Abstract More than half of the world’s rivers experience occasional, seasonal, or permanent drying, and this may increase because of climate change. Drying, i.e. severe reduction in water flow even leading to streambed desiccation, can have a profound impact on the available aquatic habitat, biodiversity, and functions of rivers. Yet, to date, it is unclear whether similar drying events in comparable climate zones result in similar changes in ecosystem processes, such as river metabolism or greenhouse gas (GHG) emissions. Here, we synthesise the detected effects of drying on gross primary production (GPP) and ecosystem respiration (ER), as well as on the emissions of GHGs (CO2, CH4, and N2O) in rivers and streams. We examined the current available scientific literature detailing the impact of drying on these variables when measured either in the field or in the laboratory. We extracted data from 30 studies analysing GPP and ER responses, and data on GHG emissions from another 35 studies. Then, we conducted a meta-analysis to determine whether the magnitude and direction of the effects varied across the systems and climate zones studied, or according to the type (natural or human-induced) and severity of drying. In general, drying enhanced GPP (under low flows) and CH4 emissions, and decreased CO2 and N2O emissions. The hydrological phases throughout streambed drying (low water flow, isolated pools, or desiccation) had differential effects on metabolism and GHG emissions. The effects of drying were generally more severe when it induced desiccation, rather than just periods of low flow. Desiccation strongly reduced GPP, likely because of the die-off of algae, while its negative effect on ER was smaller. Greater decrease in GPP than in ER under desiccation would lead to increase in CO2 emissions; our results showed accordingly that desiccation increased CO2 emissions. Furthermore, the magnitude and direction of the effects varied depending on the study type. Experimental studies conducted in micro- and mesocosms demonstrated greater effects than field studies, thus the extrapolation of results from these to real conditions should be done with caution. Overall, the effects’ direction was inconsistent across climate zones, except for the Mediterranean climate zone, where drying was showing a negative effect on both metabolism and GHG emissions. Our synthesis may contribute to identifying the worldwide trends and patterns of drying on riverine functions associated to global change impacts on river and stream ecosystems.