Development of long short-term memory models using rainfall and soil moisture to predict soil moisture dynamics
Yoon Ji Kim, Ho Jin Im, Seung Gon Wi, Min Soo Kim, Seok Min Lee, Ung Yang, Sun Woo Chung, Sang-Hyun Lee · 2026
Read Development of long short-term memory models using rainfall and soil moisture to predict soil moisture dynamics by Yoon Ji Kim, Ho Jin Im, Seung Gon Wi, Min Soo Kim, Seok Min Lee, Ung Yang, Sun Woo Chung, Sang-Hyun Lee (2026) 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/plos-pone-0353268.txt.
Description
Yoon Ji Kim and seven colleagues published this 2026 PLOS ONE study on forecasting how wet orchard soil will be hours ahead. Working in an Asian pear orchard, they recorded rainfall and soil moisture over two growing seasons. Sensors buried at three depths tracked how water moved down through the ground after each rain.
Their forecasting tool was a type of neural network built to learn patterns over time. Grouping rainfall into separate storm events, rather than feeding in hourly totals, gave clearly smaller errors. Short-range forecasts tracked the soil closely, while deeper soil stayed steadier than the surface layer. The authors argue such forecasts could help growers water only when the trees actually need it.
Catalog record
| CREATOR | Yoon Ji Kim, Ho Jin Im, Seung Gon Wi, Min Soo Kim, Seok Min Lee, Ung Yang, Sun Woo Chung, Sang-Hyun Lee |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41049 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |