Welcome!
Quick Facts About METEO 820
Course Overview
METEO 820 is a professional, graduate-level course offered by the Department of Meteorology. Analytics is the discovery, interpretation, and communication of patterns by way of statistics, computer programming, and applications that provide decision-making tools. Weather and Climate Analytics (WCA), in particular, uses weather and climate data to provide useful information for improving efficiency, mitigating risks, and optimizing productivity. Why care about WCA? Well, weather and climate affect everyone! Combining weather forecasts and climate predictions with analytics, companies can create better product placements, select efficient transportation routes, provide more lead time on health advisories, and much more.
This course provides practical guidance in the quantitative analysis of large weather and climate time series datasets for incorporation into an analytical modeling system. Students will learn a variety of methods for identifying key temporal patterns in atmospheric datasets, modeling methods based on patterns, trend analyses in climate datasets, advanced modeling methods, frequency domain analyses, and spatial-temporal visualization techniques specific to meteorology. Furthermore, data reduction techniques will be discussed for working with big weather and climate datasets. Specific emphasis will be placed on preparing environmental data for analysis, data visualization techniques, correctly selecting appropriate analyses, validating results, and realistic interpretations of results. Case studies will augment the discussion on the various time series methods, with the goal being to broaden the student’s perspective on the use of weather and climate data for forecasting and modeling as it pertains to decision making.
What will you learn in this course?
Meteo 820 seeks to provide guidance on methods used for decomposing, analyzing, and modeling environmental time series data. After successfully completing this course, you will be able to:
- break down a climate time series into parts and identify key patterns and trends for forecasting purposes;
- create weather models based on patterns using basic and advanced methods, as well as apply hypothesis tests to determine the appropriateness of each method;
- convert environmental data between the time and frequency domains when applicable, apply the correct analyses for each case, and interpret the results;
- incorporate strategies to increase computational efficiency and quantifiably reduce the number of weather and climate variables for analysis;
- visualize big data, from weather and climate observations or models, and creatively display outcomes in a manner that effectively communicates the results for decision-making purposes; and
- develop an appreciation for the extensive collection of weather and climate datasets freely available and the ability of these datasets to provide useful information to decision makers across industries.
The lessons that comprise this course are:
Lesson 1: Introduction to Time Series Data (overview of important aspects in atmospheric time series analysis, types of time series in observational weather and climate datasets (stationary, periodic, quasi-periodic, and chaotic), common patterns in weather and climate data, preparing observational data for time series analyses)
Lesson 2: Autocorrelation and ARIMA (autocorrelation techniques for analyzing meteorological patterns and seasonality in weather and climate data, autoregressive models, moving average models, ARMA and ARIMA models for environmental datasets, explanations of the difference between models)
Lesson 3: Trend Analysis (components of a trend, testing for a trend, performing a trend analysis on climate data, determining statistical significance and confidence intervals of trends in various climate observations and models)
Lesson 4: Autocorrelated Distributed Lag (ARDL) Models (overview of ARDL model and components, execution of the ARDL model on weather and climate data, hypothesis testing for significance of ARDL model)
Lesson 5: Spectrum and Cross-Spectrum Analysis (when and why to perform a spectral analysis of environmental datasets, key requirements, execution of a spectral and cross-spectrum analysis for various weather and climate data)
Lesson 6: Hovmöller Diagrams (overview of Hovmöller diagrams specific to meteorology, creation and interpretation of various environmental data sets)
Lesson 7: SOM (examples of data reduction and exploratory tools for big weather and climate data, execution methods and interpretation of results using additional time series analysis tools)
Lesson 8: Data Visualization (examples and best practices for presenting time series of weather and climate data, plotting considerations, creation of own figures and presentations of weather and or climate data used in course project)
How does this course work?
As with most graduate courses, there is a considerably higher onus on you to take responsibility for your own learning. While lessons present guidance on what you need to learn, much of your actual learning will take place as you engage in directed research and experiment with various examples presented in the text. Following through on these examples and exploring various ways to accomplish prescribed, data-procurement tasks are an absolute necessity, not only to be successful on the lesson's assessment activity, but to meet your own learning goals as well.