PSC 6123/4123
Embedded Files
Course Introduction
Course Introduction
The major theme of this course wraps around recorded lectures made at the National Central University. Most examples shown in the lectures were based on GrADS. See the advantages and disadvantages of GrADS.
There are multiple pages - select page number at the bottom.
There are multiple pages - select page number at the bottom.
Introduction 1
Introduction 1
What "climate data" are and where the data come from. And...introduction to the basic drivers of our climate system.
Introduction 2
Introduction 2
Ocean and wind currents. First, install OpenGrADS - click "grads2" and pick the latest version. Read documentation.
Install GrADS (or OpenGrADS) software.
OpenGrADS gives you a bigger suite of tools.
Traditional GrADS may integrate better with MacOS.
Mac users: You need to install quartzX first.
How to install GrADS in MacOS?
If you have other issues, try to find answers here.
You may use some basic unix commands to facilitate the operation of GrADS
Important: Please read all the text and click all the links listed on the webpage. It may not be as organized as I'd like but all the information is here, packed into one webpage.
If you encounter any issues with installing GrADS or executing commands, please refer to this Q&A document for assistance and solutions. (Special thanks to our wonderful TA of 2022, Lucy Chien)
If you encounter any issues with installing GrADS or executing commands, please refer to this Q&A document for assistance and solutions. (Special thanks to our wonderful TA of 2022, Lucy Chien)
ClimateDataAnalysisQA.docxLesson 1
Lesson 1
The atmospheric structure and pressure coordinate (important information about handling atmospheric data).
The global "reanalysis" data - concept and history.
Data link: the global precipitation climatology used in the end of lecture
The next video introduces different precipitation data sets, including station-based, objectively analyzed/gridded, and computer-generated precipitation, for both global and regional coverages.
Lesson 2
Lesson 2
Learn to access the public data resources provided by the U.S. government.
Plotting of basic climatic variables: e.g., precipitation.
_____________________
Create a folder with your name.
Upload your screenshots of exercises into the corresponding folder. Make your filename "name_#.png", like Simon_01.png
If you have never read a contour plot, here is an explanation:
See also the recap on "data structure" in the first five minutes of this video.
The next video shows similar precipitation data set with two different spatial resolutions, 2.5-deg vs. 0.5-deg over land.
For further information about the definition of time, unit, areal averaging, and getting multi-year data for the time series (or histogram), watch this video for clues.
Bonus tips!
Bonus tips!
In Lesson 2, I used a student's laptop to build a script that offers "click and display" functionality; this enables you to mouse over an area of the data map and click any location you want to display the point-wise data time series from that location. It is an interactive way to survey regional climate from anywhere with data coverage. In the following video, I re-created this script with opendap data; watch it and practice.
Technical note:
Technical note:
If your Windows OpenGrADS does not open NOAA NOMADS links (for accessing forecast data), try the following options:
Did you install the correct OpenGrADS version? It should be grads-2.2.1.oga.1-win32_superpack.exe. in OpenGrADS webpage (not gv32.exe in GMU GrADS webpage). This is because most of the OpenDAP files have been updated to NetCDF-4 format, the gv32.exe version may be too old to access OpenDAP.
If you installed the most recent version, then try to install an older version, e.g. 2.0.2 grads-2.0.2.oga.2-win32_superpack.exe
If the above did not work, and you happen to know Linux, then there is a complicated way: You can install WSL (Windows Subsystem for Linux) to simulate the Linux environment and install opengrads in WSL. Check this link http://gradsusr.org/pipermail/gradsusr/2019-February/042135.html
Lesson 4
Lesson 4
Satellite precipitation: infrared and water-vapor based w/ global coverage. The link to this data set is https://ftp.cpc.ncep.noaa.gov/precip/CMORPH_V0.x/
Data header: The key information telling GrADS how to read the data
Concept of data assimilation & introduction to the various global reanalysis data sets
_____________________
NOTE: the original path to CMORPH has changed. Watch the next video to access the data and .ctl.
Upload screenshots of the homework into the corresponding folder.
Many of you may have trouble accessing CMORPH data or displaying the hurricane as shown in the lecture. If so, watch this video for a detailed explanation of the CMORPH data .ctl file, alongside a student demonstration of plotting the 2019 hurricane Dorian using sea level pressure, surface winds, and CMORPH data ➙
The Climate Prediction Center of NOAA produced a very nice presentation (PDF) that teaches using GrADS and it specifically demonstrates how to read CMORPH precipitation data. Click and go through it - there are a lot of good stuffs!
If ( ˘︹˘ ) despite your best effort, you still can't make OpenDAP or CMORPH work, don't be discouraged.
Try to plot this "cool" debris tracking over the Pacific Ocean by using the OpenDAP data with GrADS.
This one will work, I promise... ≧◠‿◠≦✌
What makes a NETCDF file self describing?
Metadata is often referred to "data about the data." Metadata includes things like:
- The data and time a file was create
- dWhat data is in the fileUnits of variables within the file
- Information about the coordinates (e.g., lat, lon vs. x,y)
- If the data file is for model output it might include: resolution, parameterizations, time step
Metadata is super important!
Programming tips:
GrADS Commands Quick Reference Card (print for quick view)
Scripting Language Quick Reference Card (print it out!)
Daniel Lee, Software & Data Format Engineer at EUMETSAT guides you through a variety of free and open software, for you to visualize NetCDF format files. This video is divided in 4 different Chapters, feel free to skip to the following timestamps:CHAPTER 1: Viewing Data with Panoply – 4:31CHAPTER 2: Making Maps with QGIS – 9:40CHAPTER 3: Troubleshooting coordinate systems – 12:45CHAPTER 4: Workflow automation with Python – 23:15
Lesson 5
Lesson 5
Atmospheric variables: Learn to plot 3-dimensional data sets for the climate system.
Exercise: creating the 3-cell general circulation system using the global reanalysis data.
Never heard of the "general circulation"?
Well, watch this video about the 3-cell structure of the atmosphere in the Earth's climate system.
Then, watch this video ➤
to build the following concepts:
- Programming with loops ("while loop" in GrADS)
- Script language vs. commands
- Open and plot multiple variables at once
- Plotting vertical cross section
And...here is the final script shown in the end of the video demo.
Finally, watch this class recording to learn how to use the reanalysis data to plot the Earth's general circulations in a 3-dimensional sense.
If you encountered problems plotting the 3-cell general circulation exercise, then you need to watch this video that demonstrates the basic of plotting 3 variables in one script, zonal means, and a cross-section.➙
Lesson 6
Lesson 6
Plotting the monsoon systems. The largest monsoon in the world is the Asian monsoon, also referred to as the Indian monsoon. Watch these two videos to get a general concept of the monsoonal circulation.
Lesson 6.1
Lesson 6.1
My original lecture recording is lost. Here are a couple of introductions about the monsoon, including Prof. Ullrich's lecture on tropical Meteorology.
Tropical climate - the monsoon systems. Understand what drives the large-scale monsoonal circulation that shapes the climate for 1/3 of the world's population.
_____________________
Too much? Check out the next 4-min video about the monsoon system.
Lesson 6.2
Lesson 6.2
Regional climate: the West African monsoon system with a marked seasonal progression.
Understand the concurrent variations of the atmosphere and precipitation.
_____________________
Take this quiz about Data Assimilation and its role in the modern reanalysis data.
Next topic: Weather Forecasting Models
Next topic: Weather Forecasting Models
Watch these two videos of numerical weather prediction before going to Lesson 7.
Lesson 7
Lesson 7
Weather forecasting, the numerical (computer) way!
Processing of forecast data
Forecast verification
(no homework for Lesson 7)
Lesson 8
Lesson 8
Numerical weather prediction model - the basis of climate projections
Forecast verification: Case analysis for a tropical cyclone. However, it was an old case and the forecasts used in the video are long gone. Instead, go to this webpage and access the most recent forecast from the 1-deg GFS model.
Here is the recap of my instruction (Oct 5, 2021).
Up to this point, the topic of using Artificial Intelligence for weather prediction cannot be ignored. To address this, I have created an introductory video based on another expert's presentation.
AI talk - future of weather prediction
Exercise using a hurricane case
Exercise using a hurricane case
Forecast data only are archived for the most recent weeks, so you can't find Hurricane Maria. But you can practice it on current weather systems.
Here I show the script that accesses GFS winds and evaluates its forecast of Hurricane Maria (2020).
This video builds a script that plots GFS precipitation and its forecast of Hurricane Maria (2020).
There are other forecast models released by the USA other than the global model (i.e. GFS). Additional models include some regional models, air quality models, ocean models and climate forecast models. The next video introduces you where to access those model forecasts.
Catching the moment:
Catching the moment:
Precipitation forecast for Taiwan during a monsoon event (note: the exercise began from the end of Lesson 16)
Here, we follow the steps laid out in the lecture video and produce a verification for the September 8 BIG WIND event in Utah. Refer to Lecture 9 about changing the domain setting for Utah and the video "handling Grib2 and forecast data" next.
Here, we follow the steps laid out in the lecture video and produce a verification for the September 8 BIG WIND event in Utah. Refer to Lecture 9 about changing the domain setting for Utah and the video "handling Grib2 and forecast data" next.
I also included all the data and scripts HERE.
🙋🏻♂️ Handling past forecast data is not easy. It requires processing "grib2" format of the data. I made a video that shows step by step how to find the data, decode grib2 using a Perl script, use "gribmap" function, and write a script to display the spread of the forecast...and...how to line up older forecast with newer forecast.
This skill is important when it comes to analyzing climate projection data such as the IPCC models (CMIP5, CMIP6...)
watch this video to learn how-to
Programming the geostrophic wind:
Programming the geostrophic wind:
the balance between pressure gradient and Earth's rotation
the balance between pressure gradient and Earth's rotation
Equations of the geostrophic wind
Equations of the geostrophic wind
If you are mathematically minded...
Computing the geostrophic wind
Computing the geostrophic wind
A coding exercises using reanalysis data
Beginning of climate analysis:
Beginning of climate analysis:
composite, correlation map, indexing
composite, correlation map, indexing
Lesson 10.1
Lesson 10.1
After you practiced making the composite of climate anomalies based on ENSO events, you can compare notes with this NOAA website that produces ENSO composites on different variables - https://psl.noaa.gov/enso/compare/
After you practiced making the composite of climate anomalies based on ENSO events, you can compare notes with this NOAA website that produces ENSO composites on different variables - https://psl.noaa.gov/enso/compare/
Homework:
produce two comparing maps as follows:
Practice the definition of sea surface temperature index (like the Niño 3) and write a script to produce composite maps based on the Niño-3 index.
*The first minute of the video goes through last homework of La Niña precipitation anomalies.
Lesson 11
Lesson 11
Plotting of different climate indices - the ENSO mode.
Differentiating ENSO and regional climate impacts.
Lesson 12
Lesson 12
Other modes of atmospheric "teleconnection": Why does climate vary?
Large-scale modes of climate variability and oscillations
Plotting beyond weather variables
Plotting beyond weather variables
NOTE: The global NDVI data as illustrated in the following lessons is moved (new link). However, one alternative analysis is to replace the vegetation data with soil moisture (or evaporation) and compare it with precipitation in the same manner. A short video explaining how to i) obtain data, ii) rearrange data, and iii) modify to class script to plot them is given in Lesson 14.2. Watch Lesson 13 ans 14.1 first, and then watch 14.2 to practice, including other analyses not shown in the 14.2 video.
All the data and scripts are provided in the Box folder, though you are encouraged to build them up from scratch.
Lesson 13
Lesson 13
Plotting of the African climate system: mean and departure
Plotting of normalized difference vegetation index (NDVI); here is the NDVI data link.
Can't get the "color.gs" script to work? Watch this video! (for Windows and Mac OS)
When using GrADS to process monthly-interval data for year-to-year correlation analysis, the data structure needs some manipulation. Here I explain why and how:
Next, Siiri walks you through the codes reordering the time dimensions in preparation for making the one-point correlation map with yearly intervals.
Confused by tcorr, scorr, and tlag?
Confused by tcorr, scorr, and tlag?
Watch this video for the explanation about the difference between temporal correlation, point-by-point correlation, and spatial correlation.
If you can't find fcorr.gs or make it work, here is the explanation - from the Dr. Bin Guan:
Lesson 14.2
Lesson 14.2
To continue the exercise, you can replace the NDVI data with soil moisture. This demonstration will show you how to obtain soil moisture and precipitation data and conduct the same analysis as shown in the previous lecture videos. This video has no narratives, just soothing music :-)
All right, that was too fast 😛
All right, that was too fast 😛
So here is a slower and more detailed recap of the script building and plotting, demonstrated by Matthew.
Signal process: frequency-domain analysis
Signal process: frequency-domain analysis
If the concept of wave decomposition is hard to grasp, then this video may help you visualize how each wave component forms a complicated time series.
And if visualization alone doesn't help you understand wave decomposite, then maybe adding sounds will. Here's a video from Dr. Chris.
Lesson 15
Lesson 15
Signal process: time series analysis practiced on an ENSO index and a tree-ring reconstruction of precipitation in northern Utah (hyperlink goes to the data source).
Note: the webpage that computes the MTM wavelet in the lecture is no longer available. However, you can find alternative analysis tools at Prof. Michael Mann's webpage.
Besides using GrADS, there are also free software on a webpage that can help you process a time series, such as what's shown in this example.
Supplmental material
Supplmental material
Introducing the EOF analysis (see link), a common climate diagnostic tool. Unfortunately, the EOF function introduced in this lecture is no longer available. There is a new function (link) that may allow GrADS to perform the EOF analysis, if you know how to install it :-)
Further reading about the EOF analysis, and a video about the Principle Component Analysis PCA.
Bravo! NCU grad student S.-S. Lin demonstrated how to make the EOF work again in GrADS!
you will love it!
Boundary layer of the atmosphere
Boundary layer of the atmosphere
Lesson 16
Lesson 16
Analysis of the atmospheric property: the case of temperature inversion and boundary layer evolution
The next video teaches you how to show boundary layer conditions using forecast model outputs.
Here, two students demonstrated the plotting of atmospheric instability based on numerical weather prediction data.
Lesson 17
Lesson 17
Continuing on the inversion and mountain valley wind analysis. Access valley wind data and codes here. Watch the following video for accessing the station data and plotting it.
Here, NCU grad student S.-C. Chien demonstrated how to plot the station winds as shown in the recorded lecture. Good stuff!
Climate projection data!
Climate projection data!
IPCC climate model projections and downscaling
Lesson 18
Lesson 18
Entering the IPCC Climate Report and associated climate models.
What is IPCC AR6? Read https://www.ipcc.ch/assessment-report/ar6/
Lesson 19
Lesson 19
Concept of climate downscaling and bias correction. Class example: future projection of the mighty Colorado River
Assessing the massive data of downscaled climate projections can be tough.
Check the script-writing exercises given by Homa and Simon below.
Accessing downscaled climate data
Accessing downscaled climate data
re-taught by Homa
CMIP5 downscaling & projection
CMIP5 downscaling & projection
re-taught by Simon
Homework: Plot the global-averaged annual-mean air temperature using observational data and CMIP5 data by following this style:
After you have learned the basics of programming and the data structure of most climate data sources, it's time to re-visit those free software available online.
After you have learned the basics of programming and the data structure of most climate data sources, it's time to re-visit those free software available online.
It is my belief that, without having learned the programming concept and data architecture, you will be as lost in these free software websites as anyone else.
It is my belief that, without having learned the programming concept and data architecture, you will be as lost in these free software websites as anyone else.
That brings you to Lesson 20, graphic-based analysis tools.
That brings you to Lesson 20, graphic-based analysis tools.
Lesson 20
Lesson 20
Introduction of webpage-based climate data analysis tools, including displaying, composite, correlation, and animation with click-N-play functionality.
Introducing a useful software developed by NASA: Panoply - it reads NetCDF data smoothly and produces high-quality images.
Why do coding when you can click-N-play? 😏
Here is a demo about using NASA's Panoply software to plot GFS forecast for T2m and Precip from one location, as a time series. Access this webpage to download and install Panoply - https://www.giss.nasa.gov/tools/panoply/
There are other website-based data tools. This next video illustrates Reanalyses.org and through it, two handy websites: Climate Reanalyzer and KNMI Climate Explorer. Click to watch the demo!
This is a repeated demo of using KNMI Climate Explorer to plot and filter a time series / climate index.
Projection data? No problem! Here I show you how to use KNMI Climate Explorer to plot the time series of a CMIP5 variable, the evaporation. You can apply it to other variables and models.
Another demo of using KNMI Climate Explorer to plot the time series of CMIP5 variables, including temperature and evaporation. This nice video was produced by an NCHU international student.
Back to Climate Reanalyzer - a short demo to quickly plot time series from climate projections.
Here is a cool NASA software called "G.Projector" and it can conform map projections from scientific papers or web images.
.
.
.
Final Project Presentations!
Final Project Presentations!
All the exciting projects from Class 2021
.
Check out the fun projects of Class 2020!
Check out the fun projects of Class 2020!
Page updated
Google Sites
Report abuse