Successfully reported this slideshow.
Your SlideShare is downloading. ×

Summary Slide for Disentangling Climate Forcing in Multi-Model Large Ensembles Using Neural Networks

•
Dec. 31, 2020
• 0 likes • 118 views
Ad
Ad
Ad
Ad
Ad
Ad
Ad
Ad
Ad
Ad
Upcoming SlideShare
Using data visualization for accessible science (communication)
Using data visualization for accessible science (communication)
Loading in …3
×

Check these out next

Arctic climate change through the lens of data visualization
Zachary Labe
Using explainable neural networks for comparing climate model projections
Zachary Labe
Exploring climate change signals with explainable AI
Zachary Labe
Evaluating global climate models using simple, explainable neural networks
Zachary Labe
Using neural networks to explore regional climate patterns in single-forcing ...
Zachary Labe
Climate Change in the Arctic
Zachary Labe
Decadal warming slowdown predictions by an artificial neural network
Zachary Labe
Assessing climate variability and change with explainable neural networks
Zachary Labe
1 of 1 Ad

Summary Slide for Disentangling Climate Forcing in Multi-Model Large Ensembles Using Neural Networks

Dec. 31, 2020
• 0 likes • 118 views

Download to read offline

Science

The relative roles of individual forcings on large-scale climate variability remain difficult to disentangle within fully-coupled global climate model simulations. Here, we train an artificial neural network (ANN) to classify the climate forcings of a new set of CESM1 initial-condition large ensembles that are forced by different combinations of aerosol (industrial and biomass burning), greenhouse gas, and land-use/land-cover forcings. As a result of learning the regional responses of internal variability to the different external forcings, the ANN is able to successfully classify the dominant forcing for each model simulation. Using recently developed explainable AI methods, such as layerwise relevance propagation, we then compare the patterns of climate variability identified by the ANN between different external climate forcings that are learned by the neural network. Further, we apply this ANN architecture on additional climate simulations from the multi-model large ensemble archive, which include all anthropogenic and natural radiative forcings. From this collection of initial-condition ensembles, the ANN is also able to detect changes in atmospheric internal variability between the 20th and 21st centuries by training on climate fields after the mean forced signal has already been removed. This ANN framework and its associated visualization tools provide a novel approach to extract complex patterns of observable and projected climate variability and trends in Earth system models. (from https://ams.confex.com/ams/101ANNUAL/meetingapp.cgi/Paper/379553)

The relative roles of individual forcings on large-scale climate variability remain difficult to disentangle within fully-coupled global climate model simulations. Here, we train an artificial neural network (ANN) to classify the climate forcings of a new set of CESM1 initial-condition large ensembles that are forced by different combinations of aerosol (industrial and biomass burning), greenhouse gas, and land-use/land-cover forcings. As a result of learning the regional responses of internal variability to the different external forcings, the ANN is able to successfully classify the dominant forcing for each model simulation. Using recently developed explainable AI methods, such as layerwise relevance propagation, we then compare the patterns of climate variability identified by the ANN between different external climate forcings that are learned by the neural network. Further, we apply this ANN architecture on additional climate simulations from the multi-model large ensemble archive, which include all anthropogenic and natural radiative forcings. From this collection of initial-condition ensembles, the ANN is also able to detect changes in atmospheric internal variability between the 20th and 21st centuries by training on climate fields after the mean forced signal has already been removed. This ANN framework and its associated visualization tools provide a novel approach to extract complex patterns of observable and projected climate variability and trends in Earth system models. (from https://ams.confex.com/ams/101ANNUAL/meetingapp.cgi/Paper/379553)

Science
Advertisement

Recommended

Using data visualization for accessible science (communication)
Zachary Labe
32 views
113 slides
Using visualization to start climate conversations
Zachary Labe
45 views
18 slides
Temporary slowdowns in decadal warming predictions by a neural network
Zachary Labe
29 views
1 slide
Making effective and accessible figures
Zachary Labe
111 views
84 slides
Learning new climate science by thinking creatively with machine learning
Zachary Labe
60 views
110 slides
Using neural networks to predict temporary slowdowns in decadal climate warmi...
Zachary Labe
100 views
39 slides
Using data visualization for effective science (communication) - April 2022
Zachary Labe
784 views
103 slides
Communicating Arctic climate change through data-driven stories
Zachary Labe
118 views
36 slides
Advertisement

More Related Content

More from Zachary Labe (20)

Arctic climate change through the lens of data visualization
Zachary Labe
58 views
Using explainable neural networks for comparing climate model projections
Zachary Labe
42 views
Exploring climate change signals with explainable AI
Zachary Labe
156 views
Evaluating global climate models using simple, explainable neural networks
Zachary Labe
111 views
Using neural networks to explore regional climate patterns in single-forcing ...
Zachary Labe
94 views
Climate Change in the Arctic
Zachary Labe
111 views
Decadal warming slowdown predictions by an artificial neural network
Zachary Labe
119 views
Assessing climate variability and change with explainable neural networks
Zachary Labe
66 views
Exploring climate model large ensembles with explainable neural networks
Zachary Labe
99 views
Learning new climate science by opening the machine learning black box
Zachary Labe
83 views
Making Effective Figures
Zachary Labe
173 views
Climate Signals in CESM1 Single-Forcing Large Ensembles Revealed by Explainab...
Zachary Labe
78 views
Climate Summit for the ABC Network
Zachary Labe
51 views
Using data visualization for effective science (communication)
Zachary Labe
53 views
Revealing climate change signals with explainable AI
Zachary Labe
79 views
Communicating Arctic climate change through data-driven stories
Zachary Labe
98 views
Improving science (communication) through data visualization
Zachary Labe
194 views
The Pan-Arctic Impacts of Thinning Sea Ice
Zachary Labe
68 views
Disentangling Climate Forcing in Multi-Model Large Ensembles Using Neural Net...
Zachary Labe
340 views
Trends and regional variability of observed Arctic sea-ice thickness
Zachary Labe
98 views
Arctic climate change through the lens of data visualization
Zachary Labe
58 views
90 slides
Using explainable neural networks for comparing climate model projections
Zachary Labe
42 views
47 slides
Exploring climate change signals with explainable AI
Zachary Labe
156 views
102 slides
Evaluating global climate models using simple, explainable neural networks
Zachary Labe
111 views
49 slides
Using neural networks to explore regional climate patterns in single-forcing ...
Zachary Labe
94 views
43 slides
Climate Change in the Arctic
Zachary Labe
111 views
62 slides

Recently uploaded (20)

Chapter_01.ppt
jesus505978
5 views
energy 1.pdf
ssuser24a93a
2 views
(November 30, 2022) Webinar: Molecular Mechanisms Behind Lameness in Meat Chi...
Scintica Instrumentation
5 views
Photosynthesis PP.ppt
YomaBey
4 views
Trait Theories.pdf
PavanMohan10
5 views
biotechnology regulations.ppt
JhayepeejoyGuiasit
4 views
12. Meteoroids and Comets (Trexia).pptx
trixiiee
3 views
Hemophilia.pptx
Rahul kumar
2 views
8-glycolysis.ppt
MuhammadRizwan863722
1 view
Golgi apparatus.pdf
Viswa Deepika
4 views
rahul micro.pptx
Utkarsh887455
3 views
PRECIPITATION [FINAL ROUND].pptx
trixiiee
6 views
Evaporation.pptx
ANKIT915111
3 views
em_spectrum_reprt (1).pptx
JurielMarOng
1 view
carbon capture webinar
geocoach2020
3 views
2 nd International Conference on Computing and Information Technology Trends ...
ijesajournal
3 views
SchY.ppt
Priyanka Saroj
2 views
BIOASSAYS.pptx
Priyanka Saroj
1 view
finalppt phy..pptx
AbdetaBirhanu
2 views
HydrogenBomb.ppt
RajveerBatar
3 views
Chapter_01.ppt
jesus505978
5 views
67 slides
energy 1.pdf
ssuser24a93a
2 views
2 slides
(November 30, 2022) Webinar: Molecular Mechanisms Behind Lameness in Meat Chi...
Scintica Instrumentation
5 views
67 slides
Photosynthesis PP.ppt
YomaBey
4 views
22 slides
Trait Theories.pdf
PavanMohan10
5 views
15 slides
biotechnology regulations.ppt
JhayepeejoyGuiasit
4 views
24 slides
Advertisement

Summary Slide for Disentangling Climate Forcing in Multi-Model Large Ensembles Using Neural Networks

  1. 1. 1. Using explainable AI methods with artificial neural networks (ANNs) reveals climate patterns in large ensemble simulations 2. Metric proposed for quantifying the uncertainty of an ANN visualization method that extracts signals from different external forcings 3. ANN trained using a large ensemble simulation without time- evolving aerosols makes more accurate predictions of real world data Zachary M. Labe & Elizabeth A. Barnes Department of Atmospheric Science at Colorado State University 14 January 2021 – 1:15 PM to 1:20 PM (EST) 20th Conference on Artificial Intelligence for Environmental Science 101st AMS Annual Meeting (Paper 11.2) DISENTANGLING CLIMATE FORCING IN SINGLE-FORCING LARGE ENSEMBLES USING NEURAL NETWORKS Aerosol-driven Greenhouse gas-driven All forcings LRP – (LRP)

×