Determining hazard from lavas has not received
a great deal of attention from volcanologists in the past because, normally,
only property is at stake: human life rarely is. Human risk is high from
explosive volcanic activity, and most attention has focused round this style of
volcanism. However, with increasing land use and urbanisation of the flanks of
basaltic volcanoes from which the main volcanic hazard is from lavas, this has
to be a topic to be considered seriously by volcanologists.
Normally (but not always) lavas move slowly at
their fronts, in the order of metres per hour It is thus possible to remove the
most valuable items from property ahead of a flow. It is not possible to do
anything about agricultural land apart from rapid harvesting if this is
applicable. Two things are therefore required from volcanologists: first an
assessment of the most vulnerable areas from future eruptions to enable
environmental planning by government; and second an ability to give day-to-day
predictions of the likely course of a lava once an eruption has started.
The first requires knowledge of the volcano
under study. It is necessary to know the most likely areas on the volcano where
eruptions may start. The topography must be known, as well as the normal
characteristics of effusive activity. Few such studies have been made for
basaltic volcanoes. One such is Etna. Guest and Murray (1979) investigated the
hazard from this volcano based on topography, locations of vents and effusion
rates. While recognising that an eruption could take place anywhere on the
volcano, they identified the most likely vent areas based on previous activity.
Using known relations between effusion rate and altitude of vent to maximum
flow length (based on Walker, 1973, 1974), they predicted the worst-case
outcome of an eruption from these likely areas. Using topography, they also
determined the catchment areas in which an eruption had to take place in order
to threaten each of the main towns on the volcano.
Once an eruption is in progress, the source is
known, although this may change during the eruption. The important factors in
controlling the advance of the lava, topography and effusion rate are known or
can be measured. Thus, predictions can be made. However, the possibility of a
change in the vent area must always be considered, as such changes can make an
apparently benign eruption a potentially threatening one.
In the last 15 years considerable advances have
been made in understanding flow field development based on field studies (Guest
et al., 1987; Kilburn and Luongo, 1993), and on theoretical models (Kilburn and
Luongo, 1993). One of us (CRJK) has recently developed a new model that, although
taking into account the many different factors that control the development of
a flow field, provides a simple way of predicting the final maximum length of
flow, but requires that the topography is known. This method will be tested
using the Fogo data, and used to aid, together with other techniques, to
prepare hazard maps of Fogo and other volcanoes where is considerable risk from
lavas.