Data Visualization Workshop for Summer Interns

18 July 2024…
GFDL/AOS Summer Internship Lecture Series (Presentation): Data Visualization Workshop, Princeton University, NJ.

1. DATA VISUALIZATION
WORKSHOP
Zachary Labe, PhD
Research Physical Scientist in the Seasonal-to-Decadal Variability and PredictabilityDivision
[email protected]
18 July 2024 – Princeton/CIMES/NOAA Internship Programs
@ZLabehttps://zacklabe.com/methods-and-open-software/!

2. RESEARCHER
Climate signal vs. weather noise
@ZLabe
COMMUNICATOR
RESEARCHER
Climate prediction & projection
STORYTELLER
Simple, bold data visualization
ZACK LABE
Federal Climate Scientist at NOAA GFDL
@ZLabe
https://zacklabe.com/

3. The signal (climate change)
The noise (weather)
…To tell data-driven climate stories…

4. The Arctic is warming more than 3 times
faster than the global average!

5. The Arctic is warming more
than 3 times faster than
the global average!

8. NOW
Start of
satellite-era

9. DATA VISUALIZATION
IS STORY-TELLING.
Arctic temperature anomalies from 1979 to 2023

10. DON’T BE
SUCH A
SCIENTIST
WE ARE
DATA
SCIENTISTS
ART BY JILL PELTO

11. Landscape of Changeuses data
about sea level rise, glacier volume
decline, increasing global
temperatures, and the increasing use
of fossil fuels. These data lines
compose a landscape shaped by
the changing climate, a world in
which we are now living.
Jill Pelto|https://www.jillpelto.com/landscape-of-change
“
”

12. DATA VISUALIZATION
IS STORY-TELLING.

13. 1. https://betterfigures.org/
2. https://www.climate-lab-book.ac.uk/
3. http://colorbrewer2.org/
RESOURCES

14. WHAT ARE YOUR
WORST FIGURES?

15. My first map subplot?

16. LABE ET AL.
2017
, CLIMATE DYNAMICS

17. Arctic sea ice thickness variability
and the atmospheric circulation
Zachary M. Labe
Gudrun Magnusdottir
2 June 2016

19. BEGINNING WITH
OUR DATA.

21. LINE GRAPHS
Temperature Anomaly (
°C)

27. :)

31. MAP PLOTS

32. YIKES!

33. YIKES – FONT IS BLEH!

34. YIKES – COLOR!

35. MAP PROJECTION!

36. :)

37. HAVE FUN!

45. LEVERAGING ALL
DIMENSIONS OF
YOUR DATA.

46. I have a 2D-array of sea ice thickness
anomaly data [year,day]…
Each line is one day
Number of Years
Thickness Anomaly (m)

48. https://seaborn.pydata.org/examples/index.html

50. https://betterfigures.org/2012/09/15/decadal-temperatures/
Error bars and graphical clutter…

51. https://betterfigures.org/2012/09/15/decadal-temperatures/
Error bars and graphical clutter…

52. https://betterfigures.org/2012/09/15/decadal-temperatures/
Error bars and graphical clutter…

53. SCIENCE OF DESIGN.

54. cmap = ‘jet’

55. STOELZLE AND STEIN, 2021, HESS

56. Check out: https://betterfigures.org/2018/06/04/playing-hunt-the-discontinuity/

57. 1998!

58. Citations on Google Scholar…

59. 2004!

64. SCHNEIDER AND NOCKE,
2017

65. SCHNEIDER AND NOCKE,
2017

66. SCHNEIDER AND NOCKE,
2017

67. SCHNEIDER AND NOCKE,
2017

68. SCHNEIDER AND NOCKE,
2017

69. Crameri, F.(2018). Scientific colour maps.
Zenodo.http://doi.org/10.5281/zenodo.1243862
Crameri, F. (2018), Geodynamic diagnostics, scientific
visualisation and StagLab 3.0, Geosci. Model Dev., 11, 2541-
2562,doi:1 0.5194/gmd-11-2541-2018
Crameri, F., G.E. Shephard, and P.J. Heron (2020), The
misuse of colour in science communication,Nature
Communications, 11, 5444.doi:10.1038/s41467-020-19160-7
Palettable:
Color palettes for Python

70. PALETTABLE - PYTHONCubehelix

71. TURBO
https://
ai.googleblog.com
/2019/08/turbo
-improved
-rainbow
-colormap
-for.html

72. OTHER OPTIONS
THYNG ET AL. 2016; OCEANOGRAPHY

73. CMOCEAN
THYNG ET AL. 2016; OCEANOGRAPHY

74. CMASHER
“Scientific colormaps for making accessible,
informative andcmashingplots”

75. Adapted from Ed Hawkins at betterfigures.org

76. Adjusting axes (spines)

77. Changing font styles

78. 1. Seaborn.
2. Plotly.
3. ggplot.
4. Matplotlib v3.
RESOURCES

79. MAKING ACCESSIBLE
FIGURES.

85. ACCESSIBILITY
No jargon
Tell a story
Alternative text
Color contrast ratio
Label data directly
Avoid flashing GIFs
Include figure titles
Avoid data overlays
Provide data references

86. VISUALIZING
GLOBAL
TEMPERATURE
CHANGE

89. 2016 RIO OLYMPICS OPENING CEREMONY

93. PLOT BY NEIL KAYE

94. PLOT BY NASA EARTH OBSERVATORY
JOSHUA STEVENS

96. PLOT BY ED HAWKINS

97. PLOT BY ED HAWKINS
DON’T BE
SUCH A
SCIENTIST
WE ARE
DATA
SCIENTISTS

98. PLOT BY ED HAWKINS
https://showyourstripes.info/

99. SCIENCE
PRESENTATIONS.

100. LABE ET AL.
2019
, GRL
Bring drama to your data story

101. LABE ET AL. 2019, GRL

102. LABE ET AL. 2019, GRL
STRONGER
POLAR VORTEX
Zonal wind
at 10 hPa
Geopotential
at 30 hPa

103. WEAKER
POLAR VORTEX
Zonal wind
at 10 hPa
Geopotential
at 30 hPa
LABE ET AL. 2019, GRL

104. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

105. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

106. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

107. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

108. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

109. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

110. 2-m TEMPERATURE
Future minus Pre-Industrial PEINGS, LABE ET AL. 2021, JCLI

111. Outline…
•10 minutes (with about +5 minutes for questions)
•This takes practice!
•Identify the bigger story. Why do we care? So what!
•Data and methods
•For instance, what GFDL models?
•Methods.
•How did you evaluate the data?
•Findings.
•What are your results? I suggest framing it around 2-3 main points.
•Conclusions.
•What can we learn? And importantly, any future plans or suggestions
if continuing work on this topic.

112. EXAMPLE
SLIDES.

113. Explainable machine learning can
distinguish between regional patterns
of time-evolving climate change
in GFDL models.
113
SIGNIFICANCE

114. TRENDS FROM 1921 TO 1950
Fully-Coupled [Historical + SSP5-8.5]SPEAR_MED, but NO anthropogenic aerosolsSPEAR_MED, but NO anthropogenic forcings
114
WarmerColder

115. TRENDS FROM 1921 TO 1950
Fully-Coupled [Historical + SSP5-8.5]SPEAR_MED, but NO anthropogenic aerosolsSPEAR_MED, but NO anthropogenic forcings
115
WarmerColder

116. TRENDS FROM 1921 TO 1950
Fully-Coupled [Historical + SSP5-8.5]SPEAR_MED, but NO anthropogenic aerosolsSPEAR_MED, but NO anthropogenic forcings
116
WarmerColder

117. TAKEAWAYS
1.XAI can identify regional patterns of climate change & variability in GFDL large ensembles.
2.Method can identify differences in time-evolving forced climate signals between other
climate model large ensembles.
3.Framework can be adapted for monitoring and predicting patterns of climate change in
observations.
Zack Labe
[email protected]
11715 March 2023
GFDL Lunchtime Seminar

118. CHECK:
Simple.
Bold.
Stories.
@ZLabeQuestions!
ZACHARY LABE | 18 JULY 2024| SUMMER INTERNSHIP PROGRAMS IN PRINCETON, NJ

119. SHARING FIGURES
Challenge:
1 summary figure
1 #SciComm figure
@ZLabe
ZACHARY LABE | 18 JULY 2024| SUMMER INTERNSHIP PROGRAMS IN PRINCETON, NJ

120. What figure did you bring today?