Condition-level data for "Vapor depletion is not a proxy for substrate inventory: paired vapor and residue measurement on porous field substrates and the limits of air-based clearance criteria"

Paired vapor-phase and substrate-inventory measurements for sulfur mustard (HD) and soman (GD) on glass, airfield concrete and standardized Korean bare soil, measured in a 5-cm wind tunnel reproducing the near-ground wind field at a 2-cm datum. This deposit contains every condition-level value underlying the figures, tables and statistics of the associated article, together with the code that transcribes, analyzes, plots and audits them: 27 CSV files and 14 Python scripts, including three independent audit scripts that check every statistic and every numeric token in the manuscript against these data. Provenance. The measurements were acquired under a Ministry of National Defense research program conducted between 2008 and 2014 and were documented in an internal Korean-language technical report that has not been published. The CSV files are a verbatim transcription of the numbered tables of that report: printed values were transferred as printed, with no rounding, smoothing or correction. Public release has been approved by the originating institution. Values for VX are not from this laboratory; they are published US Army Edgewood Chemical Biological Center (ECBC) 5-cm wind-tunnel data, used as a comparator and identified as ECBC wherever they appear. Known limitations, stated so that reusers are not misled: the inventory series are unreplicated (one tray per time point, one condition per substance) and fix the order of magnitude rather than a precise value; mass balance is open, with mean collection efficiency of 48% (HD) and 37% (GD) on bare soil; extraction recovery was validated on fresh spikes rather than on aged residue, so inventory percentages more likely understate than overstate what remains; the loss-curve inflection point was located visually rather than by an algorithmic breakpoint criterion, with a coefficient of variation of 30% among eight replicates at one condition; relative humidity was not controlled above 32 degrees Celsius; and reported concentrations are exhaust concentrations at a stated flow over a 10.75 square centimetre source, conditional on that geometry rather than free-field breathing-zone values. Data files are licensed CC BY 4.0; code is licensed MIT. See README.md in the deposit for the file-by-file map and the run order.