Open Climate Fix

Mentors: Student: Gupta, Anshul

Open Climate Fix’s Quartz Solar forecasting model uses a simple rule-based "adjuster" that averages recent forecast errors to correct incoming solar forecasts. While effective in stable conditions, this method is rigid and might be suboptimal under rapid weather transitions or atypical error patterns. This project aims to compare this rule-based logic with TabPFN, a transformer model for tabular data. It seeks to evaluate whether TabPFN or its time-series variant, TabPFN-TS, can predict adjustments based on features such as time, forecast horizon, recent errors, and more. Through experiments, one key goal is also to determine whether TabPFN improves forecast skill—specifically the P50 quantile forecast—over the current approach, and whether it is efficient enough for deployment.

Mentors: Student: zaryab ali

This project aims to enhance PVNet, a multi-modal deep learning model for solar energy forecasting, by integrating aerosol data as an additional input feature. Aerosols like dust, smoke, and haze significantly affect solar irradiance but are not currently accounted for in PVNet’s forecasting pipeline. The work involves identifying suitable open aerosol datasets, preprocessing and aligning them with existing NWP inputs, and incorporating them into the model architecture through a dedicated encoder. By comparing baseline and aerosol-aware models, we aim to assess the impact of aerosols on forecast accuracy. The final deliverables will include updated data pipelines, modified model components, and performance evaluation. This project contributes toward more accurate, climate-aware solar forecasting using open-source tools and public datasets.

Mentors: Student: Dakshbir

This proposal describes a thorough implementation strategy for streamlining Zarr data straight from cloud storage using Ice Chunk and Zarr 3 technologies, thereby optimizing machine learning training workflows. The project's main goal is to incorporate these technologies into PVNet's training pipeline for processing NWP and satellite data. Creating a strong data access layer, intelligent prefetching techniques, flexible caching plans, and a smooth integration with PyTorch's data loading pipeline are all part of the implementation. A thorough benchmarking framework is also included in the project to gauge cost effectiveness and performance gains. Four stages make up the 175-hour schedule: preliminary setup and research, core cloud access mechanism implementation, performance benchmarking and optimization, and thorough documentation. The implementation of a cloud-aware PyTorch dataset, an adaptive prefetcher, and optimized chunk configuration based on access patterns are important technical advancements. By removing the requirement to locally download sizable datasets prior to training, the project seeks to revolutionize OpenClimatefix's use of cloud resources for machine learning training.

Mentors: Student: Tara Murphy

This project aims to improve probabilistic solar forecasting by replacing traditional quantile regression with a more expressive Gaussian Mixture Model (GMM) approach. Instead of predicting fixed quantiles, the model will learn the parameters of a GMM to represent the full predictive distribution, capturing complex and multimodal uncertainties common in solar irradiance data. Deliverables include a GMM-based forecasting model, performance comparisons with quantile baselines, robust uncertainty evaluation, and well-documented code for reproducibility and future extension.

Mentors: Student: Suvan Banerjee

Open Climate Fix recent Cloudcasting model is able to predict clouds movements very accurately but for end user it’s hard to see or visualize the results, this project aims to create a new dashboard similar to quartz-frontend that is already up and running, to visualize results from the cloudcasting model the dashboard will be build using Next.js and MapBox as base map layer, Sentry to log errors, and a minimal fastapi server written in python to preprocess zaar data and serve it to the frontend, this solution will be a proof of concept for the new MVP