ABSTRACT

The big data revolution is changing how scientists approach explaining and predicting the behavior of ecological systems. Sensor networks, ecological observatories, and open data shared from classical experiments can be brought together to provide more holistic picture. However, problems often arise when combining data from multiple sources because of both environmental variability and, importantly, subtle differences in howobservations are made as a result of human choices. These issues can be exacerbated when studying how processes that unfold over time, leading to contradictory interpretations. For example, several recent studies which examined evidence of insect populations declining at a global level have found a full range of possible conclusions, depending on the data selected and the way in which the studies? calculations were performed. The objective of this CAREER award is to understand and explain these discrepancies by building and evaluating tools for combining ecological data and to provide new approaches for ecologists to evaluate the uncertainty in the patterns they observe over time in ecological systems. An integrated education and outreach program will enable scientists-in-training to take a flexible, critical and transparent view of data and analytics, and will also offer insights into how students conceptualize data and promote public data literacy. Effective synthesis of broad scale ecological data is key to tackling a wide variety of environmental problems and has the potential to form the basis of ecological forecasting. Because ecological systems often have non-linear behavior, understanding process trajectory and changepoints in dynamical systems is a key aspect in describing their future behavior, but inadequate diagnostics for data quality and synthesis issues can mask or confound these patterns.

This CAREER award addresses both technical and cultural gaps preventing meaningful synthesis of temporal processes in ecology through three specific aims. Aim 1) develop novel approaches to data integration and tools for evaluating the reliability of trends in ecological timeseries data. This will be achieved through comparative field experiments to assess biodiversity monitoring strategies, data mining approaches, and development of novel software products for processing temporal data. Aim 2) develop a new graduate quantitative methods course which integrates critical numeracy and historical context. Aim 3) develop an interdisciplinary podcast in collaboration with sociologists and education specialists to advance societal understanding of data and information, as well as highlight the work of early-career and minoritized scientists to give diverse context to issues of "knowing" and data. Results of this project, including links to publications, curriculum, and podcast episodes can be found at https://bahlailab.org/

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Please report errors in award information by writing to: awardsearch@nsf.gov.