BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//foss4g-2026//speaker//QAZUXB
BEGIN:VTIMEZONE
TZID:JST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:JST
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-2026-JVFQTA@talks.osgeo.org
DTSTART;TZID=JST:20260902T160000
DTEND;TZID=JST:20260902T163000
DESCRIPTION:Hydrological ML requires costly upstream catchment aggregation.
  We present an efficient flow-accumulation-based method bypassing per-pixe
 l delineation\, achieving orders-of-magnitude speedups. Implemented in GRA
 SS and Python\, this open-source approach enables scalable\, high-resoluti
 on modeling\, demonstrated by a countrywide 90 m Random Forest nitrogen pr
 ediction.
DTSTAMP:20260605T004858Z
LOCATION:Conference Management Room1
SUMMARY:Efficient pixel-scale upstream covariate computation for environmen
 tal machine learning - Kajetan Chrapkiewicz
URL:https://talks.osgeo.org/foss4g-2026/talk/JVFQTA/
END:VEVENT
END:VCALENDAR