Reconstructing eruptions from historical accounts: Makaturing c. 1765, Philippines

Highlights

• Multi-disciplinary study of the c.1765 eruption of Makaturing volcano, Philippines.

• Historical record (Forrest, 1779) links eruption with far-reaching social consequences.

• Novel modelling approach filters eruptions that reproduce the reported impacts.

• We propose alternative interpretations of the historical record based on model results.

• Our study supports rapid tephra hazard assessment in the event of unrest at Makaturing.

Abstract

For some volcanoes, the only evidence for past eruption is provided by historical accounts. When interpreted carefully, these have the potential to be a rich source of information, and yet they have so far been under-utilised in reconstructing eruption histories. The navigator Thomas Forrest describes a large eruption at Makaturing volcano, southern Philippines, in approximately 1765 that he considers the catalyst for the local Iranun population transitioning from an agrarian society to long-distance piracy and slave raiding. Within the historical literature, the eruption is attributed to large scale physical impacts around the volcano and disruptions to trading routes and livelihoods that ultimately changed the course of southeast Asia's history. However, no such eruption (or impacts) are recognised in the scientific literature or in eruption databases, and fieldwork to the region remains difficult. Here, we reinterpret the account of Forrest from a multi-disciplinary perspective, with a historian and physical volcanologists working together to incorporate the greatly needed local context into identifying credible volcanic processes and impacts associated with the reported activity. We used a novel approach to eruption reconstruction by cross-referencing deposits and impacts inferred from the historical record with stochastic tephra dispersal modelling that considered multiple eruption sources and characteristics. We found that Forrest's account was best characterised by an eruption of ~VEI 4 between May and October, with plume heights in the range of 12 to 16 km. While at least one eruption of this size was required to reproduce the impacts described in the historical record, it may have formed part of a longer sequence of multiple, repeated eruptions. In this way, such an eruption could have acted as a ‘tipping point’ for a local population already on the verge of socio-political and economic collapse, disproportionately affecting regions on a much larger scale than the reported deposits suggest. There remains a disconnect between the eruption characteristics recorded in historical accounts and those reproduced by numerical modelling, for which we propose alternative interpretations of the historical record. Unfortunately, given the minimal details available about the eruption, this discordance is unlikely to be resolved, even when geological studies are possible. However, a valuable benefit of the probabilistic modelling approach presented here is that it highlights the likely direction of tephra dispersal and deposition during a future eruption of Makaturing, supporting rapid tephra hazard assessment in the event of future unrest.