Despite their global importance and poor protection, TDFS have been studied far less than other tropical forest ecosystems, particularly TDFS areas undergoing restoration. We aim to address this recently identified knowledge gap with the aim of improving the success of TDFS restoration. This project will provide the first assessment of the resilience of existing and restored TDFS to changing climate and climate extremes, through undertaking a comprehensive, community-scale assessment of traits which determine plant water-use, carbon production and nutrient-use strategies across restored TDFS sites. The information generated in this project will create a step-change in our current understanding of the function of restored and natural TDFS sites, facilitating development of state-of-the art vegetation models to improve climate prediction and the creation of new restoration policy through integrating with key stakeholders responsible for the creation and implementation of restoration strategies for Brazil. Our key aims are: Aim 1: Evaluate ecosystem function in TDFS sites restored using different strategies. Aim 2: Understand the pressures on TDFS from climate-change and climate extremes. Aim 3: Improve policy and restoration strategies for the restoration of, and long-term resilience of TDFS in collaboration with the Brazilian government. Tropical dry forests and savannas (TDFS) make up 34% of Brazil's land area and contain >50% of Brazil's plant species. More than 100 million people live in TDFS regions of Brazil and many of these people are from rural vulnerable communities who rely on essential ecosystem services TDFS provide. These services include: 1. water supply, shade and pollinators for Brazil's agricultural frontier; 2. national water security, with 43% of the surface water outside the Amazon falling in TDFS and supplying the aquifers which feed Brazil's three largest river basins; 3. a source of timber and food; 4. carbon storage for climate change mitigation; 5. areas of natural beauty, used extensively for tourism; 6. a living seed bank for >4500 woody plant species, many of which are endemic. Despite this, TDFS remain poorly protected with only 1.2% of dry forests and 7.5% of savannas in protected reserves and <10% of Brazil's dry forest and <20% of its savannahs remaining intact. Recognising the social, economic and environmental implications of the current rates of loss of TDFS, the Brazilian government has responded by committing to restoring 120,000 km2 (an area about half the UK) of natural ecosystems by 2030, with a focus on TDFS. Brazil's Ministry for the Environment (ICMBio) and Ministry for Agriculture (EMBRAPA) have started implementing this restoration plan. However, success rates of restored TDFS areas remains very low, with high variability between areas subjected to varying restoration strategies. The reasons for low success and high variability between strategies remains unknown, hampering current ability to meet national restoration targets. Until now, all TDFS restoration strategies have focused on re-creating the species composition observed in natural, undisturbed TDFS habitats. This focus has assumed that species diversity is synonymous with maximizing ecosystem productivity and resistance to climate variability, yet it ignores the suitability of these species to the new drier and disturbed environment they experience in degraded landscapes. The latest research from tropical rainforests broadly suggests that focusing only on species' diversity is too narrow. Instead, plant resource use strategies, and particularly hydraulic functional traits are likely to be the key to determining ecosystem-scale function and the resistance and resilience of TDFS ecosystems to current and future climate variability. To successfully protect and restore TDFS it is therefore vital that the current lack of understanding about ecosystem function and plant resource-use strategies in TDFS is addressed.

In the lower atmosphere ozone (O3) is an important anthropogenic greenhouse gas and is an air pollutant responsible for several billion pounds in lost plant productivity each year. Surface O3 has doubled since 1850 due to chemical emissions from vehicles, industrial processes, and the burning of forests. Tropical ecosystems are responsible for nearly half of global plant productivity and it is in these tropical regions that we are likely to see the greatest expansion of human populations this century. Alongside this growing population, we see the expansion of O3 precursor emissions from urbanization and high-intensity agricultural areas. Sugarcane is an important tropical and bioenergy crop, supplying raw material for sugar, ethanol (biofuel) and energy production and contributes to the bioeconomy of both São Paulo state (SP) and Brazil. While the São Paulo state is responsible for over half of Brazilian sugarcane production, sugarcane-derived products account for 17% of the Brazilian energy matrix. In a global context, biofuel production is one major land-based carbon-neutral approach to reduce our reliance on fossil fuels, and thus help society to achieve the challenging Paris accord of limiting climate change to below 2oC. Over the two last decades, SP state has experienced large-scale conversion of pasture (natural C4 grass) to sugarcane fields. At the same time air quality measurements demonstrate O3 concentrations across much of SP above those known to be harmful to plants. This project will make the first comprehensive set of measurements of O3 effects on plant functioning and growth in tropical grasses, both cultivated (e.g. sugarcane) and natural (e.g. used as pasture) using our unique tropical experimental facility in Cairns, Australia. Here we expose tropical grasses to different levels of ozone, to derive relationships between O3 dose and productivity loss. In addition, we investigate the role of drought on the O3 sensitivity of tropical grasses. Finally, we will use this information to assess the impact of regional O3 concentrations and changing land-cover (from natural C4 pasture to sugarcane) for southern Brazil and engage with relevant stakeholders from both policy and academia.

Montane forests in the Andes and the South-eastern Brazilian Mountain Range host the highest plant biodiversity on Earth. Current rates of warming in the Andes are three times higher than elsewhere in S. America, and higher than average warming of 5-6oC is predicted by the end of this century. Hence, the (sub)tropical mountain ranges in Latin America form a high-priority area in which to study the response of tropical trees under future environmental change. Tropical forests also play a crucial role in the global carbon budget, accounting for more than half of terrestrial net primary production and storing around 40% of plant biomass. Uncertainty in the response of tropical forests to global warming is responsible for a large uncertainty in atmospheric CO2 concentrations under any given scenario of anthropogenic CO2 emissions. However, the current generation of Dynamic Global Vegetation and Earth System Models do not include a representation of montane forest functioning, which stems from a lack of empirical understanding, leading to a consideration of only lowland tropical forests in models. We intend to address this knowledge gap by initiating a Latin America-wide network of tropical montane forest sites to gather existing understanding in order to model the contribution of these forests to the regional and global carbon and water cycles, under current and future climate change. This will be achieved via a dedicated workshop at the Uni-Campinas, Brazil, hosted by PP-FAPESP Nagy, with the participation of empirical experts across the network together with DGVM and ESM modellers.

This project addresses a key gap in understanding how tropical forests respond to drought across scales, from organ to tree and forest ecosystem. It will drive extended impact in new monitoring capability using satellite data, in advanced land surface modelling, and in drought risk mitigation planning, by engaging related stakeholders through 'Science & Impact' workshops. We propose the powerful combination of a unique large-scale field experiment in Amazônia together with detailed ecophysiological and new tower-based radar measurements to deliver new insights into drought responses across scales, both during drought, and importantly, during post-drought recovery. Water availability plays a dominant role in the global carbon cycle, with a large influence from Amazônia. However, our ability to predict the effects of changing water availability is substantially constrained by limited understanding of the ecological processes occurring in response to drought, particularly in tropical forests. These responses occur across different scales, from leaf to tree to forest ecosystem, with very large impacts on the carbon cycle observed regionally and globally. Understanding drought responses of tropical forests has proved challenging for several reasons: a lack of ecophysiological analysis at the right scales; limited capacity to deliver continuous monitoring of mechanistically-informed water stress responses at large scale, eg using satellites; and limited understanding of the ecological processes comprising drought stress and its consequences. We ask: How does drought stress affect whole-tree function, and can critical processes such as transpiration and growth recover after drought in tropical forests? Does drought stress leave a long-term legacy by limiting growth potential and by increasing the risk of possible tree mortality from future drought? And critically, how do the effects of drought on tree function affect performance at the scale of many trees, ie, that of a tropical forest? Multi-scale measurements are needed to address these questions. A combination of focused ecophysiological measurement with new tower-based radar (microwave) observations has the potential to enable large advances in understanding, scaling from tree to forest and region. This project will combine the world's only long-term drought experiment at hectare scale in tropical forest, which we have run for the past twenty years, with new radar sensors. We will use tower-based radar measurements to detect changes in vegetation water content at the scale of the experiment. This will provide higher resolution detection and mechanistic insight than was previously possible using satellite radars, and allow us to connect radar and plant ecophysiological data. Our specific hypotheses address: the links between organ-, tree- and ecosystem-scale responses to drought, and after drought; how these data advance our understanding of forest function and the risk to function and survival; and how this understanding can be used to advance satellite monitoring of drought impacts, and its wider use. In summary, we have three main goals: i) To use our ecosystem-scale drought experiment in Amazônian forest to quantify and understand the effects of drought at multiple scales, using plant physiology and tower-based radar (microwave) measurements. ii) To understand post-drought legacy effects on forest resilience by using the control enabled by our experiment to halt the drought and monitor recovery processes, and the outcomes for growth and survival. iii) To use (i) and (ii) to advance large-scale satellite detection capability in tropical forests for improved biomass and drought-response monitoring. We will lead two 'Science and Impact' workshops to rapidly multiply outcomes of the work by helping to improve prediction of land-atmosphere interactions using vegetation models, and better early-warning capability for land-use planning.

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.