Byron Adams

Nationality American
Year of selection2017
InstitutionSchool of Earth Sciences, University of Bristol
CountryUnited Kingdom
RiskEnvironmental risks

Type of support

Post-Doctoral Fellowship

Granted amount

129 964 €


2 years

Half of the 700+ people who lost their lives following the 1991 eruption of Mount Pinatubo,

Philippines were not killed during the actual event. They died in subsequent months, during the

dramatic and sustained floods and debris flows that followed. This was no one-off occurrence.

When they erupt, explosive volcanoes disturb the landscape for miles around them, creating

substantial hazards from flooding, landslides and mudflows. « Despite this, there has been little

attempt to quantitatively model how landscapes adjust after an eruption », Dr. Byron Adams points

out. This Research Fellow at the University of Bristol has set the goal to develop a quantitative

model of the physical processes that change the landscape during and following a volcanic eruption.

His objective is to provide a tool that will contribute to a better understanding and a better

mitigation of the destructive effects of secondary volcanic hazards.

Explosive volcanic eruptions have transformative impact on surrounding areas. Pyroclastic flows

obliterate forests, lava flows remodel the land, tens of cubic kilometres of tephra and ash deposit

thick layers of loose material on the ground. « The combination of these factors creates high

probabilities of secondary hazards, especially during rainfall. On steep slopes, portions of the land

can collapse and create destructive mud flows. The deposition of pyroclastic debris or lava can

create dams, which can become unstable, or overflow to trigger floods », explains Adams. « In

order to understand the short- and long-term responses of landscapes after an eruption, many

questions must be answered about local geologic, geomorphic, and climatic conditions, and of

course about how the landscape will be altered during the eruption. My study will address questions

such as: how stable are the newly-erupted volcanic deposits? How easily are the deposits eroded,

and by what processes? Has the vegetation been buried or swept away during the eruption? If so,

how quickly can it reoccupy an area? How are all these changes affected by short- and long-term variations in climate? »

The first-ever coupled volcanic deposition and landscape evolution model to assess secondary

volcanic hazards

To develop and calibrate his landscape evolution model, Adams will begin by studying recent

volcanic eruptions in populated areas within the Philippines and western North America. The aim

will be to constrain how the landscape adjusts after eruptions using aerial and satellite images. « A

key aspect of this work will be to quantify patterns of erosion across these landscapes, which are

highly dependent on the material properties of the volcanic deposits and the local climate », the

researcher specifies. This study will then feed into the development and calibration of the very first

landscape evolution model to incorporate the deposition of volcanic material, and predict the spatial

extent, duration, and severity of post eruptive secondary hazards such as floods and landslides.

« Over 800 million people live close to active volcanoes. As populations continue to rise around

them, there has never been a greater need for a tool that is capable of predicting how a landscape

will evolve after a volcanic eruption », Adams presses. Although there has been extensive analysis

of the hazards that accompany volcanic eruptions, much less is known about the long‐term

consequences of volcanic activity. Adams’ project proposes to fill that gap by coupling volcanologic

and geomorphic models – two areas where important progress has been made in recent years. This

will allow him to model the processes that unfold over years, or even decades, after a disastrous

event. His initiative is answering a long-recognised need for predicting the destructive effects of

secondary volcanic hazards.