Over billions of years of Earth's evolution, continents have drifted across the surface of the planet, colliding and breaking apart. During collisions, the continental lithosphere thickens and is uplifted to form vast mountain belts. When breaking apart, the continents are stretched and thinned.

The ancient cores of the continents have largely remained intact. However, multiple cycles of collision and rifting have probably weakened the margins of the continental plates — an idea that is difficult to investigate, because measuring the strength of the continents is not easy.

One way to assess the strength of a continental plate is to determine the degree to which it flexes under a heavy load, such as the burden of a thick mountain belt. This impact of the load on plate flexure is often inferred from the coherence between topography (as a proxy for load) and gravity (as an indicator of flexure). Any directional or azimuthal variation in this coherence is then used to identify gradients of weakness or strength — or directional fabrics — within the plate.

However, Lara Kalnins and colleagues use a series of statistical tests to show that the relationship between azimuthal variation in the coherence and strength of the plate is not as simple as previously assumed: variation in the coherence is sensitive to directionality in the gravity and topography and to other signals at short length-scales (Earth Planet. Sci. Lett. 419, 43–51; 2015). They reanalyse the strength of the North American continental lithosphere and, after statistical testing, find scant evidence for significant directional variations in the strength of the continent.

Credit: NASA

Their analysis does not rule out the idea that continental margins are weaker than their cores, or that the continental margins have a directional strength fabric that both reflects and influences supercontinent cycles, but the evidence looks less clear than before.