Evidence suggests thin ice protected subsurface lake water on frozen Mars

Small bodies of water on early Mars may have remained in a liquid state for decades at a time, even though average atmospheric temperatures were well below the freezing point.

By applying a climate model specifically adapted to Martian conditions, a research team from Rice University found that lakes located in regions such as Gale Crater, near the Martian equator, could have persisted beneath thin, seasonal ice layers for at least several decades and potentially longer, provided that surrounding climate conditions remained relatively stable. These results help address a long-standing discrepancy in Martian science: geological features indicating prolonged surface water exist alongside climate models suggesting that early Mars was generally too cold to sustain liquid water.

A concealed stabilizing mechanism

In some model scenarios, the simulated lakes froze entirely during colder seasons. In others, however, the water remained liquid beneath a thin ice cover, rather than freezing solid. This ice layer functioned as an insulating barrier, significantly limiting water loss while still allowing incoming sunlight to warm the lake during warmer periods.

Due to this seasonal freeze–thaw behavior, several simulated lakes showed minimal changes in depth over multiple decades, indicating that they could remain stable for extended periods despite mean air temperatures staying below freezing for much of the year.

“This seasonal ice cover behaves like a natural blanket for the lake,” said Kirsten Siebach, associate professor of Earth, environmental and planetary sciences and co-author of the study.

According to Siebach, the ice insulates the water during winter months while permitting partial melting in summer. Because the ice layer is thin and transient, it would leave little lasting geological evidence. This characteristic may explain why Mars rovers have not identified clear signs of persistent surface ice or glaciation associated with ancient lakes.

The findings imply that long-lived lakes on early Mars did not require continuously warm climatic conditions, challenging earlier interpretations that surface water would only be possible under sustained warmth.

Implications for future exploration

If ancient Martian lakes were protected by seasonal ice rather than thick, permanent ice sheets, several geological features that have been difficult to reconcile with cold-climate models—such as preserved shorelines, sedimentary layers, and mineral deposits—may now be more readily explained.

The researchers plan to apply the LakeM2ARS model to additional Martian basins to test whether similar lake behavior could have occurred elsewhere on the planet. They also intend to investigate how variables such as atmospheric composition changes or groundwater interactions may have influenced the stability of lake ice over time.

“If similar patterns emerge across the planet, the results would support the idea that even a relatively cold early Mars could sustain liquid water year-round,” said Eleanor Moreland, lead author of the study. Such conditions would represent an important requirement for environments potentially suitable for life.

Published in AGU Advances, the study provides a new physical explanation for how lakes could have existed on Mars without a warm global climate, as well as why ancient Martian lake beds appear so well preserved today.

“Seeing ancient lake basins on Mars without clear evidence of thick, long-lasting ice made me question whether those lakes could have held water for more than a single season in a cold climate,” Moreland said. “When our model began producing lakes that could persist for decades beneath a thin, seasonally vanishing ice layer, it was exciting to identify a mechanism that aligns with what we observe on Mars.”

Adapting terrestrial climate tools for Mars

The research team modified a climate modeling framework known as Proxy System Modeling, originally developed by Earth climate scientist Sylvia Dee to reconstruct ancient terrestrial climates using indirect indicators such as tree rings or ice cores.

Because Mars lacks biological or cryogenic proxies comparable to those on Earth, the team instead relied on rover-based measurements, using Martian rock and mineral records as substitutes for climate indicators.

Over several years, the model was reconfigured to represent Mars as it existed approximately 3.6 billion years ago, incorporating factors such as reduced solar luminosity, a carbon dioxide–dominated atmosphere, and distinct seasonal cycles.

A total of 64 simulation scenarios were conducted using the Lake Modeling on Mars with Atmospheric Reconstructions and Simulations (LakeM2ARS) framework. Each scenario modeled a hypothetical lake within Gale Crater over 30 Martian years, corresponding to about 56 Earth years, to determine whether liquid water could realistically persist under those conditions.

“It was intellectually engaging to adapt a lake model designed for Earth to a completely different planet,” said Dee, associate professor of Earth, environmental and planetary sciences and co-author of the study. She noted that substantial adjustments were required, including changes to parameters such as gravity.

The team was encouraged by the model’s sensitivity to atmospheric pressure and seasonal temperature variations. According to Dee, the results demonstrate that, with careful modification, climate models developed for Earth can generate physically plausible scenarios for ancient Mars.

More information: AGU