How full were California’s reservoirs during the drought?

End-of-month storage for reservoirs from to : the wet year before the 2012–2016 drought, its record lows in 2014–2015, the 2017 recovery, the 2020–2022 drought and the years since. Statewide totals use of them; the flood-control basins are kept empty by design. Lines link each reservoir to a river gauge in its basin. Use the timeline to move through the months, and switch to 3D to fly over real terrain where each lake shrinks and grows with its storage, with its dam at the outlet.

About the project

California Drought Impact is an interactive multimodal data artwork about the drought, built from laser-cut wooden data sculptures, projection mapping and a digital visualization. The digital part went through California’s map data, drew lines as streams and let people explore the reservoirs. It was built in 2016 with D3.js, JSON, Python and jQuery. This page rebuilds it with D3 v7 and three.js.

Concept and visual design: Yoon Chung Han (principal investigator). Development: Shankar Tiwari, student research assistant, California State University, Fullerton. Original project page · 2016 version

Publication & data

Yoon Chung Han and Shankar Tiwari. California Drought Impact: Multimodal Data Representation to Predict the Water Cycle. IEEE VISAP 2016.

Reservoir storage (sensor 15, acre-feet) and river flow (sensor 41, monthly mean of daily cfs) come from the California Data Exchange Center, to . Each monthly value is end-of-month storage; recent months are provisional until the operators revise them, and short gaps are filled and flagged. Only of the river gauges report daily flow, so the other links are drawn at a fixed width. Locations come from the original 2016 dataset; capacities do too, except where CDEC has since revised them. The totals cover storage reservoirs: are flood-control basins, kept empty by design, and CDEC has no readings for .

Snowpack is the California Cooperative Snow Surveys' snow-pillow network on CDEC (sensor 3, snow water equivalent): the reading on the 1st of each month at the stations CDEC lists for its North, Central and South Sierra sections, summed and divided by the sum of those stations' April 1 averages. A month with too few stations reporting shows "No data". The 3D snowline follows each section's figure; it is drawn, not mapped.