National Geographic News
D3YT06 Amazon River, Iquitos, Maynas Province, Loreto Region, Peru

Experts have long argued over the origin of the Amazon River, seen here in the Loreto region of Peru.

Radius Images / Alamy

Jane J. Lee

National Geographic

Published February 13, 2014

The origin of the world's largest river—by volume—has been surprisingly hard to pin down. Explorers and scientists have argued over where to locate the start of the Amazon River since at least the mid-1600s, with no fewer than five rivers in southwestern Peru given the honor over the years.

Now the authors of a study published in the journal Area say they've located the mighty river's true source: the Mantaro River in southwestern Peru. If they're right, their discovery would add 47 to 57 miles (75 to 92 kilometers) to the length of the Amazon, currently measured at about 4,000 miles (6,437 kilometers) by the U.S. Geological Survey.

Using six different methods of measurement—including GPS tracking data and satellite images—professional kayaker James Contos and his team, funded in part by a grant from the National Geographic Society, determined that the Mantaro River is about 10 percent longer than the Apurímac River, which has been considered the Amazon's source since 1971.

One reason the Mantaro may have been overlooked, say Contos and his anthropologist co-author Nicholas Tripcevich of the University of California, Berkeley: A twisting bend, or kink, in the river's lower half makes it look much shorter than it really is.

Amazon Express expedition leader West Hansen paddles the Rio Mantaro arriving at Rancho Ocac Condorvado.
Amazon Express expedition leader West Hansen paddles the Mantaro River, a possible source of the Amazon.

But whether geographers accept the claim of Mantaro as the Amazon's true origin depends on which definition of a river's source they choose to apply: Is it the farthest point upstream that provides the largest volume of water, or is it the most distant point up the longest tributary in the river's drainage basin?

The current internationally accepted definition is the most distant point of a river's longest tributary that flows continuously. In their study, Contos and Tripcevich argue that the length of a tributary should trump whether it flows year-round.

Too Many Beginnings?

Throughout the 1700s, the Marañón River in northern Peru held the distinction of being the Amazon's source, says Andrew Johnston, a geographer with the Smithsonian's National Air and Space Museum in Washington, D.C. That's because the Marañón held the largest volume of flowing water compared to other tributaries flowing into the Amazon. The Ucayali River—a lower extension of the Apurímac River—took over as the designated source for awhile because it was considered the longest tributary flowing into the Amazon.

Then, in 1971, a National Geographic expedition led by Loren McIntyre identified the snow-capped peak of Mismi (map) as the headwaters of the Apurímac River and the ultimate source of the Amazon. A follow-up National Geographic expedition in 2000 that included Johnston, further confirmed Lake Ticlla Cocha at the base of Mismi as the headwaters and the Apurímac as the longest upstream extension of the Amazon River.

A map of the Amazon River.

A Question of Flow

The trouble with Contos and Tripcevich's claim of overturning this long-held view, say some geographers, is that the Mantaro River runs dry for about five months of the year, when the Tablachaca dam, built in 1974, diverts its flow around its kink, leaving the loop devoid of water.

Continuously flowing water is important in determining the true source of a river, says Juan Valdés, the geographer at National Geographic. "You can have the longest tributary, but if it doesn't have continuous flow, then it's a moot point." Mantaro does not have water flowing through its channels year-round, says Valdés. (See: "Photos: 'Alarming' Amazon Drought—River Hits New Low.")

But the study's authors propose that short-term changes in flow or man-made alterations to a river's natural course should not be a deal-breaker in determining the source of a river. The most distant source should be defined as the farthest point "from which a drop of rain will make its way to the river's mouth"—regardless of whether it flows continuously. By that measure, they assert, the Mantaro is the most distant source.

The Smithsonian's Johnston, even though he was on the expedition that confirmed the primacy of the Apurímac, thinks the new study provides a fresh perspective. "Assuming the measurements stand up to scrutiny, yes, I think the Mantaro could be considered a new source of the Amazon," he says, but not "the source."

When water flows through the Mantaro, the river "probably has a greater flow distance than any other Amazon tributary," Johnston says. But when the Mantaro is dry, "the Apurímac is once again the place where water flows the greatest distance into the Amazon."

Quatro Canyones and the Apurímac River, in the Andes, Peru, South America
The Apurímac River is the longest tributary flowing into the Amazon when the Mantaro is dry.

It may be more straightforward to trace the source of other rivers, Johnston notes, but "the Amazon, by far the largest river in the world, is not so simple." Scientists need to look at all three of its proposed source tributaries—the Marañón, the Apurímac, and the Mantaro—he says, to truly understand the mighty Amazon.

Follow Jane J. Lee on Twitter.

Jan  Zert
Jan Zert

With all respect to all scientist and National Geographic 

- The very first theory and research about the headwaters of the Apurímac River and the ultimate source of the Amazon was done by czech Prof. Dr. Bohumir Jansky and his peruvian colleague Prof. Dr. Carlos Penaherrera del Aguila in year 1969 - two years before National Geographic expedition led by Loren McIntyre! 

-In 1995 Jansky conducted another expedition of Faculty of Science from Charles University in Prague "Peru 95", which also visited Carhuasanta river valley at the foot of Nevado Mismi, and made an initial examination of the terrain of the area. 

-During the expedition "Hatun Mayu 99" (16.06 - 03.07 1999) was measured the length of the river Carhuasanta from the confluence with the river Apacheta. Using GPS were located two springs of Carhuasanta river at the foot of Nevado Mismi: left spring flowing from the glacial lake "Laguna" and the right spring, which rises from the rock crevices and at the end of June 1999 had a yield of 20 l / s (The most of all sources). 

-During the Czech-Peruvian expedition "Hatun Mayu 2000 'were using a laser rangefinder LEM TM 30 measured the length of three other headwater streams of the river Lloqueta - Sillanque, Apacheta and Ccaccansy. Using GPS the springs were located and determined the altitude of their sources. Using hydrometric propeller C 20 companies Ott were regularly measured flow of these rivers. 

-All the field work was finished by 10. of July, 2000. On 11. of July 2000  at the bottom of the river valley Lloqueta was expedition of Mr.Jansky visited by protagonists of the National Geographic Society expedition - Mr Andrew Pietowski and Andrew Johnston, who came to the site by helicopter and who just got to know about success of the Czech - Peruvian expedition from the daily press in Lima and Arequipa! Sounds funny no?

-Although both groups agreed that each will contribute their results for common work, Pietowski and Johnston later didn't hesitate to announced that the discovery and the priority accurate assessment of the sources belong to them. This fundamentally reject a claim that the first assessment of the lengths of streams and springs were localizated by Czech expedition "Hatun Mayu 99" and Czech-Peruvian expedition "Hatun Mayu 2000". 

- Johnston and Pietowski apologized to Jansky for their unfair behavior after ten years!!!on conference in Washington DC - but National Geographic instead of putting shame on them is  still using their names - in this article just name of Mr. McIntyre and Mr. Johnston... 

Can Jane J. Lee pls verified my words - call to Mr. Johnston and ask him for a public responsibility before due to global warming some other ice caps will melt down and Amazon river gonna gets a new more distant point of its beginning. Because this article is not part of the true about searching for sources of Amazon river!    

Expresses doubts over setting the spring at the foot of Nevado Mismi at the height of 5509 m at this altitude during the period of June, the peak of the winter season in the southern hemisphere, taken there is no permanent water flow, but only long-term snow cover. In another part of the spring can not provide because they are due to snow melting occurs even number of periodic flows. 

When determining the main stream in the hydrography - the criteria are used in the following order even in US, or Poland: the length of a steady flow along with the altitude source, catchment area, flow.

National Geographic Society expedition failed to identify the main source account or a catchment area or river flows. 

And the results of Hatun Mayu expeditions can explain a lot even for this article:

The longest river is permanent Ccaccansa, which also has the highest altitude of the spring, but of all headwater streams has the smallest flow. 

Farthest point from the mouth of the Amazon River Basin in the Atlantic is top of the mountain Nevado Mismi from which water flows into the basin Carhuasanta. Carhuasanta also has the largest catchment area of all tributaries in the area. 

The largest flow before the confluence has Apacheta river.

The source is not just one!

Thank You for reading!

Sebastian Muglia
Sebastian Muglia

Where tha Amazon river begins? What a question!  "In his spring!"

West Hansen
West Hansen

Good questions Rafael. Here's the answer for this specific article: the farthest source of the Amazon River from which a drop of rain will travel down a consistent stream bed to the mouth of the river is above Lago Acucocha north of Lake Junin, which spills into the Mantaro River stream bed. There are stream beds with larger volumes of water, bu the Mantaro is the longest.

Rafael Madrigal
Rafael Madrigal

it's over 500 rivers that make the Amazon river so where does the first drop of water begins?  Scientists need to get together to put this debate to rest.  Does it matter?  Not really.  But, ii will be nice to know.  And a good question on a georgraphy test.

West Hansen
West Hansen

Actually, the source is at Lago Acucocha, which is about 40 or so miles further than Lake Junin.

George Stiller
George Stiller

If this article is correct, then William Lewis Herndon's 1851 exploration of Amazon Valley was correct. Wikipedia lists the source of the Mantaro River as lake Chinchayqucha,(also known as Lake Junin). My Google Map of Herndon's exploration indicates he passed Lake Junin on July 3, 1851 and followed  the river north from there, reaching the Rio Huallaga which became the Amazon River.

Check it out at:

Rocky Contos
Rocky Contos

I checked out what Herndon did and published. Although he passed through the Lago Junin area (and Cerro de Pasco), he DID NOT MAKE ANY DISCOVERY with regard to the source of the Amazon, nor did he think the source of the river to be on Río Mantaro.  So I don't understand the comment that he was "correct". Herndon was surveying the area of the Amazon drainage, and was supposed to cover the area from the "source" down. Through the Lago Junin area he was trying to get close to the "source", which he sates in his book is at Lago Lauricocha up the Río Marañon. He happened to traverse through the Andes on the common route at the time, which went through Cerro de Pasco by the Río Mantaro. I append direct quotations from his book below to support my statements: 

"I felt that, under my instructions requiring me to explore the Amazon from its source to its mouth, I could not neglect the route I finally determined to take. This route would enable me to form a judgment respecting the practicability of a transitable connexion between Lima and the navigable headwaters of the tributaries of the Amazon — would lead me through the richest and most productive mineral district of Peru — would put under my observation nearly all the course of the Amazon — and would enable me to gather information regarding the Pampa del Sacramento, or great plain, shut in between four great rivers, and concerning which the " Viagero Universal " says " that the two continents of America do not contain another country so favorably situated, or so fertile."

"The last and most commonly-used route to the Montana is through

the cities of Truxillo, Caxamarca, Chachapoyas, Moyobamba, &c. The Andes here break into many chains, sending off spurs in all directions, but none of any great height, so that there is a tolerably good mule road all the way to Moyobamba; and almost all articles of foreign manufacture — such as cloths and the necessary household articles used in the small towns that border the Huallaga and the Marafion — are  supplied by this route. The climate and productions of this country are, on account of its precipitous elevations, and, consequently, deep valleys, very various ; and here the sugar-cane and the pine-apple may be seen growing by a spectator standing in the barley field and the potato patch."

"This route crosses the Amazon, or rather the Maranon, where, according to Lieut. Maw, it is sixty yards wide, and rushes between mount ains whose summits are hid in the clouds. This point is about three degrees north of its source, in Lake Lauricocha ; but the river is no where navigable until Tomependa, in the province of Jaen de Braca Moros, is reached ; whence it may be descended, but with great peril and difficulty, on rafts. There are twenty-seven " pongos," or rapids, to pass, and the water rushes over these with frightful velocity. Four days of such navigation passes the last, called the Pongo de "Manse- riehi" near the village of San Borja, and I am satisfied that an un broken channel, of at least eighteen feet in depth, may be found thence to the Atlantic Ocean."

George Stiller
George Stiller

I forgot to mention that there is another aspect to Herndon's expedition that is not plotted in my map. It had been plotted but had to be removed due to memory problems that made it hard to load. Herndon's partner in the expedition was Lardemer Gibbon who led part of the expedition south from Lake Junin along the river that eventually ran into other rivers that came out into the Amazon River. You can read about it in the eBook linked to the map. They both worked for the US Navy and were to determine if there was a navigatable route up the Amazon to the silver mines in Peru.

They both reached the Amazon by different routes. The only thing common was they split at Lake Junin and met again on the Amazon River.


Popular Stories

  • Lost City Found in Honduras

    Lost City Found in Honduras

    A joint Honduran-American expedition has confirmed the presence of extensive pre-Columbian ruins in Mosquitia in eastern Honduras, a region rumored to contain ruins of a lost "White City" or "City of the Monkey God."

  • Astronomers Find a Galaxy That Shouldn't Exist

    Astronomers Find a Galaxy That Shouldn't Exist

    Small, young galaxies should be free of interstellar dust, but an object called A1689-zD1 is breaking all the rules.

  • Cool Polar Bear Pictures

    Cool Polar Bear Pictures

    Take a peek at polar bears playing, swimming, and sleeping in their changing habitat.

The Future of Food

  • Why Food Matters

    Why Food Matters

    How do we feed nine billion people by 2050, and how do we do so sustainably?

  • Download: Free iPad App

    Download: Free iPad App

    We've made our magazine's best stories about the future of food available in a free iPad app.

See more food news, photos, and videos »