This notebook explores daily precipitation patterns near Temecula, California using ERA5 climate reanalysis data. The analysis focuses on identifying extreme precipitation events and visualizing their spatial impact across the continental United States.
The project is part of ATMS 523: Weather and Climate Data Analytics and demonstrates memory-safe workflows using xarray, dask, and cloud-hosted datasets.
Extract and analyze daily precipitation from 2005–2015 over a 5° × 5° box around Temecula.
Identify extreme precipitation days using the 95th percentile threshold and visualize the cumulative distribution.
Create composite and anomaly maps of precipitation across the U.S. on those extreme days, compared to the 1981–2010 climatological mean.
ERA5 hourly total precipitation data accessed via Google Cloud and processed with xarray and dask.
Daily totals computed by shifting time and summing hourly values.
Extreme days identified using the 95th percentile of daily precipitation.
Composite mean maps built using a memory-safe loop over selected days.
Climatology sourced from a monthly ERA5 dataset (1981–2020) provided by Prof. Nesbitt.
Anomaly maps created by subtracting climatological means from composite precipitation.
Assignment-2.ipynb: Main analysis notebook with step-by-step code and markdown explanations.
temecula_daily_precip_googlecloud.nc: NetCDF file containing daily mean precipitation time series near Temecula.
README-Assignment-2.md: This file.
Python 3.13.7
xarray, dask, matplotlib, cartopy, numpy, pandas
JupyterLab in a cloud VS Code environment
CDF Plot: Shows the distribution of daily precipitation and highlights the 95th percentile threshold.
Composite Map: Displays average precipitation across the U.S. on Temecula’s extreme days.
Anomaly Map: Reveals deviations from the 1981–2010 climatological mean, centered on Temecula.
ERA5 data provided by ECMWF via Google Cloud.
ERA5 data accessed via Google Cloud Public Datasets.
Google Research. (2022). Analysis-Ready, Cloud Optimized ERA5 (ARCO ERA5). Zenodo. https://doi.org/10.5281/zenodo.7253877 DOI: 10.5281/zenodo.7253877
Monthly climatology dataset courtesy of Prof. Steve Nesbitt.
Assignment developed for ATMS 523, Fall 2025.