Greenhouse gas

A greenhouse gas absorbs and emits planetary thermal radiation and therefore contributes to the greenhouse effect. Its climatic impact depends on absorption spectrum, concentration, lifetime and chemical effects. Carbon dioxide, methane and water vapour are important examples.

The incremental radiative forcing per increment of a specified atmospheric gas burden. State whether the abundance unit is emitted mass, mole fraction or ppb, and specify the radiation-adjustment convention. For equal emitted masses in a fixed atmospheric air inventory, a per-ppb coefficient must be divided by the gas molar mass before comparing efficiencies. Spectral overlap with existing absorption makes efficiency depend on the background state.

Overlap of a gas absorption spectrum with existing atmospheric absorption changes its incremental radiative effect. A strong band in an already opaque spectral interval can contribute less additional outgoing-flux reduction than a weaker band in a window. Pressure broadening, altitude, temperature and the source spectrum also matter.

Global warming potential compares the time-integrated radiative forcing of equal-mass emission pulses of a greenhouse gas and carbon dioxide, over a stated time horizon $H$. It is not an instantaneous temperature ratio, and its numerical value depends on the horizon and included chemical and climatic responses.

For a gas pulse with unchanged greenhouse gas radiative efficiency and decay $e^{-t/\tau }$, its integrated forcing through $H$ is proportional to $\tau (1-e^{-H/\tau })$. Comparing two such lifetimes at fixed carbon-dioxide reference gives the ratio of these two factors. A fractional lifetime increase is not generally the same fractional GWP increase at finite $H$.

A global warming potential compares equal-mass pulses. If ${ϵ}_i$ is the greenhouse gas radiative efficiency per ppb, an emitted mass produces a mixing-ratio increment inversely proportional to molar mass $M_i$. For two gases with exponential pulse removal, their GWP ratio at horizon $H$ is $({ϵ}_i/{ϵ}_j)(M_j/M_i){\tau }_i(1-e^{-H/{\tau }_i})/[{\tau }_j(1-e^{-H/{\tau }_j})]$. The common carbon-dioxide reference integral cancels. Comparing only the per-ppb coefficients would instead compare unequal emitted masses.