EDS 232: Machine Learning in Environmental Science

“Everything that civilisation has to offer is a product of human intelligence; we cannot predict what we might achieve when this intelligence is magnified by the tools that AI may provide, but the eradication of war, disease, and poverty would be high on anyone’s list. Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last.” – Stephen Hawking.

Course Schedule

* NOTE: Readings in the Course Schedule reference articles in Author (Year) format and textbooks with acronyms in italics. PDFs are in a Google Drive folder requiring a UCSB Google login for access to prevent copyright infringement. The full references are listed below under Readings.

Course Description

Machine learning enables predictions from big, complex data into actionable insights. It forms one of the foundations in data science. This course provides a broad introduction to machine learning and statistical pattern recognition with environmental data. Supervised learning (logistic regression, decision trees and neural networks) and unsupervised learning (clustering and ordination) techniques will be applied to the environmental tasks of mapping species distributions and ecological community analyses, respectively. Integer linear programming will be used to optimize biodiversity in planning a reserve. Deep learning techniques (convolutional neural networks) will enable species classification from photos and landcover classification from satellite imagery using Google Earth Engine. The first half of the course will use R and Rmarkdown on the desktop and second half will use Python and Jupyter notebooks in the cloud. This class teaches a broad sweep of advanced scientific programming skills for environmental problem solving.

Grading

The majority of points (100 total) will come from the labs. The Mid-Term exam will draw questions from lecture, lab and reading materials. For the Group Project you will select a Kaggle competition, submit an entry, and present on it for the last day of class. Participation will be based on class attendance.

Type Item Points
Lab 1. Species Distributions 20
Lab 2. Communities 10
Lab 3. Reserves 10
Exam Mid-Term 10
Lab 4. Species Images 15
Lab 5. Satellite Imagery 10
GP Group Project 20
Other Participation 5

Readings

Textbooks

Articles

Elith, J. and J. R. Leathwick (2009). “Species Distribution Models: Ecological Explanation and Prediction Across Space and Time”. In: Annu. Rev. Ecol. Evol. Syst. 40.1, pp. 677-697. ISSN: 1543-592X. DOI: 10.1146/annurev.ecolsys.110308.120159.

Evans, J. S., M. A. Murphy, Z. A. Holden, et al. (2011). “Modeling Species Distribution and Change Using Random Forest”. In: Predictive Species and Habitat Modeling in Landscape Ecology: Concepts and Applications. Ed. by C. A. Drew, Y. F. Wiersma and F. Huettmann. New York, NY: Springer, pp. 139-159. ISBN: 978-1-4419-7390-0. DOI: 10.1007/978-1-4419-7390-0_8.

Andrade, A. F. A. de, S. J. E. Velazco, and P. De Marco Júnior (2020). “ENMTML: An R Package for a Straightforward Construction of Complex Ecological Niche Models”. In: Environmental Modelling & Software 125, p. 104615. ISSN: 1364-8152. DOI: 10.1016/j.envsoft.2019.104615.

Zhong, S., K. Zhang, M. Bagheri, et al. (2021). “Machine Learning: New Ideas and Tools in Environmental Science and Engineering”. In: Environ. Sci. Technol. 55.19, pp. 12741-12754. ISSN: 0013-936X. DOI: 10.1021/acs.est.1c01339.

Oksanen, J. (2022). “Multivariate Analysis of Ecological Communities in R: Vegan Tutorial” , p. 43.