TY - ABST

T1 - deadtrees.earth - an open, dynamic database for accessing, contributing, analyzing, and visualizing remote sensing-based tree mortality data

AU - Kattenborn, Teja

AU - Mosig, Clemens

AU - Pratima, KC

AU - Frey, Julian

AU - Perez-Priego, Oscar

AU - Schiefer, Felix

AU - Cheng, Yan

AU - Potts, Alastair

AU - Jehle, Janusch

AU - Mälicke, Mirko

AU - Mahecha, Miguel D.

PY - 2024/3/9

Y1 - 2024/3/9

N2 - Excessive tree mortality rates prevail in many regions of the world. Understanding tree mortality dynamics remains elusive as this multifaceted phenomenon is influenced by an interplay of abiotic and biotic factors including, but not limited to, global warming, climate extremes, pests, pathogens, and other environmental stressors. Earth observation satellites, coupled with machine learning, present a promising avenue to unravel map standing dead trees and lay the foundation for explaining the underlying dynamics.However, the lack of globally comprehensive, georeferenced training data spanning various biomes and forest types has hindered the development of a unified global product detailing tree mortality patterns. Present ground-based observations, e.g., sourced from national inventories, are often sparse, lack standardization, and spatial specificity. Alternatively, aerial imagery captured via drones or airplanes in concert with computer vision methods offers a potent resource for mapping standing deadwood with high precision and efficiency on local scales. Such products can subsequently be used to train models based on satellite data to infer standing deadwood on large spatial scales.In pursuit of harnessing this potential to enhance our global comprehension of tree mortality patterns, we initiated the development of a dynamic database (https://deadtrees.earth), which enables to 1) upload and download aerial imagery with optional labels on standing deadwood, 2) automatically detect (semantic segmentation) standing dead trees in uploaded aerial imagery through a generic detection computer vision model, 3) Visualization and download of extensive spatiotemporal tree mortality products derived from extrapolating standing deadwood using Earth observation data.This presentation provides an in-depth overview of the deadtrees.earth database, outlining its motivation, current status, and future perspectives. By integrating Earth observation, machine learning, and ground-based data sources, this initiative aims to bridge the existing gaps in understanding global tree mortality dynamics, fostering a comprehensive and accessible resource for researchers and stakeholders alike.

AB - Excessive tree mortality rates prevail in many regions of the world. Understanding tree mortality dynamics remains elusive as this multifaceted phenomenon is influenced by an interplay of abiotic and biotic factors including, but not limited to, global warming, climate extremes, pests, pathogens, and other environmental stressors. Earth observation satellites, coupled with machine learning, present a promising avenue to unravel map standing dead trees and lay the foundation for explaining the underlying dynamics.However, the lack of globally comprehensive, georeferenced training data spanning various biomes and forest types has hindered the development of a unified global product detailing tree mortality patterns. Present ground-based observations, e.g., sourced from national inventories, are often sparse, lack standardization, and spatial specificity. Alternatively, aerial imagery captured via drones or airplanes in concert with computer vision methods offers a potent resource for mapping standing deadwood with high precision and efficiency on local scales. Such products can subsequently be used to train models based on satellite data to infer standing deadwood on large spatial scales.In pursuit of harnessing this potential to enhance our global comprehension of tree mortality patterns, we initiated the development of a dynamic database (https://deadtrees.earth), which enables to 1) upload and download aerial imagery with optional labels on standing deadwood, 2) automatically detect (semantic segmentation) standing dead trees in uploaded aerial imagery through a generic detection computer vision model, 3) Visualization and download of extensive spatiotemporal tree mortality products derived from extrapolating standing deadwood using Earth observation data.This presentation provides an in-depth overview of the deadtrees.earth database, outlining its motivation, current status, and future perspectives. By integrating Earth observation, machine learning, and ground-based data sources, this initiative aims to bridge the existing gaps in understanding global tree mortality dynamics, fostering a comprehensive and accessible resource for researchers and stakeholders alike.

U2 - https://doi.org/10.5194/egusphere-egu24-15502

DO - https://doi.org/10.5194/egusphere-egu24-15502

M3 - Conference abstract for conference

T2 - EGU General Assembly 2024

Y2 - 15 April 2024 through 19 April 2024

ER -