Beneath the Surface: Exploring GLOBGM High Resolution Global Groundwater Model

This week explore how a gridded global groundwater model reduced global groundwater dataset resolution by a factor of 10 and delivered a computed water table depth for a 58 yr period between 1958-2015

The GLOBGM v1.0 dataset marks a significant milestone in global groundwater modeling, offering a parallel implementation of a 30 arcsec PCR-GLOBWB-MODFLOW model. Developed by Jarno Verkaik et al., this dataset presents a comprehensive understanding of global groundwater dynamics at a spatial resolution of approximately 1 km at the Equator. Leveraging unstructured grids and a prototype version of MODFLOW 6, the research team has meticulously partitioned the Earth's surface into independent grids, totaling a staggering 278 million active cells.

Computed steady-state groundwater head [m] for the lower model layer

The dataset utilizes available 30′′ PCR-GLOBWB data to drive simulations, enabling researchers to explore groundwater flow dynamics on a global scale. The focus of the paper lies with the computational implementation is parallelized using the message-passing interface, facilitating efficient processing on distributed memory parallel clusters and you can read it here.

GLOBGM: Data Details!

GLOBGM offers a massive dataset split into two main categories: transient (changing over time) and steady state (constant). The transient data spans an impressive 58 years, from 1958 to 2015, giving you a glimpse into historical groundwater trends. ⏳

Getting to the dataset is relatively simple though exporting the data seems to be throughput limited and can take sometime even with a multigigabit line. The overall folder with all steady state and transient files and collections equal about 1.31 TB of data.

Model raster output (30 arcsec; WGS84 latitude longitude)

Steady State Rasters
- globgm-heads-lower-layer-ss.tif: computed steady-state groundwater head [m] for the lower model layer
- globgm-heads-lower-layer-ss.tif: computed steady-state groundwater head [m] for the upper model layer
- globgm-wtd-ss.tif: computed water table depth [m] (sampled from upper to lower layer)

Transient Rasters 1958-2015
globgm-wtd-<date>.tif: computed water table depth [m] (sampled from upper to lower layer)
globgm-wtd-bot-<date>*.tif: computed water table depth [m] (lower layer only)

Finding and Exploring GLOBGM in the Community Catalog

The dataset can be downloaded from links provided in the source paper and also available in the Awesome GEE Community Catalog page here. The transient outputs were ingested into collection and the filename were parsed to get dates and these were then attached to the collection objects. Each image a global snapshot with monthly snapshots. I created five different assets representing the three steady state layers and two transient raster collections. The images dates that are sepearated by a month were added to the image collection as start and end dates.

Computed water table depth [m] (sampled from upper to lower layer) 1958-01-01

Bonus tip: Want to compare specific periods? Simply filter by date within the catalog. While visual analysis might not reveal major differences across large areas, the monthly timesteps in the transient data allow for in-depth, continuous analysis.

Montly Time series animation of WBT computed water table depth [m] (sampled from upper to lower layer) only plotted for 6 months starting at 2000-01-01

Feeling inspired? Check out the sample code provided and get coding! Explore the data yourself and uncover hidden insights into groundwater changes over the past 58 years (1958-2015).

Example code from Awesome GEE Community Catalog here

📬 Sign up for the newsletter so you can get exciting updates and insightful deep dives 🔍 like these delivered straight to your inbox 💌 and Substack app! 📱Get involved and help shape the future of the GEE Community Catalog? Let's connect! Reach out on Linkedin and Github to share dataset ideas, provide feedback, and join the conversation.

If you appreciate these efforts, please consider giving the GitHub repository a star ⭐️This simple gesture helps increase the visibility of our work and spreads awareness about the community catalog.

As always check the changelog for direct links and more updates.💡 Help out and become a sponsor of the community catalog and join me in building a vibrant community🌍