The Maritime Continent (MC) is the archipelago of tropical islands that lies between the Indian and Pacific Oceans, with a population of over 400 million. It comprises large (Sumatra, Java, Borneo, and New Guinea) and many smaller islands, with high mountains. High solar input warms the surrounding seas, which supply an abundance of moisture to the atmosphere, turning the whole region into an atmospheric "boiler box". Deep convective clouds rise up over the islands every day, leading to average rainfall rates in excess of 10 mm per day, approximately three times the rainfall rate over the UK. As well as supplying local agriculture, rain that falls over the MC has a far-reaching, global effect on weather and climate. Tremendous heat energy is released by condensation into the atmosphere in these convective clouds. This heat source drives giant, overturning circulations in the atmosphere: the Hadley and Walker cells, which feed into the jet streams and lead to weather and climate changes far downstream, even over the UK. For example, the origins of the infamous cold winter of 1962/63 and the recent very cold March of 2013 have been traced to atmospheric convection over the MC. For these reasons, the MC has been described as the engine room of the global climate system. Due to the complex nature of the distribution of the islands, and fundamental inadequacies in current models of the atmosphere (mainly related to their representation of convection), both climate predictions and weather forecasts show serious errors over the MC, particularly in their simulation of rainfall. Up until now, these errors have been extremely difficult to address, as there has been a lack of suitable observations over this region. Computing power, and the atmospheric modelling expertise to harness the advances in computing resources, has been inadequate to run computer models with sufficient detail to resolve the convective processes and their interactions, which are the building blocks of atmospheric circulation, for long enough to allow interactions with larger scales. However, we now stand on the cusp of transforming our understanding of atmospheric processes over the MC. Computer power and modelling expertise have progressed to the point where we have the capability to run simulations of the atmosphere at sufficient resolution to accurately capture the complex distribution of islands, and to accurately model the convective processes themselves. In response to this, the international Years of the Maritime Continent (YMC) field experiment (2017-2020) will make the measurements of the atmosphere and ocean at the very small scales that are needed to evaluate and understand the outputs of these new model simulations. Through TerraMaris the UK will take a leading role in YMC, by making observations of convective processes over the MC using the UK meteorological research aircraft, atmospheric radars, balloon and land-based measurements on the islands, and observing the surrounding seas using autonomous underwater and surface vehicles. This unprecedented suite of coordinated observations will complement measurements being taken by our international partners. The UK and the TerraMaris research team has led the way in developing high-resolution atmospheric modelling over recent years. We will apply the skills and knowledge learned to understand the complex mechanisms behind the multiple scales of convection and atmospheric circulations that have made the weather over the MC such a tough problem to crack. This knowledge will enable ground-breaking advances in atmospheric modelling, to improve weather forecasts and climate prediction over the MC region, with direct benefit to the substantial regional population. The downstream effects will see these benefits extend to the far corners of the globe, improving global and regional medium-range weather prediction, and climate projections.

The NCEO NC-ODA programme is focussed on a series of generic science issues that are particularly relevant to development challenges: characterisation and forecasting of land surface state including vegetation change and soil moisture; the evolution of forest carbon and characterisation of carbon fluxes arising from deforestation and degradation; the dynamic nature of fires, their emissions into the atmosphere and the development of large-scale air pollution; the development of a cadre of researchers and applications specialists trained in state-of-the-art Earth Observation (EO). We will address specific problems faced by DAC countries: the vulnerability of crop yields in semi-arid regions in Africa to drought, the challenge of protecting and enhancing Kenya's forest resources to mitigate climate change, the forecast skill necessary to capture hazardous air quality in South-East Asia stemming from open biomass burning, and the current lack of capacity of many African nations to make effective use of satellite EO data. The programme is structured into four WPs. Relevant UN Sustainable Development Goals are: 2 (Zero hunger), 3 (Good health and well-being), 13 (Climate action), 15 (Life on Land), and 17 (Partnership for the goals). WP 1 will improve crop yield modelling in Ghana and potentially Ethiopia through data assimilation of multiple EO data streams, for example effective leaf area index and soil moisture. The research will yield new knowledge on the value of accurate EO data parameters in a data-model system, to better characterise crop change and increase predictive skill, to examine upscaling from landscape to country scale, and improve soil moisture forecast skill [SGDs 2, 15]. WP 2 will establish a baseline of carbon emissions from deforestation in Kenya, identify different types of deforestation and degradation from synthetic aperture radar, optical and laser ranging (LiDAR) data, and establish areas that are suitable for afforestation to support the Vision 2030 of the Kenyan Government that aims to increase forest cover from 6 to 10 per cent by 2030. The work will establish forest reference emission levels and above-ground carbon stocks. This research is key to understanding carbon cycling estimates in a REDD+ policy context. [SGDs 13, 15]. WP 3 will develop and demonstrate new data sources that can improve forecast accuracy for large-scale air pollution during fire events. Currently, forecast models use estimates of fires that fail to capture the magnitude and variability of dynamic large forest and peatland fires and fires due to agricultural residue burning. The research will improve pollutant emissions estimates from fires, EO-based retrievals of smoke plume aerosols and auto-identification of biomass-burning plumes. We will work with stakeholders in the ASEAN countries to co-develop and demonstrate new systems, characterise improvements and train staff in the interpretation of complex EO data [SDGs 3, 13]. WP 4 will build capacity through international EO-related initiatives, including the Group on Earth Observations (GEO) AfriGEOSS initiative and the Committee on Earth Observation Satellites (CEOS) Working Group on Capacity Building & Data Democracy, to improve access to and use of contemporary EO datasets in African nations and other DAC nations. Work will include scoping of UK-related EO projects and experts related to AfriGEOSS identified needs, extension of the training of WP1-3 to wider DAC countries and to strategic capacity building, co-ordinated work with the relevant GEO initiatives of GFOI and GEOGLAM, and support of access to EO data for DAC countries. These actions will also benefit UK national priorities such as the monitoring of projects supported by the GNU partnership (Germany, Norway and UK), which is making US$5 billion available between 2015 and 2020 for REDD+ early movers, and the Biocarbon Fund's Initiative for Sustainable Forest Landscapes (ISFL) [SDG 17].