The DS421 program is designed to equip students to tackle a variety of research challenges and priorities aligned with the Grand Challenges that guide our overall program. Here, we outline examples of current research being conducted by various DS421 faculty, illustrating the kinds of problems trainees may pursue in team projects as part of the training program, as well as subsequent dissertation research.

• Attribution of rare events: Where controlled experiments are impossible, separating association from causation to inform policy presents an important, engaging challenge for statistics and data science, particularly in rare events, where the chances of spurious associations become more likely. Prof. Philip Stark is working with ecologists in Yosemite National Park to evaluate the impacts of pack horses, used by back-country travellers, on a protected amphibian that breeds in meadows favored by campers. While there is anecdotal evidence that pack horses trample the Yosemite toads and tadpoles, it is plausible that puddles in hoofprints enhance toad breeding opportunities. Should the Park Service reduce packstock use to protect toads? Addressing this policy question requires combining (noisy and potentially incomplete) historical data sets on stock use with fieldwork on toad occupancy and novel statistical techniques for inference in sparse data sets. The methods under development in this study are broadly applicable to the challenge of inferring the natural and societal impacts of rare events, and represent an interdisciplinary research frontier for students in statistics who have a deep appreciation for the systems under study.

• Environmental restoration and adaptive management: Large-scale environmental restoration projects require quick turnaround and adaptive decision-making in response to dynamics of complex systems that are being managed for diverse outcomes. A classic example, in which Prof. Laurel Larsen has deep expertise, is the $10.5 billion restoration of the Florida Everglades. Central to this restoration is “getting the water right,” managing flows through a complex hydrologic network to achieve multiple outcomes: sustaining recharge to aquifers for urban water supply, curtailing saltwater intrusion, providing flood control, sustaining ecological communities, and maintaining agricultural productivity. The Everglades has one of the most comprehensive data collection networks of any ecosystem, with spatially dense data on water levels, salinity, and water quality available in real time (e.g., the USGS Everglades Depth Estimation Network). A key challenge in management of the Everglades is closing the real-time feedback loop between data generation, analysis, and decision-making for water routing in a way that balances both short-term water needs and long-term sustainability in the face of climate change. This system provides a case study at the intersection of natural and social science with a strong reliance on sophisticated data management and quantitative analysis. Trainees will have opportunities to contribute to research and take on new projects as part of the DS421 training program and their subsequent PhD research. Collaborations with US Geological Survey as an external partner will support this research.

• Regional land use planning: Land use planning represents a critical aspect of public policy influencing regional economies, housing and population, disaster planning, transportation and carbon footprint, wildfire risk, habitat conservation, provision of ecosystem services and, collectively, quality of life. Under a stationary climate, land use decisions based on historic and current conditions may work for the future as well. Environmental change upends these fundamental assumptions for public policy and regional planning: sea level rise threatens coastal development; changing flood regimes impact communities in floodplains; and landscapes across the globe used for agriculture, residential development, and biodiversity conservation are affected by changes in climatic conditions. Advances in remote sensing, economic data collection, citizen science, and dynamics of human activity derived from mobile technology present enormous opportunities in data assimilation and analysis supporting integrated spatial modeling. Land use planning will serve as a major training and research theme in the DS421 program; Profs. David Ackerly, Maggi Kelly, and Kristina Hill all maintain active research programs in this area, and the collaborative problem solving class and co-mentoring of graduate students will create additional opportunities for collaboration and new research directions across our departments.

• Climate change adaptation and agriculture: The 2014 5^th Assessment Report of the IPCC highlights how historical climate change already has had detectable impacts on crop yields for a number of crops and regions. One major concern is the global agricultural sector’s ability to improve yields and area harvested to feed a growing global population with higher quality (and often more water- and energy-intensive foods) as incomes rise. This is complicated by projected changes in the relevant dimensions of climate due to anthropogenic emissions of greenhouse gases. It is difficult to project future yields because we do not know how to quantify change in area planted, adoption of new technologies, and average yields. The research of Prof. Max Auffhammer addresses these questions on several fronts. His lab uses remote sensing data and algorithms from Facebook’s facial recognition programs to look at changes in irrigation technology by scanning for the occurrence of circles, consistent with center pivot irrigation. Similar tools provide insight into changing cropping patterns after weather shocks and the impacts of biofuels production. These studies will test hypotheses regarding societal responses to changing climatic and economic conditions. We are ideally positioned to exploit the rapidly growing spatial data from new micro-satellites, and we are engaging with the WorldBank on this important research priority. Study of the systemic interconnections between population, agriculture, land use, and climate provide numerous opportunities to develop a global perspective as part of the DS421 training program.