Page 153 - Fundamentals of Physical Volcanology

9780632054435_4_009.qxd 12/10/2007 12:31PM Page 130

130 CHAPTER 9

it may happen that the magma reservoir feeding
the flow is small enough that it becomes depleted
of eruptible magma before a flow has reached its
cooling-limited length. In that case the flow stops
when the available magma is used up, and has a
smaller length and volume than it might have had –
this is a volume-limited flow.
If magma continues to emerge from a vent after
the first cooling-limited flow unit has stopped grow-
ing, one of three things can happen. First, a new
flow lobe may begin to grow from the original vent
along the side of the first flow. Presumably the first
lava flow will have flowed down the steepest topo-
graphic gradient leading away from the vent, and
Fig. 9.9 A basaltic sheet flow forming from a breakout from the new flow will therefore take whatever is now
a pahoehoe lava flow. (Photograph by Richard Hoblitt,
the line of least resistance over the new terrain.
courtesy U.S. Geological Survey, Hawaiian Volcano
Second, a new flow may begin to form on top of
Observatory.)
the previous flow unit. Normally this only happens
if the pre-eruption topography around the vent is
flows are more commonly of basaltic, rather than such as to confine the new flow and prevent it
any other, composition and, as shown in Chapter 1, finding a path alongside the older unit. Third, some
most eruptions on the ocean floor are of basaltic part of the boundary of the original flow unit may
lava. give way and allow lava from its interior to spill out
An immediately obvious feature of all lava flows to start to form a new flow.
is that as they spread away from a vent they initially This third process, called a lava breakout, can
move downhill and grow in width, but soon the happen because all lava flows are constantly cool-
lateral spreading ceases and the flow maintains a ing along all their boundaries; the base cools by
fairly constant width thereafter, unless significant conduction into the cold ground under the flow
changes in ground slope occur. This is one con- and the top and sides lose heat both by radiation
sequence of the characteristic cooling-induced and as a result of convection currents in the sur-

rheology of the lava and is in marked contrast to rounding air. However, liquid rock is a very poor
the spreading of liquids such as water: a water flood conductor of heat, and so the information that the
will continue to spread sideways as well as down- outside of the flow has become cool does not reach
hill more or less indefinitely, at least until it be- the center of the flow very quickly. It can be shown
comes so thin that surface tension forces become that the wave of cooling penetrates the flow in such
important. Lava too has a surface tension, but its a way that the greatest depth below the surface or
effects are confined to influencing the sizes of the above the base that experiences significant cooling
gas bubbles it contains, as was seen in Chapter 5, after the flow has been traveling for a time t is λ
and are quite irrelevant at the scale of even small given by
lava flows.
Another striking feature of lava flows is that any λ =∼2.3 (κ t) 1/2 (9.1)
one flow will not continue to lengthen indefinit-
ely but, for a given set of eruption conditions, will where κ is the thermal diffusivity of the lava, com-
2 −1
reach a well defined maximum length. This max- monly about 10 −6 m s for all magma composi-
imum length is ultimately controlled by the cooling tions. Consider two flows, one of which has been
of the flow, as discussed later, and flows which growing for an hour and the other for a day; cooling
cease to advance just because they have cooled too will have penetrated a distance of only ∼0.14 m into
much are called cooling-limited flows. Of course the first and 0.68 m into the second. Thus although