A microcosm or closed incubation is a sealed bottle with known liquid and gas headspace volumes. Because gases partition between both phases, the headspace gas percentage alone does not give the total amount in the bottle.

Here, GC peak areas or measured gas percentages are converted to gas amounts using calibration, pressure, temperature, bottle volumes and gas solubility. pH can be included for CO₂ and H₂S speciation. NaCl-equivalent concentration is optional.

What this tool does

From GC peak area to bottle amount

Example: convert a CO peak area into headspace %, then calculate the amount present in the headspace, liquid and complete bottle.

From a target concentration to a gas dose

Example: calculate how much O₂ to add to reach 150 µM dissolved O₂ at the entered bottle conditions.

Excel batch analysis

One row per bottle and time point; gas columns contain GC peak areas. Supported gas headers are recognized automatically, and blank gas cells are ignored.

Example measurement input

experiment_idsample_idtime_h pressure_bar_abstemperature_C bottle_volume_mLliquid_volume_mL salinity_g_L_NaClpH COCO2CH4O2 H2SN2ONON2C2H6
WK3_example1_CO3.63331.58325 2312039.5607.27 829.056274.382.40
WK3_example1_CO16.16671.54525 2312039.0607.27 408.20455.307.50

Recognized gas columns: CO, CO2, CH4, O2, H2S, N2O, NO, N2 and C2H6.

Example calibration input

gas_idcalibration_idgas_percentpeak_area
COCO_0.935_rep10.93541100.6
COCO_4.516_rep14.51637530.0

Single calculation

Gas dosing

Calculate the amount of gas required to reach a target dissolved concentration after gas-liquid equilibration. NaCl-equivalent concentration is optional.

Calculations

This page describes how a GC response or measured headspace gas percentage is converted into a gas amount for the complete bottle. The same equations are used for single calculations, batch analysis and gas dosing.

Res ipsa loquitur.

GC peak areagas % partial pressureheadspace amount dissolved amounttotal bottle amount

Parameters and units

ParameterSymbolMeaning
GC peak areaAInstrument response measured for a standard or sample
Gas concentrationxPercentage of the headspace made up by the selected gas
Gas fractionyiGas concentration expressed as a fraction from 0 to 1
Absolute pressurePabsTotal pressure in the bottle
Partial pressurepiPressure contributed by the selected gas
Bottle volumeVbottleTotal internal bottle volume
Liquid volumeVliquidVolume occupied by liquid
Headspace volumeVheadspaceGas-filled bottle volume
TemperatureTBottle temperature when pressure and gas composition apply
Compressibility factorZFixed at 1 in v0.1
Gas constantR8.314462618 J mol-1 K-1
Henry solubilityHscpDissolved concentration per gas partial pressure
Gas amountnAmount of gas; results are reported in mmol

Internal calculation units are Pa, m³, K and mol.

1. GC peak area to gas concentration

For a multipoint calibration, every calibration observation is retained, including replicates. A straight line is fitted:

A=mx+b

The sample gas concentration is then:

xsample=Asample-bm

The calibration summary reports slope, intercept, R², number of observations and calibration range. Samples outside that range are still calculated but are flagged as extrapolations.

For a multipoint calibration, the normal model includes an intercept. If that unconstrained fit would give a negative intercept - and therefore a physically impossible negative predicted peak area at 0% gas - the intercept is constrained to 0 and the slope is refitted by least squares through the origin.

With only one non-zero standard, a proportional calibration through zero is used because slope and intercept cannot both be estimated.

2. Gas concentration to partial pressure

yi=xi100
Pabs[Pa]=Pabs[bar]×105
pi=yiPabs

Pressure is absolute pressure. Water vapour is neglected in v0.1, so the entered pressure is treated as the pressure of the modelled dry gas mixture. Because partial pressure changes with gas fraction and bottle pressure, it is reported per gas and time point in Results rather than as a static metadata value.

3. Headspace amount

Vheadspace=Vbottle-Vliquid
T[K]=T[°C]+273.15
nheadspace=piVheadspaceZRT

v0.1 uses Z = 1 and is intended for normal incubation pressures.

4. Temperature-dependent Henry solubility

The Henry convention used here is:

Hscp=cp

with units mol m-3 Pa-1.

Reference values are stored at 298.15 K and corrected to the bottle temperature using Sander (2023):

Hscp(T)=Hscp,*exp[B(1T-1298.15)]

Stored temperature-range metadata are used for warnings where a supporting range is available. For entries without such a range, the two-parameter correction should be regarded as a local temperature approximation.

Gas solubility constants used

GasHscp,* at 298.15 K B (K)Stored temperature range
CO9.70 × 10-61300278-323 K
CO₂3.40 × 10-42300-
CH₄1.40 × 10-51600275-328 K
NO1.90 × 10-51600-
N₂O2.40 × 10-42600-
H₂S1.00 × 10-32100-
C₂H₆1.90 × 10-52400275-328 K
N₂6.40 × 10-61300278.15-323.15 K
O₂1.30 × 10-51500274-328 K

All Henry solubility values, temperature coefficients and stored ranges are selected from Sander (2023).

5. Optional salting-out correction

NaCl-equivalent concentration is optional. If it is blank or 0, the pure-water Henry solubility is used. When a value is supplied, the Weisenberger-Schumpe salting-out correction is applied where a gas-specific parameter is available:

log10(cwatercsalt)=KscNaCl

This is an NaCl-equivalent approximation, not a full mixed-electrolyte model. CO has no parameter in this model, so CO is left uncorrected and flagged.

6. Dissolved amount

cdissolved=Hscp(T)pi
ndissolved=cdissolvedVliquid

This is the molecular dissolved gas amount. For CO₂ and H₂S, pH-dependent pools are calculated separately.

7. Optional pH-dependent speciation

CO₂ / carbonate

CO₂*H⁺+HCO₃⁻
HCO₃⁻H⁺+CO₃²⁻

At 25 °C, the implemented values are:

  • pKa1 = 6.352
  • pKa2 = 10.329

Both are recalculated for the sample temperature.

H₂S / sulfide

H₂SH⁺+HS⁻
HS⁻H⁺+S²⁻

At 25 °C, the implemented values are:

  • pKa1 ≈ 6.980
  • pKa2 = 14.00

H₂S, HS⁻ and S²⁻ are included. Around neutral pH the S²⁻ fraction is normally extremely small.

8. Total bottle amount

ntotal=nheadspace+nmolecular,dissolved

This definition is the same for every gas. Estimated DIC and estimated dissolved total sulfide are separate outputs.

CO₂: physical CO₂ and estimated DIC

CO₂ has two useful outputs and they are deliberately kept separate.

Physical CO₂

nCO₂,physical=nCO₂,headspace+nCO₂*,dissolved

This is the same physical total reported as total bottle amount.

Estimated dissolved inorganic carbon

DIC=CO₂*+HCO₃⁻+CO₃²⁻

Estimated DIC is only added when pH is available. It does not replace the physical CO₂ result.

Detailed carbonate species remain in Extended_data.

Gas dosing

The target dissolved concentration is converted back to the required gas partial pressure. For CO₂ and H₂S, the target can mean molecular dissolved gas or the estimated total dissolved pool. pH is required for DIC or dissolved total sulfide targets.

If the target is already reached, gas to add is reported as 0.

Current assumptions and scope

References

Henry solubility and temperature dependence
Sander, R. (2023). Compilation of Henry's law constants (version 5.0.0) for water as solvent. Atmospheric Chemistry and Physics, 23, 10901-12440. doi:10.5194/acp-23-10901-2023

Optional salting-out correction
Weisenberger, S. & Schumpe, A. (1996). Estimation of gas solubilities in salt solutions at temperatures from 273 K to 363 K. AIChE Journal, 42, 298-300. doi:10.1002/aic.690420130

Carbonate and sulfide speciation are documented here by the equations and pKa values implemented in the calculation.

Metrics & citation

(E)Gasboard is an open tool for gas calculations in closed incubations.

It runs in the browser and includes GC calibration, gas-liquid partitioning, Henry-law calculations, optional salting-out correction, CO₂/H₂S speciation, batch analysis, plots and Excel output.

Tool uses - this browser 0
Gas calculations - this browser 0
Love to connect? Bluesky @raegas.bsky.social

Site analytics count visits and completed actions. Uploaded files and calculation values are not sent.

Software output

Open research software with documented calculations and a public web interface.

Citation

(E)Gasboard, Reinier A. Egas, 2026, v0.1. DOI pending.

Requests, suggestions or ideas for improvements are very welcome - please reach out.