BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.earthmonitor.org//soil-health-now-2026//Y8VGBY
BEGIN:VTIMEZONE
TZID:Europe/Athens
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:EET
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:EEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-soil-health-now-2026-UZLZ3G@pretalx.earthmonitor.org
DTSTART;TZID=Europe/Athens:20261209T141500
DTEND;TZID=Europe/Athens:20261209T143000
DESCRIPTION:The main methodologies governing the voluntary carbon market (V
 CM) and Scope 3 accounting now allow the use of Digital Soil Mapping (DSM)
  for soil organic carbon (SOC) stock estimation. Verra’s VT0014 has quic
 kly become the industry standard for DSM-based SOC stock estimation and pr
 ovides step-by-step guidance on uncertainty estimation. This guidance is l
 argely based on Wadoux & Heuvelink (2023) and is applicable for a large cl
 ass of models. However\, the generality of the approach has the cost of su
 boptimality: uncertainty of the resulting SOC stock estimates is larger th
 an necessary\, which directly influences project economics. \n\nIn this wo
 rk\, we empirically study the uncertainty estimator from Wadoux & Heuvelin
 k (2023) and compare it to a Random Forest Residual Kriging (RFRK) approac
 h (Hengl et al.\, 2015\; Szatmári et al.\, 2024). We also include the cla
 ssical Horvitz-Thompson estimator (HTE) calculated from soil samples only.
  For a fair comparison\, we apply the Wadoux & Heuvelink (2023) estimator 
 on top of a Quantile Regression Forest (QRF) model that uses the same cali
 bration data\, covariates and hyperparameters as the Random Forest model f
 rom the RFRK approach. We benchmark all estimators across three agricultur
 al projects. Special attention is given to the uncertainty of the estimato
 rs across differing project sizes and sampling intensities.\n\nBoth DSM ap
 proaches are compliant with VT0014 and passed the model validation require
 ments with comparable validation metrics. RFRK consistently showed slightl
 y higher R² scores\, with empirical coverage probabilities closer to the 
 nominal coverage. Throughout the benchmark\, RFRK was consistently more pr
 ecise than HTE\, reducing uncertainty of the SOC stock estimate substantia
 lly across all scenarios. In contrast\, QRF improved over HTE only in the 
 largest-sample project\, while inflating the uncertainty in the other two.
  DSM uncertainty estimates showed diminishing returns as sample sizes incr
 eased\, and dropped sharply as project area increased. \n\nThese findings 
 confirm that a DSM-based SOC stock estimate can be substantially more prec
 ise than a classical soil sampling approach: RFRK roughly halved the uncer
 tainty of the estimate\, translating to reduced uncertainty deductions. Ho
 wever\, a DSM-based estimate can also inflate uncertainties\, despite bein
 g validated with state-of-the-art performance metrics\, as demonstrated by
  the QRF-Wadoux approach. The exact choice of the DSM model appears to gov
 ern precision more than the model accuracy in terms of R² score. Therefor
 e\, project proponents should carefully select the most suitable modelling
  approach for their project.
DTSTAMP:20260825T180502Z
LOCATION:Amphitheater I
SUMMARY:A comparison of DSM-based SOC stock estimators: Uncertainty reducti
 ons and practical insights - Erik Scharwächter\, Ahmad Awad
URL:https://pretalx.earthmonitor.org/soil-health-now-2026/talk/UZLZ3G/
END:VEVENT
END:VCALENDAR