In this section, I will be sharing the teaching materials I have been creating and using during my classes. It can include commented R scripts or slides for presentation.
MIT Course 1.845: Terrestrial carbon cycle and ecosystem ecology
Graduate course led by Prof. Cesar Terrer at the Department of Civil and Environmental Engineering, Massachusetts Institute of Technology. The course included lectures, paper discussions as well and practical sessions. The course targets graduate students interested in earth systems and environmental sciences, willing to acquire basic skills and knowledge about the terrestrial carbon cycle and global ecosystem ecology.
Lecture
This lecture emphasizes the role of nutrients as a cofactor of the land C sink as well as an important variable in understanding ecosystem functioning.
CONTENT AND INTENDED LEARNING OUTCOMES
The PDF presentation includes:
-
From molecules to ecosystems
-Explain the importance of nutrients in plants and how they affect biological processes
-Describe the concept of nutrient limitation and nutrient resorption -
Global scale patterns
-Describe the biogeochemical cycles of nitrogen and phosphorus
-Enumerate N and P cycle differences and how those translate to a global scale -
Biogeochemical disruption in the Anthropocene
-Describe how humans are affecting biogeochemical cycles and their implications in natural ecosystems
-How N, P, and nutrients in general affect the land C sink
Practical sessions
DESCRIPTION: Their goal is to provide students with tools and good practices to face data analysis applied to earth sciences. The students will practice those skills by using the provided data and following the process of data curation, data analysis, and interpretation. The students are going to learn the basics of R programming, the highlights of C cycle modeling and modeling in ecology, how to work with maps and geographical information systems, the basics and how to use data synthesis techniques such as metanalysis, and how to use machine learning to upscale the results. No previous knowledge of programming is required.
CONTENT and INTENDED LEARNING OUTCOMES
1 - Introduction to R
a. Intro, data types, vectors, matrices, data frames, and lists
b. Import and export data, play with data, and basic statistic models.
c. Loops, functions, good manners, and tricks
Section available on OpenCourseWare or YouTube
Describe what R is and its characteristics, the different types of data, and data structures. Be able to use R to load and store data and use essential functions such as “subset”, “unique”, “hist”, “colnames”, “order” or “cbind”. Create and interpret basic linear models. Formulate simple loops and functions and familiarize yourself with good practices to maintain a tidy workflow.
2 - Modeling the C cycle
a. Introduction to land systems modeling
b. One-box model Special acknowledge to Prof. Benjamin Stocker
c. Two-box model Special acknowledge to Prof. Benjamin Stocker
Explain the uses of C modeling and its scientific implications. Use a one-box carbon model and a two-box carbon model in R.
3 - Geographical Information Systems (GIS)
a. Lecture with basic concepts
b. Vectorial maps, rasters, projection and resolution concepts, crop, stack, and brick
c. Extract information from maps using spatial data points
Section available on OpenCourseWare or YouTube
Explain the differences between raster and vectorial maps and the importance of the resolution and projection concepts. Use R to read and visualize raster and vectorial maps, and change projections and resolution. Use R to create stacks and crop maps.
4 - Metanalysis - Cesar Terrer
5 - Machine learning algorithms
a. Introduction
b. Subsets selection
c. Select the best model
d. Model implementation and tunning
e. Visualization and upscaling
Discuss the advantages and limitations of machine learning algorithms. Create subsets for the train-test-validation and explain why they are necessary. Choose the best machine learning model based on your data. Tune, display, and understand the model.
6 - Final project, summary, and conclusion