019) and post-fire (2020–2024) periods.

Methodology and Data

• Models used: OpenET ensemble product (Alexi/DisAlexi, eeMETRIC, geeSEBAL, PT-JPL, SIMS, SSEBop) for evapotranspiration estimation.
• Data sources:
  • Satellite: Landsat 8 (Optical Land Imager (OLI) and Thermal Infrared Sensors (TIRS)) for Normalized Difference Vegetation Index (NDVI) and Difference Normalized Burn Ratio (dNBR). OpenET also uses data from GOES, Sentinel-2, Suomi NPP, Terra, Aqua.
  • Observation/Reanalysis: Google Earth Engine (GEE) and Python 3.10.18 for data access and visualization. National Land Cover Database (NLCD) for vegetation classes. Monitoring Trends of Burn Severity (MTBS) for fire boundary. United Nations Office for Outer Space Affairs UN-Spider Knowledge Portal for dNBR thresholds.

Main Results

• Burn Severity Area Changes: By the end of 2024, High Severity (HS) areas dropped from 15% to 0%, Moderate Severity (MS) areas from 44% to 10%, and Low Severity (LS) areas returned to 25% after peaking at 49% in May 2022. Unburned (UB) and Enhanced Regrowth (ER) areas increased to 52% and 13%, respectively.
• Vegetation Type Conversion: Pre-fire Evergreen Forest and Mixed Forest areas, especially in HS and MS zones, predominantly converted to Shrubs or Grasslands by 2024. Shrubs showed significant recovery in HS areas (82.2% re-classification to shrubs) but poor recovery in LS areas (15% re-classification to shrubs), with most LS shrub areas becoming Grasslands.
• NDVI Recovery: NDVI values increased across all burn severity levels post-fire. HS areas, despite the largest initial drop, showed the most rapid growth, reaching a peak of 0.6 after 4 years (from pre-fire 0.7). LS areas were the only severity level to match pre-fire NDVI conditions. Wet season median NDVI decreased from 2023 to 2024 across all burn severities and vegetation classes.
• Evapotranspiration (ET) Recovery: ET values increased across all burn severity levels post-fire. HS areas showed the most rapid increase, with peak summer ET reaching 146 millimeters in 2024, more than double its 2021 post-fire value of 67 millimeters. ET values for all three severity levels are approaching pre-fire levels after 4 years.
• NDVI and ET Correlation: A moderate positive and statistically significant Spearman correlation coefficient (0.43) was found between monthly NDVI and ET in HS areas post-fire, indicating a linkage between vegetation greenness and transpiration.
• Seasonal Trends: Mann–Kendall tests showed statistically significant increasing trends for seasonal median NDVI and ET post-fire across most burn severities and vegetation classes, except for wet season ET in Evergreen Forest, which showed no significant trend.

Contributions

• Provided a comprehensive assessment of post-fire vegetation recovery by simultaneously analyzing both structural (NDVI) and functional (ET) vegetation health using OpenET and Landsat 8 data.
• Expanded the study period to 2016–2024, offering a longer-term perspective on pre-fire conditions and post-fire recovery dynamics in Southern California's Mediterranean climate.
• Detailed the variation in vegetation growth between wet and dry seasons across different burn severity and NLCD vegetation classes, highlighting the influence of climatic conditions on recovery.
• Demonstrated the accessibility and utility of analytical tools like Google Earth Engine and Python, along with publicly available satellite imagery, for wide-ranging wildfire vegetation studies.
• Identified a potential ecosystem shift from forests to grasslands and shrubs, providing crucial insights for targeted forest management strategies to support biodiversity and reduce future fire vulnerability.

Funding

• U.S. Department of Energy (Grant No. DE-SC0024604)
• NASA MOSAICS program (Grant No. 80NSSC24K1074)

Citation

@article{Alamillo2025PostFire,
  author = {Alamillo, Andrew and Li, Jingjing and Farahmand, Alireza and Pascolini‐Campbell, Madeleine and Lee, Christine},
  title = {Post-Fire Vegetation Recovery Response: A Case Study of the 2020 Bobcat Fire in Los Angeles, California},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs17244023},
  url = {https://doi.org/10.3390/rs17244023}
}