Climate Data

with open tools

- learn to decode climate data without having to write codes

(for better viewing, use a laptop)

This course provides an introduction to the sources of climate data used in climate diagnostics and prediction. The topics covered include data visualization, simple statistical analysis, and future climate projections. Students will gain practical skills in applying these methods to real-world climate data sets.

The course is designed to accommodate students with diverse backgrounds who are interested in gaining hands-on experience with real data. No prior knowledge of climate science is required.

I have provided two recorded lectures as an introduction to this course. Most of the content will apply, except that we won't need to install any software (i.e., GrADS) or require you to write any programming code. The interfaces introduced here are mostly click-N-play and graphical.

This class is designed for students from various backgrounds, providing essential skills in handling data, online interfaces, and data interpretation. Discover different types of climate data, such as vegetation index and lightning frequency, and learn how to process and analyze them effectively. Whether you're an undergraduate, master's, or PhD student, this class will broaden your perspective on climate data collection and usage.

Understanding wind patterns and their impact on ocean currents is crucial for studying climate dynamics. Despite its sparsity, sailors have been collecting data on wind and ocean for centuries. An expedition to the North Pole by Nansen revealed the fascinating Ekman spiral and drift, explaining how sea ice and ships deviate from wind direction. Learn about how nature's forces shape our planet.

The basic data source for this assignment includes station (pointwise) and ASCII (text) data. Typically, these are provided in a spreadsheet format. While this data is easy to interpret, it only applies to a single location. I recommend watching this NOAA video on how to download their publicly available station-origin data.

Your assignment is to follow this other demonstration of obtaining station data, select any station/city of your choice, and then open the data file in Excel. Each of you will then post the screenshot of the data file in the comment section, similar to the image posted at the end of the video. Note: your column F should list Your-City/Station...

Introduction to the atmosphere and global reanalysis data

The lecture taught us the atmospheric structure and how climate data is layered accordingly. Solar radiation, conduction, and convection influence temperature changes. The boundary layer extends deeper over deserts, while the troposphere is crucial for studying climate change. Surface temperature and pressure help determine altitude using pressure levels.

The demo video shows how to use NOAA ESRL's free website tools to display precipitation maps and plot time series for the USA.

The video explores the concept of reanalysis, which involves analyzing data or information again to gain fresh insights and enhance understanding. It likely delves into the numerous applications and advantages of reanalysis across various fields.

Take the quiz about the atmosphere and global reanalysis data: https://forms.gle/ZpBvQNS4fyfnbLHj8

Next lecture: Concept of data assimilation

Data rescue provides digitized measurements for understanding and predicting past weather events. Global weather reconstruction requires combining local measurements with complex models and employing assimilation strategies like particle and Kalman filters. This process refines the reconstructions by comparing three-dimensional models with two-dimensional photographs. Reanalysis enables the generation of three-dimensional continuous weather information despite limited data availability.

Check out how ECMWF, the world's most recognized weather forecasting agency, explains Continuous Data Assimilation

Take the quiz about data assimilation: https://forms.gle/NrEvi8bbxcrTH6mV6

We are building blocks, so bear with me on these seemingly piecewise steps.

Next: The role of weather forecast computer models

Forecast Factory: Lewis Fry Richardson, an English physicist and mathematician, conducted experiments as a child and later pioneered numerical weather forecasting. He manually created the first numerical weather forecast, envisioning a "forecast factory" with mathematicians calculating real-time forecasts. His work set the foundation for modern weather forecasting using powerful supercomputers.

This lecture focuses on the analysis of data to understand the Earth's climate system. Data assimilation is utilized to incorporate observational data into computer models, resulting in accurate outcomes. The history of numerical weather prediction dates back to the late 1900s. During that time, telegraph was used for disseminating forecast information. The lecture emphasizes the importance of extensive observations and computers in achieving reliable weather forecasting.

The fun starts! Let's plot some data!

Accessing gridded climate data!

First, watch this video about contours on horizontal plane with evenly distributed data. This will help you interpret what's plotted by some of the software we will experience later.

A demo for the basics of accessing and plotting data from the reanalysis

This video shows you how to download and practice NASA's freeware Panoply.

Another demo of using the 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.

Take the quiz about this subject: https://forms.gle/wP9M9oVQTe9RLjtS8

Your homework is to replicate the mean sea level pressure time series and post your figure in the comment section, similar to the example shown near the end of the third video, also shown here. You can select the time and location of your choice and explain where you are plotting for and why.

Precipitation data comes in different formats: station data and gridded data. Gridded data extrapolates station records onto grids, while station data requires analysis. Gridded data sets include high-resolution options like "daymat" and global gridded point data sets such as "GPCC." Merged data sets combine rain gauge and satellite data, like GPCP and CMAP. Other types include satellite-only data sets (TRMM, GPM), regional data sets (Stage IV, CRU), and re-analysis data sets from modeling centers.

Here's an exercise on plotting the time series of precipitation or temperature using another tool called WRIT.

Watch the video demo here.

Recreate the associated bar graphs based on the location of your choice. (Do you notice that both precipitation and evaporation are enhancing? Why?)

Apply the hydrological concept introduced on this website to explain your findings. For example, what does it mean when you observe both rainfall and evaporation increasing?

Merge your figures of Prec and Evap using PowerPoint, then crop a screenshot, upload it, and write about your interpretation of the figures.

Let's pause and dive into the concept left out in the previous lectures - that is, data format - for its tediousness, yet it's necessary.

I purposely delayed the introduction to data formats until now because it can be tedious and potentially boring.

Intro: data format and header

However, understanding data formats is crucial as you work with climate data. In this lecture, I used the example of specific satellite precipitation data (infrared and water-vapor-based with global coverage) to introduce two key concepts: (1) data headers, which provide necessary information for GrADS to read the data, and (2) data assimilation and an introduction to various global reanalysis datasets—a revisit of some online videos shown earlier.

Visualizing NetCDF format files

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 into 4 different Chapters; feel free to skip to the following timestamps:

CHAPTER 1: Viewing Data with Panoply – 4:31

CHAPTER 2: Making Maps with QGIS – 9:40

CHAPTER 3: Troubleshooting coordinate systems – 12:45

CHAPTER 4: Workflow automation with Python – 23:15

AI vs. NWP - the future talk

Finally, the trending 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.

Moving onto horizontal maps

The video shows spatial interpolation in geostatistical analysis through a practical rainfall estimation example. The presenter uses ArcGIS's Inverse Distance Weighting (IDW) technique to focus on geostatistical analysis and spatial interpolation. It demonstrates how to estimate rainfall values across a spatial domain by assigning more weight to nearby known values. It explains data preparation, interpolation parameter setup, and executing the example's interpolation.

Watch this video to learn how to plot precipitation data. The video has no narration, only visuals, so feel free to pause and rewind to follow the steps. Your homework is to plot the rainfall for the India and Bangladesh floods of June 2022. Save the image and post it to me.

If you want to access temperature data, it works the same way.

Continuing to Plot 2D Climate Variables on a Map

In this lesson, we will practice plotting the Western Pacific Warm Pool (WPWP) on a map. The WPWP is a significant driver of Earth's climate variability and is responsible for the largest population of tropical cyclones in the region, known as typhoons.

Homework: Please email me your SST map along with a concise description of the Western Pacific Warm Pool, detailing any changes it has undergone during the past 50 years or more.

First, watch this demo on plotting 2D variables using Panoply: https://youtu.be/dTvxE5zcY08 (↓)

Second, follow this demo to learn how to download sea surface temperature (SST) data: https://youtu.be/zerL3eYy_J0 (↓)

Third, use the downloaded SST data to create a mean SST map in Panoply. You can choose your own style for the visualization. (↓)

The end of this 3rd video is a set of plots of precipitable water using Climate Reanalyzer from 1970 to 2020, divided into 10-year segments.

Your assignment is to create similar plots, but for the 2-meter air temperature (T2m). I have provided a demonstration in the left (top) video. Follow it and produce your set of figures for T2m, send it to me, and share your interpretation.

Need GIS?

This video shows you how to output drought map as KML file to be opened and displayed in Google Earth.

Advanced Plotting I: Composite Maps

First, watch these two lectures. You can skip the parts where I teach GrADS coding, and focus primarily on understanding the concepts:

This video discusses finding the ENSO index through a Google search and explains how to construct an El Niño index using historical data. It demonstrates creating a composite map of sea surface temperature anomalies during El Niño winters. The methodology can be applied to precipitation data as well.

The professor discussed constructing precipitation composite differences for El Nino and La Nina using precipitation data. Students were commended on their homework. The lecture emphasized the importance of understanding circulation systems and provided code for visualizing circulation changes during El Nino using geopotential height data.

Then, take this quiz: https://forms.gle/EUpRWJJJFBztRsVi9

In this video, which has no narratives, you can learn to use the online tool to construct and plot composite maps for different climate variables. Follow the instruction and produce the same example as you watch it.

Your homework, similar to the one shown here, is to produce temperature and precipitation anomaly maps for a region of your choice. Place these figures into a single PowerPoint slide, which should consist of six panels as follows:

[ SSTA for El Niño ] [ Precip. for El Niño ] [ T2m for El Niño ]

[ SSTA for La Niña ] [ Precip. for La Niña ] [ T2m for La Niña ]

Make a screenshot and upload it, supplied with your interpretation.

Continuing with this homework: Follow my demonstration to produce TWO figures of composite "mean sea level pressure" for El Niño and La Niña using the same years indicated in my demo. Upload your figures in the comments and provide an explanation. MSLP patterns can infer the tropospheric circulation anomalies.

Examining ENSO impacts: Try to plot the composite precipitation differences between El Niño and La Niña years following the style of these CPC maps.

In this video, devoid of narration, you will discover how to employ Google Earth Engine to generate a map featuring the Normalized Difference Vegetation Index (NDVI) for any chosen region, alongside the creation of a corresponding time series.

On this website, you can produce an NDVI map for your selected area during the year 2020, coupled with the added benefit of crafting a time series. Whether opting for a rectangle (akin to the example) or forming a polygon for a specific locale, the resulting time series is downloadable in PNG, SVG, or CSV formats, facilitating personalized combinations and analyses at designated intervals.

Advanced Plotting II: Correlation Maps

First, watch these two lectures. You can skip the parts where I teach GrADS coding, and focus primarily on understanding the concepts:

This silent video explains the concept of the correlation map analysis, a common diagnostic approach in climate science for understanding climate variability patterns. The one-point correlation map is a graphical representation of the correlation between two variables, which can be anything that can be measured, such as drought severity index as illustrated in the video.

Wait...

If you're unfamiliar with the correlation coefficient or correlation analysis, I recommend watching this video first!

This silent video demonstrates how to use a webpage-based tool to compute a correlation map that is compatible with your own time series. For example, if you are studying tree rings, you can upload your chronology index and then select the climatic fields of interest to correlate with. The most commonly used fields include SSTA (Sea Surface Temperature Anomalies), air pressure, and air temperature.

Sometimes, we need to process time series data to reduce noise or highlight specific signals embedded within it. This video demonstrates how to perform simple signal processing using the KNMI webpage. You can also upload your own data series and process its signals.

Advanced Plotting III: Climate Extremes

Climate extremes, such as heat waves, droughts, floods, and wildfires, are becoming more frequent and severe under global warming. This is due to the fact that the Earth's atmosphere is trapping more heat, which is causing the planet to warm. As the planet warms, it becomes more likely that these extreme events will occur.

(1) Watch the following lecture on the making of extreme weather/climate events. (2) Follow the next video on how to plot the statistics of climate extremes.

🌍 Join me in the lecture video where we explore the causes and impacts of climate extremes. Discover why understanding the threats and risks of climate change is vital for shaping future strategies and what it means for Climate-related Financial Disclosures (TCFD).

(3) After watching this lecture, please take the quiz: https://forms.gle/iVmHL2ofw9Pok2o88

Here, I provide demonstrations of using two online tools to plot for extreme events in the changing climate. The first tool provides all U.S. statistics on the historical change in extreme events. The second tool allows you to plot for future (and simulated historical) extreme events across the world.

Here is another short demo about plotting the projection of "annual icing days" using CMIP5 for a subtropical island. If you change the location, the curve will change.

For your homework, please select two related extreme metrics. Choose a region of interest and plot how these metrics have changed in the past and/or might change in the future. Be sure to include an explanation of what these changes signify. Email me your result.

Advanced Plotting IV: The EOF analysis

In climate data analysis, Empirical Orthogonal Function (EOF) analysis serves as a powerful tool for deciphering spatial and temporal patterns in large climate datasets. By reducing the dimensionality of complex meteorological variables, EOF analysis helps isolate dominant modes of variability, such as seasonal cycles or teleconnection patterns. This enables more accurate forecasting and a nuanced understanding of climate phenomena, thereby aiding in both short-term weather prediction and long-term climate studies.

A lecture on Utah's drought variability

This presentation serves as an invaluable resource for students studying EOF (Empirical Orthogonal Function) analysis in the context of the American West's hydroclimate. It delves into local modes of drought variability, correlation maps, and climate "forcing," offering a comprehensive understanding of factors that influence water availability and climate patterns in the region.

The Lecture: making the Empirical Orthogonal Function (EOF) analysis

This video provides a step-by-step guide on utilizing KNMI to perform an EOF analysis, specifically focusing on drought data. You are encouraged to experiment with alternative datasets and variables to broaden your understanding of EOF applications.

Entering Climate Projections

This lecture provides an overview of the evolution of weather prediction and its connection to climate change research. →

The instructor delves into the historical development of weather prediction, dating back to the late 1900s, with a focus on early attempts at numerical weather prediction by scientists like Bjerknes and Richardson. It briefly mentions the significance of climate change research and the evolution of climate models. The importance of model intercomparison is highlighted. The discussion then shifts to regional climate modeling and the need for downscaling techniques. The transcript concludes with examples of event attribution analysis, showcasing how climate change can impact seemingly unrelated weather events.

Climate models, such as those in the CMIP (Coupled Model Intercomparison Project) and CMIP6, are complex computer codes used to understand Earth's system and project future climate. If you want to understand the future of the Earth's climate, these computer models that simulate climate are the only tools available.

Let's learn about numerical climate simulations by watching these three lectures:

The lecture discusses the IPCC assessment report for climate change. It highlights the establishment of the IPCC and its warnings about climate change causing conflicts. Model-based simulations are crucial for assessing future climate, and model comparison is essential for evaluation.

The lecture discusses the use of regional downscaling of climate projections. It highlights assessments of precipitation extremes in Canada and the impact on the Colorado River Basin, discussing the uncertainty in climate models and the importance of considering snow hydrology changes.

The UN's CMIP involves over 30 groups worldwide, setting standards and allowing collective analysis of climate outputs. CMIP6 improves models with higher resolutions, additional processes, and greater flexibility, contributing to scientific advancements in climate research.

You can also use Panoply, the free software NASA introduced, to access and plot climate model data, as shown in this video. Take a look at this video and learn how to do it.

After you watch the videos, take the quiz: https://forms.gle/KDTujjBeyZVKhkyJ8

Next, Leveraging Climate Projections for Future Insight.

Evaporation - the transformation of water from liquid to gas - is being affected by our warming climate. This alters heat distribution and rainfall, impacting our climate system. To explore the change in evaporation for your region of interest, watch these two lectures.

Here is a quick demo about using KNMI to plot evaporation, an important variable for monitoring and predicting change under climate warming.

Here I will show you how to begin with a question about the future and then find appropriate data to project and interpret the result.

Your homework is to use CMIP6 data to plot evaporation trends at a location of your choice, providing a practical insight into future climate change from 2000 to 2100. Bonus is given if you also add temperature (T2m) and precipitation.

Page updated

Google Sites

Report abuse