13 January 2023…
Scientific Machine Learning Community (Presentation): Explainable AI for identifying regional climate change patterns, University of Leeds, UK. Remote Presentation.
Explainable AI for identifying regional climate change patterns
1. Explainable AI for identifying
regional climate change patterns
@ZLabe
Zachary M. Labe
Postdoc at Princeton University and NOAA GFDL
13 January 2023 – University of Leeds
Scientific Machine Learning Community (SciML)
https://zacklabe.com/
2. • Do it better
• e.g., parameterizations in climate models are not
perfect, use ML to make them more accurate
• Do it faster
• e.g., code in climate models is very slow (but we
know the right answer) - use ML methods to speed
things up
• Do something new
• e.g., go looking for non-linear relationships you
didn’t know were there
Very relevant for
research: may be
slower and worse,
but can still learn
something
WHY SHOULD WE CONSIDER
MACHINE LEARNING?
5. Today’s weather or climate
scientist is far more likely to be
debugging code written in
Python… than to be poring over
satellite images or releasing
radiosondes.
“
D. Irving| Bulletin of the American Meteorological Society| 2016
6. Machine learning for weather
IDENTIFYING SEVERE THUNDERSTORMS
Molina et al. 2021
Toms et al. 2021
CLASSIFYING PHASE OF MADDEN-JULLIAN OSCILLATION
SATELLITE DETECTION
Lee et al. 2021
DETECTING TORNADOES
McGovern et al. 2019
7. Machine learning for climate
FINDING FORECASTS OF OPPORTUNITY
Mayer and Barnes, 2021
PREDICTING CLIMATE MODES OF VARIABILITY
Gordon et al. 2021
TIMING OF CLIMATE CHANGE
Barnes et al. 2019
8. Machine learning for oceanography
CLASSIFYING ARCTIC OCEAN ACIDIFICATION
Krasting et al. 2022
TRACK AND REVEAL DEEP WATER MASSES
Sonnewald and Lguensat, 2021
ESTIMATING OCEAN SURFACE CURRENTS
Sinha and Abernathey, 2021
14. We know some metadata…
+ What year is it?
+ Where did it come from?
[Labe and Barnes, 2022; ESS]
TEMPERATURE
15. TEMPERATURE
Neural network learns nonlinear
combinations of forced climate
patterns to identify the year
We know some metadata…
+ What year is it?
+ Where did it come from?
[Labe and Barnes, 2022; ESS]
16. ----ANN----
2 Hidden Layers
10 Nodes each
Ridge Regularization
Early Stopping
[e.g., Barnes et al. 2019, 2020]
[e.g., Labe and Barnes, 2021]
TIMING OF EMERGENCE
(COMBINED VARIABLES)
RESPONSES TO
EXTERNAL CLIMATE
FORCINGS
PATTERNS OF
CLIMATE INDICATORS
[e.g., Rader et al. 2022]
Surface Temperature Map Precipitation Map
+
TEMPERATURE
We know some metadata…
+ What year is it?
+ Where did it come from?
[Labe and Barnes, 2022; ESS]
17. ----ANN----
2 Hidden Layers
10 Nodes each
Ridge Regularization
Early Stopping
[e.g., Barnes et al. 2019, 2020]
[e.g., Labe and Barnes, 2021]
TIMING OF EMERGENCE
(COMBINED VARIABLES)
RESPONSES TO
EXTERNAL CLIMATE
FORCINGS
PATTERNS OF
CLIMATE INDICATORS
Surface Temperature Map Precipitation Map
+
TEMPERATURE
[e.g., Rader et al. 2022]
We know some metadata…
+ What year is it?
+ Where did it come from?
[Labe and Barnes, 2022; ESS]
19. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
Anomaly is relative to 1951-1980
20. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
Let’s run a
climate model
21. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
Let’s run a
climate model
again
22. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
Let’s run a
climate model
again & again
23. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
CLIMATE MODEL
ENSEMBLES
24. What is the annual mean temperature of Earth?
THE REAL WORLD
(Observations)
CLIMATE MODEL
ENSEMBLES
Range of ensembles
= internal variability (noise)
Mean of ensembles
= forced response (climate change)
25. What is the annual mean temperature of Earth?
Range of ensembles
= internal variability (noise)
Mean of ensembles
= forced response (climate change)
But let’s remove
climate change…
26. What is the annual mean temperature of Earth?
Range of ensembles
= internal variability (noise)
Mean of ensembles
= forced response (climate change)
After removing the
forced response…
anomalies/noise!
27. What is the annual mean temperature of Earth?
• Increasing greenhouse gases (CO2, CH4, N2O)
• Changes in industrial aerosols (SO4, BC, OC)
• Changes in biomass burning (aerosols)
• Changes in land-use & land-cover (albedo)
28. What is the annual mean temperature of Earth?
• Increasing greenhouse gases (CO2, CH4, N2O)
• Changes in industrial aerosols (SO4, BC, OC)
• Changes in biomass burning (aerosols)
• Changes in land-use & land-cover (albedo)
Plus everything else…
(Natural/internal variability)
30. Greenhouse gases fixed to 1920 levels
All forcings (CESM-LE)
Industrial aerosols fixed to 1920 levels
[Deser et al. 2020, JCLI]
Fully-coupled CESM1.1
20 Ensemble Members
Run from 1920-2080
Observations
31. So what?
Greenhouse gases = warming
Aerosols = ?? (though mostly cooling)
What are the relative responses
between greenhouse gas
and aerosol forcing?
34. INPUT LAYER
HIDDEN LAYERS
OUTPUT LAYER
Surface Temperature Map
“2000-2009”
DECADE CLASS
“2070-2079”
“1920-1929”
ARTIFICIAL NEURAL NETWORK (ANN)
35. INPUT LAYER
HIDDEN LAYERS
OUTPUT LAYER
Surface Temperature Map
“2000-2009”
DECADE CLASS
“2070-2079”
“1920-1929”
BACK-PROPAGATE THROUGH NETWORK = EXPLAINABLE AI
ARTIFICIAL NEURAL NETWORK (ANN)
36. INPUT LAYER
HIDDEN LAYERS
OUTPUT LAYER
Layer-wise Relevance Propagation
Surface Temperature Map
“2000-2009”
DECADE CLASS
“2070-2079”
“1920-1929”
BACK-PROPAGATE THROUGH NETWORK = EXPLAINABLE AI
ARTIFICIAL NEURAL NETWORK (ANN)
[Barnes et al. 2020, JAMES]
[Labe and Barnes 2021, JAMES]
37. LAYER-WISE RELEVANCE PROPAGATION (LRP)
Volcano
Great White
Shark
Timber
Wolf
Image Classification LRP
https://heatmapping.org/
LRP heatmaps show regions
of “relevance” that
contribute to the neural
network’s decision-making
process for a sample
belonging to a particular
output category
Neural Network
WHY
WHY
WHY
Backpropagation – LRP
38. LAYER-WISE RELEVANCE PROPAGATION (LRP)
Volcano
Great White
Shark
Timber
Wolf
Image Classification LRP
https://heatmapping.org/
LRP heatmaps show regions
of “relevance” that
contribute to the neural
network’s decision-making
process for a sample
belonging to a particular
output category
Neural Network
WHY
WHY
WHY
Backpropagation – LRP
39. LAYER-WISE RELEVANCE PROPAGATION (LRP)
Volcano
Great White
Shark
Timber
Wolf
Image Classification LRP
https://heatmapping.org/
LRP heatmaps show regions
of “relevance” that
contribute to the neural
network’s decision-making
process for a sample
belonging to a particular
output category
Neural Network
Backpropagation – LRP
WHY
WHY
WHY
40. LAYER-WISE RELEVANCE PROPAGATION (LRP)
Image Classification LRP
https://heatmapping.org/
NOT PERFECT
Crock
Pot
Neural Network
Backpropagation – LRP
WHY
41. [Adapted from Adebayo et al., 2020]
EXPLAINABLE AI IS
NOT PERFECT
THERE ARE MANY
METHODS
42. [Adapted from Adebayo et al., 2020]
THERE ARE MANY
METHODS
EXPLAINABLE AI IS
NOT PERFECT
44. Neural
Network
[0] La Niña [1] El Niño
[Toms et al. 2020, JAMES]
Input a map of sea surface temperatures
45. Visualizing something we already know…
Input maps of sea surface
temperatures to identify
El Niño or La Niña
Use ‘LRP’ to see how the
neural network is making
its decision
[Toms et al. 2020, JAMES]
Layer-wise Relevance Propagation
Composite Observations
LRP [Relevance]
SST Anomaly [°C]
0.00 0.75
0.0 1.5
-1.5
46. INPUT LAYER
HIDDEN LAYERS
OUTPUT LAYER
Layer-wise Relevance Propagation
Surface Temperature Map
“2000-2009”
DECADE CLASS
“2070-2079”
“1920-1929”
BACK-PROPAGATE THROUGH NETWORK = EXPLAINABLE AI
ARTIFICIAL NEURAL NETWORK (ANN)
[Barnes et al. 2020, JAMES]
[Labe and Barnes 2021, JAMES]
47. 1960-1999: ANNUAL MEAN TEMPERATURE TRENDS
Greenhouse gases fixed
to 1920 levels
[AEROSOLS PREVAIL]
Industrial aerosols fixed
to 1920 levels
[GREENHOUSE GASES PREVAIL]
All forcings
[STANDARD CESM-LE]
DATA
48. 1960-1999: ANNUAL MEAN TEMPERATURE TRENDS
Greenhouse gases fixed
to 1920 levels
[AEROSOLS PREVAIL]
Industrial aerosols fixed
to 1920 levels
[GREENHOUSE GASES PREVAIL]
All forcings
[STANDARD CESM-LE]
DATA
49. 1960-1999: ANNUAL MEAN TEMPERATURE TRENDS
Greenhouse gases fixed
to 1920 levels
[AEROSOLS PREVAIL]
Industrial aerosols fixed
to 1920 levels
[GREENHOUSE GASES PREVAIL]
All forcings
[STANDARD CESM-LE]
DATA
50. 1960-1999: ANNUAL MEAN TEMPERATURE TRENDS
Greenhouse gases fixed
to 1920 levels
[AEROSOLS PREVAIL]
Industrial aerosols fixed
to 1920 levels
[GREENHOUSE GASES PREVAIL]
All forcings
[STANDARD CESM-LE]
DATA
51. CLIMATE MODEL DATA PREDICT THE YEAR FROM MAPS OF TEMPERATURE
AEROSOLS
PREVAIL
GREENHOUSE GASES
PREVAIL
STANDARD
CLIMATE MODEL
[Labe and Barnes 2021, JAMES]
52. OBSERVATIONS PREDICT THE YEAR FROM MAPS OF TEMPERATURE
AEROSOLS
PREVAIL
GREENHOUSE GASES
PREVAIL
STANDARD
CLIMATE MODEL
[Labe and Barnes 2021, JAMES]
53. OBSERVATIONS
SLOPES
PREDICT THE YEAR FROM MAPS OF TEMPERATURE
AEROSOLS
PREVAIL
GREENHOUSE GASES
PREVAIL
STANDARD
CLIMATE MODEL
[Labe and Barnes 2021, JAMES]
61. Higher LRP values indicate greater relevance
for the ANN’s prediction
AVERAGED OVER 1960-2039
Aerosol-driven
Greenhouse gas-driven
All forcings
Low High
[Labe and Barnes 2021, JAMES]
91. N
Y
HIDDEN LAYERS
INPUT LAYER
INPUT LAYER
SAI WORLD?
or
or
map of near-surface temperature
map of near-surface temperature
map of total precipitation
map of total precipitation
Years Since
SAI Injection
OUTPUT
LOGISTIC
REGRESSION
ARTIFICAL
NEURAL
NETWORK
softmax
[Labe et al. 2023, EarthArXiv]
92. N
Y
HIDDEN LAYERS
INPUT LAYER
INPUT LAYER
SAI WORLD?
or
or
map of near-surface temperature
map of near-surface temperature
map of total precipitation
map of total precipitation
Years Since
SAI Injection
OUTPUT
LOGISTIC
REGRESSION
ARTIFICAL
NEURAL
NETWORK
softmax
[Labe et al. 2023, EarthArXiv]
93. CAN WE DETECT A SAI WORLD?
[Labe et al. 2023, EarthArXiv]
99. CAN WE DETECT A SAI WORLD?
[Labe et al. 2023, EarthArXiv]
100. N
Y
HIDDEN LAYERS
INPUT LAYER
INPUT LAYER
SAI WORLD?
or
or
map of near-surface temperature
map of near-surface temperature
map of total precipitation
map of total precipitation
Years Since
SAI Injection
OUTPUT
LOGISTIC
REGRESSION
ARTIFICAL
NEURAL
NETWORK
softmax
[Labe et al. 2023, EarthArXiv]
105. WE CAN LEARN NEW SCIENCE
FROM EXPLAINABLE AI.
3)
106. KEY POINTS
1. Machine learning is just another tool to add to our scientific workflow
2. We can use explainable AI (XAI) methods to peer into the black box of machine learning
3. We can learn new science by using XAI methods in conjunction with existing statistical tools
Zachary Labe
zachary.labe@noaa.gov
Labe, Z.M. and E.A. Barnes (2021), Detecting climate signals using explainable AI with single-forcing
large ensembles. Journal of Advances in Modeling Earth Systems, DOI: 10.1029/2021MS002464
Po-Chedley, S., J.T. Fasullo, N. Siler, Z.M. Labe, E.A. Barnes, C.J.W. Bonfils, and B.D. Santer (2022). Internal
variability and forcing influence model-satellite differences in the rate of tropical tropospheric
warming. Proceedings of the National Academy of Sciences, DOI: 10.1073/pnas.2209431119
Labe, Z.M., E.A. Barnes, and J.W. Hurrell (2023). Identifying the regional emergence of climate patterns
in a simulation of stratospheric aerosol injection. EarthArXiv, DOI: 10.31223/X5394Z