Course materials for the Data Science connector course PyEarth: A Python Introduction to Earth Science being taught in the Earth and Planetary Science Department at UC Berkeley in Spring 2023 by Nick Swanson-Hysell (https://github.com/Swanson-Hysell).
If you have a UC Berkeley account, you can launch these materials on Datahub using this link: https://datahub.berkeley.edu/hub/user-redirect/git-pull?repo=https%3A%2F%2Fgithub.com%2Fds-connectors%2FEPS88-Spring-2023&urlpath=tree%2FEPS88-Spring-2023%2F
The contents of this repository are licensed for reuse under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.
| Meeting Date | Topic | Learning objective | Technical skills |
|---|---|---|---|
| Week 1 : Jan 23 | What is your birthquake? | Be introduced to using the Jupyter notebook environment and using it to access and inspect data. | Defining variables and accessing tabular data through an API. Use Python code within a Jupyter notebook to access an online API to pull in and inspect tabular data using pandas. |
| Week 1 assignment | Maps | Develop an understanding of map projections and visualizations of geospatial data. | Map generation. Learn to generate maps using cartopy using different projections and to plot locations. |
| Week 2 : Jan 30 | Earth's continents and oceans | Compare the distribution of Earth's topography (hypsometry) to a normal distribution. Explain the distribution in terms of continental crust vs. oceanic crust density. | Simple statistics and data visualization. Use numpy to load data and do simple statistical calculations such as np.mean. Use the matplotlib to visualize distributions of data as histograms and cumulative distributions. |
| Week 2 assignment | Seafloor age and plate tectonics | Use global seafloor age distribution to make inferences about plate tectonic processes. | Tabular data wrangling and visualization. Get further introduced to the pandas library. Use cartopy functions to visualize geospatial data in order to identify patterns. |
| Week 3 : Feb 6 | Global earthquakes and plate tectonics | Illuminate plate boundaries by plotting global earthquakes. Visualize a subducting plate by examining earthquakes in the Andes. | Data wrangling and visualization. Continue use of the pandas library as well as matplotlib and cartopy functions. |
| Week 3 assignment | Global earthquakes and seismograph interpretation | Make connections between earthquake magnitude/depth and plate boundary type. Analyze a seismograph and relate differential wave travel times to distance using established time-travel curves. | Continued data wrangling and visualization; calculations. Continued use of the pandas library as well as matplotlib and cartopy functions to visualize data. Use Python to make calculations related to Earthquake distance. |
| Week 4 : Feb 13 | Seafloor spreading and crustal magnetism | Use bathymetry data in combination with marine magnetic anomaly data to evaluate seafloor spreading. Interpret the distance from the spreading ridge where the last geomagnetic reversal happened across different spreading ridges and use this distance to calculate spreading rate. | Data filtering and detrending Indexing, filtering and cleaning data using pandas. Declare a function to detrend and filter magnetic anomaly data. |
| Week 4 assignment | Seafloor spreading | Use expanded age pick data sets from multiple ocean basins and determine seafloor spreading rates through applying linear regression. | Correlation and regression Use the numpy polyfit function to develop linear least-squares regressions for data and the polyval data to use these regressions to model a process for two sets of measurements. Use the scipy stats.linregress function to calculate related summary parameters. |
| Week 5 : Feb 20 | No class (President's Day) | ||
| Week 6 : Feb 27 | Probabilities applied to Earth and planetary processes | Discuss different statistical distributions and the concepts of theoretical versus empirical distributions. Consider probability distributions applied to meteor showers and earthquake occurence and the geomagnetic field . | Monte Carlo simulation Use the module numpy.random to generate examples of simulated data sets. |
| Week 6 assignment | Earth's reversing magnetic field | Earth's geomagnetic polarity timescale. Consider the time series of geomagnetic reversals and how reversal frequency could be represented with probability distributions. | Implement additional probability distributions using scipy including the gamma distribution |
| Week 7 : March 6 | Ice sheets | Use directional data to make inferences about flow directions of the Cordillera ice sheet | Bootstrap resampling Utilize bootstrap resampling as a means make inferences about data. |
| Week 7 assignment | Bay Area wind | Gain insight into Bay Area climatology as a function of season through analysis of hourly wind data at SFO | Apply the bootstrap percentile method Utilize bootstrap resampling to develop estimates of the 95% confidence interval associated with statistical summary parameters. |
| Week 8 : March 13 | The age of the universe and the age of the Earth | Use the retreat velocity of galaxies and supernova as a function of their distance to gain insight into the Hubble constant and the age of the universe | Regression Determining regression lines through least squares to gain insight into data and enable prediction |
| Week 8 assignment | Atmospheric CO2 | Use the Mauna Loa Observatory atmospheric carbon dioxide concentration data set to quantify the rate of CO2 rise over the past 60 years. Compare trends inferred from regression to RCP emission scenarios | Regression Assess residuals in order to evaluate regressions. Use regression models for prediction. |
| Week 9 : March 20 | Ice core paleoclimate archives | Determining rate of change of atmospheric CO2 levels from merged ice core data sets | Data wrangling Gain experience wrangling data from different formats for analysis. Get set-up with a local Python installation. |
| Week 9 assignment | Ice core CO2 | Compare rates of change in the ice core and Mauna Loa CO2 datasets | LOWESS Apply locally weighted scatterplot smoothing using the scipy statsmodels in order to develop non-parametric fits to data. Use these fits to gain insight into rates of change. |
| March 27 | No class — Spring Break | ||
| Week 10 : April 3 | Glacial-interglacial cycles | Investigate time series data to gain insight into the drivers of glacial-interglacial cycles | Conduct spectral analysis Use tools within the scipy package in order to calculate power spectral density from time series data |
| Week 10 assignment | Paleoclimate proxies and the mid-Pleistocene transition | Evaluate when there were changes in the period of glacial-interglacial cycles | Continuing conducting spectral analysis to assess the dominant frequencies within data sets using scipy with functions such as signal.periodogram() |
| Week 11 : April 10 | Lava flow geochemistry | Explore compositional differences between mafic, intermediate, and felsic volcanics rocks | Visualizing distributions of data and comparing populations Apply different techniques to visualize and analyze distributions of data applying tools from the seaborn library including kernel density estimates. |
| Week 11 assignment | Cobalt resources | Evaluate what lithologies have the highest concentrations of cobalt to gain insight as pertains to mineral exploration | A/B testing Gain further experience visualizing distributions of data and conducting A/B testing to evaluate the significance of compositional differences |
| Week 12 : April 17 | Tectonic setting of volcanos | Assess differences in the geochemistry of basalts based on the tectonic setting in which they formed. Evaluate whether these data give the ability to confidently differentiate between settings | Classification algorithms Use scikit-learn library to apply SVC and Decision Tree supervised machine learning algorithms to categorized igneous geochemistry datasets |
| Week 12 assignment | Tectonic setting of volcanos continued | Utilize more recent data rich compilations to further evaluate discrimination between tectonic setting on the basis of basalt geochemistry | Gain further experience using scikit-learn to split data into training and test sets in order to train and evaluate classifiers |
| Week 13 : April 24 | Geodetic data | Review concepts while analyzing a GPS data set | |
| Week 13 assignment | Final project plan | ||
| Finals week | Presentations on final projects during exam period |