Tropical climate-smart agriculture could be a cost-effective and scalable nature-based solution toward achieving net-zero greenhouse gas (GHG) emissions in the coming decades. Nevertheless, the lack of integrated on-farm CO2, CH4, and N2O flux measurements remains a barrier to including these systems in certification programs and the carbon (C) market. This systematic literature review critically examines field-measured GHG emission studies within climate-smart agri-systems in Brazil, aiming to: (i) integrate GHG-measured data, (ii) explore geographical representativeness and connections between organizations and authors, (iii) analyze distributions of studied GHG molecules, climate-smart system types, and productive components, (iv) conduct a meta-analysis on the adoption of climate-smart agriculture, and (v) identify methodological and scope gaps, providing insights for future research directions. The systems considered include no-till, well-managed pasture, integrated crop-livestock, and integration with trees (i.e., crop-forestry, livestock-forestry, crop-livestock-forestry, and multistrata agroforestry). Out of 1200 papers mentioning GHGs and climate-smart systems, 74 contained reliable on-farm flux measurements. Most authors were Brazilians affiliated with Embrapa and public Universities. Studies concentrated on the Atlantic Forest and Cerrado, with scarce data for other biomes. Over half of the trials assessed individual GHG molecules and only one climate-smart system. Urochloa spp. and Avena spp. were the most studied forage genera, while soybeans and corn dominated annual crops. Eucalyptus spp. and beef cattle were common trees and animals. No-till, integrated crop-livestock, and agricultural integrations involving trees showed soil C–CH4 influx from 0.8 to 1.0 kg ha−1 y−1. Transitioning from pasture monoculture to integrated systems decreased N–N2O emissions by up to 1.63 kg ha−1 y−1. Converting extensive grazed land to intensified-integrated systems reduced animal CH4 intensity, with cuts of up to 122 g CH4 per kg of average daily weight gain. Future research must urgently address methodological issues, such as de-standardization (e.g., 8–10 flux units found per GHG molecule) and the absence of crucial ancillary variables. In conclusion, adopting diversified climate-smart agri-systems shows potential to mitigate GHG emissions, but caution is advised when extrapolating our numerical findings due to limited data across Brazilian edaphoclimatic conditions. We urge scientists to prioritize frequent on-farm measurements of all three GHGs (CO2, N2O, and CH4) along with agroecosystem C sequestration across multiple climate-smart systems, enabling reliable and comparable net GHG flux calculations to identify the most sustainable option. Advancing and standardizing GHG measurements in Brazil offer an opportunity to enhance GHGs inventory accuracy, calibrate models, and promote policies and investments that expand refined tropical climate-smart agriculture.
