The Iberian Pyrite Belt is the richest massive sulfide province in the world. The massive sulfide ore deposits occur in a felsic volcanic and sedimentary succession (VS Complex) of late Famennian (Upper Devonian) to late Visean (Middle Carboniferous) age. Volcanic facies analysis has been carried out on three areas in Portugal, including the Neves Corvo mine, and five sections in Spain. In all sections studied, the depositional setting during accumulation of the VS Complex was submarine and below wave base. The principal felsic volcanic facies are: (1) coherent rhyolite and dacite, associated with monomictic breccia; (2) fiamme-rich breccia (with variable amounts of dense volcanic and sedimentary clasts), fiamme-rich sandstone and fiamme-bearing mudstone; and (3) crystal-rich sandstone and mudstone. Mafic units are minor, dominated by coherent facies and have uncertain mode of emplacement (intrusions or lavas). Fiamme typically have lenticular shape and quartz- or quartz- and feldspar-phyric texture, and are interpreted to be altered and compacted pumice clasts. The volcanic facies are typically interleaved with, and regionally less voluminous than, the non-volcanic facies, which are dominated by mudstone. The felsic volcanic facies are interpreted to be the products of numerous, relatively small intrabasinal volcanic centres that generated abundant lavas, domes and pyroclastic units. Some volcanic centres are dominated by lavas, whereas others have similar proportions of lavas and pyroclastic units. The domes and lavas are more voluminous but less laterally extensive than the pyroclastic units. A sediment-matrix breccia typically occurs at the top contact of the felsic lavas with sedimentary units. This sediment-matrix breccia formed from the infiltration of fine sediment into interclast spaces in previously formed hyaloclastite, and could be misinterpreted as peperite. Felsic intrusions are less voluminous than lavas, and were emplaced as cryptodomes and partly extrusive cryptodomes, late in the evolution of the VS Complex. The architecture of the different study areas reflects differences in the eruption style, emplacement processes and proximity to source. Parts of the succession interpreted to be proximal are dominated by thick lavas/domes and intrusions, and coarse pyroclastic deposits. Medial parts comprise resedimented autoclastic facies derived from the lavas and domes, and relatively thin pyroclastic units. Distal parts comprise relatively thinly bedded crystal-rich sandstone and siliceous mudstone. Regional correlations in the VS Complex are impossible, as none of the volcanic facies are regionally extensive and each of the volcanic centres has a unique internal architecture. At Neves Corvo mine, the massive sulfide ore deposits are close to one of the felsic volcanic centre(s), occurring immediately above the rhyolitic lavas/domes. IV

Ground deformation due to volcanic magma intrusion is recognised as an important precursor of eruptive activity at a volcano. The Global Positioning System (GPS) is ideally suited for this application by being able to measure three-dimensional coordinate changes of the monitoring points over time. Due to the highly disturbed ionosphere in equatorial regions, particularly during times of maximum solar activity, a deformation monitoring network consisting entirely of single-frequency GPS receivers cannot deliver baseline solutions at the desired accuracy level. In this thesis, a mixed-mode GPS network approach is proposed in order to optimise the existing continuous single-frequency deformation monitoring system on the Papandayan volcano in West Java, Indonesia. A sparse network of dual-frequency GPS receivers surrounding the deformation zone is used to generate empirical 'correction terms' in order to model the regional ionosphere. These corrections are then applied to the single-frequency data of the inner network to improve the accuracy of the results by modelling the residual atmospheric biases that would otherwise be neglected. This thesis reviews the characteristics of existing continuously operating GPS deformation monitoring networks. The UNSW-designed mixed-mode GPS-based volcano deformation monitoring system and the adopted data processing strategy are described, and details of the system's deployment in an inhospitable volcanic environment are given. A method to optimise the number of observations for deformation monitoring networks where the deforming body itself blocks out part of the sky, and thereby significantly reduces the number of GPS satellites being tracked, is presented. The ionosphere and its effects on GPS signals, with special consideration for the situation in equatorial regions, are characterised. The nature of the empirically-derived 'correction terms' is investigated by using several data sets collected over different baseline lengths, at various geographical locations, and under different ionospheric conditions. Data from a range of GPS networks of various sizes, located at different geomagnetic latitudes, including data collected on Gunung Papandayan, were processed to test the feasibility of the proposed mixed-mode deformation monitoring network approach. It was found that GPS baseline results can be improved by up to 50% in the mid-latitude region when the 'correction terms' are applied, although the performance of the system degrades in close proximity to the geomagnetic equator during a solar maximum.

The most common hazards for communities surrounding mountain‐forming stratovolcanoes are mass flows of a range of types. Determining their frequency,characteristics and distribution is a major focus of hazard mapping efforts. Recent improvements in computer power and numerical models have meant that simulation of mass flow scenarios is a new tool available for hazard analysis. Its application to hazard mapping, land use planning and emergency management awaits robust evaluation of the conditions under which simulation tools are effective. This study focuses on this question in attempting to improve mass‐flow hazard assessments at the typical stratovolcanoes of Mts. Taranaki and Ruapehu in New Zealand. On Mt. Ruapehu, Titan2D modelling was applied to forecast behaviour of non‐cohesive lahars in the Whangaehu River, primarily produced by Crater Lake break‐outs, such as on 18 March 2007. The simulations were accurate in predicting inundation area, bifurcation, super‐elevation, hydraulic ponding, velocity and travel times of the lahar to 9‐10 km. A 6 x 10[exponent 6] m³ simulated granular flow had a minimum discharge of 1800‐2100 m³/s at the apex of the Whangaehu Fan, 9‐10 km from source, comparable to all historic information. The modelling implied that it was highly unlikely for a flow of this nature to overtop a lahar training dyke (bund) at the fan‐apex location and avulse northward into a more vulnerable catchment. Beyond this point, the model could not cope with the rapid and complex changes in rheology of these non‐cohesive lahars. At Mt. Taranaki chronostratigraphic grouping of mapped past lahar deposits often clouds the actual series of landscape forming processes and hence variations in hazard that occurred over time. Here, patterns of mass flows following emplacement of a 7 km³ debris avalanche deposit were examined from field geology and Titan2D modelling to define a three‐stage recovery process, where lahars of different types and sources were focused initially beside and later on top of the debris avalanche deposit for up to 10 000 years. Results from Titan2D were used to identify source areas of mass flows at different stages and their probable rheologies. Debris avalanche emplacement at Mt. Taranaki was investigated on the c. 7 ka B.P. Opua Formation with the help of Titan2D simulations to identify initial collapse parameters and major flow paths. Once again, the simulations were reliable in proximal reaches, but could not reproduce the rheological transformations from an initial collapsing/sliding pile through to a cohesive clay‐rich flow with long runout. In a further example, past block‐and‐ash flows (BAFs) and dense pyroclastic flow deposits northwest of the current crater were analysed to define the range of realistic model parameters for Titan2D simulations. These could be incorporated inside aGeographic Information System to produce a gradational map of relative probabilities of inundation by future BAF events that took both modelling and geological variability into account. This study highlights that computational models are now reaching the stage where a holistic approach can be taken to hazard analysis that combines both geological mapping and simulation of mass flow scenarios in a probabilistic framework to provide better tools for decision makers and land‐use planners.

Andesitic volcanoes are notorious for their rapid and unpredictable changes in eruptive style between and during volcanic events, a feature normally attributed to shallow crustal and intra-edifice magmatic processes. Using the example of eruptions during the last 1000 yrs at Mt. Taranaki (the Maero Eruptive Period), deposit sequences were studied to (1) understand lava dome formation and destruction, (2) interpret the causes of rapid shifts from extrusive to explosive eruption styles, and (3) to build a model of crustal magmatic processes that impact on eruption style. A new detailed reconstruction of this period identifies at least 10 eruptive episodes characterised by extrusive, lava dome- and lava flow-producing events and one sub- Plinian eruption. To achieve this, a new evaluation procedure was developed to purge glass datasets of contaminated mineral-glass analyses by using compositional diagrams of mineral incompatible-compatible elements. Along with careful examination of particle textures, this procedure can be broadly applied to build a higher degree of resolution in any tephrostratigraphic record. Geochemical contrasts show that the products of the latest Mt. Taranaki eruption, the remnant summit dome (Pyramid Dome) was not formed during the Tahurangi eruptive episode but extruded post-AD1755. Its inferred original maximum volume of 4.9×106 m3 (DRE) was formed by simultaneous endogenous and exogenous dome growth within days. Magma ascent and extrusion rates are estimated at =0.012 ms-1 and =6 m3s-1, respectively, based on hornblende textures. Some of the Maero-Period dome effusions were preceded by a vent-clearing phase producing layers of scattered lithic lapilli around the edifice [Newall Ash (a), Mangahume Lapilli, Pyramid Lapilli]. The type of dome failure controlled successive eruptive phases in most instances. The destruction of a pressurised dome either caused instantaneous but short-lived magmatic fragmentation (Newall and Puniho episodes), or triggered a directed blast-explosion (Newall episode), or initiated sustained magmatic fragmentation (Burrell Episode). The transition from dome effusion to a sustained, sub- Plinian eruption during the Burrell Lapilli (AD1655) episode was caused by unroofing a conduit of stalled magma, vertically segregated into three layers with different degrees of vesiculation and crystallisation. The resultant ejecta range from brown, grey and black coloured vesicular clasts to dense grey lithics. Bulk compositional variation of erupted clasts can be modelled by fractionation of hornblende, plagioclase, clinopyroxene, and Fe-Ti oxides. Pre-eruption magma ascent for the Maero Period events is assumed to begin at depths of c.9.5 km.

The stratigraphic record of lahars in the Whangaehu River reveals that in the past 180,000 years this route has been one of the main conduits for lahars from Mount Ruapehu, the highest active andesitic stratovolcano in the Central North Island of New Zealand. Both debris flows and hyperconcentrated flows have engulfed surfaces up to 160 km distance from the Volcano. Eight episodes of laharic activity are recognized by the distinctive lithology and similar age of their deposits. The newly defined upper Pleistocene Whangaehu Formation provides evidence for the earliest lahar event in the Valley, c. 180,000- 140,000 years ago. There is only meagre evidence for laharic activity following this event until the Ohakean and Holocene, although two new informally named deposits - the Mangatipona pumice sand (c. 37,000 years B.P.) and Apitian lahars (c. 32,000-25,500 years B.P) - are recognized, of minor extent. The formerly defined late Quaternary Te Heuheu (c. 25,500- 14,700 years B.P.), Tangatu (c. 14,700-5,370 years B.P.), Manutahi (c. 5 ,370-3,4600 years B.P.), Mangaio (c. 4,600 years B.P.) and Onetapu (< c. 1,850 years B.P.) Formations are here described and interpreted. Triggering mechanisms for lahar deposits are distinguished based on lithological criteria. (a) Bouldery deposits in the Whangaehu Formation are interpreted to have been emplaced by a single highly competent debris flow triggered by a southerly-directed flank collapse at Mount Ruapehu. This debris flow was competent enough to transport boulders up to 2 m in diameter over 140 km from the Volcano. Bouldery deposits are also recognized in the Onetapu Formation, but are restricted to higher gradient surfaces on the Mount Ruapehu ring plain. The Onetapu Formation deposits are interpreted to have been emplaced by lahars resulting from catastrophic drainage of Crater Lake, which occupies the active crater on Mount Ruapehu. (b) Pebbly and sandy deposits are interpreted to have been emplaced by low competence debris flows and hyperconcentrated flows. These lahar deposits are recognized in all formations described. The lithology in these deposits is commonly pumice and they are interpreted to have been triggered by eruptions and/or high rainfall events at the Volcano. Formations, and individual members within Formations, were dated by radiocarbon dating of organic material found below, within or above lahar deposits, or by coverbed stratigraphy. Both rhyolitic and andesitic tephras provided recognizable time planes in the late Quaternary coverbeds overlying lahar deposits. In this study quantitative analysis of quartz abundance, which is shown to vary between loesses and palaeosols, is used as an indirect means of establishing a surrogate for past climate changes which have been correlated to the deep sea oxygen isotope curve. A minimum age for the newly defined Whangaehu Formation is established by this method. The accumulation rate for lahars in the Whangaehu River has accelerated from 1 km3 every c. 23,000 years in the past c. 160,000 years to 1 km3 in 589 years in the past c. 2,000 years. This acceleration probably results from the increased frequency of lahars in the River following the development of Crater Lake c. 2,000 years B.P. According to this pattern an estimated 0.17 km3 volume of lahars could be anticipated over the next 100 years. If the 2,000 year accumulation rate were to be met over the next 100 years there would be 170 lahars of l0[superscript]6 m3 in this time interval , or 17 lahars of 10[superscript]7 m3 (or 1.7 lahars of 10[superscript]8 m3). The largest reported volume for an historic lahar is 10[superscript]6 m3 and these have occurred on average once every 30 years. The accumulation rate for historic lahars is 0.0054 km3 in 100 years. Therefore, although the accumulation rate appears to have slowed down, further large lahars with magnitudes 10 or 100 times greater than those witnessed could be expected.

The stratigraphic record of lahars in the Whangaehu River reveals that in the past 180,000 years this route has been one of the main conduits for lahars from Mount Ruapehu, the highest active andesitic stratovolcano in the Central North Island of New Zealand. Both debris flows and hyperconcentrated flows have engulfed surfaces up to 160 km distance from the Volcano. Eight episodes of laharic activity are recognized by the distinctive lithology and similar age of their deposits. The newly defined upper Pleistocene Whangaehu Formation provides evidence for the earliest lahar event in the Valley, c. 180,000- 140,000 years ago. There is only meagre evidence for laharic activity following this event until the Ohakean and Holocene, although two new informally named deposits - the Mangatipona pumice sand (c. 37,000 years B.P.) and Apitian lahars (c. 32,000-25,500 years B.P) - are recognized, of minor extent. The formerly defined late Quaternary Te Heuheu (c. 25,500- 14,700 years B.P.), Tangatu (c. 14,700-5,370 years B.P.), Manutahi (c. 5 ,370-3,4600 years B.P.), Mangaio (c. 4,600 years B.P.) and Onetapu (< c. 1,850 years B.P.) Formations are here described and interpreted. Triggering mechanisms for lahar deposits are distinguished based on lithological criteria. (a) Bouldery deposits in the Whangaehu Formation are interpreted to have been emplaced by a single highly competent debris flow triggered by a southerly-directed flank collapse at Mount Ruapehu. This debris flow was competent enough to transport boulders up to 2 m in diameter over 140 km from the Volcano. Bouldery deposits are also recognized in the Onetapu Formation, but are restricted to higher gradient surfaces on the Mount Ruapehu ring plain. The Onetapu Formation deposits are interpreted to have been emplaced by lahars resulting from catastrophic drainage of Crater Lake, which occupies the active crater on Mount Ruapehu. (b) Pebbly and sandy deposits are interpreted to have been emplaced by low competence debris flows and hyperconcentrated flows. These lahar deposits are recognized in all formations described. The lithology in these deposits is commonly pumice and they are interpreted to have been triggered by eruptions and/or high rainfall events at the Volcano. Formations, and individual members within Formations, were dated by radiocarbon dating of organic material found below, within or above lahar deposits, or by coverbed stratigraphy. Both rhyolitic and andesitic tephras provided recognizable time planes in the late Quaternary coverbeds overlying lahar deposits. In this study quantitative analysis of quartz abundance, which is shown to vary between loesses and palaeosols, is used as an indirect means of establishing a surrogate for past climate changes which have been correlated to the deep sea oxygen isotope curve. A minimum age for the newly defined Whangaehu Formation is established by this method. The accumulation rate for lahars in the Whangaehu River has accelerated from 1 km3 every c. 23,000 years in the past c. 160,000 years to 1 km3 in 589 years in the past c. 2,000 years. This acceleration probably results from the increased frequency of lahars in the River following the development of Crater Lake c. 2,000 years B.P. According to this pattern an estimated 0.17 km3 volume of lahars could be anticipated over the next 100 years. If the 2,000 year accumulation rate were to be met over the next 100 years there would be 170 lahars of l0[superscript]6 m3 in this time interval , or 17 lahars of 10[superscript]7 m3 (or 1.7 lahars of 10[superscript]8 m3). The largest reported volume for an historic lahar is 10[superscript]6 m3 and these have occurred on average once every 30 years. The accumulation rate for historic lahars is 0.0054 km3 in 100 years. Therefore, although the accumulation rate appears to have slowed down, further large lahars with magnitudes 10 or 100 times greater than those witnessed could be expected.

The stratigraphic record of lahars in the Whangaehu River reveals that in the past 180,000 years this route has been one of the main conduits for lahars from Mount Ruapehu, the highest active andesitic stratovolcano in the Central North Island of New Zealand. Both debris flows and hyperconcentrated flows have engulfed surfaces up to 160 km distance from the Volcano. Eight episodes of laharic activity are recognized by the distinctive lithology and similar age of their deposits. The newly defined upper Pleistocene Whangaehu Formation provides evidence for the earliest lahar event in the Valley, c. 180,000- 140,000 years ago. There is only meagre evidence for laharic activity following this event until the Ohakean and Holocene, although two new informally named deposits - the Mangatipona pumice sand (c. 37,000 years B.P.) and Apitian lahars (c. 32,000-25,500 years B.P) - are recognized, of minor extent. The formerly defined late Quaternary Te Heuheu (c. 25,500- 14,700 years B.P.), Tangatu (c. 14,700-5,370 years B.P.), Manutahi (c. 5 ,370-3,4600 years B.P.), Mangaio (c. 4,600 years B.P.) and Onetapu (< c. 1,850 years B.P.) Formations are here described and interpreted. Triggering mechanisms for lahar deposits are distinguished based on lithological criteria. (a) Bouldery deposits in the Whangaehu Formation are interpreted to have been emplaced by a single highly competent debris flow triggered by a southerly-directed flank collapse at Mount Ruapehu. This debris flow was competent enough to transport boulders up to 2 m in diameter over 140 km from the Volcano. Bouldery deposits are also recognized in the Onetapu Formation, but are restricted to higher gradient surfaces on the Mount Ruapehu ring plain. The Onetapu Formation deposits are interpreted to have been emplaced by lahars resulting from catastrophic drainage of Crater Lake, which occupies the active crater on Mount Ruapehu. (b) Pebbly and sandy deposits are interpreted to have been emplaced by low competence debris flows and hyperconcentrated flows. These lahar deposits are recognized in all formations described. The lithology in these deposits is commonly pumice and they are interpreted to have been triggered by eruptions and/or high rainfall events at the Volcano. Formations, and individual members within Formations, were dated by radiocarbon dating of organic material found below, within or above lahar deposits, or by coverbed stratigraphy. Both rhyolitic and andesitic tephras provided recognizable time planes in the late Quaternary coverbeds overlying lahar deposits. In this study quantitative analysis of quartz abundance, which is shown to vary between loesses and palaeosols, is used as an indirect means of establishing a surrogate for past climate changes which have been correlated to the deep sea oxygen isotope curve. A minimum age for the newly defined Whangaehu Formation is established by this method. The accumulation rate for lahars in the Whangaehu River has accelerated from 1 km3 every c. 23,000 years in the past c. 160,000 years to 1 km3 in 589 years in the past c. 2,000 years. This acceleration probably results from the increased frequency of lahars in the River following the development of Crater Lake c. 2,000 years B.P. According to this pattern an estimated 0.17 km3 volume of lahars could be anticipated over the next 100 years. If the 2,000 year accumulation rate were to be met over the next 100 years there would be 170 lahars of l0[superscript]6 m3 in this time interval , or 17 lahars of 10[superscript]7 m3 (or 1.7 lahars of 10[superscript]8 m3). The largest reported volume for an historic lahar is 10[superscript]6 m3 and these have occurred on average once every 30 years. The accumulation rate for historic lahars is 0.0054 km3 in 100 years. Therefore, although the accumulation rate appears to have slowed down, further large lahars with magnitudes 10 or 100 times greater than those witnessed could be expected.

The long-term behaviour of andesitic stratovolcanoes is characterised by a repetition of edifice growth and collapse phases. This cyclic pattern may represent a natural frequency at varying timescales in the growth dynamics of stratovolcanoes, but is often difficult to identify because of long cycle-timescales, coupled with incomplete stratigraphic records. The volcaniclastic ring-plain succession surrounding the 2 518 m Mt. Taranaki, New Zealand, comprises a wide variety of distinctive volcanic mass-flow lithofacies with sedimentary and lithology characteristics that can be related to recurring volcanic cycles over >190 ka. Debrisflow and monolithologic hyperconcentrated-flow deposits record edifice growth phases while polylithologic debris-avalanche and associated cohesive debris-flow units were emplaced by collapse. Major edifice failures at Mt. Taranaki occurred on-average every 10 ka, with five events recognised over the last 30 ka, a time interval for which stratigraphic records are more complete. The unstable nature of Mt. Taranaki mainly results from its weak internal composite structure including abundant saturated pyroclastic deposits and breccia layers, along with its growth on a weakly indurated and tectonically fractured basement of Tertiary mudstones and sandstones. As the edifice repeatedly grew beyond a critical stable height or profile, large-scale collapses were triggered by intrusions preceding magmatic activity, major eruptions, or significant regional tectonic fault movements. Clasts within debris-avalanche deposits were used as a series of windows into the composition of previous successive proto-Mt Taranaki edifices in order to examine magmatic controls on their failure. The diversity of lithologies and their geochemical characteristics are similar throughout the history of the volcano, with the oldest sample suites displaying a slightly broader range of compositions including more primitive rock types. The evolution to a narrower range and higher-silica compositions was accompanied by an increase in K2O. This shows that later melts progressively interacted with underplated amphibolitic material at the base of the crust. These gradual changes imply a long-term stability of the magmatic system. The preservation of similar internal conditions during the volcano’s evolution, hence suggests that external processes were the main driving force behind its cyclic growth and collapse behaviour and resulting sedimentation pattern.

The long-term behaviour of andesitic stratovolcanoes is characterised by a repetition of edifice growth and collapse phases. This cyclic pattern may represent a natural frequency at varying timescales in the growth dynamics of stratovolcanoes, but is often difficult to identify because of long cycle-timescales, coupled with incomplete stratigraphic records. The volcaniclastic ring-plain succession surrounding the 2 518 m Mt. Taranaki, New Zealand, comprises a wide variety of distinctive volcanic mass-flow lithofacies with sedimentary and lithology characteristics that can be related to recurring volcanic cycles over >190 ka. Debrisflow and monolithologic hyperconcentrated-flow deposits record edifice growth phases while polylithologic debris-avalanche and associated cohesive debris-flow units were emplaced by collapse. Major edifice failures at Mt. Taranaki occurred on-average every 10 ka, with five events recognised over the last 30 ka, a time interval for which stratigraphic records are more complete. The unstable nature of Mt. Taranaki mainly results from its weak internal composite structure including abundant saturated pyroclastic deposits and breccia layers, along with its growth on a weakly indurated and tectonically fractured basement of Tertiary mudstones and sandstones. As the edifice repeatedly grew beyond a critical stable height or profile, large-scale collapses were triggered by intrusions preceding magmatic activity, major eruptions, or significant regional tectonic fault movements. Clasts within debris-avalanche deposits were used as a series of windows into the composition of previous successive proto-Mt Taranaki edifices in order to examine magmatic controls on their failure. The diversity of lithologies and their geochemical characteristics are similar throughout the history of the volcano, with the oldest sample suites displaying a slightly broader range of compositions including more primitive rock types. The evolution to a narrower range and higher-silica compositions was accompanied by an increase in K2O. This shows that later melts progressively interacted with underplated amphibolitic material at the base of the crust. These gradual changes imply a long-term stability of the magmatic system. The preservation of similar internal conditions during the volcano’s evolution, hence suggests that external processes were the main driving force behind its cyclic growth and collapse behaviour and resulting sedimentation pattern.

The long-term behaviour of andesitic stratovolcanoes is characterised by a repetition of edifice growth and collapse phases. This cyclic pattern may represent a natural frequency at varying timescales in the growth dynamics of stratovolcanoes, but is often difficult to identify because of long cycle-timescales, coupled with incomplete stratigraphic records. The volcaniclastic ring-plain succession surrounding the 2 518 m Mt. Taranaki, New Zealand, comprises a wide variety of distinctive volcanic mass-flow lithofacies with sedimentary and lithology characteristics that can be related to recurring volcanic cycles over >190 ka. Debrisflow and monolithologic hyperconcentrated-flow deposits record edifice growth phases while polylithologic debris-avalanche and associated cohesive debris-flow units were emplaced by collapse. Major edifice failures at Mt. Taranaki occurred on-average every 10 ka, with five events recognised over the last 30 ka, a time interval for which stratigraphic records are more complete. The unstable nature of Mt. Taranaki mainly results from its weak internal composite structure including abundant saturated pyroclastic deposits and breccia layers, along with its growth on a weakly indurated and tectonically fractured basement of Tertiary mudstones and sandstones. As the edifice repeatedly grew beyond a critical stable height or profile, large-scale collapses were triggered by intrusions preceding magmatic activity, major eruptions, or significant regional tectonic fault movements. Clasts within debris-avalanche deposits were used as a series of windows into the composition of previous successive proto-Mt Taranaki edifices in order to examine magmatic controls on their failure. The diversity of lithologies and their geochemical characteristics are similar throughout the history of the volcano, with the oldest sample suites displaying a slightly broader range of compositions including more primitive rock types. The evolution to a narrower range and higher-silica compositions was accompanied by an increase in K2O. This shows that later melts progressively interacted with underplated amphibolitic material at the base of the crust. These gradual changes imply a long-term stability of the magmatic system. The preservation of similar internal conditions during the volcano’s evolution, hence suggests that external processes were the main driving force behind its cyclic growth and collapse behaviour and resulting sedimentation pattern.