Rivers and streams contribute substantially to global emissions of carbon dioxide (CO₂) and methane (CH₄). The Amazon River plays a pivotal role in fluvial greenhouse emissions as it is the largest river on Earth in terms of freshwater discharge (6,600 cubic kilometers per year) and drainage basin (6,300,000 km2 which compares to the surface area of the USA of 9,834,000 km²). The Amazon river drains the largest rain forest on Earth which supplies large amounts of organic carbon to the rivers that are transformed by microbes into CO₂ and CH₄ and then emitted from surface waters to the atmosphere. As such, the Amazon River emits to the atmosphere large quantities of these two potent greenhouse gases.

The Amazon River discharges to the Atlantic Ocean on the Eastern side of South America, and its source (headwaters) is on the Western side bordered by the Andes Cordillera. Rock erosion in the Andes is the main source of mineral particles that are transported about 3,000 km across the South American continent to the mouth of the River in Belem (Brazil) and then discharged to the Atlantic Ocean.

Yet, all of the studies, so far, on the emissions to the atmosphere of CO₂ and CH₄ from the Amazon rivers and streams have carried out in the lowlands of Central Amazonia, at least 1,000 km away from the Andes Cordillera. However, mountainous streams have very different emission rates of CO₂ and CH₄ than lowland rivers.

If you picture yourself a mountainous stream, it should be small and fast flowing in a steep rocky terrain. This promotes a vigorous physical exchange of gases with the atmosphere. Conversely, the steep terrain does not allow a large accumulation of soils that sustain the production of CO₂ and CH₄.

If you picture yourself a lowland river, it should be large and meandering in a flat terrain. The more sluggish water flow does not promote as vigorously the physical exchange of gases with the atmosphere as in mountainous streams. But the higher temperature (lower elevation) allows the growth of more abundant vegetation (forests) and the flat terrain promotes the accumulation of thicker soils than in mountainous terrain. This should sustain more production and transport of CO₂ and CH₄ to the lowland rivers. Finally, the flat terrain promotes the occurrence of floodplains connected to lowland rivers that additionally subsidize rivers in CO₂ and CH₄.

There is a third type of river system located in the plains at the base of mountain chains, called piedmont streams. From a physical point of view they resemble lowland rivers, but they receive massive amounts of particles from the upstream mountainous rivers. These particles will settle temporarily and then be re-suspended and transported further downstream until they eventually reach the ocean. But when the particles settle as sediments, this promotes the production of CH₄ by fermentation. So piedmont rivers should in theory be CH₄ factories.

So, based on these theoretical considerations, we should expect emissions of CO₂ and CH₄ to be very different in mountainous streams, piedmont rivers and lowland rivers. As mentioned above, CO₂ and CH₄ emissions have only be measured so far in lowland rivers of Central Amazonia, so we were missing potentially important pieces of the jigsaw, which is critical for the largest river on Earth. This was addressed by our recent publication that reports data in the mountainous and piedmont rivers in Ecuador along a transect of elevation from 175 m to 3990 m above sea level. We find that Andean mountainous headwater and piedmont streams are hotspots of CO₂ and CH₄ emission, with flux intensities substantially higher than in lowland streams of Central Amazonia. This is substantial because together, Andean mountainous headwater and piedmont streams and rivers represent 35% CO₂ and 72% CH₄ of basin scale integrated fluvial emissions.