Global River Topology (GRIT): A bifurcating river hydrography

Hydraulic modeling 15.05.2025
  • Wortmann, M.,
  • Slater, L.,
  • Hawker, L.,
  • Liu, Y.,
  • Neal, J.,
  • Zhang, B.,
  • Schwenk, J.,
  • Allen, G.,
  • Ashworth, P.,
  • Boothroyd, R.,
  • Cloke,, H.,
  • Delorme, P.,
  • Gebrechorkos, S.,
  • Griffith, H.,
  • Leyland, J.,
  • McLelland, S.,
  • Nicholas, A.,
  • Sambrook-Smith, G.,
  • Vahidi, E.,
  • Parsons, D.,
  • Darby, S.

Researchers have created the first branching global river network that represents bifurcations, multi-threaded channels and canals. It has the potential to significantly improve flood modeling and management.

Accurately mapping river and stream networks is crucial to global and regional flood modeling. Most global river maps, however, overlook the complex, multi-threaded river systems that exist in reality.

River networks often feature multiple channels that split and rejoin (known as bifurcations), which can form through both natural processes and human intervention. These branching systems are important because they are often found in densely populated, flood-prone regions such as river basins and deltas, and they are crucial for understanding water movement across the Earth’s surface.

However, global river maps that exist today assume downstream run-off flows in a single channel. To overcome this, researchers have developed the Global RIver Topology (GRIT), the first global data set of river bifurcations. With a total length of 19.6m km and 67,495 bifurcations, GRIT has the potential to significantly improve applications in fields from hydrology to ecology and flood management.

The research: data input and validation

A common limitation of existing global river network maps is that they are based on outdated digital elevation models (DEMs). Since many of them were created, a newer, more accurate, DEM called Copernicus DEM has been made publicly available by the European Space Agency.

Fathom researchers used machine learning to remove forests and buildings from Copernicus DEM and released it as FABDEM, which has a 30m spatial resolution and was found in numerous independent studies to be the most accurate global terrain data set available.

For this project, the researchers used the ∼30 m FABDEM together with satellite data from the Global River Widths of Landsat (GRWL) database as the primary input data, and validated it using a global collection of river gauging stations, reference river networks and randomized manual inspections of high-resolution satellite imagery.

Learn more about FathomDEM+, Fathom’s best-in-class global terrain data

The results: The first global bifurcating river network

The result of the research project was the first globally consistent, bifurcating river network hydrography. The GRIT river network offers river maps that cover 19.6 m kilometres and comprise:

  • 30m raster layers showing flow accumulation, drainage direction, catchments and other derivatives. 
  • Vector maps showing river segments, reaches, nodes and catchments that give information on flow directions, widths and bifurcation points as well as main channels for rivers wider than 30 m (2.87 million km). 

What does this mean for flood modeling?

This new river network map has a variety of potential uses. GRIT is well-suited for data-driven or machine learning applications such as large-sample streamflow modeling and forecasting. 

A second application is more accurate global hydrodynamic modeling in regions with bifurcating river systems and deltas, which often tend to be densely populated and flood-prone. 

arrow-light-short-left
arrow light short right
Read more
Composite image comprising two graphs. The figure on the right shows the results from applying the EQD and TAILS methods to every GESLA record used in this study, showing the distributions of quantile probabilities of the selected thresholds. On the left are spatial plots of (a) the quantile probabilities of selected thresholds using the TAILS methods, and (b) the difference in the quantile probabilities of the selected thresholds between the TAILS and EQD approaches.
New Research paper

Automated tail-informed threshold selection for extreme coastal sea levels

A new method, TAILS, sharpens how we identify extreme events by focusing on the most critical data, giving clearer, more reliable estimates of future coastal hazards worldwide.

11/17/2025
Read more
BURGER rainfall intensity derived with regionalized SMEV parameters for a 1 hr, 200 years rainfall event. The gray rectangles correspond with 13 areas of interest which have been assessed individually in Section 3.2.3 to evaluate BURGER's similarity with GPEX.
Research paper

Introducing BURGER, Fathom’s new extreme rainfall dataset

  • Hoch, J.,
  • Probyn, I.,
  • Marra, F.,
  • Lucas, C.,
  • Bates, J.,
  • Cooper, A.,
  • Fowler, H.,
  • Hatchard, S.,
  • Lewis, E.,
  • Savage, J.,
  • Addor, N.,
  • Sampson, C.

New research by Fathom’s Senior Hydrologist, Dr Jannis Hoch, and other Fathom scientists describes Fathom’s new dataset, Bottom Up Regionalised Global Extreme Rainfall, or BURGER for short. It’s the first of its kind to be created from rainfall gauge data and the first to contain extreme estimates for durations as short as an hour.

10/13/2025
Read more
Satellite to capture water propogation of Earth's river
Research paper

Satellite Requirements to Capture Water Propagation in Earth’s Rivers

  • Cerbelaud, A.,
  • David, C.,
  • Pavelsky, T.,
  • Biancamaria, S.,
  • Garambois, P.,
  • Kittel, C.,
  • Papa, F.,
  • Bates, P.,
  • Tourian, M.,
  • Durand, M.,
  • Frasson, R.,
  • Oubanas, H.,
  • Simard, M.,
  • Schumann, G.,
  • Allen, G.,
  • Tarpanelli, A.,
  • Wongchuig, S.,
  • Wade, J.,
  • Tom, M.,
  • Malaterre, P.,
  • Crétaux, J.,
  • Gleason, C.,
  • Nielsen, K.,
  • Elmi, O.,
  • Schwatke, C.,
  • Scherer, D.,
  • Gal, L.,
  • Paris, A.,
  • Dasgupta, A.,
  • Munier, S.,
  • Feng, D.,
  • Bauer-Gottwein, P.,
  • Hossain, F.,
  • Yamazaki, D.,
  • Kitambo, B.,
  • Mischel, S.,
  • Bonnema, M.,
  • Verma, K.,
  • Andreadis, K.,
  • Picot, N.,
  • Benveniste, J.,
  • Dourand, M.

07/22/2025
Read more
Research paper

Global River Topology (GRIT): A bifurcating river hydrography

  • Wortmann, M.,
  • Slater, L.,
  • Hawker, L.,
  • Liu, Y.,
  • Neal, J.,
  • Zhang, B.,
  • Schwenk, J.,
  • Allen, G.,
  • Ashworth, P.,
  • Boothroyd, R.,
  • Cloke,, H.,
  • Delorme, P.,
  • Gebrechorkos, S.,
  • Griffith, H.,
  • Leyland, J.,
  • McLelland, S.,
  • Nicholas, A.,
  • Sambrook-Smith, G.,
  • Vahidi, E.,
  • Parsons, D.,
  • Darby, S.

Researchers have created the first branching global river network that represents bifurcations, multi-threaded channels and canals.

05/15/2025
Read more
FathomDEM AI tiles terrain
Research paper

FathomDEM Global Terrain Map

A new research paper published in Environmental Research Letters outlines the creation of FathomDEM, a new global 30 m DEM produced using a novel machine-learning approach.

02/11/2025
Read more
Research paper

Increasingly seasonal jet stream raises risk of co-occurring flooding and extreme wind in Great Britain

  • Hillier, J.,
  • Bloomfield, H.,
  • Manning, C.,
  • Garry, F.,
  • Shaffrey, L.,
  • Bates, P.,
  • Kumar, D.

New research examines the role of the North Atlantic jet stream in driving co-occurring weather events. This type of multi-hazard risk is difficult to quantify for insurers, and – as this study shows – promises to worsen as our climate changes.

02/03/2025