It is estimated that 5% of the world's population lives on land which is less than 5 metres above current sea level, in communities that are vulnerable to the impacts of sea level rise, either from direct loss of land, or increased flood risk. Society more broadly may be impacted by disruption to key infrastructure which is located on the coast e.g. power stations, and by the movement of displaced communities. The Antarctic ice sheet is the largest potential contributor to future sea level rise and projections of Antarctic ice sheet change in the future also have the largest range of estimates. This makes it difficult to accurately determine the risks of future sea level rise. Because sea level rise from Antarctic ice loss is not evenly distributed across the oceans, retreat of the Antarctic Ice Sheet will disproportionately affect coastlines that are furthest away, such as those in Europe and North America. In this proposal we will improve projections of Antarctic ice sheet change by reconstructing how the ice sheet changed during past warm intervals during the mid-Pliocene (approximately 3 million years ago). The mid-Pliocene is the last geological interval when atmospheric CO2 was similar to present day. The proposal will focus on reconstructing the amplitudes of mid-Pliocene sea level change between colder glacial stages and warmer interglacial states. We will use these data as a constraint for two types of ice sheet models. Recent work has used Pliocene interglacial sea level maxima as a constraint for Antarctic ice sheet models and has led to much higher projections of future sea level rise from Antarctica under anthropogenic warming. However, subsequent work has suggested that it may not be possible to accurately determine absolute Pliocene sea level maxima, such that the value of using these data has been questioned. The main source of uncertainty on these estimates comes from attempts to quantify them relative to a modern-day reference (i.e. as metres above present). An alternative approach that we will propose and one that can greatly improve past sea level estimates is to focus on the Pliocene glacial-interglacial sea level amplitude. We will reconstruct the glacial-interglacial sea level amplitude for 3 intervals in the mid-Pliocene using analysis of sediments recovered from the drilling of ocean sediment cores. Specifically, we will measure the geochemical composition (the isotopes of oxygen, magnesium and calcium) of calcite microorganisms (benthic foraminifera) to reconstruct past ice volume. In the absence of a modern-day reference we will simulate both the Pliocene glacial (cooler climate intervals) and interglacial (warmer climate intervals) extent of the Antarctic and Northern Hemisphere Ice Sheets (principally the Greenland Ice Sheet) and compare this with the sea level data that we will produce. We will then be able to determine what was the magnitude of Antarctic ice sheet melt during the past. Combining two groups based in the UK and US, the ice sheet models used will include the Penn State Ice Sheet Model (PSU-ISM) and the BISICLES ice sheet model. The treatment of the grounding line physics (the point at which grounded ice becomes floating ice shelf) is very different in these two models. The PSU-ISM requires additional processes (ice shelf hydrofracture and ice cliff failure) to simulate Antarctic retreat that was consistent with Pliocene sea level maxima. By using the BISICLES model, which has much higher resolution at the grounding line, we will be able to test whether these processes are needed to simulate ice retreat consistent with our measured Pliocene sea level amplitudes. Finally, we will use what we learn to produce a new set of future sea level estimates that are constrained using the palaeoclimate data. These will have tighter constraints than previous future sea level projections, enabling a more accurate estimate of the risk of future sea level rise from Antarctica.

The Paris Agreement presents humanity with an ambitious and critical goal: to keep global warming well below 2 degrees above pre-industrial levels. All 197 countries have signed up, and 189 have formally approved it. But there is no doubt that these targets present a massive challenge, and a certain amount of environmental change is already inevitable. We know that sea level will rise 10s of centimetres over the next several decades, displacing many millions of people living in low lying coastal areas. But we don't yet know just how much more our seas will rise through the coming centuries. Will our efforts to curb emissions stop the collapse of Antarctica's ice shelves and loss of the West Antarctic Ice Sheet? Under which conditions does collapse occur? And which part of the ice sheet will react first? Computer models yield conflicting results on these questions, partly because they simulate the past (and our future) using different environmental conditions and model physics. To figure out which of these are right, we need to obtain observational data from the geological past to test the models. Our project will embark in a detective story to provide some long-searched for evidence. We will exploit two geological records to reconstruct West Antarctic Ice Sheet history under temperatures only slightly elevated above modern levels (i.e. late Pleistocene interglacials). The first of these records comes from a recent ship-based drilling campaign (International Ocean Discovery Program Expedition 374) that recovered mud and sand from the Ross Sea, an area right next to the West Antarctic Ice Sheet. The second record will be retrieved from an ice shelf-based drilling rig that will recover the first extended record of sub-seafloor mud and sand from far beneath the Ross Ice Shelf, at a location where the West Antarctic ice sheet detaches from the seafloor and starts to float into the Ross Sea (Siple Coast drilling). For our first work package, we will analyse the chemical composition of mud, sand and organic particles to reconstruct critical environmental conditions. Firstly, the mud and sand will uncover where on the continent the pieces of rocks came from. Knowledge of the location of erosion can then tell us in turn whether the West Antarctic Ice Sheet melted during past times when temperatures were just a little bit warmer than today, or not. Secondly, the chemical composition of organic particles in the same samples will reveal prevailing ocean temperatures at the time of deposition. Thirdly, the presence/absence of certain types of marine algae will tell us whether floating ice was present or not. The combination of the three different sets of data will help us unravel where geographically ice melting started in West Antarctica. For our second work package, we will utilise our new data to test coupled ice sheet-climate models, which are also used to predict future sea level. Assessing how well these models perform in simulating the geological past is a key way of determining how accurate their projections of the future are. In detail we will test two such models, called PSUICE3D and BISICLES. We will analyse existing model simulations that led to collapse of the West Antarctic Ice Sheet during past warm times, and perform new simulations using a more realistic environmental framework constrained by our new data. The comparison of predicted places of ice retreat and modelled places of ice retreat has never been realised before and will allow us to pinpoint which parts of Antarctica are most vulnerable to moderate levels of global warming, providing vital information towards mitigation and adaptation of sea level rise for settlements in coastal areas around the globe.

Following heavy rainfall on the 4/1/14, a major debris flow at Slip Stream (44.59 S 168.34 E) introduced >10^6 m^3 of sediment to the Dart River valley floor in NZ Southern Alps (S. Cox, pers. comm). Runout over the existing fan dammed the Dart River causing a sudden drop in discharge downstream. This broad dam was breached quickly, however the loss of conveyance has since impounded a 4 x 1 km lake with depths that exceed 20 m. This event presents a rare and unprecedented opportunity to study the impacts of a discrete, high magnitude 'sediment pulse', remarkable in its capacity to dam a large river in flood (peak discharge during the event was recorded at ~790 m^3/s). Quantifying the impact of this disturbance on the form and stability of the receiving body, the Dart River, will advance our understanding of how such low frequency geophysical events shape the evolution of large alpine rivers and will create a vital baseline for future research that seeks to test theories of how such large bed wave propagates and disperse sediment downstream. The impact of this pulse also elevates the risks posed by natural hazards in the region. Enhanced sediment transport has the potential to raise riverbed levels, destabilise floodplain assets, reduce standards of flood protection, increase the risk of channel avulsion and impact on freshwater and riparian ecology with a legacy that long outlasts the initial disturbance. Locally, this event may result in rapid advance of the Dart-Rees delta into Lake Wakatipu threatening the lakeshore communities of Glenorchy and Kinloch. The assessment of how large fluvial sediment pulses migrate, disperse and condition such hazards will offer key insights that may be transferable to other dynamic alpine settings. However, in order to constrain this event effectively, an initial topographic and sedimentological survey must be undertaken urgently, in the immediate aftermath of the event, to enable robust quantification of the sediment pulse and the existing channel morphology. This research aims to advance this goal by seeking to: develop a unique baseline dataset that will be used to quantify the delivery and dispersal of sediment inputs from the Slip Stream landslide, from its source at Te Koroka to its sedimentary sink in Lake Wakatipu. Using a combination of aerial, terrestrial and bathymetric surveying, we will acquire two synoptic, system-wide snapshots of this highly charged sediment cascade that record the 3d morphology and sedimentology of the interlinked components of the sediment transfer system. Surveys will be undertaken in April 2014 and then one year later in March 2015, following the annual summer floods that dominate fluvial sediment transport in the region. The first survey will establish the initial state of the system and so create the opportunity to quantify the downstream pattern of sediment storage and transport through comparison with the second and any subsequent re-surveys direct differencing of Digital Elevation Models. The simultaneous bathymetric surveys of the upstream impounded lake and the delta morphology will provide constraints on sediment flux across the boundaries of the study area, enabling closure of the coarse sediment budget. The combined results of these two survey campaigns will create an unparalleled dataset to help frame and test hypotheses that seek to explain the dispersal of major sediment pulses within rivers.

This proposal is the UK component of a major international campaign, the Deep Fault Drilling Project (DFDP) to drill a series of holes into the Alpine Fault, New Zealand. The overarching aim of the DFDP to understand better the processes that lead to major earthquakes by taking cores and observing a major continental fault during its build up to a large seismic event. The next stage of this project will be to drill and instrument a 1.5 km hole into the Alpine Fault. Earthquakes are major geohazards. Although scientists can predict where on the Earth's surface earthquakes are most likely to occur, principally along plate boundaries, we have only imperfect knowledge. We also don't know when earthquakes will occur. This is well illustrated by recent events on the South Island of NZ. Two earthquakes in Christchurch in Sept 2010 and Feb 2011 caused 181 deaths and £7-10 billion of damage (~10% of NZ GDP). Yet Christchurch had previously been considered of relatively low seismic risk. In contrast, the western side of the South Island is defined by the Southern Alps, a major mountain chain (>3700 m) formed along the Australian-Pacific Plate boundary. Until a few million years ago this plate boundary was a strike-slip fault like the San Andreas Fault in California, but subtle changes in plate motion has led to the collision of the Pacific and Australian Plates. This caused uplift of the mountains and due to very high rates of rainfall and erosion, rapid exhumation of rocks that until recently had been deep within the Earth. Although these plates are moving past each other at ~30 mm/y and the uplift rate in the Southern Alps approaches 10 mm/y, there has not been a major earthquake along the Alpine Fault in NZ's, albeit short, written history. However, there is palaeo-seismic evidence that major earthquakes do occur along the Alpine Fault with magnitude ~8 earthquakes occurring every 200-400 years, with the latest event in 1717 AD. Earthquake occur because stresses build-up within the relatively strong brittle upper crust. At greater depths (>15 km) rocks can flow plastically and plates can move past each other without building up dangerous stresses. On some faults, the brittle crust "creeps" in numerous small micro-earthquake events and this inhibits the build up of stress. Unfortunately there are few even micro-earthquake events along the Alpine Fault or surface evidence for deformation, suggesting that the stresses along this plate boundary have been building up since 1717 - if that stress was released in a single earthquake it would result in a horizontal offset across the fault of >8m! A major hindrance to earthquake research is a lack of fault rock samples from the depths where stresses build up before an earthquake. Fault rocks exposed at the surface tend to be strongly altered. The strength of fault rocks will depend on a number of factors include pressure, temperature and the nature of the materials, but also whether there are geothermal fluids present. The geometry of the Alpine Fault is special in that the fault rocks that were recently deforming at depth within the crust are exposed close to the surface. Also because of rapid uplift and erosion the local geothermal gradients are high and relatively hot rocks are near the surface. This results in a relatively shallow depth (5-8 km) for the transition from brittle to plastic behaviour. This provides a unique opportunity to drill into the fault zone to recover cores of the fault, to undertake tests of the borehole strata, and to install within the borehole instruments to measure temperature, fluid pressures, and seismic activity. Once core samples are recovered we will perform geochemical and microstructural analyses on the fault rocks to understand the conditions at which they were deformed. We will subject them to geomechanical testing to see how changes in their environment affects the strength of the rocks and their ability to accommodate stresses before breaking.

On January 15th, 2022, a month into the eruption of Hunga Tonga - Hunga Ha'apai volcano, located in the Tonga Island arc in the southwest Pacific, there was a short lived (two hour), extremely high intensity explosive eruption that destroyed most of the subaerial parts of the volcanic edifice. The resulting pressure wave and tsunami impacts were local and global, and the most far-reaching since the eruption of Krakatau volcano in 1883. Tsunami waves striking the nearby (65 km) low-lying coasts of Tongatapu Island, were up to 15 m high. Because of well organised early warning and evacuation, there were very few (three) fatalities, although there was significant destruction along coastal areas. The low-lying islands of the more distant Nomuka Group were completely overwhelmed, and villages destroyed. Farther afield from the Tongan islands, the tsunamis were caused by the massive atmospheric pressure wave, that is the first instrumentally recorded eruption-generated event of this scale, which affected the entire global atmosphere and ionosphere, causing the observed infrasound waves and unusual long-period seismic resonances. The Hunga Tonga - Hunga Ha'apai eruption was a major surprise and there remains a major uncertainty over the mechanism(s) and the generation of the local tsunamis striking the nearby Tongan islands coasts, with three possibilities considered 1) pyroclastic density currents resulting from the collapse of the 50 km high ash column resulting from the explosion, 2) submarine mass sediment movements associated with the destruction of the volcanic edifice, and/or, 3) the massive shock wave resulting from a phreatomagmatic explosion as sea water entered the fractured volcanic caldera. The objectives of the proposed research, therefore, are to identify from very high resolution (one metre) satellite images the volcanic mechanism (s) of the eruption and the impact (inundation, height, and destruction) of the tsunami on the nearby Tonga islands. To achieve these aims, we will use the satellite imagery, 1) to map the morphological changes of Hunga Tonga - Hunga Ha'apai volcano in the build-up to and during the eruption; 2) as basis for observations on the eruption mechanism and its impacts, particularly information on the timing of the events, and 3) identify the height and inundation of the tsunami on the Tongan islands together the resulting destruction. After previous recent tsunamis, such as those of 2018 in Indonesia, the impacts on coastal areas were mapped soon after the event from field work by multinational teams that recorded the destruction and inundation. A major challenge with the Hunga Tonga - Hunga Ha'apai event, however, is the ongoing lack of information on the tsunami impact from Tonga because the eruption fractured the internet communications cable connecting the country to the outside world - so since the eruption the Island Kingdom has been largely isolated. It is also because Tonga is in Covid lockdown, with no access for foreign (non-Tongan) visitors. Our use of satellite imagery to map the tsunami impact, therefore, is a novel approach, not used previously in mapping tsunami inundation immediately after an event. The eruption has resulted in a major programme of scientific research carried out by many scientific organisations with who we are co-ordinating. We are also co-ordinating as far as possible, with local scientists, who will provide observational information on the tsunami impact to validate the interpretations of satellite data. These interpretations, in addition to validating the tsunami modelling, will also be used to underpin mitigation of the tsunami impact.