Measurements of spectrally resolved outgoing longwave radiation recorded in 1970, 1997 and 2003 are compared to determine the change in radiative forcing over that period. The changes are shown to be in agreements with that simulated by MODTRAN, a band model, using the known changes in atmospheric temperature and greenhouse gas concentrations when the effects of noise in the observed spectra are considered. The only region where the simulations are unable to reproduce the observations is in the v4 band of methane around 1306cm-1. The methane profiles used to simulate this region of the spectrum are shown to be in good agreement with all available data and the noise levels on the spectra are small. Therefore, it is proposed that the inability to model this region lies in the model formulation. Genln2, a line-by-line model, is shown to give very different results in this particular band to those obtained using MODTRAN. Sensitivity studies show that Genln2 is also not able to fully reproduce the spectrum observed. Errors in the spectroscopic parameters are shown to be smaller than the observed discrepancy and line mixing in methane is suggested as a possible cause of the discrepancy.
Rose publication type: Journal article
Sponsorship: This work was supported by a NERC studentship.
Author final version (often known as postprint) , 299 KB, PDF-document
Author final version (often known as postprint) , 4 KB, text/plain
Author final version (often known as postprint) , 38 KB, text/plain
- outgoing longwave radiation, spectrum, greenhouse gases, simulation, infrared, methane, HITRAN