Editing Umbriel (section)

== Origin and evolution ==
[[File:Umbriel usgsx2.jpg|thumb|[[False color]] image of Umbriel showing polygons|alt=A spherical blueish body with its surface covered by craters and polygons. The lower right part is smooth.]]

Umbriel is thought to have formed from an [[accretion disc]] or subnebula; a disc of gas and dust that either existed around Uranus for some time after its formation or was created by the giant impact that most likely gave Uranus its large [[Axial tilt|obliquity]].<ref name="Mousis 2004" /> The precise composition of the subnebula is not known, but the higher density of Uranian moons compared to the [[moons of Saturn]] indicates that it may have been relatively water-poor.{{efn|For instance, [[Tethys (moon)|Tethys]], a Saturnian moon, has a density of 0.97&nbsp;g/cm<sup>3</sup>, which suggests that over 90% of its composition is water.<ref name="Grundy Young et al. 2006" />}}<ref name="Smith Soderblom et al. 1986" /> Significant amounts of [[nitrogen]] and [[carbon]] may have been present in the form of [[carbon monoxide]] (CO) and molecular nitrogen (N<sub>2</sub>) instead of [[ammonia]] and methane.<ref name="Mousis 2004" /> The moons that formed in such a subnebula would contain less water ice (with CO and N<sub>2</sub> trapped as clathrate) and more rock, explaining the higher density.<ref name="Smith Soderblom et al. 1986" />

Umbriel's accretion probably lasted for several thousand years.<ref name="Mousis 2004" /> The impacts that accompanied accretion caused heating of the moon's outer layer.<ref name="Squyres Reynolds et al. 1988" /> The maximum temperature of around 180&nbsp;K was reached at the depth of about 3&nbsp;km.<ref name="Squyres Reynolds et al. 1988" /> After the end of formation, the subsurface layer cooled, while the interior of Umbriel heated due to decay of [[radioactivity|radioactive elements]] in its rocks.<ref name="Smith Soderblom et al. 1986" /> The cooling near-surface layer contracted, while the interior expanded. This caused strong [[Stress (mechanics)|extensional stresses]] in the moon's crust, which may have led to cracking.<ref name="Hillier & Squyres 1991" /> This process probably lasted for about 200&nbsp;million years, implying that any endogenous activity ceased billions of years ago.<ref name="Smith Soderblom et al. 1986" />

The initial [[accretion (astrophysics)|accretional heating]] together with continued decay of radioactive elements may have led to melting of the ice<ref name="Squyres Reynolds et al. 1988" /> if an [[antifreeze]] like ammonia (in the form of [[hydrate|ammonia hydrate]]) or some salt was present.<ref name="Hussmann Sohl et al. 2006" /> The melting may have led to the separation of ice from rocks and formation of a rocky core surrounded by an icy mantle.<ref name="Plescia 1987" /> A layer of liquid water (ocean) rich in dissolved ammonia may have formed at the core–mantle boundary. The [[eutectic temperature]] of this mixture is 176&nbsp;K. The ocean is likely to have frozen long ago.<ref name="Hussmann Sohl et al. 2006" /> Among Uranian moons Umbriel was least subjected to endogenic resurfacing processes,<ref name="Plescia 1987" /> although it may, like other Uranian moons, have experienced a very early resurfacing event.<ref name="Helfenstein Thomas et al. 1989" />