The lake is now being asked to file its own heat record
the lake is now being asked to file its own heat record: a physics informed transfer network pre trains on satellite lakes, retrains on process simulations and fine tunes on local buoys to generalize beyond any single site, while an unconstrained lstm trained on cloud affected data overpredicts warmest month temperatures and misreads ice duration, and a process guided model penalizes impossible values above 35 degrees.
the lake now files its own heat ledger.
Context
A paper in Water Resources Research, volume 62, issue 4, published April 2026 by Muyuan Liu, R. Iestyn Woolway and colleagues, presents Transfer-PIDL, a three-stage training strategy for lake surface temperature: pre-training on large-scale satellite observations, re-training with process-based model simulations, and fine-tuning with local measurements. Given pre-training on 40 source lakes it beat local physics-informed models by 20 to 39 percent in validation RMSE. Across 43 additional lakes it reached a mean validation RMSE of 1.2 C against 1.6 C for local PIDL, 1.8 C for deep learning and 1.9 C for process-based models, and across 869 lakes the poorest validation scenario had an RMSE of 1.5 C. A separate preprint, Mind the Cloud (May 2025), shows that cloud gaps in optical satellite lake temperature are not random, and that an LSTM trained on cloud-affected synthetic data had significantly amplified biases, particularly in warm-season temperature and ice duration, while the physically based air2water model stayed relatively stable.
The transfer network matches the Transfer-PIDL paper. The LSTM result comes from a different, synthetic-data study, and it describes amplified bias in warm-season temperature and ice duration without the direction overpredicts, so the note's overpredicts warmest month wording is stronger than the abstract. The process guided model that penalizes values above 35 degrees was not located in either paper, and air2water is a physically based model and not a process-guided one. The lake now files its own heat ledger is the author's line.
Watch next
- The direction of the cloud-induced bias in the full Mind the Cloud text, and the 35 degree penalty source.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 22:16 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →