View this email in your browser

Read on for details on upcoming events, recent developments, research, and new opportunities from across the climate change and machine learning community.

Do you have anything you would like to share in a future newsletter or an event you would like to add to the community calendar?  Get in touch at and/or fill out this short intake formFor discussion with fellow readers, follow @ClimateChangeAI on Twitter and join us on Circle.

Join the discussion with fellow readers:

Follow CCAI on Twitter
Join the CCAI community platform

  CCAI Announcements

Join the CCAI Community Directory 

We are excited to launch two new interactive tools to support the Climate Change AI community!

Our new Community Directory allows you to share your profile and contact details with the CCAI network. You can state your areas of expertise and interests, as well as indicate ways in which you are seeking or offering support. We hope that the directory will make it easier to find potential collaborators and mentors. 

Join the CCAI Community Platform

Our new Community Platform, hosted on Circle, provides a space to chat with ML and climate enthusiastsbuild your network, explore opportunities and discuss current developments around climate change and AI. Some highlights from the last weeks include a discussion on the process of finding research ideas a channel dedicated to finding grant collaborators.  and the communities’ thoughts on the recent IPCC report

CCAI Browsable Paper Repository

Trying to find cool research ideas in climate change and machine learning? CCAI has archived all papers presented at our past workshops and events and built an easily searchable web tool to access them. You can filter by venue or subject area, as well as do a free text search for keywords most relevant to your interests!

CCAI Innovation Grants

With the support of Schmidt Futures and the Quadrature Climate Foundation, Climate Change AI will fund year-long research projects at the intersection of climate change and machine learning for up to USD 150K per project, for a total of USD 1.8M. The program will be focused on fostering pathways to deployment for data-driven work on climate change, and on funding the creation of impactful datasets that can spur further work in this area. Proposals are due October 15. 

We will be holding informational webinars on September 23rd and 27th for those interested in learning more about the program.  For those looking to find collaborators to augment a proposal, we recommend using the new CCAI community platform and directory.

CCAI Webinars

On September 22nd at 11 AM ET / 15:00 UTC we will be holding a webinar on how  “International organizations working on climate change are using machine learning”.  Register here

The following webinar will be on October 19th at 11 AM ET / 15:00 UTC, focusing on machine learning applications for climate science and Earth observation. Register here.

Recordings of our past webinars are available on the CCAI webinars site.

Contribute to the CCAI Wiki! 

The Climate Change AI Wiki is a community-knowledge hub summarizing state-of-the-art knowledge and resources on various topics related to machine learning applications for tackling climate change. We invite you to contribute to the Wiki by sharing resources and improving existing content! Please refer to the Wiki guidelines for instructions on how to contribute.

Open positions within CCAI

CCAI is seeking a volunteer to lead the management of two important communities: the editors of our new Wiki and the users of our broader CCAI community platforms. The position description is available here and the link to apply is here.

Upcoming events

Energy, AI and Climate Change (organized by Pie & AI), September 20th.
Noam Chomsky will share his views on a variety of topics related to advances in technology including AI in the fight against climate change.

Applying AI to Tackle the Climate Crisis (organized by The Good AI), September 22nd. 
A virtual panel involving leaders from AI startups working on managing the climate crisis,  moderated by CCAI chair Priya Donti.

Data Science for Society Seminar Series (hosted by the University of Pretoria), September 23rd. 
This virtual talk series is featuring a talk by Emily Mueller (Imperial College, London) on measuring the urban environment using satellite imagery.

AI and Electric Power Summit (organized by EPRI), September 28 and 29th.
This virtual event will bring together leading innovators in the electric power and AI industries to advance AI technologies for electric power utilities.

Energy and AI (organized by Agoria), October 1st
This virtual event will feature CCAI chair Priya Donti, discussing ways AI can be used to tackle climate change. This event will also feature an opening talk by the Belgian Federal Minister for Energy, Tinne Van der Straeten

AI and Climate Science (organized by ITU), multiple dates
This virtual talk series will feature talks from Noah Brenowitz/Chris Bretherton (Vulcan Inc./UW), Gavin Schmidt/Claire Monteleoni (NASA/University of Colorado) and Alison Lowndes (NVIDIA) on various strategies to use AI for climate science research.
Digital Transformation for the Built Environment (organized by, October 26 and 28th.
This online workshop features speakers from the industry and academia on building tools for effective digital transformation for all built environments. 

CCAI Happy Hour

CCAI is hosting virtual happy hours twice a month. The happy hour is an opportunity to meet others with similar interests in an informal and welcoming environment. To accommodate different time zones, we are alternating between two different times:

  CCAI News

We are excited to welcome  Dr. Claire Monteleoni (Associate Professor at CU Boulder and co-founder of Climate Informatics) and Dr. Catherine Nakalembe (Associate Research Professor at the University of Maryland and NASA Harvest Africa Program Director) to the Climate Change AI Advisory Board!

Wired Magazine featured CCAI’s feedback piece on the EU-AI policy. Story here

  News on AI and Climate Change

Climate scientists at ETH Zurich perform high-resolution simulations (2 km) of the entire European and central Atlantic Ocean region, revealing new findings

Climate scientists discuss the shortcomings and opportunities in using AI for climate science research. (Full Nature Machine Intelligence paper here).

A massive open online course (MOOC) is being offered on the applications of AI to Earth monitoring using Copernicus Earth observation data and the AI and machine learning techniques that can be used to work with it.

The world’s biggest plant to capture CO2 from the air just opened in Iceland. 

The World Economic Forum and its partners have released a white paper on Harnessing AI to accelerate the energy transition

A new study creates a deep convolutional neural network using global satellite imagery to detect sustainable roofscapes—a promising strategy for climate mitigation.

A new study maps the links between climate change and health using machine learning by creating computer-assisted systematic maps between climate variability, and weather (CCVW), and health.

National Science Foundation (NSF) announces the setting up of the Center for Learning the Earth with Artificial Intelligence and Physics (LEAP).

Allen Institute of AI (A2I) Climate Modeling group has open-sourced its climate modeling repositories

An Arctic AI model developed by scientists from Alan Turing Institute predicts sea ice loss (blog featured by NVIDIA).

  Dataset of the Month

If you would like your dataset or paper to be highlighted, feel free to email a short pitch to: or fill out this form.

Title:  Climate TRACE Emissions Inventory

Description: Climate TRACE has released a  comprehensive accounting of global greenhouse gas (GHG) emissions based primarily on direct, independent observation. Check out the Emissions inventory and explore emissions across different sectors.  



  Data and Challenges


Climate TRACE has released a  comprehensive accounting of global greenhouse gas (GHG) emissions based primarily on direct, independent observation. Check out the Emissions inventory and explore emissions across different sectors

The Sustainable Eateries is a crowdsourced open database to capture sustainability practices of eateries across the globe. The code is available here.

Global Flood Database contains data and code related to flood management using satellite imagery. The associated paper describing results and mapping algorithm can be accessed here.

Challenges and Hackathons

Microsoft AI for Earth is hosting a challenge to map floodwater from radar imagery using Sentinel-1 imagery data. The competition's end date is September 29th. 

AI for Earth Observation (AI4EO) has an open challenge to map cultivated land using Copernicus Sentinel imagery and to develop solutions to extract as much information as possible from the native 10-meter per pixel resolution. Due September 30th.

xView3, an ML challenge that focuses on building ML models to detect illegal, unregulated, and unreported (IUU) fishing on synthetic aperture radar (SAR) is open. More here. Submission is due November 30th  


Funding opportunities

The Alan Turing Institute is accepting applications for proposals on digital twins and AI. Applications due September 24.

The Defense Advanced Research Projects Agency (DARPA) is issuing an Artificial Intelligence Exploration (AIE) Opportunity inviting submissions of innovative basic or applied research concepts in the technical domain of AI-assisted Climate Tipping-point Modeling

AI for Green by the Austrian Ministry for Climate Action is accepting applications from companies that aim to use AI to support the goal of climate neutrality in 2040 and solve ecological problems. Only open to European companies. Applications due November 2nd.

Call for Nominations

The U.S. Congress has directed the National Academies to establish a Climate Security Roundtable that will convene experts from academia, the private sector, and civil society to provide support to the Climate Security Advisory Council (CSAC). There is an open call to nominate experts to the committee. Deadline September 30

Call for Contributions is requesting contributions from researchers, users, and stakeholders to write, or talk, about their work on “life on land”, #15 on the UN Sustainable Development Goals (SDGs). The deadline for contributions is October 15th.


PaleoHack2 hackathon focusing on using  Python tools for the analysis of paleoclimate data, chiefly the Pyleoclim package, is now open.

Summer School

The Caltech Resnick Sustainability Institute is accepting applications for a summer school on Computer Vision Methods for Ecology (CV4Ecology). Applications are due December 1st.

The Activate Fellowship is accepting applications for the 2022 cohort. This two-year fellowship embeds innovators in a world-class research lab, supported with funding, mentorship, entrepreneurial education, and connections with a network of industry leaders, investors, and philanthropists. 

Carbon13 is an eight-month venture builder program that supports developers, machine learning specialists, and data scientists to become founders of high-growth startups which can credibly, measurably, and significantly reduce CO2e emissions. The next cohort begins their journey on September 21st. More information here

Calls for Papers and speakers

First ACM SIGEnergy Workshop on Fair, Accountable, Transparent and Ethical (FATE) AI for Smart Environments and Energy Systems is accepting papers. Submissions due October 3rd

The 11th International Conference on Climate Informatics is inviting submissions for the 2022 edition of the conference.  Submission deadline November 2021

Energy and AI has a special issue on AI and digital technology for energy conservation in buildings. Research topics currently accepted include anomaly detection and digital twins. Submission deadline: November 15th

Energy and AI is accepting manuscripts for original research work at the intersection of energy and machine learning. Submission deadline December 31st

Artificial Intelligence for the Earth Systems (AIES) will begin accepting original research on the development and application of methods in Artificial Intelligence (AI), Machine Learning (ML), data science, and statistics that are relevant to meteorology, atmospheric science, hydrology, climate science, and ocean science

The Current Opinion in Environmental Sustainability Journal (COSUST) is accepting articles for an upcoming issue Environmental Sustainability Science, Artificial Intelligence and Digitization for articles that shall explore opportunities and risks of digitization, automation and AI in the field of environmental sustainability. 

Applied Energy is accepting original research in the area of building climate resiliency for energy systems in a special issue. Submission deadline January 15th, 2022

The Environmental Data Science (EDS) journal is inviting papers using AI and Data Science for Environment, including topics such as hybrid models, and  Environmental Informatics.
Calls for Speakers

PAW Climate, a conference featuring commercial climate tech companies using ML, is accepting proposals for speakers for the 2022 edition. Applications due November 11


A data fusion approach to optimize compositional stability of halide perovskites

by Shijing Sun, Armi Tiihonen, Felipe Oviedo, Zhe Liu, Janak Thapa, Yicheng Zhao, Noor Titan P. Hartono, Anuj Goyal, Thomas Heumueller, Clio Batali, Alex Encinas, Jason J.Yoo, Ruipeng Li, Zekun Ren, Marius Peters, Christoph J.Brabec, Moungi G.Bawendi, Vladan Stevanovic, John Fisher III,Tonio Buonassisi 
What: This paper uses ML to experimentally develop a 17-fold more stable perovskite solar cell composition.

Why: Perovskite solar cells are the most promising technology for cheaper and high-efficiency next generation photovoltaics. This new technology can have a substantial and timely impact on climate change. However, environmental stability of perovskites is still very limited compared to mainstream technologies such as silicon solar cells. Finding perovskite compositions whose performance degrades at a slow rate when exposed to air and sunlight remains a major challenge in the field.

How: The authors perform closed-loop optimization of perovskite stability. By using constrained Bayesian optimization, the authors integrate first-principles calculation with experimental data to iteratively suggest more stable perovskites compositions in a data-efficient manner.

Key findings: The machine learning approach was able to identify a 17-fold more stable perovskite solar cell, experimentally exploring less than 1.3% of the compositional space. The authors were able to fuse thermodynamic simulations with experimental results to account for multiple causes of limited stability. This approach can have a substantial impact in other material optimization problems relevant to climate change, including developing more stable batteries or carbon capture materials.

CCAI perspective: Using ML to accelerate the development of functional materials can have a substantial impact on climate change. Most climate change mitigation technologies, such as batteries or solar cells, rely on complex functional materials which take years to develop and optimize. Machine learning can largely simplify this process by allowing theoretical screening of vast variable spaces and guiding experimental processes. Many companies developing new climate change mitigation technologies are starting to use data science and machine learning to accelerate their learning cycles.

Check out full paper here
African burned area and fire carbon emissions are strongly impacted by small fires undetected by coarse resolution satellite data

by Ruben Ramo, Ekhi Roteta, Ioannis Bistinas, Dave van Wees, Aitor Bastarrika, Emilio Chuvieco and Guido R. van der Werf

What: The authors compare a new Burned Area (BA) dataset generated from medium-resolution satellite images and its resulting fire carbon emissions with existing information based on global BA datasets derived from coarse-resolution data. The study focuses on the Sub-Saharan Africa region, which accounts for about 70% of the biomass burning worldwide.

Why: Current BA global data products use coarse spatial resolution satellite data, which hinders the detection of small fires. However, in regions with high anthropogenic fire activity, e.g., associated with agricultural and hunting practices, small fires have a significant role, both for land use transformation and fire emissions. 

How: The FireCCISFD11 product was generated from Sentinel-2 MSI data at 20-m resolution in combination with active fire observation from MODIS at 1-km resolution. The BA algorithm performs an initial classification of burned pixels using fixed thresholds which is overlaid with MODIS active fires. The final BA classification uses a two-stage procedure, by first detecting high burned probability seeds and then extending the burned regions.

Key findings: Small fires are the cause of an 80% increase in BA in Sub-Saharan Africa Small fires (observed from satellite products) have a high impact on BA, while alternative BA datasets widely used for estimation of CO2 emissions are very conservative.

CCAI perspective: Burned area products are extensively used in atmospheric and climate modelling, particularly fire carbon emissions estimates, an important contribution to global CO2 emissions. With higher-resolution satellite data becoming available, the development of improved algorithms for burned area estimation can have a significant contribution towards deriving realistic fire emissions estimates for a largely under-observed part of fire activity.

Check out full paper here
An OpenAI-Gym environment for the Building Optimization Testing (BOPTEST) framework

by Javier Arroyo, Carlo Manna, Fred Spiessens, and L. Helsen

What: The purpose of the proposed framework is to rigorously benchmark different reinforcement learning algorithms against model predictive control methods in building energy control applications. 

Why: The heating, ventilation, and air conditioning (HVAC) of buildings corresponds roughly to 40% of the global energy use. Advanced controllers, like model predictive control (MPC) or reinforcement learning (RL), can substantially enhance the energy efficiency of buildings while maintaining or even improving indoor thermal comfort. However, systematic comparison of data-driven and model-based approaches on realistic scenarios has proved to be difficult due to the lack of accessible high-fidelity emulator models and standardized performance indicators approved by the domain experts. 

How: The paper presents an OpenAI Gym environment for a set of high-fidelity building physics emulator models and a set of key performance indicators for a rigorous comparison. The environment leverages the BOPTEST project funded by the US Department of Energy (DOE) focusing on the development of open-source high-fidelity building models.

Key findings: This work fosters a novel interface that bridges the gap between the latest innovations of machine learning and the field of building energy management. The presented environment can be used to assess the performance of reinforcement learning algorithms when implemented in detailed and reliable building emulator models.

CCAI perspective: This paper represents a promising trend in combining methods and tools from machine learning, control engineering, and building physics. In the coming years, initiatives such as this one will drive the emergence of strong multidisciplinary research in domains such as energy-efficient buildings.

Check out full paper here


Do you have a job opportunity relevant to the Climate Change AI community? Share it with us by filling out this form and we will feature it in our next newsletter!


Ember is hiring a Energy and Climate Data Analyst [Location: Anywhere in Europe]. Due September 26th

Apollo Agriculture is hiring a Senior Data Scientist for their commercial farming platform. [Location: Nairobi/Amsterdam]

Gaiascope Inc., is hiring a Sr. Software Engineer. Gaiascope is led by CCAI’s Lauren Kuntz. [Location: Remote]

German Climate Computing Centre is hiring a Computational Research Scientist. [Location: Hamburg, Germany]

Comon Solutions is hiring for a Machine Learning Engineer. [Location: US Remote]

ClimateAI is hiring a Chief Product Officer. [Location: San Francisco, CA]

MitraChem is hiring for multiple positions including Data Scientist [Location: SF Bay Area, CA]

IntronHealth (servicing customers in Africa) is hiring for multiple positions including Data Scientists [Location: SF Bay Area, CA or Remote]

Winscape.AI is hiring for AI engineers [Location: Berkeley, CA]
Stanford Research Computing Center (SRCC) is looking for a Research Software Engineer. [Location: Palo Alto, CA]

Common Solutions is hiring a Machine Learning Engineer. [Location: Remote/San Francisco Bay Area, CA]

AMP Robotics is hiring for multiple positions in Machine Learning, Software Engineering and  Robotics. [Location: Louisville, CO, USA]

The Extreme Scale Computing and data platform for cloud-resolving weather and climate modeling project (EXCLAIM) based out of ETH Zurich is looking for software engineers. [Location: Zurich, Switzerland]

Jupiter Intelligence is looking for a Data Scientist and Geospatial Data Engineer. [Location: Remote/CO, USA]

Neuraspace is hiring a MLOps/Data Engineer. [Location: Coimbra, Portugal]

Google X is hiring an Algorithms Engineer. [Location: San Francisco, CA]

PARC, a Xerox Company, is hiring a Data Scientist and a Senior Mechanical Engineer for their cleantech projects. [Location: Palo Alto, CA]

Booz Allen is hiring a Chemist/Chemical Engineer. [Location: Washington D.C., USA]

Fluence Energy, a Siemens and AES company, is hiring for multiple positions including Software Engineering, Data Science, and Product Management across North America, Asia, Europe and Australia. 

Connect Earth is hiring a Data Scientist. [Location: Remote/London, UK]

PhD positions

Multiple PhD positions at Technical University of Denmark available in Data-driven Modeling and Energy Flexibility for Buildings. [Location:  Kongens Lyngby, Capital Region, Denmark]

PhD position available at Technical University of Denmark in AI for Electricity Market Design. [Location:  Kongens Lyngby, Capital Region, Denmark]

Multiple PhD positions available at the University of Texas at Austin in optimization for energy systems. [Location: Austin, TX, USA]

Fully funded PhD positions available at University of California, Irvine in the area of climate dynamics and land-atmosphere interactions, to begin in Fall 2022. [Location: Irvine, CA, USA] Due December 31

Multiple PhD positions available at the University of Massachusetts Amherst in the area of fluid dynamics, and energy systems including offshore wind turbines. [Location: Amherst, MA, USA]

Multiple PhD positions available at the University of Massachusetts Amherst for the NSF-funded ELEVATE program in the area of energy transition and climate change. [Location: Amherst, MA, USA]

PhD position available at the Center of Energy, Austrian Institute of Technology in  modeling short-term electricity markets using agent-based modeling and machine learning [Location: Seibersdorf, Austria]

PhD position available at Vrije Universiteit Amsterdam in the area of Climate Dynamics and Data Science. [Location: Amsterdam, Netherlands]

PhD position available at TU Berlin on Uncertainty Quantification in Energy Management. [Location: Berlin, Germany]

Fully funded PhD positions at University of California, Irvine in the area of climate dynamics and land-atmosphere interactions, to begin in Fall 2022. Due December 31

Postdoc positions

Postdoc position available at ETH Zurich in AI for Remote Sensing. [Location: Zurich, Switzerland]. Due September 30 

Multiple postdoc positions available at M²LInES (New York University/Princeton University) in the area of Machine Learning for Climate Extremes and ocean models, [Location: Princeton, NJ or New York, NY, USA]

Postdoc position available at Stanford University in the area of advanced energy systems modeling. [Location: Palo Alto, CA, USA]

Postdoc position available at Karlsruhe Institute of Technology (KIT) in the area of material science for battery/energy applications. [Location: Karlsruhe, Germany]

Multiple postdoc positions (five in total) available at the Los Alamos National Laboratory in the area of Computational Earth Sciences and Machine Learning [Location: Los Alamos, NM, USA]

Postdoc position available at Princeton University in the area of earth sciences modeling and machine learning. [Location: Princeton, NJ, USA]

Postdoc position available at the University of Washington in the area of machine learning and fisheries/ocean acoustics. [Location: Seattle, WA, USA]

Multiple postdoc positions available at the Center for Global Sustainability, University of Maryland in the area of Climate and Energy Innovation Policy. [Location: College Park, MD, USA]

Postdoc position available at PARC, a Xerox company, in the area of IoT technologies with climate applications. [Location: Palo Alto, CA]

Postdoc position available in the Department of Global Ecology, Stanford University on conducting basic and applied research into the interactions among Earth’s ecosystems and people using AI. [Location: Stanford, CA, USA]

Postdoc position available at Iowa State University in the area of atmospheric/climate science. [Location: Ames, IA, USA]. 

Postdoctoral position available at the Department of Earth and Atmospheric Sciences, Université du Québec à Montréal (UQAM) to conduct very high-resolution simulations of winter precipitation types. [Location: Montréal, Québec, Canada]

Multiple postdoc, research assistant positions available at the University of Minnesota Institute on the Environment (IonE) in the area of geospatial modeling, science policy, machine learning. [Location: Twin Cities, MN, USA]

Postdoctoral fellowship in climate science and machine learning available at Cotton University. [Location: Guwahati, India]

Postdoctoral research associate position available at Oak Ridge National Laboratory is in the area of developing AI/ML tools for scientific research (ex. in climate science) and for computational hydrology [Location: Oak Ridge, TN, USA]

Multiple Postdoctoral positions available in the Urban Analytics Lab at the National University of Singapore (NUS) to work on ML and geospatial data. [Location: Singapore]

Postdoctoral position in AI and energy systems available at the Energy Power Research Institute. [Location(s): Charlotte, NC or Knoxville, TN or Palo Alto, CA, USA]

Multiple postdoc positions available at Stanford University in the area of energy systems modeling. [Location: Palo Alto, CA, USA]

Two postdoc positions available at Stanford University in the areas of remote sensing, geospatial analyses, computer science, the oil and gas sector, and biogeoscience. [Location: Stanford, CA, USA]

Postdoc position available at Lawrence Berkeley Laboratory in the area of machine learning and earth sciences. [Location: Berkeley, CA, USA]

Faculty positions

Tenure-track faculty position available at Syracuse University in the area of fluid mechanics, and statistics [Location: Syracuse, NY]

For other job resources (not specifically ML), please see the list of recommended links on our community platform.

  CCAI Newsletter Feedback

We would like to ask you, our CCAI newsletter audience, to provide feedback (~2 mins) that will help us improve your experience. Please visit this link and/or write to us at with ‘NEWSLETTER’ in the subject line.
Copyright © 2021 Climate Change AI, All rights reserved.

Dates and deadlines reported in this newsletter may change, and CCAI is not responsible for any inadvertent inaccuracies.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Icons made by Nhor Phai, Payungkead, Becris, Freepik, and surang.

Email Marketing Powered by Mailchimp