Undergraduate students Maria Villarreal Simon (left) and Iman Ali (right) presented their PURM research at the Penn Center for Undergraduate Research and Fellowships (CURF)'s Fall Research Expo 2024.

As our planet continues to shatter temperature records year after year, the urgency of preparing for the escalating impacts of climate change grows clearer. Rising global heat is driving more frequent and severe weather extremes—heatwaves, floods, storms, and wildfires—that threaten communities worldwide. To better anticipate and adapt to these hazards, it is crucial to connect the human experience of climate-related disasters with the underlying physical data that track changes in the Earth system.

Through this tutorial (available via GitHub: https://github.com/xueke-li/climate-data-analysis), students engage directly with that connection. They learn to collect, analyze, and visualize climate data, link it to real-world impacts, and develop the systems-level perspective needed to bridge science and policy. The goal is not just technical skill-building, but also cultivating the analytical and research capacities essential for shaping effective, equitable responses to a changing climate.

Li et al. (2021); Li et al. (2019a); Li et al. (2019b)

ARIMA is a widely used statistical model for time series analysis and forecasting. It integrates three key components: autoregression (AR), which captures dependencies on past values; integration (I), which ensures stationarity through differencing; and moving average (MA), which incorporates past forecast errors. By capturing temporal dependencies and trends, ARIMA models are particularly effective for detecting and predicting evolving patterns in data. 

Our previous work has demonstrated the model's scalability across multiple data sources, timescales, and disciplines, extending its utility well beyond traditional applications in finance and economics to serve as a robust tool for climate and environmental research, including aerosol science, hydrology, and ecology. R code for ARIMA implementation is available on GitHub: https://github.com/xueke-li/ARIMA.