ecoscape-earth

EcoScape

The EcoScape project aims at modeling animal (and perhaps, in the future, plant) habitats, study their connectivity, and more.

Publications

• Natalia Ocampo-Peñuela, Coen Adler, Artie Nazarov, Jasmine Tai, Natalie Valett, Luca de Alfaro. Diffusion Models For Species-Specific Functional Habitat Connectivity. Technical Report UCSC-SOE-23-10, 2023. Reproduction material.

Code Repositories

• scgt, the UC Santa Cruz Geographical Toolkit, contains functions for working with geotiff files.
• ecoscape-layers contains functions for generating habitat and landcover layers for bird species.
• ecoscape-connectivity is the tool for computing habitat connectivity.

The above code can be installed via pip, e.g., pip install ecoscape-connectivity.

Sample Colab Notebooks

Some of the EcoScape software (such as connectivity computation) benefits from GPUs. On Google Colab, you can use GPUs for free, or at low cost. The following Python notebooks, on Google Colab, demonstrate the EcoScape tools:

• EcoScape Connectivity Computation. The notebook shows how to run the EcoScape Connectivity computation given input habitat and landcover layers.
• EcoScape Layer Generation and Connectivity Computation. The notebook does the process of studying the connectivity for a species from beginning to end: it computes habitat and landcover layers from IUCN data, and analyzes the connectivity using the EcoScape algorithm.

Project Members

(In alphabetical order)

Faculty

• Luca de Alfaro (luca@ucsc.edu)
• Natalia Ocampo-Peñuela (nocampop@ucsc.edu)
• Tyler Sorensen (tyler.sorensen@ucsc.edu)

Contributors

• Coen Adler (ctadler@ucsc.edu)
• Artie Nazarov (anazarov@ucsc.edu)
• Jasmine Tai (cjtai@ucsc.edu)
• Ian Terry (imterry@ucsc.edu)
• Natalie Valett (nvalett@ucsc.edu)