Global Geomorphic Process Domain Map

Harrison K. Martin and Michael P. Lamb

Division of Geological and Planetary Sciences, Caltech

Background

We made a map of Earth’s geomorphic process domains. There is a paper associated with the map: 

Martin, H.K. and Lamb, M.P., (accepted). The Earth is mostly diffusive: a global analysis of landscape evolution: Science Advances.

The paper describes a spatially continuous high-resolution (250 meter) global map of Earth’s geomorphic process domains – the dominant set of sediment transport functions that predict how Earth’s surface will evolve over the coming 10²-10⁵ years. Each land pixel is classified as one of eight process domains. The ocean is shown as transparent.

There are two ways to access the data. The first is to view it using this webmap, and the second is to download and work with it using our figshare repository (below).

Webmap

For casual or reference use, this webmap is the best way to quickly look at the data. First, it requires no special software or technical knowledge to view the data, whereas the original dataset (below, via figshare) requires QGIS, ArcMap, or similar. Second, accessing the data is far more efficient: similar to Google Maps, it transfers only the required data to users and renders it quickly at only the necessary resolution. This makes it good for users with limited data speeds or bandwidth caps. The downside is that, as is standard for webmaps, the data are rendered in the Web Mercator projection. Unlike the Equal Earth projection used to create our dataset, the webmap does not preserve equal area (making higher latitudes look bigger and lower latitudes look smaller). The other major issue is that, when zoomed in far enough, the webmap contains reprojection errors where square pixels have been distorted and resampled, that do not exist in the original dataset. For this reason, the online map is intended for quick reference, but all actual data analysis has been (and should be in the future!) done on the original dataset as available through figshare below.

Download data (via figshare)

The data are available as a series of 12,000 by 12,000 pixel tiles in the Equal Earth projection. Each tile is optimized for download and use, as described in the figshare repository and in the supplementary information of the manuscript. We also included the code used to generate the map, and all of the intermediate processed input datasets, so that you can reproduce it at home with the click of a button. All files are available at the following figshare link:

https://doi.org/10.6084/m9.figshare.29902013

Methods

In brief, we created this map by harmonizing a series of interpreted, remotely sensed, mapped, and machine-learned global data products. All datasets were downloaded, rasterized to 250 m pixels, and unified to a single projected coordinate reference system. There were many datasets used to refine the final results, but the overarching philosophy was to leverage previous work by experts in each geomorphic process domain in a way that emphasizes the dominant processes for each location. Flatlands are the “everything else” category, marking places on Earth that lack obvious active sediment transport features and are <2.5 degrees steep. The full methodology, citations, and error estimations are provided in the paper and supplementary information.

Citation

If you use the data in your own research or projects, please include references to the paper and to the figshare dataset.

Contact

Please feel free to reach out at hkm@caltech.edu, or via the contact form at https://harrison.studies.rocks. Thanks for stopping by!