BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.earthmonitor.org//global-workshop-2026//CTMG7H
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-QEER9E@pretalx.earthmonitor.org
DTSTART;TZID=Europe/Amsterdam:20261008T184500
DTEND;TZID=Europe/Amsterdam:20261008T190000
DESCRIPTION:Accurate crop field boundary delineation is foundational for ag
 ricultural mapping\, yield estimation\, and decision support systems. Yet 
 existing AI models\, trained predominantly on data from the Global North\,
  perform poorly in underrepresented farming systems such as those in Sub-S
 aharan Africa (typically under 2 hectares\, irregularly shaped) and South 
 America (characterized by shifting cultivation and complex morphologies). 
 This data gap misleads agricultural statistics\, weak policies\, and ineff
 icient resource allocation. \n\nWe leveraged AI and open-source remote sen
 sing data to automatically delineate field boundaries in both regions usin
 g transfer learning\, adapting pretrained global models to local contexts.
  In South America\, we annotated over 46\,000 field boundaries for model t
 raining and generated more than 10 million boundaries continent-wide. In E
 ast Africa's Great Rift Valley\, we automatically detected over 400\,000 f
 arms from just 6\,000 samples\, incorporating multi-stakeholder annotation
  workflows and quality assurance pipelines refined from lessons learned in
  South America. \n\nOur results show that models trained on limited but hi
 gh-quality local annotations scale effectively to out-of-sample regions. I
 n Africa\, delineated fields have enabled field level crop type and yield 
 data collection\, in preparation for field level crop type mapping\, yield
  estimation and monitoring of agroecological and regenerative agriculture 
 practices. In South America\, they have supported supply chain auditing fo
 r deforestation-free commitments\, EUDR compliance\, country-level crop fo
 recasting\, and scope 3 emissions estimation. Across both regions\, the ap
 proach has strengthened national and subnational agricultural data systems
  and climate resilience frameworks. \n\nBy demonstrating AI model transfer
 ability across contrasting geographies\, this work charts a pathway toward
  open\, inclusive\, and scalable Earth observation systems that close crit
 ical data gaps in the Global South\, positioning AI as a core enabler of s
 ustainable agricultural monitoring at national and subnational scales.
DTSTAMP:20260624T070020Z
LOCATION:Aula Magna
SUMMARY:Democratizing Field Boundary Delineation in the Global South with A
 I. - Christine Muthee\, Tristan Grupp
URL:https://pretalx.earthmonitor.org/global-workshop-2026/talk/QEER9E/
END:VEVENT
END:VCALENDAR