Agricultural methane and nitrous oxide emissions represent around 10–12% of total anthropogenic GHG emissions and have a key role to play in achieving a 1.5 °C (above pre-industrial) climate stabilization target. Using a multi-model assessment approach, we quantify the potential contribution of agriculture to the 1.5 °C target and decompose the mitigation potential by emission source, region and mitigation mechanism. The results show that the livestock sector will be vital to achieve emission reductions consistent with the 1.5 °C target mainly through emission-reducing technologies or structural changes. Agriculture may contribute emission reductions of 0.8–1.4 Gt of CO₂-equivalent (CO₂e) yr⁻¹ at just US$20 per tCO₂e in 2050. Combined with dietary changes, emission reductions can be increased to 1.7–1.8 GtCO₂e yr⁻¹. At carbon prices compatible with the 1.5 °C target, agriculture could even provide average emission savings of 3.9 GtCO2e yr⁻¹ in 2050, which represents around 8% of current GHG emissions.

Despite its frequent use in policy discussions on future agricultural production, both the concept of the yield gap and its determinants are understood differently by economists and agronomists. This study provides a micro-level framework that disentangles and integrates agronomic and economic approaches to yield gap measurement. It decomposes the conventional yield gap indicator into four components that together provide a better understanding of why actual farm yield falls below potential: (1) the technical efficiency yield gap, (2) the allocative yield gap, (3) the economic yield gap and (4) the technology yield gap. The results can be used to inform targeted policy and farming recommendations at plot, farm household, local and national level. The framework is operationalised and tested by combining results from crop models with detailed farm and plot level survey data for maize production in Tanzania