BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.earthmonitor.org//global-workshop-2026//LEM9GS
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-global-workshop-2026-CDLLLB@pretalx.earthmonitor.org
DTSTART;TZID=Europe/Amsterdam:20261007T160000
DTEND;TZID=Europe/Amsterdam:20261007T161500
DESCRIPTION:This study explores the application of high-dimensional embeddi
 ngs derived from Sentinel-2 imagery for automated anomaly detection in env
 ironmental monitoring. By utilizing the SSL4EO self-supervised learning fr
 amework\, we transform raw satellite data into compact\, informative repre
 sentations that capture essential spatial and temporal features. The entir
 e workflow is integrated within the Copernicus Data Space Ecosystem (CDSE)
 \, ensuring efficient access to large-scale\, analysis-ready archives and 
 enabling rapid processing of planetary-scale datasets. \n\nThe core of our
  approach lies in leveraging SSL4EO to bypass the need for massive labeled
  datasets\, creating a standardized intermediate format that bridges raw i
 magery with advanced AI tasks. These embeddings serve as the foundation fo
 r an anomaly detection pipeline designed to pinpoint deviations from expec
 ted seasonal or spatial trends\, such as flooding\, wildfires\, or shifts 
 in vegetation health. To ensure interoperability and ease of use in geospa
 tial analytics\, results are stored in the GeoParquet format\, which suppo
 rts both reproducibility and high-performance data handling. \n\nTo confir
 m the framework's robustness\, we conducted extensive validation across di
 verse geographical regions and seasonal cycles\, including challenging win
 ter conditions with snow cover and low solar illumination. The pipeline de
 monstrated high resilience\, producing consistent embeddings even in the p
 resence of partial cloud cover. Furthermore\, we evaluated the system’s 
 portability across heterogeneous computing environments. Testing on the CR
 EODIAS cloud platform (using both CPU and GPU nodes) alongside high-perfor
 mance computing (HPC) infrastructures like EOHPC and SpaceHPC proved that 
 the solution scales effectively and maintains functional integrity across 
 different hardware architectures. \n\nThe results indicate that combining 
 self-supervised embeddings with anomaly detection creates a powerful tool 
 for environmental intelligence. The proposed framework is suitable for a w
 ide range of operational applications\, from tracking urban growth to moni
 toring climate-induced environmental changes. By providing a portable and 
 scalable bridge between raw Copernicus data and actionable insights\, this
  study highlights the transformative potential of GeoAI in supporting glob
 al and regional decision-making processes.
DTSTAMP:20260624T065129Z
LOCATION:Aula Magna
SUMMARY:Satellite-Based Anomaly Detection using GeoAI Embeddings: A Scalabl
 e Workflow - Marcin Kluczek
URL:https://pretalx.earthmonitor.org/global-workshop-2026/talk/CDLLLB/
END:VEVENT
END:VCALENDAR