Biomass burning emissions factors are vital to quantifying trace gas release from vegetation fires. Here we evaluate emissions factors for a series of savannah fires in Kruger National Park (KNP), South Africa using ground-based open path Fourier transform infrared (FTIR) spectroscopy and an IR source separated by 150–250 m distance. Molecular abundances along the extended open path are retrieved using a spectral forward model coupled to a non-linear least squares fitting approach. We demonstrate derivation of trace gas column amounts for horizontal paths transecting the width of the advected plume, and find for example that CO mixing ratio changes of ~0.01 μmol mol<sup>−1</sup> [10 ppbv] can be detected across the relatively long optical paths used here. Though FTIR spectroscopy can detect dozens of different chemical species present in vegetation fire smoke, we focus our analysis on five key combustion products released preferentially during the pyrolysis (CH<sub>2</sub>O), flaming (CO<sub>2</sub>) and smoldering (CO, CH<sub>4</sub>, NH<sub>3</sub>) processes. We demonstrate that well constrained emissions ratios for these gases to both CO<sub>2</sub> and CO can be derived for the backfire, headfire and residual smouldering combustion (RSC) stages of these savannah fires, from which stage-specific emission factors can then be calculated. Headfires and backfires often show similar emission ratios and emission factors, but those of the RSC stage can differ substantially. The timing of each fire stage was identified via airborne optical and thermal IR imagery and ground-observer reports, with the airborne IR imagery also used to derive estimates of fire radiative energy (FRE), allowing the relative amount of fuel burned in each stage to be calculated and "fire averaged" emission ratios and emission factors to be determined. These "fire averaged" metrics are dominated by the headfire contribution, since the FRE data indicate that the vast majority of the fuel is burned in this stage. Our fire averaged emission ratios and factors for CO<sub>2</sub> and CH<sub>4</sub> agree well with those from prior studies conducted in the same area using e.g. airborne plume sampling. We also concur with past suggestions that emission factors for formaldehyde in this environment appear substantially underestimated in widely used databases, but see no evidence to support suggestions by Sinha et al. (2003) of a major overestimation in the emission factor of ammonia in works such as Andreae and Merlet (2001) and Akagi et al. (2011). We also measure somewhat higher CO and NH<sub>3</sub> emission ratios and factors than are usually reported for this environment, which is interpreted to result from the OP-FTIR ground-based technique sampling a greater proportion of smoke from smouldering processes than is generally the case with methods such as airborne sampling. Finally, our results suggest that the contribution of burning animal (elephant) dung can be a significant factor in the emissions characteristics of certain KNP fires, and that the ability of remotely sensed fire temperatures to provide information useful in tailoring modified combustion efficiency (MCE) and emissions factor estimates maybe rather limited, at least until the generally available precision of such temperature estimates can be substantially improved. One limitation of the OP-FTIR method is its ability to sample only near-ground level smoke, which may limit application at more intense fires where the majority of smoke is released into a vertically rising convection column. Nevertheless, even in such cases the method potentially enables a much better assessment of the emissions contribution of the RSC stage than is typically conducted currently.