Geologists broadly agree the modern Grand Canyon (the deep, river-integrated version we see today) is relatively “young” in geologic terms—primarily carved starting ~5–6 million years ago. Earlier ideas of a 70-million-year-old canyon (from the Laramide orogeny era) have largely been set aside in favor of this younger timeline.

The big question has always been the mechanism during that ~5-million-year gap (after the river existed upstream ~11 Ma but before it exited the canyon area ~5.6 Ma).

Proposed ideas included:

• Gradual headward erosion.
• Groundwater sapping/karst collapse.
• River capture.
• Or this lake-spillover model.

New research article published in Science on April 16, 2026, titled “Late Miocene Colorado River arrival in the Bidahochi basin supports spillover origin of Grand Canyon” by John J. Y. He and colleagues (including Ryan Crow of the USGS and others from UCLA, Arizona Geological Survey, and additional institutions).

The study provides new geochemical and sedimentological data showing that by ~6.6 million years ago (late Miocene), the ancestral Colorado River (already present upstream in western Colorado by ~11 Ma) was delivering water and sediment into the Bidahochi basin in northeastern Arizona (largely on Navajo Nation land, east/southeast of the modern Grand Canyon).

Key lines of evidence include:

Detrital zircon U-Pb geochronology: Thousands of zircon grains from upper Bidahochi Formation sandstones show a distinctive age “fingerprint” (notably a population of 40–25 Ma grains) that matches sediments from the upper Colorado–Green River system, including the Browns Park Formation. This signature appears abruptly around 6.6 Ma and is absent or rare in older deposits below.

Increased sedimentation rate: An order-of-magnitude jump in deposition at that time, consistent with a major river input.

Strontium isotopes (⁸⁷Sr/⁸⁶Sr): Shifts in carbonate ratios indicating a new water source.

Sedimentology and fossils: Rippled layers suggesting a strong river flowing into standing lake water; appearance of large fish species typical of fast-flowing river environments (not just a closed lake).

Together, these indicate the proto-Colorado River fed into a large paleolake (Lake Bidahochi or Hopi Lake) for hundreds of thousands to over a million years, gradually filling it.

Once the lake rose high enough, it overtopped a topographic barrier (associated with the Kaibab Arch/plateau) to the west. This spillover established (or accelerated) the river’s integration through the future Grand Canyon, with the river reaching its modern outlet to the Gulf of California around 5.6–5 Ma.

The authors argue this makes lake spillover the simplest and most likely primary mechanism for integrating the Colorado River across the Colorado Plateau and initiating major canyon incision—rather than purely headward erosion, groundwater sapping, or river capture alone. Other processes may have played supporting roles, but the new data strengthen the spillover model.

Context in the broader debate

The modern Grand Canyon (the deep, fully integrated gorge) is considered relatively young geologically (~5–6 Ma for its final integration phase). Earlier ideas of a much older (70+ Ma) canyon have been largely discounted.

A 5-million-year “gap” existed between the river’s upstream presence (11 Ma) and its downstream exit (~5.6 Ma). This study fills part of that gap by confirming the river reached the Bidahochi basin by 6.6 Ma.

The spillover idea isn’t entirely new (proposed decades ago), but it has been contested. This paper provides direct sedimentary “fingerprint” evidence that was previously missing, prompting a reevaluation.

Not all geologists are fully convinced—some note that hybrid models (spillover plus other erosional or tectonic factors) may still apply, and debates about exact timing, lake size, and overflow mechanics continue.

This is a solid incremental advance in understanding one of North America’s most iconic landscapes. The work relies heavily on field sampling from the Bidahochi Formation (with permissions on Navajo land) and high-precision lab analysis of mineral grains.

_____________________________________________________________________________________

Late Miocene Colorado River arrival in the Bidahochi basin supports spillover origin of Grand Canyon

Title: Late Miocene Colorado River arrival in the Bidahochi basin supports spillover origin of Grand Canyon

Authors: John J. Y. He et al. (lead author John J. Y. He; key co-authors include Ryan S. Crow of the USGS, with contributions from researchers at UCLA, Arizona Geological Survey, University of Washington, and others)

Journal: Science

Publication date: 16 April 2026

Volume/Issue: Vol. 392, Issue 6795

Pages: 289–295

DOI: DOI: 10.1126/science.adz6826

This is the full research article (not a perspective or news piece) that appeared on the cover of that issue of Science. It presents new detrital zircon U-Pb geochronology data from the Bidahochi Formation, along with supporting sedimentological, isotopic (⁸⁷Sr/⁸⁶Sr), depositional rate, and paleontological evidence.

Quick recap of the core contribution

• The ancestral Colorado River (already present upstream by 11 Ma) began delivering its distinctive sediment (with a clear “fingerprint” matching the Browns Park Formation) into the Bidahochi paleolake basin by **6.6 Ma**.
• This arrival coincides with:
  • A sharp increase in sedimentation rate (order-of-magnitude jump).
  • Shifts in strontium isotopes.
  • Sediment structures indicating a strong river inflow into standing lake water.
  • Appearance of large river-adapted fish fossils.
• The lake filled over time and eventually spilled westward over the Kaibab Arch/plateau, integrating the river through the future Grand Canyon and reaching its outlet to the Gulf of California by ~5.6–5 Ma.

The authors conclude that lake spillover was the primary (though not necessarily exclusive) mechanism linking the upper and lower river systems, providing the simplest explanation supported by the new “fingerprint” data. Other processes (e.g., some headward erosion or karst features) may have contributed locally, but the Bidahochi connection fills a critical gap in the ~5-million-year interval that had been debated.

The work involved extensive fieldwork on Navajo Nation land (with proper permissions) and high-precision lab analysis of thousands of zircon grains.

This paper has generated significant coverage because it strengthens one of the leading hypotheses for one of geology’s classic puzzles. The article is behind the Science paywall, but abstracts and many news summaries are freely available.

PDF science.org/doi/pdf/10.1126/science.adz6826