Analyzing Spatio-Temporal Dynamics of Dissolved Oxygen for the River Thames using Superstatistical Methods and Machine Learning
Hankun He 1 ∗ \,{}^{1*} , Takuya Boehringer 2 \,{}^{2} , Benjamin Schäfer 3 \,{}^{3} , Kate Heppell 4 , 5 \,{}^{4,5} , Christian Beck · 2025
Read Analyzing Spatio-Temporal Dynamics of Dissolved Oxygen for the River Thames using Superstatistical Methods and Machine Learning by Hankun He 1 ∗ \,{}^{1*} , Takuya Boehringer 2 \,{}^{2} , Benjamin Schäfer 3 \,{}^{3} , Kate Heppell 4 , 5 \,{}^{4,5} , Christian Beck (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-07599.txt.
Description
Hankun He and co-authors analyzed dissolved oxygen in the River Thames with statistics and machine learning in a 2025 paper. Oxygen levels drive river health, and their swings over time and place are complex. After removing slow trends, the wobbles in oxygen showed heavy tails that a special distribution fit well. One detrending method stood out as best across nearly all fits.
The width of the fitted distribution narrowed with distance from the sea, a sign that geography shapes water quality. For same time prediction of oxygen, a boosting model beat the rivals. Temperature, acidity, and time of year were its key inputs. For forecasting ahead, a transformer variant built for long series did best of all. It used a sparse attention trick to catch long range links in the data, the authors say.
Catalog record
| CREATOR | Hankun He 1 ∗ \,{}^{1*} , Takuya Boehringer 2 \,{}^{2} , Benjamin Schäfer 3 \,{}^{3} , Kate Heppell 4 , 5 \,{}^{4,5} , Christian Beck |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-07599