Learning Outcomes What you will know

Workshop Description

This workshop will cover the fundamentals of descriptive models and methods for exploring characteristics of time series. Attendees will learn how to decompose and analyze a time series’ secular trend, seasonal, cyclical, and irregular variability components; create descriptive additive/multiplicative models of time series; and explore characteristics of stationarity, autocorrelation, and cross-correlation with other time series. The workshop aims to investigate the practical application of Exploratory Data Analysis (EDA) using NOAA/GML climate change data, including factors like Annual Surface Temperature Change, CO2 concentration, and the frequency of climate-related disasters. The session will involve hands-on EDA using the R programming language.

Learning Outcomes

Grasp the process of decomposing time series utilizing climate change data available on the R-Studio server during the workshop.
Comprehend the exploratory techniques applied to time series analysis by breaking down climate change trends and variations to discern underlying causes and contributing factors.
Acquire practical experience using R as a tool for exploratory time series analysis.
Recognize the significance of exploratory data analysis in time series, acknowledging its pivotal role in achieving a more precise understanding of time series data.

Primary Outcomes

Understand the workflow for decomposing time-series using climate change data provided in the workshops R-Studio server.
Understand key methods for additive/multiplicative models of time-series.
Understand exploratory methods of time-series by decomposing climate change trends and variations to identify causes and factors.
Experience with R as a tool for exploratory time-series analysis.
Understand why exploratory data analysis of a time-series is a crucial tool for a more accurate understanding of data.

Secondary Outcomes

Explore evidence of climate change via time series analysis
- Climate change as evidenced by increasing global temperatures
- Solar activity does not explain increasing global temperatures
- A significant factor in increasing global temperatures is increasing CO2 concentrations in the atmosphere (greenhouse effect)
- Increasing CO2 concentrations are due to human generated CO2 emissions
- Increasing human generated CO2 emissions has increased frequency of climate disasters
- What is the expected impact of decreasing human generated CO2 emissions

You understand how to succeed in the HI-DSI Data Fellows program

You understand the goals, structures, and procedures for success as a participant in HI-DSI Data Fellows program.
Your computer and its software is adequate to support the experiences you will have in this program.
You know how to successfully onboard as an HI-DSI Data Fellow.

Referencing modules: Introduction to HI-DSI Data Fellows

You understand basic Python and Jupyter

You understand the basic syntax of Python.
You can install and use a Python interpreter on your computer.
You can install and use Jupyter Notebooks on your computer.

Referencing modules: Python and Jupyter

You can perform basic command line functions

You have an account on Koa and can login to it.
You can navigate the Koa linux environment using shell commands
You can create, copy, and delete files.
You can combine commands using pipes.
You can loop over commands
You can create simple shell scripts.
You can use the find and grep commands to locate files containing strings

Referencing modules: Scientific Software Basics

Assessed by: Help us assess this workshop

You understand basic HPC concepts

How to access Open OnDemand and use Jupyter Lab on Mana
How to request GPUs and utilize them for deep learning tasks
The performance difference between CPU and GPU for deep learning
Different types of file system found on HPC clusters
How correct file system choice affect deep learning performance

Referencing modules: High Performance Computing

Assessed by: Help us assess this workshop

You understand the challenges and options in moving scientific data over the network

Demonstrate understanding of advantages and disadvantages of various bulk data transfer tools.
Understand appropriate solutions for a data transfer workflow, taking into account current computing, storage and network infrastructure, and end-point sites.
Demonstrate the ability to identify, communicate, and mitigate potential bottlenecks in collaboration with campus cyberinfrastructure and network operators.
Transfer data using Rclone.
Transfer data using Globus.

Referencing modules: Data Movement

Assessed by: Help us assess this workshop

You understand basic data wrangling concepts

Understand how to use notebooks to clean and wrangle climatological data
Understand how to combined code and documentation to allow for reproducible analyses

Referencing modules: Data Wrangling, Part 1, Data Wrangling, Part 2

Assessed by: Help us assess this workshop, Help us assess this workshop