homehome Home chatchat Notifications


New Global River Map Is the First to Include River Bifurcations and Canals

GRIT provides a much more detailed look at how rivers merge and split, which could enhance hydrological modeling, flood forecasting, and water management efforts.

Rebecca Owen
May 16, 2025 @ 12:58 pm

share Share

Global River Topology (GRIT) is the first global dataset to map branching river systems, such as the Mekong, seen here. Credit: ESA, CC BY-SA 3.0 IGO

Global river datasets represent rivers that flow downstream in single paths that follow surface elevation, but they often miss branching river systems found in areas such as floodplains, canals, and deltas. Forked, or bifurcated, rivers also often exist in densely populated areas, so mapping them at scale is crucial as climate change makes flooding more severe.

Wortmann et al. aimed to fill the gaps in existing global river maps with their new Global River Topology (GRIT) network, the first branching global river network that includes bifurcations, multithreaded channels, river distributaries, and large canals. GRIT uses a new digital elevation model with improved horizonal resolution of 30 meters, 3 times finer than the resolution of previous datasets, and incorporates high-resolution satellite imagery.

The GRIT network focuses on waterways with drainage areas greater than 50 square kilometers and bifurcations on rivers wider than 30 meters. GRIT consists of both vector maps, which use vertices and pathways to display features such as river segments and catchment boundaries, and raster layers, which are made up of pixels and capture continuously varying information, such as flow accumulation and height above the river.

In total, the effort maps approximately 19.6 million kilometers of waterways, including 818,000 confluences, 67,000 bifurcations, and 31,000 outlets—6,500 of which flow into closed basins. Most of the mapped bifurcations are on inland rivers, with nearly 30,000 in Asia, more than 12,000 in North and Central America, nearly 10,000 in South America, and nearly 4,000 in Europe.

GRIT provides a more precise and comprehensive view of the shape and connectivity of river systems than did previous reference datasets, the authors say, offering potential to improve hydrological and riverine habitat modeling, flood forecasting, and water management efforts globally. (Water Resources Researchhttps://doi.org/10.1029/2024WR038308, 2025)

This article originally appeared in EOS Magazine.

share Share

Science Just Debunked the 'Guns Don’t Kill People' Argument Again. This Time, It's Kids

Guns are the leading cause of death of kids and teens.

It Looks Like a Ruby But This Is Actually the Rarest Kind of Diamond on Earth

One of Earth’s rarest gems finally reveals its secrets at the Smithsonian.

ChatGPT Got Destroyed in Chess by a 1970s Atari Console. But Should You Be Surprised?

ChatGPT’s chess skills falter against a 46-year-old video game in a quirky AI test.

This Self-Assembling Living Worm Tower Might Be the Most Bizarre Escape Machine

The worm tower behaves like a superorganism.

Big Tech Said It Was Impossible to Create an AI Based on Ethically Sourced Data. These Researchers Proved Them Wrong

A massive AI breakthrough built entirely on public domain and open-licensed data

This Is How the Wheel May Have Been Invented 6,000 Years Ago

The wheel may have a more surprising origin story than you'd think.

So, Where Is The Center of the Universe?

About a century ago, scientists were struggling to reconcile what seemed a contradiction in Albert Einstein’s theory of general relativity. Published in 1915, and already widely accepted worldwide by physicists and mathematicians, the theory assumed the universe was static – unchanging, unmoving and immutable. In short, Einstein believed the size and shape of the universe […]

Dehorning Rhinos Looks Brutal But It’s Slashing Poaching Rates by 78 Percent

Removing rhino horns drastically cuts poaching, new study reveals.

A Chemical Found in Acne Medication Might Help Humans Regrow Limbs Like Salamanders

The amphibian blueprint for regeneration may already be written in our own DNA.

Everyone Thought ChatGPT Used 10 Times More Energy Than Google. Turns Out That’s Not True

Sam Altman revealed GPT-4o uses around 0.3 watthours of energy per query.