Concentrations of both greenhouse gases (GHG) and aerosols (tiny particles suspended in the atmosphere) have increased considerably since pre-industrial time. Whilst anthropogenic emissions of GHG warm the planet, aerosol emissions exert a significant, yet poorly quantified cooling that acts to offset a significant fraction of global warming from GHG. Despite decades of research, the Intergovernmental Panel on Climate Change Assessment Report continues to highlight the climate sensitivity and aerosol-cloud-interactions (ACI) as the two key uncertainties limiting our understanding of climate change. Improving model estimates of climate change sensitivity (global temperature change per unit climate forcing) to greenhouse gas emissions is primarily driven by inter-model differences how climate models represent the impacts of feedbacks between low-level clouds and the climate system as temperature increases. Reducing these inter-model differences is severely hampered by the accuracy by which low level marine boundary layer (MBL) clouds, key modulators of the net radiation budget, are represented in the Earth System Models (ESMs) we use to provide estimates of future climate scenarios. Due to computational limitations these ESMs cannot explicitly represent small-scale atmospheric processes key for the formation of MBL at the scale at which they occur in nature (down to the size of aerosols). Instead, atmospheric physical processes related to cloud formation have to be parameterised (a simplified form of the complex process). Creating simplified representations of complex cloud processes that occur over a wide range of temporal/spatial scales is a challenging undertaking for climate scientists. Uncertainties in these parameterisations propagates through to our ability to accurately represent MBL in ESMs. The focus of this project will be to improve understanding of small-scale MBL processes by addressing current deficiencies in ESM parameterisations of cloud droplet formation, the direct microphysical link between aerosols and clouds. This will be achieved by using new modelling frameworks to capitalise on detailed flight measurements of MBL clouds from the NASA Earth Venture Suborbital mission called ACTIVATE (Aerosol Cloud meTeorology Interactions oVer the western ATlantic Experiment). ACTIVATE represents a novel measurement campaign of unprecedented scope for understanding MBL clouds as it will involve the deployment of two aircraft with well-matched groundspeeds. This strategy will allow for co-location of radiative properties of clouds from an aircraft flying above the MBL with an aircraft performing in-situ aerosol and cloud measurements within the MBL. This will provide a unique dataset with which we can constrain both process-scale cloud models, and large-scale ESMs to improve current small-scale ACI parameterisations, and subsequently the accuracy by which MBL clouds are represented in ESMs. To reach these goals the CLOSURE will use a new modelling framework in which a computationally fast cloud model known as a cloud parcel model (CPM). has been embedded in an ESM for the first time. These types of cloud models can accurately simulate the growth of a population of aerosol particles into cloud droplets in an ascending parcel of air. This embedded CPM framework will crucially allow for a detailed investigation of ACI in ESMs against measurements from ACTIVATE by providing additional model information for evaluation, e.g. droplet spectra. Furthermore, it will provide an efficient and seamless integration of process knowledge gained at the process scale from offline simulation to the large-scale when embedded in the ESM. This will be used to provide better understanding on the role of key small-scale processes involved in ACI for the representation of MBL clouds. The resulting improved theoretical descriptions of MBL cloud processes will reduce current uncertainties in future climate scenarios estimates.

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Forests are a critical component of the global carbon cycle because they take carbon dioxide out of the atmosphere through photosynthesis, and store the carbon in wood and soil. All living things in forests also produce carbon dioxide through respiration as an inevitable consequence of sustaining themselves and growing. At present, forests take in more carbon dioxide than they release, helping to reduce the amount of carbon dioxide present in the atmosphere, but this 'free gift' from forests is not guaranteed to continue at its current rate indefinitely under climate change. As well as the carbon cycle, forests are also crucial in the water cycle as trees pump water from the soil into the atmosphere. Leaves are the key part of the plant that regulates the exchange of gases (water, carbon dioxide) with the atmosphere. The pores in the leaf surface (stomata) are important for water loss and temperature control as well as the entry of carbon dioxide. Leaves exposed to direct sunlight can be more than ten degrees hotter than the air, even in temperate latitudes. Leaf temperature is important because many biological processes, including photosynthesis and respiration, are sensitive to temperature; very high temperatures can cause immediate and acute damage to leaves. Over the coming century, we expect carbon dioxide concentrations and air temperatures to continue to rise. When trees are grown in higher carbon dioxide concentrations, stomata close and limit water loss; this prevents the plant dehydrating but also reduces how much leaves can cool down. However, there is limited monitoring on forest canopy temperatures, and limiting understanding on how different species and forests in different climate zones are responding to climate change. This project will build a global network of researchers working to measure forest canopy temperatures using thermal infrared cameras, which will provide both greater understanding and also a crucial data resource for scientists in other disciplines to utilise. The network will ensure that the data collected by separate groups are comparable, and aid data processing and analysis by providing clear guidance and tools. This is will encourage other researchers to take up use of thermal infrared cameras, the analysis of which can be challenging. Our network will monitor canopy temperatures at fourteen sites in tropical and temperate forests and savannah, in UK, China, India, Australia, Brazil, Peru, Panama, USA, and Ghana. The sites in the UK and Peru will be newly established by this project. Ten sites already have established data collection, while the final two sites (Australia, Ghana) are in development. Having data collected using cameras will allow us to understand not only how forests in different locations are behaving, but also whether and how different species within sites respond. The long-term nature of the project means that seasonal variation will be included, and the forest response to extreme events such as heat waves and droughts will be quantified. Future work will establish in more detail how changes to canopy temperature link to changes in forest carbon and water cycling. Our work providing insight into the response of forest canopies to climate change will inform models produced to assess the impacts of greenhouse gas emissions on the planet, which are used to inform global climate change policies. Further, the current global emphasis on mitigating climate change through tree planting makes it crucial to assess how these trees will cope under future conditions.

Earth's climate has changed considerably in the past, and is predicted to change in the future. By studying past climates we gain a broader understanding of what climates are possible and likely in the future. In this proposal we focus on the very warm climates of the past and their relationship to global warming. In the far past, some 60 million years ago, the planet was very warm. However, the warming was not distributed uniformly over the globe. Rather, the high latitudes warmed much more than low latitudes, to the extent that palm trees grew in Wyoming and crocodile-like animals roamed Northern Canada. The evidence for this is very robust, since fossil remains are unambiguous. Crocodilians are intolerant to cold, meaning there were no long periods of very cold weather, even in winter, in northern North America. This is a complete mystery that current climate models cannot explain. We will study this problem using a novel suite of models, and apply what we learn to better understand the global warming ahead of us.

Modern society depends massively on the generation, processing and transmission of vast amounts of data. It is predicted that by 2025, 175 zettabytes (175 trillion gigabytes) of data will be generated around the globe, with so-called 'edge computing' devices creating more than 90 zettabytes alone. Processing such huge amounts of data demands ever increasing computational power, memory and communication bandwidth - demands that cannot be sustainably met by conventional digital electronic technologies. The growing gap between the needs and the capabilities of today's information technology is exemplified if we consider the historical trend in total number of computations (in units of #days of calculating at a rate of 1 PetaFLOP/s) needed to train various artificial intelligence (AI) systems. The trend followed Moore's Law (doubling approximately every two years) until 2012, after which the doubling time reduced to a mere 3.4 months! This trend is compounded by the breakdown in Koomey's Law, which states that the number of computations per Joule of energy doubles around every 1.5 years. This law was also followed until quite recently, but we are now approaching a widely accepted computing efficiency-wall at around 10 GMAC/Joule (a MAC is a multiply-accumulate operation) for CMOS electronics and the von-Neumann architecture. As a result, the energy consumption used in training modern AI systems is truly staggering, with consequent adverse effects for sustainability. This has led to a move away from standard CPU designs in AI towards the use of co-processors - GPUs, ASICs, FPGAs - with superior parallelism. However, even here the limitations of electrical signalling lead to massive levels of energy consumption. It was recently estimated, for example, that the training of a large GPU-based natural language processing system used for accurate machine translation resulted in carbon dioxide emissions equivalent to lifetime use of 5 cars! Clearly, a new approach is needed. Thus, in the APT-NuCOM project we will develop a highly efficient novel non-von Neumann co-processor that exploits clear advantages offered by photonic computation, but at the same time links seamlessly with the electronic domain to enable integration with existing electronic computing infrastructure. The APT-NuCOM co-processor will exploit novel phase-change photonic in-memory computing concepts to deliver massively parallel computation at PetaMAC/s speeds and, ultimately, an energy budget approaching that of the human brain.