Societal Adaptation & Resilience

Blog Posts

Discussion Seminars and Webinars

Innovation Grants

Talks

Workshop Papers

Venue Title
NeurIPS 2025 Spatial Uncertainty Quantification in Wildfire Forecasting for Climate-Resilient Emergency Planning (Papers Track)
Abstract and authors: (click to expand)

Abstract: Climate change is intensifying wildfire risks globally, making reliable forecasting critical for adaptation strategies. While machine learning shows promise for wildfire prediction from Earth observation data, current approaches lack uncertainty quantification essential for risk-aware decision making. We present the first systematic analysis of spatial uncertainty in wildfire spread forecasting using multimodal Earth observation inputs. We demonstrate that predictive uncertainty exhibits coherent spatial structure concentrated near fire perimeters. Our novel distance metric reveals high-uncertainty regions form consistent 20–60 meter buffer zones around predicted firelines - directly applicable for emergency planning. Feature attribution identifies vegetation health and fire activity as primary uncertainty drivers. This work enables more robust wildfire management systems supporting communities adapting to increasing fire risk under climate change.

Authors: Aditya Chakravarty (Texas A&M University)
NeurIPS 2025 Quantifying Climate Policy Action and Its Links to Development Outcomes: A Cross-National Data-Driven Analysis (Papers Track)
Abstract and authors: (click to expand)

Abstract: Addressing climate change effectively requires more than cataloguing the number of policies in place; thus it calls for tools that can predict their themes or subject, and analyze their tangible impacts on development outcomes. Existing assessments often rely on qualitative descriptions or composite indices, which can mask crucial differences between key domains such as mitigation, adaptation, disaster risk management, and loss and damage. To bridge this gap, we develop a quantitative indicator of climate policy orientation by applying a multilingual transformer-based language model to official national policy documents, achieving a classification accuracy of 0.90 (F1-score). Linking these indicators with World Bank development data in panel regressions reveals that mitigation policies are associated with higher GDP and GNI; disaster risk management correlates with greater GNI and debt but reduced foreign direct investment; adaptation and loss and damage show limited measurable effects. This integrated NLP–econometric framework enables comparable, theme-specific analysis of climate governance, offering a scalable method to monitor progress, evaluate trade-offs, and align policy emphasis with development goals. The code and datasets used in this study are publicly available at: https://github.com/booktrackerGirl/climate_change_policy_analysis.

Authors: Aditi Dutta (University of Exeter)
NeurIPS 2025 Emulating Climate Across Scales with Conditional Spherical Fourier Neural Operators (Papers Track)
Abstract and authors: (click to expand)

Abstract: Estimating local impacts of climate change is critical for informing adaptation methods. The ACE2 climate emulator successfully reproduces changes in historically observed climate, but poorly represents variability of key variables, such as surface precipitation, at small scales. We demonstrate that by adapting ACE2 to use conditional layer normalization and conditioning on isotropic Gaussian noise with a probabilistic loss function, we can successfully reproduce these small-scale features. This is a crucial step towards the goal of applying climate emulator predictions to inform real-world decisions.

Authors: Jeremy McGibbon (Allen Institute for Artificial Intelligence); Troy Arcomano (Allen Institute for Artificial Intelligence); Spencer Clark (Allen Institute for Artificial Intelligence); James Duncan (Allen Institute for Artificial Intelligence); Brian Henn (Allen Institute for Artificial Intelligence); Anna Kwa (Allen Institute for Artificial Intelligence); W. Andre Perkins (Allen Institute for Artificial Intelligence); Oliver Watt-Meyer (Allen Institute for Artificial Intelligence); Elynn Wu (Allen Institute for Artificial Intelligence); Christopher Bretherton (Allen Institute for Artificial Intelligence)
NeurIPS 2025 Detection and Simulation of Urban Heat Islands Using a Fine-Tuned Geospatial Foundation Model for Microclimate Impact Prediction (Papers Track)
Abstract and authors: (click to expand)

Abstract: As urbanization and climate change progress, urban heat island effects are becoming more frequent and severe. To formulate effective mitigation plans, cities require detailed air temperature data, yet conventional machine learning models with limited data often produce inaccurate predictions, particularly in underserved areas. Geospatial foundation models trained on global unstructured data offer a promising alternative by demonstrating strong generalization and requiring only minimal fine-tuning. In this study, an empirical ground truth of urban heat patterns is established by quantifying cooling effects from green spaces and benchmarking them against model predictions to evaluate the model’s accuracy. The foundation model is subsequently fine-tuned to predict land surface temperatures under future climate scenarios, and its practical value is demonstrated through a simulated in-painting that highlights its role for mitigation support. The results indicate that foundation models offer a powerful way for evaluating urban heat island mitigation strategies in data-scarce regions to support more climate-resilient cities.

Authors: Jannis Fleckenstein (IBM); David Kreismann (IBM); Tamara Rosemary Govindasamy (IBM); Thomas Brunschwiler (IBM); Etienne Vos (IBM); Mattia Rigotti (IBM)
NeurIPS 2025 Coupled Climate Simulations with ACE and Samudra (Papers Track)
Abstract and authors: (click to expand)

Abstract: We present a coupling of the Ai2 Climate Emulator (ACE) 3D global atmosphere emulator to the Samudra 3D global ocean emulator, both of which are large autoregressive ML models with a combined total of nearly 600 million parameters. A coupled emulator has the potential advantage of accelerating climate change projections under different future scenarios, enabling more iterative and insightful strategies for climate policy, adaptation, and mitigation. The coupled emulator facilitates the exchange of boundary conditions between separate models of the atmosphere and ocean, with prognostic sea ice included among Samudra’s outputs. The coupled emulator produces a stable climate with remarkably small climate biases, a good seasonal cycle of sea ice, insignificant temporal climate drift, and realistic ENSO variability. The coupled emulator marks a significant step toward enabling fully coupled climate modeling with emulators.

Authors: Elynn Wu (Ai2); James Duncan (Ai2); Troy Arcomano (Ai2); Spencer Clark (Ai2); Brian Henn (Ai2); Anna Kwa (Ai2); Jeremy McGibbon (Ai2); Andre Perkins (Ai2); Oliver Watt-Meyer (Ai2); Christopher Bretherton (Ai2); Surya Dheeshjith (NYU); Adam Subel (NYU); Laure Zanna (NYU); William Hurlin (GFDL); William Gregory (Princeton University); Alistair Adcroft (Princeton University)
NeurIPS 2025 Scalable Geospatial Data Generation Using AlphaEarth Foundations Model (Papers Track)
Abstract and authors: (click to expand)

Abstract: High-quality labeled geospatial datasets are essential for extracting insights and understanding our planet. Unfortunately, these datasets often do not span the entire globe and are limited to certain geographic regions where data was collected. Google DeepMind’s recently released AlphaEarth Foundations (AEF) provides an information-dense global geospatial representation designed to serve as a useful input across a wide gamut of tasks. In this article we propose and evaluate a methodology which leverages AEF to extend geospatial labeled datasets beyond their initial geographic regions. We show that even basic models like random forests or logistic regression can be used to accomplish this task. We investigate a case study of extending LANDFIRE's Existing Vegetation Type (EVT) dataset beyond the USA into Canada at two levels of granularity: EvtPhys (13 classes) and EvtGp (80 classes). Qualitatively, for EvtPhys, model predictions align with ground truth. Trained models achieve 81% and 73% classification accuracy on EvtPhys validation sets in the USA and Canada, despite discussed limitations.

Authors: Luc Houriez (X, The Moonshot Factory); Sebastian Pilarski (X, The Moonshot Factory); Behzad Vahedi (X, The Moonshot Factory); Ali Ahmadalipour (X, The Moonshot Factory); Teo Honda Scully (X, The Moonshot Factory); Nicholas Aflitto (X, The Moonshot Factory); David Andre (X, The Moonshot Factory); Caroline Jaffe (X, The Moonshot Factory); Martha Wedner (X, The Moonshot Factory); Rich Mazzola (X, The Moonshot Factory); Josh Jefferey (X, The Moonshot Factory); Ben Messinger (X, The Moonshot Factory); Sage McGinley-Smith (X, The Moonshot Factory); Sarah Russell (X, The Moonshot Factory)
NeurIPS 2025 A Graph Neural Network Approach for Localized and High-Resolution Temperature Forecasting (Papers Track)
Abstract and authors: (click to expand)

Abstract: Heatwaves are intensifying worldwide and are among the deadliest weather disasters. The burden falls disproportionately on marginalized populations and the Global South, where under-resourced health systems, exposure to urban heat islands, and the lack of adaptive infrastructure amplify risks. Yet current numerical weather prediction models often fail to capture micro-scale extremes, leaving the most vulnerable excluded from timely early warnings. We present a Graph Neural Network framework for localized, high-resolution temperature forecasting. By leveraging spatial learning and efficient computation, our approach generates forecasts at multiple horizons, up to 48 hours. For Southwestern Ontario, Canada, the model captures temperature patterns with a mean MAE of 1.93°C across 1–48h forecasts and MAE@48h of 2.93°C, evaluated using 24h input windows on the largest region. While demonstrated here in a data-rich context, this work lays the foundation for transfer learning approaches that could enable localized, equitable forecasts in data-limited regions of the Global South.

Authors: Joud El-Shawa (Western University / Vector Institute); Elham Bagheri (Western University / Vector Institute); Sedef Akinli Kocak (Vector Institute); Yalda Mohsenzadeh (Western University / Vector Institute)
NeurIPS 2025 Geospatial Chain of Thought Reasoning for Enhanced Visual Question Answering on Satellite Imagery (Papers Track)
Abstract and authors: (click to expand)

Abstract: Geospatial chain of thought (CoT) reasoning is essential for advancing Visual Question Answering (VQA) on satellite imagery, particularly in climate related applications such as disaster monitoring, infrastructure risk assessment, urban resilience planning, and policy support. Existing VQA models enable scalable interpretation of remote sensing data but often lack the structured reasoning required for complex geospatial queries. We propose a VQA framework that integrates CoT reasoning with Direct Preference Optimization (DPO) to improve interpretability, robustness, and accuracy. By generating intermediate rationales, the model better handles tasks involving detection, classification, spatial relations, and comparative analysis, which are critical for reliable decision support in high stakes climate domains. Experiments show that CoT supervision improves accuracy by 34.9% over direct baselines, while DPO yields additional gains in accuracy and reasoning quality. The resulting system advances VQA for multispectral Earth observation by enabling richer geospatial reasoning and more effective climate use cases.

Authors: Shambhavi Shanker (IIT Bombay); Manikandan Padmanaban (IBM Research India); Jagabondhu Hazra (IBM Research India)
NeurIPS 2025 Reflexive Evidence-Based Multimodal Learning for Clean Energy Transitions: Causal Insights on Cooking Fuel Access, Urbanization, and Carbon Emissions (Papers Track)
Abstract and authors: (click to expand)

Abstract: Achieving Sustainable Development Goal 7 (Affordable and Clean Energy) requires not only technological innovation but also a deeper understanding of the socio-economic factors that influence energy access and carbon emissions. Despite growing attention to these drivers, key questions remain, particularly regarding how to quantify socio-economic impacts, how these impacts interact across domains such as policy, technology, and infrastructure, and how feedback processes shape energy systems. To address these gaps, this study introduces ClimateAgents, an AI-based framework that combines large language models with domain-specialized agents to support hypothesis generation and scenario exploration. Leveraging 20 years of socio-economic and emissions data from 265 economies, countries and regions, and 98 indicators drawn from the World Bank database, the framework applies a machine learning–based causal inference approach to identify key determinants of carbon emissions in an evidence-based, data-driven manner. The analysis highlights three primary drivers: (1) access to clean cooking fuels in rural areas, (2) access to clean cooking fuels in urban areas, and (3) the percentage of population living in urban areas. These findings underscore the critical role of clean cooking technologies and urbanization patterns in shaping emission outcomes. In line with growing calls for evidence-based AI policy, ClimateAgents offers a modular and reflexive learning system that supports the generation of credible and actionable insights for policy. By integrating heterogeneous data modalities, including structured indicators, policy documents, and semantic reasoning, the framework contributes to adaptive policymaking infrastructures that can evolve with complex socio-technical challenges. This approach aims to support a shift from siloed modeling to reflexive, modular systems designed for dynamic, context-aware climate action.

Authors: Shan Shan (Zhejiang University)
NeurIPS 2025 A modular framework to run AI-based models from high-resolution climate projections (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Recent advances in AI-based weather and climate models promise transformative improvements in forecasting, yet their integration with state-of-the-art climate simulations remains constrained by data heterogeneity. High-resolution projections, such as those from the Destination Earth Climate Digital Twin, are produced at 5 km resolution with specialized grids, formats, and variable sets that are incompatible with most AI models. Current integration efforts are ad-hoc, model-specific, and difficult to reproduce, slowing progress and limiting large-scale evaluation. Taking advantage of decades of experience in climate workflows, we propose a modular, source-agnostic framework that enables systematic and reproducible execution of AI-based climate models across diverse datasets and high-performance computing environments. The framework standardizes data preparation, automates model execution through containerized workflows, and provides built-in post-processing and evaluation tools. Preliminary experiments show that the framework reproduces and extends recent studies on AI model robustness in future climates with minimal technical overhead.

Authors: Aina Gaya-Àvila (Barcelona Supercomputing Center); Amirpasha Mozaffari (Barcelona Supercomputing Center); Amanda Duarte (Barcelona Supercomputing Center); Oriol Tintó Prims (Barcelona Supercomputing Center)
NeurIPS 2025 From Sparse to Representative: Machine Learning to Densify IAM Scenario Ensembles for Policy Insight (Proposals Track)
Abstract and authors: (click to expand)

Abstract: This research addresses the challenge of extracting policy-relevant insights from Integrated Assessment Model (IAM) scenario ensembles, which are often sparse, non-representative, and inaccessible to non-experts. We propose a machine learning framework preserving high-dimensional dependencies between variables, enabling generation of plausible in-gap scenarios when one or more outputs are constrained. The intended output is a simplified exploration space for policymakers concerned with crucial climate policy exploration.

Authors: Georgia Ray (Imperial College London)
NeurIPS 2025 Low-Power Weakly-Supervised Audio Detection for Real-World Mosquito Surveillance (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Climate change is expanding the geography and altering habitats of mosquitoes that transmit Dengue, Zika, etc. For example, China’s first large-scale chikungunya outbreak in August 2025 underscores the urgency for scalable and adaptive mosquito surveillance. Existing approaches often rely on expert entomologists or extensive labeled datasets, making them poorly suited for rapid deployment in low-resource settings. Our approach uses weak labels from time-aligned video recordings to train audio classifiers via Multi Instance Learning (MIL). This approach combines local data, citizen labelers and low power audio classification to enable rapid adaptation to climate driven outbreaks. Early proof points suggest potential for MIL in audio classification from mosquito video labels.

Authors: Danika Gupta (The Harker Upper School); Ming Zhao (Arizona State University); Neha Rajendra Vadnere (Arizona State University)
NeurIPS 2025 Scalable Country-Level Crop Yield Modeling for Food Security and Risk Mitigation (Proposals Track)
Abstract and authors: (click to expand)

Abstract: We propose a scalable framework for generating regional crop yield predictions at the country level using coarse-resolution data, designed to support food security monitoring, policy planning, and climate adaptation. While much recent work focuses on fine-grained, field-level yield prediction, such models often require high-resolution data unavailable for many low-income countries. Our approach leverages coarser signals from satellites and household surveys, demonstrating through a Uganda proof-of-concept that machine learning models trained on regional averages can approach theoretical accuracy limits for smallholder yield prediction. We propose to extend this method by building country-specific models across all LSMS-ISA partner countries (Burkina Faso, Ethiopia, Malawi, Mali, Niger, Nigeria, Tanzania, and Uganda), and then constructing a pooled multi-country model. Comparing country-specific and multi-country models will allow us to assess generalizability in data-scarce settings. These models could provide actionable early warning signals of food shortages, enabling governments, NGOs, and international agencies to better target interventions and allocate resources under climate stress.

Authors: Andrew Hobbs (University of San Francisco); Jesse Anttila-Hughes (University of San Francisco)
NeurIPS 2025 GeoWaste: Leveraging GIS and Machine Learning for Urban Waste Management in African Cities (Proposals Track) Best Pathway to Impact
Abstract and authors: (click to expand)

Abstract: Waste collection in many cities in Africa remains ineffective, with little reliable data, fragmented reporting, and static truck routes, all of which lead to increased greenhouse gas emissions and overflowing bins. We present GeoWaste, a GIS and machine learning-driven system that integrates open-source geospatial datasets, GPS-enabled collection trucks, citizen geo-tagged reports, and fill-level sensors to deliver an integrated spatio-temporal waste database. GeoWaste generates optimised routes and makes forecasts using a Random Forest regressor to predict waste volumes, heuristic solutions for routing, and clustering (K-Means, DBSCAN) to locate hotspots. A pilot was done in Yenagoa, the state capital of Bayelsa State, covering 120 bins and 8540 households. Service coverage was increased from 62% to 87%, average collection time was reduced by 32%, and truck fuel use was reduced by 28%. GeoWaste demonstrates a scalable pathway for data-driven, climate-smart urban resilience.

Authors: Bakumor Yolo (University of Calabar)
NeurIPS 2025 Prioritization Learning for Equitable Residential Decarbonization Investments (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Upgrading residential buildings with energy-efficient heating and electrical equipment, including heat pumps, electric cooktops, solar panels, and efficient insulation, is a crucial component of transitioning to a more efficient, climate-resilient, and affordable energy system. The residential and commercial building sector is responsible for 31 percent of greenhouse gas emissions in the United States. Moreover, to advance energy equity, priority populations, defined by California Climate Investments (CCI) as the category of homes particularly vulnerable to energy poverty and the impacts of climate change, must be brought to the forefront of transition plans. Doing so will reduce national energy costs and decrease the average energy burden, or the percentage of income spent on utility bills. The California Energy Commission (CEC) has implemented efforts through the Federal Inflation Reduction Act (IRA) to provide affordable retrofitting programs. However, the process of manually selecting which homes to upgrade is too inefficient to meet the pace required for the rapid and robust transition needed to combat climate change. I propose leveraging the emission and savings results from the National Renewable Energy Laboratory's ResStock database to develop a prioritization algorithm that optimizes home selection for residential decarbonization efforts. Together, this ranking system uses algorithmic intelligence and user input to optimize energy affordability, emission reduction, electricity usage reduction, and cost savings, thus advancing residential electrification schemes.

Authors: Eva Geierstanger (Stanford University)
ICLR 2025 Predicting extreme weather impacts on physical activity and sleep patterns using real-world data from wrist-worn accelerometers (Papers Track)
Abstract and authors: (click to expand)

Abstract: The increasing frequency of extreme weather events, such as heat waves, is among the most pressing consequences of climate change, with profound implications for human health and well-being. Despite increasing incidence of extreme weather events globally, there is a lack of understanding on the impact of hot weather on health outcomes. In this study, we utilized machine learning techniques to explore how variations in outdoor temperature influence physical activity and sleep patterns, two critical determinants of physical and mental health. Using data from 90,434 participants in the UK Biobank, recorded via wrist-worn accelerometers, linked with meteorological data from the UK Met Office, we analysed the relationship between outdoor temperature (5°C to 30°C) and daily magnitudes and durations of a) physical activity and b) sleep, whilst adjusting for sociodemographic, clinical, lifestyle, seasonality, precipitation, and regional variables. Our results reveal that moderate-to-vigorous physical activity (MVPA) increases with temperature, reaching its peak at 25°C, but plateaus thereafter. Conversely, sedentary behaviour and sleep disturbances significantly intensify as temperatures reach 30. Here tested in UK settings, our approach is generalisable to other climatic regions and determinants of health and should be further investigated in regions with high climate-vulnerability. These findings emphasize the role of machine learning in identifying health risks associated with climate change and underscore the necessity of climate-adaptive public health strategies to mitigate these effects.

Authors: Sara khalid (University of Oxford)
ICLR 2025 Towards Flood Extent Forecasting: Evaluating a Weather Foundation Model and U-Net for Flood Forecasting. (Papers Track)
Abstract and authors: (click to expand)

Abstract: This study explores a data-driven approach that combines flood forcing factors from observation and reanalysis datasets, antecedent flood extent maps, and deep learning to forecast daily flood extents in Rwanda. We extend the architecture used in ClimaX (transformer weather and climate foundation model), investigate its pretrained representations for flood forecasting, and compare performance against a U-Net baseline. Our results demonstrate that a ClimaX variant trained from scratch with a linear projection decoder outperforms the U-Net and other ClimaX variants, highlighting its potential as an effective tool for flood extent forecasting. This work underscores the potential of data-driven deep learning models for flood extent forecasting with implications for improving disaster preparedness and flood risk assessment in vulnerable regions.

Authors: Eric Wanjau (UCL); Samuel Maina (Microsoft)
ICLR 2025 CLIMGEN: Learning the Forcing-Response Relationship in Climate System (Papers Track)
Abstract and authors: (click to expand)

Abstract: Solar Radiation Management (SRM) is emerging as a potential geoengineering strategy to address the anthropogenic impact on climate, but its effective implementation requires an iterative and large ensemble of highly accurate and efficient climate projections. Traditional climate projections rely on executing computationally demanding and time-consuming numerical climate models. Recent advances in machine learning (ML) aim to enhance these approaches by emulating traditional methods. In this work, we propose a novel framework for directly learning the relationship between solar radiation flux at the top of the atmosphere and the corresponding surface temperature response. To evaluate the feasibility of this direct ML-based projection, we developed a dataset using an intermediate complexity model, incorporating a comprehensive suite of different forcing patterns and evaluation metrics to rigorously assess the ML model’s performance. We introduce a Conditional Denoising Diffusion Probabilistic Model (cDDPM) for this task, which demonstrates encouraging skill in representing climate statistics under previously unseen forcing patterns. This approach provides a promising pathway for direct climate projections by accurately learning the forcing-response relationship, with a wide range of applications in impact mitigation, emissions policy design, and SRM strategies.

Authors: Tse-Chun Chen (Pacific Northwest National Laboratory); Parvathi Kooloth (Pacific Northwest National Laboratory); Jian Lu (Pacific Northwest National Laboratory); Jason Z. Hou (Pacific Northwest National Laboratory)
ICLR 2025 Populous: A Multimodal Geospatial AI Model for Understanding the Climate-Driven Insurance Crisis in the U.S. (Papers Track)
Abstract and authors: (click to expand)

Abstract: The U.S. home insurance market is experiencing growing instability due to rising climate risks, surging premiums, and widespread non-renewals. Despite the urgent need for data-driven insights, publicly available datasets remain limited, making it difficult to assess and mitigate climate-driven insurance risk. This study introduces an AI-driven modeling framework that predicts home insurance premiums and non-renewals at high spatial resolution. By integrating multimodal data—including Google’s Population Dynamics Foundation Model embeddings (leveraging transfer learning), socioeconomic indicators, historical loss records, and physical climate risk—the model first predicts county-level trends before applying embedding-based super-resolution to refine predictions at the zipcode level. Results show strong predictive performance, achieving 93% top-2 accuracy in premium classification and explaining 61% of the variability in non-renewals at the county level. Feature importance analysis highlights climate hazards, such as hurricane and wildfire risk, as key drivers of recent insurance market shifts. By offering fine-scale, data-driven insights, this study not only enhances understanding of how climate change is reshaping the insurance industry but also provides actionable guidance to incentivize resilient building, discourage high-risk development, and inform policies aimed at mitigating climate-related insurance volatility.

Authors: Tristan Ballard (Sust Global); Gopal Erinjippurath (Sust Global); Michael Sierks (Sust Global); Peter Sousounis (Sust Global)
ICLR 2025 Large Language Models for Monitoring Dataset Mentions in Climate Research (Papers Track)
Abstract and authors: (click to expand)

Abstract: Effective climate change research relies on diverse datasets to inform mitigation and adaptation strategies and policies. However, the ways these datasets are cited, used, and distributed remain poorly understood. This paper presents a machine learning framework that automates the detection and classification of dataset mentions in climate research papers. Leveraging large language models (LLMs), we generate a weakly supervised dataset through zero-shot extraction, quality assessment via an LLM-as-a-Judge, and refinement by a reasoning agent. The Phi-3.5-mini instruct model is pre-fine-tuned on this dataset, followed by fine-tuning on a smaller manually annotated subset to specialize in extracting data mentions. At inference, a ModernBERT-based classifier filters for dataset mentions, optimizing computational efficiency. Evaluated on a held-out manually annotated sample, our fine-tuned model outperforms NuExtract-v1.5 and GLiNER-large-v2.1 in dataset extraction accuracy. As a framework for monitoring dataset mentions in research papers, this approach enhances transparency, identifies data gaps, and enables researchers, funders, and policymakers to improve data discoverability and usage, leading to more informed decision-making.

Authors: Aivin Solatorio (The World Bank); Rafael Macalaba (The World Bank); James Liounis (The World Bank)
ICLR 2025 Using Reinforcement Learning to Integrate Subjective Wellbeing into Climate Adaptation Decision Making (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Subjective wellbeing is a fundamental aspect of human life, influencing life expectancy and economic productivity, among others. Mobility plays a critical role in maintaining wellbeing, yet the increasing frequency and intensity of both nuisance and high-impact floods due to climate change are expected to significantly disrupt access to activities and destinations, thereby affecting overall wellbeing. Addressing climate adaptation presents a complex challenge for policymakers, who must select and implement policies from a broad set of options with varying effects while managing resource constraints and uncertain climate projections. In this work, we propose a multi-modular framework that uses reinforcement learning as a decision-support tool for climate adaptation in Copenhagen, Denmark. Our framework integrates four interconnected components: long-term rainfall projections, flood modeling, transport accessibility, and wellbeing modeling. This approach enables decision-makers to identify spatial and temporal policy interventions that help sustain or enhance subjective wellbeing over time. By modeling climate adaptation as an open-ended system, our framework provides a structured framework for exploring and evaluating adaptation policy pathways. In doing so, it supports policymakers to make informed decisions that maximize wellbeing in the long run.

Authors: Arthur Vandervoort (Technical University of Denmark); Miguel Costa (Technical University of Denmark); Morten W. Petersen (Technical University of Denmark); Martin Drews (Technical University of Denmark); Sonja Haustein (Technical University of Denmark); Karyn Morrissey (University of Galway); Francisco C. Pereira (Technical University of Denmark)
ICLR 2025 From Rumors to Risk: Mapping and Modeling Climate-Disaster Misinformation (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Recent years have seen a surge in climate-disaster misinformation, with social media amplifying unfounded claims in the lead-up to and aftermath of major disasters. This misinformation has hindered disaster preparation and recovery while fueling harassment against meteorologists and government officials, eroding trust in scientific institutions. While tools exist for analyzing general climate-change misinformation, current datasets often overlook the rapidly shifting narratives tied to specific events like wildfires, floods, or hurricanes. This proposal addresses that gap by developing a dynamic, evolving dataset on climate-disaster misinformation. Built through targeted social media data collection and rigorous labeling, the dataset will adapt alongside AI/ML advancements through iterative feedback from model performance and emerging trends. This openly accessible resource will enable researchers and practitioners to refine detection algorithms, design interventions, and inform crisis communication strategies—ensuring both data and models remain aligned with the shifting misinformation landscape. Ultimately, this work seeks to clarify key drivers of misinformation propagation and support more effective climate disaster response.

Authors: Tristan Ballard (Independent)
NeurIPS 2024 A Water Efficiency Dataset for African Data Centers (Papers Track)
Abstract and authors: (click to expand)

Abstract: AI computing and data centers consume a large amount of freshwater, both directly for cooling and indirectly for electricity generation. While most attention has been paid to developed countries such as the U.S., this paper presents the first-of-its-kind dataset that combines nation-level weather and electricity generation data to estimate water usage efficiency for data centers in 41 African countries across five different climate regions. We also use our dataset to evaluate and estimate the water consumption of inference on two large language models (i.e., Llama-3-70B and GPT-4) in 11 selected African countries. Our findings show that writing a 10-page report using Llama-3-70B could consume about 0.7 liters of water, while the water consumption by GPT-4 for the same task may go up to about 60 liters. For writing a medium-length email of 120-200 words, Llama-3-70B and GPT-4 could consume about 0.13 liters and 3 liters of water, respectively. Interestingly, given the same AI model, 8 out of the 11 selected African countries consume less water than the global average, mainly because of lower water intensities for electricity generation. However, water consumption can be substantially higher in some African countries with a steppe climate than the U.S. and global averages, prompting more attention when deploying AI computing in these countries.

Authors: NOAH SHUMBA (Carnegie Mellon University Africa); Opelo Tshekiso (Carnegie Mellon University Africa); Pengfei Li (UCR); Giulia Fanti (CMU); Shaolei Ren (UC Riverside)
NeurIPS 2024 Advancing Heat Demand Forecasting with Attention Mechanisms: Opportunities and Challenges (Papers Track)
Abstract and authors: (click to expand)

Abstract: Global leaders and policymakers are unified in their unequivocal commitment to decarbonization efforts in support of Net-Zero agreements. District Heating Systems (DHS), while contributing to carbon emissions due to the continued reliance on fossil fuels for heat production, are embracing more sustainable practices albeit with some sense of vulnerability as it could constrain their ability to adapt to dynamic demand and production scenarios. As demographic demands grow and renewables become the central strategy in decarbonizing the heating sector, the need for accurate demand forecasting has intensified. Advances in digitization have paved the way for Machine Learning (ML) based solutions to become the industry standard for modeling complex time series patterns. In this paper, we focus on building a Deep Learning (DL) model that uses deconstructed components of independent and dependent variables that affect heat demand as features to perform multi-step ahead forecasting of head demand. The model represents the input features in a time-frequency space and uses an attention mechanism to generate accurate forecasts. The proposed method is evaluated on a real-world dataset and the forecasting performance is assessed against LSTM and CNN-based forecasting models. Across different supply zones, the attention-based models outperforms the baselines quantitatively and qualitatively, with an Mean Absolute Error (MAE) of 0.105 0.06kWh and a Mean Absolute Percentage Error (MAPE) of 5.4% 2.8%, in comparison the second best model with a MAE of 21 0.10 0.06kWh and a MAPE of 5.6% 3%.

Authors: Adithya Ramachandran (Pattern Recognition Lab, Friedrich Alexander University, Erlangen); Andreas Maier (Pattern Recognition Lab, FAU Erlangen-Nuremberg); Siming Bayer (Pattern Recognition Lab, Friedrich-Alexander University)
NeurIPS 2024 Climate Impact Assessment Requires Weighting: Introducing the Weighted Climate Dataset (Papers Track)
Abstract and authors: (click to expand)

Abstract: High-resolution gridded climate data are readily available from multiple sources, yet climate research and decision-making increasingly require country and region-specific climate information weighted by socio-economic factors. Moreover, the current landscape of disparate data sources and inconsistent weighting methodologies exacerbates the reproducibility crisis and undermines scientific integrity. To address these issues, we have developed a globally comprehensive dataset at both country (GADM0) and region (GADM1) levels, encompassing various climate indicators (precipitation, temperature, SPEI, wind gust). Our methodology involves weighting gridded climate data by population density, night-time light intensity, cropland area, and concurrent population count – all proxies for socio-economic activity – before aggregation. We process data from multiple sources, offering daily, monthly, and annual climate variables spanning from 1900 to 2023. A unified framework streamlines our preprocessing steps, and rigorous validation against leading climate impact studies ensures data reliability. The resulting Weighted Climate Dataset is publicly accessible through an online dashboard at https://weightedclimatedata.streamlit.app/.

Authors: Marco Gortan (University of Basel); Lorenzo Testa (Carnegie Mellon University); Giorgio Fagiolo (Sant'Anna School of Advanced Studies); Francesco Lamperti (Sant'Anna School of Advanced Studies)
NeurIPS 2024 WildfireGPT: Tailored Large Language Model for Wildfire Analysis (Papers Track)
Abstract and authors: (click to expand)

Abstract: Recent advancement of large language models (LLMs) represents a transformational capability at the frontier of artificial intelligence. However, LLMs are generalized models, trained on extensive text corpus, and often struggle to provide context-specific information, particularly in areas requiring specialized knowledge, such as wildfire details within the broader context of climate change. For decision-makers focused on wildfire resilience and adaptation, it is crucial to obtain responses that are not only precise but also domain-specific. To that end, we developed WildfireGPT, a prototype LLM agent designed to transform user queries into actionable insights on wildfire risks. We enrich WildfireGPT by providing additional context, such as climate projections and scientific literature, to ensure its information is current, relevant, and scientifically accurate. This enables WildfireGPT to be an effective tool for delivering detailed, user-specific insights on wildfire risks to support a diverse set of end users, including but not limited to researchers and engineers, for making positive impact and decision making.

Authors: Yangxinyu Xie (University of Pennsylvania); Bowen Jiang (University of Pennsylvania); Tanwi Mallick (Argonne National Laboratory); Joshua Bergerson (Argonne National Laboratory); John Hutchison (Argonne National Laboratory); Duane Verner (Argonne National Laboratory); Jordan Branham (Argonne National Laboratory); M. Ross Alexander (Argonne National Laboratory); Robert Ross (Argonne National Laboratory); Yan Feng (Argonne National Laboratory); Leslie-Anne Levy (Argonne National Laboratory); Weijie Su (University of Pennsylvania); Camillo Jose Taylor (University of Pennsylvania)
NeurIPS 2024 Critical misalignments between climate action and sustainable development goals revealed (Papers Track)
Abstract and authors: (click to expand)

Abstract: A mere 12 percent of the Sustainable Development Goals (SDGs) is currently on track to meet the 2030 deadline in a world under climate change. Since their launch in 2015, the 2030 Agenda for Sustainable Development and the Paris Agreement have suffered persistent mismatches, which limit the potential for mutual gains. We use Artificial Intelligence (AI) to assess the degree and type of alignment between the Nationally Determined Contributions (NDCs) and the SDGs. While high income countries tackle the energy-infrastructure-community nexus in term of opportunity, lower income countries make climate impacts more explicit and center their trade-offs around the water-energy-food nexus. These two approaches mark different development trajectories and have non-negligible implications on international financial flow architecture and climate governance.

Authors: Francesca Larosa (Royal Institute for Technology); Sergio Hoyas (Universitat Politècnica de València); Fermin Mallor Franco (Royal Institute of Technology); J. Alberto Conejero (Universitat Politècnica de València); Javier García-Martinez (University of Alicante); Francesco Fuso Nerini (Royal Institute of Technology); Ricardo Vinuesa (KTH Royal Institute of Technology)
NeurIPS 2024 Exploring Climate Awareness and Anxiety in Teens: An Expert-Driven AI Perspective (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Climate awareness and climate anxiety often go hand in hand.The growing awareness of climate change among young people is increasingly shadowed by climate anxiety, a condition marked by profound stress, fear, and a sense of helplessness stemming from overwhelming information on environmental crises and perceived inaction by authorities. Our proposal is an innovative approach using Large Language Model (LLM)-based chatbots to support children and adolescents in fostering climate awareness, managing climate anxiety, and promoting sustainable practices. By collaborating with schools and engaging a multidisciplinary experts and the young people themselves, we seek to co-create an impactful educational intervention. Starting with a comprehensive survey of school counsellors to map the current state of climate awareness and anxiety, and to understand their expectations for our AI solution, this project is poised for global scalability, addressing the pressing mental health challenges associated with climate change, particularly in vulnerable and resource-constrained regions.

Authors: Sruthi Viswanathan (University of Oxford and The Spaceship Academy); Omar Mohammed (Independant Researcher); Craig Vezina (The Spaceship Academy); Mia Doces (Commitee For Children)
NeurIPS 2024 Flood Prediction in Kenya - Leveraging Pre-Trained Models to Generate More Validation Data in Sparse Observation Settings (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Kenya has lacked a national flood risk management framework and also has sparse flood observation data, which makes developing deep learning flood prediction models on a national scale challenging. Flood prediction models are critical to operationalize Early Warning Systems (EWS). We propose two different models to feed into an EWS. The first model will leverage statistical machine learning approaches to predict flood or no flood events on a 0.25 x 0.25 degree scale (approximately 30 km x 30 km in Kenya) using ERA5 features as well as land cover and Digital Terrain Model (DTM) data. This first model will also be used to create a baseline prediction benchmark across the entire country of Kenya. The second model will leverage pre-trained remote sensing based models to generate segmented flood or no flood data on a fine spatial scale. This will increase the number of validation points by a factor of 1000, which opens the door to deep learning approaches to predict flood or no-flood events on a 30 meter x 30 meter spatial scale. We hope that this approach of leveraging pre-trained models to generate fine scale validation data can eventually be used widely in other extreme climate event forecasting scenarios given the scarcity of historical extreme climate events compared to normal weather events.

Authors: Alim Karimi (Self / Purdue University); David Quispe (University of Toronto); Hammed Akande (Concordia University); Nicole Mong'are (Athi Water Works Development Agency); Valerie Brosnan (Mitga Solutions); Asbina Baral (Ministry of Education, Science and Technology)
NeurIPS 2023 Combining deep generative models with extreme value theory for synthetic hazard simulation: a multivariate and spatially coherent approach (Papers Track)
Abstract and authors: (click to expand)

Abstract: Climate hazards can cause major disasters when they occur simultaneously as compound hazards. To understand the distribution of climate risk and inform adaptation policies, scientists need to simulate a large number of physically realistic and spatially coherent events. Current methods are limited by computational constraints and the probabilistic spatial distribution of compound events is not given sufficient attention. The bottleneck in current approaches lies in modelling the dependence structure between variables, as inference on parametric models suffers from the curse of dimensionality. Generative adversarial networks (GANs) are well-suited to such a problem due to their ability to implicitly learn the distribution of data in high-dimensional settings. We employ a GAN to model the dependence structure for daily maximum wind speed, significant wave height, and total precipitation over the Bay of Bengal, combining this with traditional extreme value theory for controlled extrapolation of the tails. Once trained, the model can be used to efficiently generate thousands of realistic compound hazard events, which can inform climate risk assessments for climate adaptation and disaster preparedness. The method developed is flexible and transferable to other multivariate and spatial climate datasets.

Authors: Alison M Peard (University of Oxford); Jim Hall (University of Oxford)
NeurIPS 2023 Can Reinforcement Learning support policy makers? A preliminary study with Integrated Assessment Models (Papers Track)
Abstract and authors: (click to expand)

Abstract: Governments around the world aspire to ground decision-making on evidence. Many of the foundations of policy making — e.g. sensing patterns that relate to societal needs, developing evidence-based programs, forecasting potential outcomes of policy changes, and monitoring effectiveness of policy programs — have the potential to benefit from the use of large-scale datasets or simulations together with intelligent algorithms. These could, if designed and deployed in a way that is well grounded on scientific evidence, enable a more comprehensive, faster, and rigorous approach to policy making. Integrated Assessment Models (IAM) is a broad umbrella covering scientific models that attempt to link main features of society and economy with the biosphere into one modelling framework. At present, these systems are probed by by policy makers and advisory groups in a hypothesis-driven manner. In this paper, we empirically demonstrate that modern Reinforcement Learning can be used to probe IAMs and explore the space of solutions in a more principled manner. While the implication of our results are modest since the environment is simplistic, we believe that this is a stepping stone towards more ambitious use cases, which could allow for effective exploration of policies and understanding of their consequences and limitations.

Authors: Theodore LM Wolf (Carbon Re); Nantas Nardelli (CarbonRe); John Shawe-Taylor (University College London); Maria Perez-Ortiz (University College London)
NeurIPS 2023 Towards Recommendations for Value Sensitive Sustainable Consumption (Papers Track)
Abstract and authors: (click to expand)

Abstract: Excessive consumption can strain natural resources, harm the environment, and widen societal gaps. While adopting a more sustainable lifestyle means making significant changes and potentially compromising personal desires, balancing sustainability with personal values poses a complex challenge. This article delves into designing recommender systems using neural networks and genetic algorithms, aiming to assist consumers in shopping sustainably without disregarding their individual preferences. We approach the search for good recommendations as a problem involving multiple objectives, representing diverse sustainability goals and personal values. While using a synthetic historical dataset based on real-world sources, our evaluations reveal substantial environmental benefits without demanding drastic personal sacrifices, even if consumers accept only a fraction of the recommendations.

Authors: Thomas Asikis (University of Zurich)
ICLR 2023 A simplified machine learning based wildfire ignition model from insurance perspective (Papers Track)
Abstract and authors: (click to expand)

Abstract: In the context of climate change, wildfires are becoming more frequent, intense, and prolonged in the western US, particularly in California. Wildfires cause catastrophic socio-economic losses and are projected to worsen in the near future. Inaccurate estimates of fire risk put further pressure on wildfire (re)insurance and cause many homes to lose wildfire insurance coverage. Efficient and effective prediction of fire ignition is one step towards better fire risk assessment. Here we present a simplified machine learning-based fire ignition model at yearly scale that is well suited to the use case of one-year term wildfire (re)insurance. Our model yields a recall, precision, and the area under the precision-recall curve of 0.69, 0.86 and 0.81, respectively, for California, and significantly higher values of 0.82, 0.90 and 0.90, respectively, for the populated area, indicating its good performance. In addition, our model feature analysis reveals that power line density, enhanced vegetation index (EVI), vegetation optical depth (VOD), and distance to the wildland-urban interface stand out as the most important features determining ignitions. The framework of this simplified ignition model could easily be applied to other regions or genesis of other perils like hurricane, and it paves the road to a broader and more affordable safety net for homeowners.

Authors: Yaling Liu (OurKettle Inc); Son Le (OurKettle Inc.); Yufei Zou (Our Kettle, Inc.); mojtaba Sadgedhi (OurKettle Inc.); Yang Chen (University of California, Irvine); Niels Andela (BeZero Carbon); Pierre Gentine (Columbia University)
ICLR 2023 Mining Effective Strategies for Climate Change Communication (Papers Track)
Abstract and authors: (click to expand)

Abstract: With the goal of understanding effective strategies to communicate about climate change, we build interpretable models to rank tweets related to climate change with respect to the engagement they generate. Our models are based on the Bradley-Terry model of pairwise comparison outcomes and use a combination of the tweets’ topic and metadata features to do the ranking. To remove confounding factors related to author popularity and minimise noise, they are trained on pairs of tweets that are from the same author and around the same time period and have a sufficiently large difference in engagement. The models achieve good accuracy on a held-out set of pairs. We show that we can interpret the parameters of the trained model to identify the topic and metadata features that contribute to high engagement. Among other observations, we see that topics related to climate projections, human cost and deaths tend to have low engagement while those related to mitigation and adaptation strategies have high engagement. We hope the insights gained from this study will help craft effective climate communication to promote engagement, thereby lending strength to efforts to tackle climate change.

Authors: Aswin Suresh (EPFL); Lazar Milikic (EPFL); Francis Murray (EPFL); Yurui Zhu (EPFL); Matthias Grossglauser (École Polytechnique Fédérale de Lausanne (EPFL))
ICLR 2023 Robustly modeling the nonlinear impact of climate change on agriculture by combining econometrics and machine learning (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Climate change is expected to have a dramatic impact on agricultural production; however, due to natural complexity, the exact avenues and relative strengths by which this will happen are still unknown. The development of accurate forecasting models is thus of great importance to enable policy makers to design effective interventions. To date, most machine learning methods aimed at tackling this problem lack a consideration of causal structure, thereby making them unreliable for the types of counterfactual analysis necessary when making policy decisions. Econometrics has developed robust techniques for estimating cause-effect relations in time-series, specifically through the use of cointegration analysis and Granger causality. However, these methods are frequently limited in flexibility, especially in the estimation of nonlinear relationships. In this work, we propose to integrate the non-linear function approximators with the robust causal estimation methods to ultimately develop an accurate agricultural forecasting model capable of robust counterfactual analysis. This method would be a valuable new asset for government and industrial stakeholders to understand how climate change impacts agricultural production.

Authors: Benedetta Francesconi (Independent Researcher); Ying-Jung C Deweese (Descartes Labs / Georgia Insititute of Technology)
ICLR 2023 On the impact of small-data diversity on forecasts: evidence from meteorologically-driven electricity demand in Mediterranean zones. (Papers Track)
Abstract and authors: (click to expand)

Abstract: In this paper, we compare the improvement of probabilistic electricity demand forecasts for three specific coastal and island regions using raw and pre-computed meteorological features based on empirically-tested formulations drawn from climate science literature. Typically for the general task of time-series forecasting with strong weather/climate drivers, go-to models like the Autoregressive Integrated Moving Average (ARIMA) model are built with assumptions of how independent variables will affect a dependent one and are at best encoded with a handful of exogenous features with known impact. Depending on the geographical region and/or cultural practices of a population, such a selection process may yield a non-optimal feature set which would ultimately drive a weak impact on underline demand forecasts. The aim of this work is to assess the impact of a documented set of meteorological features on electricity demand using deep learning models in comparative studies. Leveraging the defining computational architecture of the Temporal Fusion Transformer (TFT), we discover the unimportance of weather features for improving probabilistic forecasts for the targeted regions. However, through experimentation, we discover that the more stable electricity demand of the coastal Mediterranean regions, the Ceuta and Melilla autonomous cities in Morocco, improved the forecast accuracy of the strongly tourist-driven electricity demand for the Balearic islands located in Spain during the time of travel restrictions (i.e., during COVID19 (2020))--a root mean squared error (RMSE) from ~0.090 to ~0.012 with a substantially improved 10th/90th quantile bounding.

Authors: Reginald Bryant (IBM Research - Africa); Julian Kuehnert (IBM Research)
NeurIPS 2022 Forecasting European Ozone Air Pollution With Transformers (Papers Track)
Abstract and authors: (click to expand)

Abstract: Surface ozone is an air pollutant that contributes to hundreds of thousands of premature deaths annually. Accurate short-term ozone forecasts may allow improved policy to reduce the risk to health, such as air quality warnings. However, forecasting ozone is a difficult problem, as surface ozone concentrations are controlled by a number of physical and chemical processes which act on varying timescales. Accounting for these temporal dependencies appropriately is likely to provide more accurate ozone forecasts. We therefore deploy a state-of-the-art transformer-based model, the Temporal Fusion Transformer, trained on observational station data from three European countries. In four-day test forecasts of daily maximum 8-hour ozone, the novel approach is highly skilful (MAE = 4.6 ppb, R2 = 0.82), and generalises well to two European countries unseen during training (MAE = 4.9 ppb, R2 = 0.79). The model outperforms standard machine learning models on our data, and compares favourably to the published performance of other deep learning architectures tested on different data. We illustrate that the model pays attention to physical variables known to control ozone concentrations, and that the attention mechanism allows the model to use relevant days of past ozone concentrations to make accurate forecasts.

Authors: Seb Hickman (University of Cambridge); Paul Griffiths (University of Cambridge); Alex Archibald (University of Cambridge); Peer Nowack (Imperial College London); Elie Alhajjar (USMA)
NeurIPS 2022 Temperature impacts on hate speech online: evidence from four billion tweets (Papers Track)
Abstract and authors: (click to expand)

Abstract: Human aggression is no longer limited to the physical space but exists in the form of hate speech on social media. Here, we examine the effect of temperature on the occurrence of hate speech on Twitter and interpret the results in the context of climate change, human behavior and mental health. Employing supervised machine learning models, we identify hate speech in a data set of four billion geolocated tweets from over 750 US cities (2014 – 2020). We statistically evaluate the changes in daily hate tweets against changes in local temperature, isolating the temperature influence from confounding factors using binned panel-regression models. We find a low prevalence of hate tweets in moderate temperatures and observe sharp increases of up to 12% for colder and up to 22% for hotter temperatures, indicating that not only hot but also cold temperatures increase aggressive tendencies. Further, we observe that for extreme temperatures hate speech also increases as a percentage of total tweeting activity, crowding out non-hate speech. The quasi-quadratic shape of the temperature-hate tweet curve is robust across varying climate zones, income groups, religious and political beliefs. The prevalence of the results across climatic and socioeconomic splits points to limits in adaptation. Our results illuminate hate speech online as an impact channel through which temperature alters societal aggression.

Authors: Annika Stechemesser (Potsdam Insitute for Climate Impact Research); Anders Levermann (Potsdam Institute for Climate Impact Research); Leonie Wenz (Potsdam Institute for Climate Impact Research)
NeurIPS 2022 Evaluating Digital Tools for Sustainable Agriculture using Causal Inference (Papers Track)
Abstract and authors: (click to expand)

Abstract: In contrast to the rapid digitalization of several industries, agriculture suffers from low adoption of climate-smart farming tools. Even though AI-driven digital agriculture can offer high-performing predictive functionalities, they lack tangible quantitative evidence on their benefits to the farmers. Field experiments can derive such evidence, but are often costly and time consuming. To this end, we propose an observational causal inference framework for the empirical evaluation of the impact of digital tools on target farm performance indicators. This way, we can increase farmers' trust via enhancing the transparency of the digital agriculture market, and in turn accelerate the adoption of technologies that aim to increase productivity and secure a sustainable and resilient agriculture against a changing climate. As a case study, we perform an empirical evaluation of a recommendation system for optimal cotton sowing, which was used by a farmers' cooperative during the growing season of 2021. We leverage agricultural knowledge to develop the causal graph of the farm system, we use the back-door criterion to identify the impact of recommendations on the yield and subsequently we estimate it using several methods on observational data. The results showed that a field sown according to our recommendations enjoyed a significant increase in yield 12% to 17%.

Authors: Ilias Tsoumas (National Observatory of Athens); Georgios Giannarakis (National Observatory of Athens); Vasileios Sitokonstantinou (National Observatory of Athens); Alkiviadis Marios Koukos (National Observatory of Athens); Dimitra A Loka (Hellenic Agricultural Organization ELGO DIMITRA); Nikolaos S Bartsotas (National Observatory of Athens); Charalampos Kontoes (National Observatory of Athens); Ioannis N Athanasiadis (Wageningen University and Research)
NeurIPS 2022 Forecasting Global Drought Severity and Duration Using Deep Learning (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Drought detection and prediction is challenging due to the slow onset of the event and varying degrees of dependence on numerous physical and socio-economic factors that differentiate droughts from other natural disasters. In this work, we propose DeepXD (Deep learning for Droughts), a deep learning model with 26 physics-informed input features for SPI (Standardised Precipitation Index) forecasting to identify and classify droughts using monthly oceanic indices, global meteorological and vegetation data, location (latitude, longitude) and land cover for the years 1982 to 2018. In our work, we propose extracting features by considering the atmosphere and land moisture and energy budgets and forecasting global droughts on a seasonal and an annual scale at 1, 3, 6, 9, 12 and 24 months lead times. SPI helps us to identify the severity and the duration of the drought to classify them as meteorological, agricultural and hydrological.

Authors: Akanksha Ahuja (NOA); Xin Rong Chua (Centre for Climate Research Singapore)
NeurIPS 2022 Personalizing Sustainable Agriculture with Causal Machine Learning (Proposals Track) Best Paper: Proposals
Abstract and authors: (click to expand)

Abstract: To fight climate change and accommodate the increasing population, global crop production has to be strengthened. To achieve the "sustainable intensification" of agriculture, transforming it from carbon emitter to carbon sink is a priority, and understanding the environmental impact of agricultural management practices is a fundamental prerequisite to that. At the same time, the global agricultural landscape is deeply heterogeneous, with differences in climate, soil, and land use inducing variations in how agricultural systems respond to farmer actions. The "personalization" of sustainable agriculture with the provision of locally adapted management advice is thus a necessary condition for the efficient uplift of green metrics, and an integral development in imminent policies. Here, we formulate personalized sustainable agriculture as a Conditional Average Treatment Effect estimation task and use Causal Machine Learning for tackling it. Leveraging climate data, land use information and employing Double Machine Learning, we estimate the heterogeneous effect of sustainable practices on the field-level Soil Organic Carbon content in Lithuania. We thus provide a data-driven perspective for targeting sustainable practices and effectively expanding the global carbon sink.

Authors: Georgios Giannarakis (National Observatory of Athens); Vasileios Sitokonstantinou (National Observatory of Athens); Roxanne Suzette Lorilla (National Observatory of Athens); Charalampos Kontoes (National Observatory of Athens)
NeurIPS 2021 Being the Fire: A CNN-Based Reinforcement Learning Method to Learn How Fires Behave Beyond the Limits of Physics-Based Empirical Models (Papers Track)
Abstract and authors: (click to expand)

Abstract: Wildland fires pose an increasing threat in light of anthropogenic climate change. Fire-spread models play an underpinning role in many areas of research across this domain, from emergency evacuation to insurance analysis. We study paths towards advancing such models through deep reinforcement learning. Aggregating 21 fire perimeters from the Western United States in 2017, we construct 11-layer raster images representing the state of the fire area. A convolution neural network based agent is trained offline on one million sub-images to create a generalizable baseline for predicting the best action - burn or not burn - given the then-current state on a particular fire edge. A series of online, TD(0) Monte Carlo Q-Learning based improvements are made with final evaluation conducted on a subset of holdout fire perimeters. We examine the performance of the learned agent/model against the FARSITE fire-spread model. We also make available a novel data set and propose more informative evaluation metrics for future progress.

Authors: William L Ross (Stanford)
ICML 2021 Climate-based ensemble machine learning model to forecast Dengue epidemics (Papers Track)
Abstract and authors: (click to expand)

Abstract: Dengue fever is one of the most common and rapidly spreading arboviral diseases in the world, with major public health and economic consequences in tropical and sub-tropical regions. Countries such as Peru, 17.143 cases of dengue were reported in 2019, where 81.4% of cases concentrated in five of the 25 departments. When predicting infectious disease outbreaks, it is crucial to model the long-term dependency in time series data. However, this is challenging when performed on a countrywide level since dengue incidence varies across administrative areas. Therefore, this study developed and applied a climate-based ensemble model using multiple machine learning (ML) approaches to forecast dengue incidence rate (DIR) by department. The ensemble combined the outputs from Long Short-Term Memory (LSTM) recurrent neural network and Categorical Boosting (CatBoost) methods to predict DIR one month ahead for each department in Peru. Monthly dengue cases stratified by Peruvian departments were analysed in conjunction with associated demographic, geographic, and satellite-based meteorological data for the period January 2010–December 2019. The results demonstrated that the ensemble model was able to forecast DIR in low-transmission departments, while the model was less able to detect sudden DIR peaks in some departments. Air temperature and wind components demonstrated to be the significant predictors for DIR predictions. This dengue forecast model is timely and can help local governments to implement effective control measures and mitigate the effects of the disease. This study advances the state-of-the-art of climate services for the public health sector, by informing what are the key climate factors responsible for triggering dengue transmission. Finally, this project summarises how important it is to perform collaborative work with complementary expertise from intergovernmental organizations and public health universities to advance knowledge and address societal challenges.

Authors: Rochelle Schneider (European Space Agency); Alessandro Sebastianelli (European Space Agency); Dario Spiller (Italian Space Agency); James Wheeler (European Space Agency); Raquel Carmo (European Space Agency); Artur Nowakowski (Warsaw University of Technology); Manuel Garcia-Herranz (UNICEF); Dohyung Kim (UNICEF); Hanoch Barlevi (UNICEF LACRO); Zoraya El Raiss Cordero (UNICEF LACRO); Silvia Liberata Ullo (University of Sannio); Pierre-Philippe Mathieu (European Space Agency); Rachel Lowe (London School of Hygiene & Tropical Medicine)
ICML 2021 Wildfire Smoke Plume Segmentation Using Geostationary Satellite Imagery (Papers Track)
Abstract and authors: (click to expand)

Abstract: Wildfires have increased in frequency and severity over the past two decades, especially in the Western United States. Beyond physical infrastructure damage caused by these wildfire events, researchers have increasingly identified harmful impacts of particulate matter generated by wildfire smoke on respiratory, cardiovascular, and cognitive health. This inference is difficult due to the spatial and temporal uncertainty regarding how much particulate matter is specifically attributable to wildfire smoke. One factor contributing to this challenge is the reliance on manually drawn smoke plume annotations, which are often noisy representations limited to the United States. This work uses deep convolutional neural networks to segment smoke plumes from geostationary satellite imagery. We compare the performance of predicted plume segmentations versus the noisy annotations using causal inference methods to estimate the amount of variation each explains in Environmental Protection Agency (EPA) measured surface level particulate matter <2.5μm in diameter (PM2.5).

Authors: Jeffrey L Wen (Stanford University); Marshall Burke (Stanford University)
NeurIPS 2020 Formatting the Landscape: Spatial conditional GAN for varying population in satellite imagery (Papers Track)
Abstract and authors: (click to expand)

Abstract: Climate change is expected to reshuffle the settlement landscape: forcing people in affected areas to migrate, to change their lifeways, and continuing to affect demographic change throughout the world. Changes to the geographic distribution of population will have dramatic impacts on land use and land cover and thus constitute one of the major challenges of planning for climate change scenarios. In this paper, we explore a generative model framework for generating satellite imagery conditional on gridded population distributions. We make additions to the existing ALAE [30] architecture, creating a spatially conditional version: SCALAE. This method allows us to explicitly disentangle population from the model’s latent space and thus input custom population forecasts into the generated imagery. We postulate that such imagery could then be directly used for land cover and land use change estimation using existing frameworks, as well as for realistic visualisation of expected local change. We evaluate the model by comparing pixel and semantic reconstructions, as well as calculate the standard FID metric. The results suggest the model captures population distributions accurately and delivers a controllable method to generate realistic satellite imagery.

Authors: Tomas Langer (Intuition Machines); Natalia Fedorova (Max Planck Institute for Evolutionary Anthropology); Ron Hagensieker (Osir.io)
NeurIPS 2020 FireSRnet: Geoscience-driven super-resolution of future fire risk from climate change (Papers Track)
Abstract and authors: (click to expand)

Abstract: With fires becoming increasingly frequent and severe across the globe in recent years, understanding climate change’s role in fire behavior is critical for quantifying current and future fire risk. However, global climate models typically simulate fire behavior at spatial scales too coarse for local risk assessments. Therefore, we propose a novel approach towards super-resolution (SR) enhancement of fire risk exposure maps that incorporates not only 2000 to 2020 monthly satellite observations of active fires but also local information on land cover and temperature. Inspired by SR architectures, we propose an efficient deep learning model trained for SR on fire risk exposure maps. We evaluate this model on resolution enhancement and find it outperforms standard image interpolation techniques at both 4x and 8x enhancement while having comparable performance at 2x enhancement. We then demonstrate the generalizability of this SR model over northern California and New South Wales, Australia. We conclude with a discussion and application of our proposed model to climate model simulations of fire risk in 2040 and 2100, illustrating the potential for SR enhancement of fire risk maps from the latest state-of-the-art climate models.

Authors: Tristan C Ballard (Sust Global, Stanford University); Gopal Erinjippurath (Sust Global)
NeurIPS 2020 The Human Effect Requires Affect: Addressing Social-Psychological Factors of Climate Change with Machine Learning (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Machine learning has the potential to aid in mitigating the human effects of climate change. Previous applications of machine learning to tackle the human effects in climate change include approaches like informing individuals of their carbon footprint and strategies to reduce it. For these methods to be the most effective they must consider relevant social-psychological factors for each individual. Of social-psychological factors at play in climate change, affect has been previously identified as a key element in perceptions and willingness to engage in mitigative behaviours. In this work, we propose an investigation into how affect could be incorporated to enhance machine learning based interventions for climate change. We propose using affective agent-based modelling for climate change as well as the use of a simulated climate change social dilemma to explore the potential benefits of affective machine learning interventions. Behavioural and informational interventions can be a powerful tool in helping humans adopt mitigative behaviours. We expect that utilizing affective ML can make interventions an even more powerful tool and help mitigative behaviours become widely adopted.

Authors: Kyle Tilbury (University of Waterloo); Jesse Hoey (University of Waterloo)
NeurIPS 2020 Privacy Preserving Demand Forecasting to Encourage Consumer Acceptance of Smart Energy Meters (Proposals Track)
Abstract and authors: (click to expand)

Abstract: In this proposal paper we highlight the need for privacy preserving energy demand forecasting to allay a major concern consumers have about smart meter installations. High resolution smart meter data can expose many private aspects of a consumer’s household such as occupancy, habits and individual appliance usage. Yet smart metering infrastructure has the potential to vastly reduce carbon emissions from the energy sector through improved operating efficiencies. We propose the application of a distributed machine learning setting known as federated learning for energy demand forecasting at various scales to make load prediction possible whilst retaining the privacy of consumers’ raw energy consumption data.

Authors: Christopher Briggs (Keele University); Zhong Fan (Keele University); Peter Andras (Keele University, School of Computing and Mathematics, Newcastle-under-Lyme, UK)
ICLR 2020 Understanding the dynamics of climate-crucial food choice behaviours using Distributional Semantics (Papers Track)
Abstract and authors: (click to expand)

Abstract: Developed countries must make swift movements toward plant-based diets in order to mitigate climate change and maintain food security. However, researchers currently lack clear insight into the psychological dimensions that influence food choice, which is necessary to encourage the societal adaptation of new diets. In this project, we use Skip-gram word embeddings trained on the ukWaC corpus as a lens to study the implicit mental representations people have of foods. Our data-driven insights expand on findings from traditional, interview-based studies by uncovering implicit mental representations, allowing a better understanding the complex combination of conscious and sub-conscious processes surrounding food choice. In particular, our findings shed light on the pervasiveness of meat as the ‘centre’ of the meal in the UK.

Authors: Claudia Haworth (University of Sheffield); Gabriella Viglioco (University College London)
ICLR 2020 Non-linear interlinkages and key objectives amongst the Paris Agreement and the Sustainable Development Goals (Papers Track)
Abstract and authors: (click to expand)

Abstract: The United Nations' ambitions to combat climate change and prosper human development are manifested in the Paris Agreement and the Sustainable Development Goals (SDGs), respectively. These are inherently inter-linked as progress towards some of these objectives may accelerate or hinder progress towards others. We investigate how these two agendas influence each other by defining networks of 18 nodes, consisting of the 17 SDGs and climate change, for various groupings of countries. We compute a non-linear measure of conditional dependence, the partial distance correlation, given any subset of the remaining 16 variables. These correlations are treated as weights on edges, and weighted eigenvector centralities are calculated to determine the most important nodes. We find that SDG 6, clean water and sanitation, and SDG 4, quality education, are most central across nearly all groupings of countries. In developing regions, SDG 17, partnerships for the goals, is strongly connected to the progress of other objectives in the two agendas whilst, somewhat surprisingly, SDG 8, decent work and economic growth, is not as important in terms of eigenvector centrality.

Authors: Felix Laumann (Imperial College London); Julius von Kügelgen (MPI for Intelligent Systems, Tübingen & University of Cambridge); Mauricio Barahona (Imperial College London)
ICLR 2020 A CONTINUAL LEARNING APPROACH FOR LOCAL LEVEL ENVIRONMENTAL MONITORING IN LOW-RESOURCE SETTINGS (Papers Track)
Abstract and authors: (click to expand)

Abstract: An increasingly important dimension in the quest for mitigation and monitoring of environmental change is the role of citizens. The crowd-based monitoring of local level anthropogenic alterations is essential towards measurable changes in different contributing factors to climate change. With the proliferation of mobile technologies here in the African continent, it is useful to have machine learning based models that are deployed on mobile devices and that can learn continually from streams of data over extended time, possibly pertaining to different tasks of interest. In this paper, we demonstrate the localisation of deforestation indicators using lightweight models and extend to incorporate data about wildfires and smoke detection. The idea is to show the need and potential of continual learning approaches towards building robust models to track local environmental alterations.

Authors: Arijit Patra (University of Oxford)
ICLR 2020 Machine Learning Approaches to Safeguarding Continuous Water Supply in the Arid and Semi-arid Lands of Northern Kenya (Proposals Track)
Abstract and authors: (click to expand)

Abstract: Arid and semi-arid regions (ASALs) in developing countries are heavily affected by the effects of global warming and climate change, leading to adverse climatic conditions such as drought and flooding. This paper explores the problem of fresh-water access in northern Kenya and measures being taken to safeguard water access despite these harsh climatic changes. We present an integrated water management and decision-support platform, eMaji Manager, that we developed and deployed in five ASAL counties in northern Kenya to manage waterpoint access for people and livestock. We then propose innovative machine learning methods for understanding waterpoint usage and repair patterns for sensor-instrumented waterpoints (e.g., boreholes). We explore sub-sequence discriminatory models and recurrent neural networks to predict water-point failures, improve repair response times and ultimately support continuous access to water.

Authors: Fred Otieno (IBM); Timothy Nyota (IBM); Isaac Waweru (IBM); Celia Cintas (IBM Research); Samuel C Maina (IBM Research); William Ogallo (IBM Research); Aisha Walcott-Bryant (IBM Research - Africa)
ICLR 2020 MACHINE LEARNING APPLICATIONS THAT CAN HELP PASTORAL COMMUNITIES IN NORTHERN KENYA AND ELSEWHERE ADAPT TO CLIMATE CHANGE (Proposals Track)
Abstract and authors: (click to expand)

Abstract: I propose the use of Machine Learning techniques such as Active Learning(AL) and Transfer Learning(TL) to translate climate information and adaption technique from major Western and Asian languages to thousands of low resource languages in the developing world. Studies have shown that access to information can help people assess the magnitude of the climate change challenge, possible options and those feasible within the relevant context (Nyahunda & Tiri-vangasi, 2019) I endeavor to demonstrate that if this information was available in a language the locals can understand, it would result in local empowerment and as a result inspire action.

Authors: Jefferson Sankara (Lori Systems)
ICLR 2020 USING MACHINE LEARNING TO ANALYZE CLIMATE CHANGE TECHNOLOGY TRANSFER (CCTT) (Proposals Track)
Abstract and authors: (click to expand)

Abstract: The objective of the present paper is to review the climate change technology transfer. This research proposes a method for analysing CCTT using patent analysis and topic modelling. A collection of climate change mitigation related technology (CCMT) patents from patent databases would be used as input to group patents in several relevant topics for climate change mitigation using the topic exploration model in this research. The research questions we want to address are: how have the patenting activities changed over time in CCMT patents? And who are the technological leaders? The investigation of these questions can offer the technological landscape in climate change-related technologies at the international level. We propose a hybrid Latent Dirichlet Allocation (LDA) approach for topic modelling and identification of relationships between terms and topics related to CCMT, enabling better visualizations of underlying intellectual property dynamics. Further, we propose predictive modelling for CCTT and competitor analysis to identify and rank countries with a similar patent landscape. The projected results are expected to facilitate the transfer process associated with existing and emerging climate change technologies and improve technology cooperation between governments.

Authors: Shruti Kulkarni (Indian Institute of Science (IISc))
NeurIPS 2019 Establishing an Evaluation Metric to Quantify Climate Change Image Realism (Papers Track)
Abstract and authors: (click to expand)

Abstract: With success on controlled tasks, generative models are being increasingly applied to humanitarian applications. In this paper, we focus on the evaluation of a conditional generative model that illustrates the consequences of climate change-induced flooding to encourage public interest and awareness on the issue. Because metrics for comparing the realism of different modes in a conditional generative model do not exist, we propose several automated and human-based methods for evaluation. To do this, we adapt several existing metrics, and assess the automated metrics against gold standard human evaluation. We find that using Frechet Inception Distance (FID) with embeddings from an intermediary Inception-V3 layer that precedes the auxiliary classifier produces results most correlated with human realism. While insufficient alone to establish a human-correlated automatic evaluation metric, we believe this work begins to bridge the gap between human and automated generative evaluation procedures.

Authors: Sharon Zhou (Stanford University); Sasha Luccioni (Mila); Gautier Cosne (Mila); Michael Bernstein (Stanford University); Yoshua Bengio (Mila)
NeurIPS 2019 A User Study of Perceived Carbon Footprint (Papers Track)
Abstract and authors: (click to expand)

Abstract: We propose a statistical model to understand people’s perception of their carbon footprint. Driven by the observation that few people think of CO2 impact in absolute terms, we design a system to probe people’s perception from simple pairwise comparisons of the relative carbon footprint of their actions. The formulation of the model enables us to take an active-learning approach to selecting the pairs of actions that are maximally informative about the model parameters. We define a set of 18 actions and collect a dataset of 2183 comparisons from 176 users on a university campus. The early results reveal promising directions to improve climate communication and enhance climate mitigation.

Authors: Victor Kristof (EPFL); Valentin Quelquejay-Leclere (EPFL); Robin Zbinden (EPFL); Lucas Maystre (Spotify); Matthias Grossglauser (École Polytechnique Fédérale de Lausanne (EPFL)); Patrick Thiran (EPFL)
NeurIPS 2019 Quantifying the Carbon Emissions of Machine Learning (Papers Track)
Abstract and authors: (click to expand)

Abstract: From an environmental standpoint, there are a few crucial aspects of training a neural network that have a major impact on the quantity of carbon that it emits. These factors include: the location of the server used for training and the energy grid that it uses, the length of the training procedure, and even the make and model of hardware on which the training takes place. In order to approximate these emissions, we present our Machine Learning Emissions Calculator, a tool for our community to better understand the environmental impact of training ML models. We accompany this tool with an explanation of the factors cited above, as well as concrete actions that individual practitioners as well as organizations can take to mitigate their carbon emissions.

Authors: Sasha Luccioni (Mila); Victor Schmidt (Mila); Alexandre Lacoste (Element AI); Thomas Dandres (Polytechnique Montreal)
ICML 2019 Unsupervised Temporal Clustering to Monitor the Performance of Alternative Fueling Infrastructure (Research Track)
Abstract and authors: (click to expand)

Abstract: Zero Emission Vehicles (ZEV) play an important role in the decarbonization of the transportation sector. For a wider adoption of ZEVs, providing a reliable infrastructure is critical. We present a machine learning approach that uses unsupervised temporal clustering algorithm along with survey analysis to determine infrastructure performance and reliability of alternative fuels. We illustrate this approach for the hydrogen fueling stations in California, but this can be generalized for other regions and fuels.

Authors: Kalai Ramea (PARC)