The overarching goal of the proposed research is the sustainable management of water resources in coastal regions with diverse geological, hydro-technical and governance settings. Pressures on water resources in coastal regions are already great and are expected to intensify due to increasing populations, standards of living and impacts from climate change and sea level rise (SLR). We will focus on coastal areas where aquifer over-drafting has caused seawater intrusion (SWI), thus deteriorating groundwater quality, and where SLR is expected to further reduce availability of fresh groundwater. Solutions to these problems will involve combinations of more efficient pumping schemes, demand reduction, and technological interventions such as desalination. However, determining optimal solutions for these problems poses extreme computational demands. This project will greatly advance the development and application of simulation-optimization (SO) by developing computationally efficient, robust, and accurate surrogate models for coastal groundwater systems. The limited literature on SO and surrogate modeling in SWI problems has focused on simplified hydrogeological settings and mathematical representations of management strategies. However, realistic SWI problems involve hydrogeological complexities, including discrete lithological facies, faults and fractures, saltwater-freshwater mixing zone dynamics, and surface-water groundwater interactions, as well as nonlinear objective functions and continuous and discrete decision variables to represent a wide range of engineering components. We hypothesize that these hydrogeologic and management features determine the building of accurate and efficient surrogates; and accurate surrogate SO models for SWI problems can be at least an order of magnitude faster than full-scale models. The reduced computational cost allows to investigate a broader range of SLR and climate change impacts and a wider range of management responses to these impacts. The innovative aspects of this research are: (a) development of a systematic approach for building robust surrogates by testing against full-scale SO models on simple to complex problems; (b) assessment of tradeoffs between surrogate model accuracy and computational efficiency across a range of hydrogeologic and management settings; (c) identification of robust management schemes for managing coastal groundwater resources in three "end-member" case study aquifers; and (d) collaboration with water management agencies to develop useful scenarios, optimization frameworks, and model output. The three test aquifers (Santa Barbara, California; Biscayne, Florida; and San Salvador Island, The Bahamas) have diverse hydrogeologic and management characteristics and well-calibrated groundwater flow models. The project objectives are: (a) develop SO-SWI-SLR test problems to provide robust evaluation of model surrogates; (b) formulate management objectives and constraints based on management of the test case aquifers, and identify scenarios relevant to the test cases; (c) program, train, and evaluate the performance of "data-driven" and "model-driven" surrogates to identify optimal management schemes for the test case aquifers, a range of SLR rates, climatology, and groundwater demand scenarios. This work will build on our US-UK group's complementary experience simulating SLR and climate impacts on SWI and in developing SO models for other groundwater problems.

An eruption of Fuego volcano, Guatemala, on 3rd June 2018, had tragic outcomes when an entire village was inundated by pyroclastic flows. The eruption has prompted evacuations of around 12,000 people. This event resulted in changes to hazard, exposure and vulnerability, demonstrating the complex and dynamic nature of ongoing and future risk. This proposal seeks to characterise this dynamic risk observed in the natural environment, and understand the interactions between dynamic risk and society. Following the 3rd June eruption of Fuego, evacuations have resulted in reduced exposure in some regions, however, vulnerability (physical, systemic, functional, social, economic and political) remains high and is a key component of the evolving risk. In particular, systemic and functional vulnerability are believed to be highly dynamic. This provides an opportunity to investigate how the evolving hazard situation at Fuego, combined with changes in exposure and highly dynamic systemic and functional vulnerability, play out to affect risk in a context where both recovery and continued eruption risk management are ongoing. This opportunity is urgent: we must characterise changing hazard, exposure and vulnerability through time. Although the nature of the hazard can be investigated retrospectively, documenting changes to exposure (evacuations and reoccupations) and vulnerability as they respond to changing hazard and socio-economic conditions needs to be done as it occurs. For example, it is important to document physical vulnerability on buildings already impacted by the pyroclastic flows before further damage by weather or heavy machinery occurs, or document road closures next to affected drainages which can constitute a major element of the systemic vulnerability to lahars or pyroclastic flows of a community isolated by that road closure. Information on systemic vulnerability at this level of granularity is not normally documented in Guatemala, thus will not be available for later study. Through this proposed work, we will collect an unprecedented dataset on vulnerability, documenting physical vulnerability of buildings impacted by pyroclastic flows before any further damage. When considering risk to life by volcanic flow hazards and lahars however, physical vulnerability of infrastructure can be reduced to a binary effect (impacted or not. It is actually systemic and functional vulnerability that are the more important, and harder to ascertain, unknowns. A key research component, therefore, is to test the hypothesis that for volcanic flow related hazards, in contrast to tephra hazards, it is widespread systemic vulnerability and not physical vulnerability of the footprint of potential impact that is the root cause of risk. This is important because much of the work currently undertaken on risk in volcanology is led by frameworks used for tephra fall hazards, yet flow impacts and risk are very different. The project is will-aligned with the UN Sendai Framework for Disaster Risk Reduction, as well as recent initiatives in the wider volcanology community to engage and improve our capacity to do risk well. We will use a combination of volcanology field approaches, forensic approaches, and interviews to gather the information.

The eruption of Nabro in Eritrea is of great scientific interest and has had substantial impacts even in the remote part of Afar in which it is located. It is sited in the extensional zone of Afar, but just south of the Mesozoic crustal block of the Danakil Alps. It is a predominantly trachytic edifice, with an 8-km-diameter caldera and associated ignimbrites. This is the first eruption of Nabro on record, highlighting the potential of caldera systems to erupt without warning. It is also the first seismicity of note recorded in this particular part of the rift. Comparatively little is known about magma differentiation, storage and transport mechanisms and eruptive processes in such tectonic settings. The urgency reflects the hope of reaching the site before cessation of lava and gas emissions, while seismicity rates remain high, and to sample undisturbed tephra and lavas. We are the only international group to be invited to visit and study the eruption site. The field data will complement satellite observations of deformation and gas emission. The eruption resulted in seven fatalities and has displaced 5000 people, requiring an ongoing humanitarian response. Aviation was disrupted by ash clouds, adding considerably to work of the Toulouse Volcanic Ash Advisory Centre, which was also tracking the Puyehue plume above southern Africa. Based on satellite observations, the eruption began shortly before 20:42 UTC on 12 June following intense seismicity. The eruption was detected via infrasound records in Djibouti, and already (emissions continue at the time of writing) ranks among the largest sources of SO2 to the atmosphere since the 1991 Pinatubo eruption (~2 Tg of SO2 released according to retrievals of Ozone Monitoring Instrument data). The main aims of the project are to arrive at a detailed synthesis of the nature and causes of the eruption, to evaluate the events in the context of understanding restless calderas worldwide, and to compare and contrast activity of Nabro with the fissural basaltic systems that have been the focus of research by the NERC Afar Consortium (http://www.see.leeds.ac.uk/afar/).

This project will result in methods to detect boreal recruitment failure (RF) due to fire, an explanatory model of RF, and quantification of climate feedbacks from RF that are not currently accounted for in any climate or vegetation model. The associated data collection and research outputs will benefit models of climate-fire-vegetation feedbacks. Presently all models that incorporate fire disturbance assume forest recovery. Research Questions 1. When and where does boreal RF occur? 2. What are the factors that cause boreal RF? 3. What climate feedbacks are likely to result from boreal RF? Forest loss due to the failure of new trees to survive (recruitment failure) post-fire occurs in boreal forests in Eurasia and North America. The existence of ecological thresholds, or "tipping points" that cause abrupt ecological shifts, is well-known in ecosystems theory but where and when ecosystems are approaching such dramatic changes is difficult to predict. One such extreme ecological shift has been observed in boreal forests that fail to recover after multiple fires within a short time interval (< 10 years). These areas are dominated by grass and are similar to steppe vegetation. Transition from forest to steppe is consistent with predicted changes in vegetation composition in response to regional climate change, and is consistent with global observations of forest loss in response to climate. Preliminary analyses of these sites indicate causes related to changing fire regimes effected by climate. Firstly, although vegetation indices have been used to identify forest loss, there is currently no method to detect RF using remotely sensed data. We address here the likelihood that RF produces a unique signature that can be detected remotely. The total area affected by RF in Eurasia and North America is at present unknown. Using RF locations provided by the Sukachev Institute (see letter of support), we have developed preliminary methods to differentiate between successfully recovery from fire and RF using remotely-sensed vegetation indices. The proposed research would refine these methods and develop an automated approach to detect RF. The lengthening satellite data record permits a new focus on the impact of climate change on boreal forests (the largest terrestrial biome) and its potential consequences. Remotely sensed imagery to date have yielded "snapshots" of ecosystems and disturbance events. With more than a decade of daily imagery from the MODIS sensors, we can begin to monitor processes like disturbance-recovery cycles. This new focus is critically important to the study of climate-ecosystem interactions and climatic "tipping points". Secondly, the causes of RF have not been identified. RF has been observed in areas of Siberia where the length of time between fire disturbances was extremely low. Initial field observations of RF sites indicate that high soil temperature and low moisture create a seedbed unsuitable for recruitment of trees following a fire. Additional field data will provide the inputs for an explanatory model of RF that includes characteristics of the fire (such as intensity and fire weather), pre-fire vegetation (e.g., stand age and density), and post-fire environment (e.g., soil temperature and moisture). Thirdly, the effect of RF on carbon, water and energy fluxes that impact climate has not been quantified. The replacement of forests with steppe vegetation results in carbon losses to the atmosphere from combustion and post-fire decomposition. The net climate impact of RF is presently unknown. Albedo is initially low following a fire and then may become higher due to the higher albedo of replacement vegetation. Changes in evapotranspiration rates affect latent and sensible heat fluxes. The area of RF is likely to grow in response to increasing fire frequency and severity, but the dynamics of recovery from wildfire and RF have not been incorporated into any coupled climate-vegetation models.

Tropical forests play a critical role in global water, carbon and nutrient cycles, and currently absorb billions of tonnes of carbon, thus reducing rates of climate change. For this reason, computer models that are used to predict future climate change and the impacts of climate on plants and ecosystems, need to be able to represent tropical forest very well. In fact, the response of tropical forests to changes in temperature is one of the greatest uncertainties in climate change prediction. However, currently, scientists do not understand how these forests will respond to increasing temperatures. This is worrying because temperatures are increasing faster today than in the past, forcing forests to respond to unprecedented rates of warming. Critically, the lack of seasonal changes in temperature may mean that trees growing in these regions have a reduced capacity to deal with rapid climate change compared with more temperate and high-latitude species. If this is the case, then global warming may represent a considerable threat to these forests, the amazing amounts of biodiversity that they contain, and their role in reducing current rates of climate change. However, this suggestion is yet to be tested formally. The lack of understanding is even more worrying for tropical forest growing in mountains, as in these areas temperatures are increasing faster than in the lowlands. For example, scientists studying Andean forest in Colombia and Peru have observed that some tree species native to high elevations are dying out while others are moving to higher elevations. These scientists have suggested that these observations may be explained by the fact that trees are already seeing the impacts of climate change and are not able to withstand current temperatures. However, this explanation remains controversial and has not been tested formally. The major goal of this project is to determine if tropical Andean species can tolerate current temperatures and adjust to withstand the higher temperatures expected for the future. To answer this question, we will plant trees from high elevations in the Colombian Andes in their home environment but also at two lower elevations where temperatures are 5oC and 9oC higher, respectively. Our trees will be all planted in common soils and will have access to plenty of water, eliminating potential differences in water and nutrient access. We will monitor photosynthesis, respiration and growth at the three locations in other to understand how they respond to temperature. Compared to other experiments, our study is unique as it will: i) be the first to investigate the ability of large 3 - 4m tall trees planted in a common soil to respond to long-term (3 year) changes in temperature, ii) investigate a much greater number of species than all other field studies on this subject, and iii) measure a more complete set of key physiological and growth responses than in any other experiment. The measurements taken will be used to the derive mathematical equations that can represent the response of these tree montane species to elevated temperatures. Furthermore, to predict the response of tropical forest everywhere in the world to higher temperatures, we need data from high and low elevations in as many locations as possible. Scientists around the world are now starting to collect some of these measurements in forests from Costa Rica, Puerto Rico, Panama, Brazil, Peru, Rwanda and Australia. Although, no one of these investigations is as detailed as our study, by teaming up with all these groups we can use their data to test and extrapolate our equations across all tropics globally. We will then introduce these mathematical equations into a computer model to predict future behavior of the tropical forest under warming conditions. The outcome will represent a step change in our ability to accurately predict how this critically important biome will respond to global warming.