Dataset for Physics-Informed Geo-AI for Irrigation Quantification_CA Central Valley

Read full text Summarise with VIRAT
Authors: Esmaeel Adrah
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This data element contains the foundational datasets required to execute the Physics-Informed Geo-AI irrigation mapping framework. These files provide the spatial and hydrological constraints necessary for the TabNet-parameterized SM2RAIN model. Included Files: Master_ML_Ready_Data.csv: The primary dataset used for model training and inference. This file integrates AlphaEarth satellite embeddings with hydrological forcing data (Soil Mosture from SMAP, Precipitation, ET from OpenET) and ground-truth irrigation from USGS, pre-processed for immediate use in the TabNet architecture. Grid_updated.csv: The geospatial index for the California Central Valley study area, containing 1km-resolution grid cell identifiers and coordinates. Usage Note:To ensure successful execution of the submitted Jupyter Notebook, these files should be placed in a directory labeled /Data within the project root. You may directly unzip the folder to have the folder Data. The notebook uses relative pathing to reference these elements for seamless reproducibility.

Fields of Study & Subjects

Soil Moisture and Remote Sensing Hydrological Forecasting Using AI Remote Sensing in Agriculture

Publication details

Journal / source
Figshare
Publication date
25 March 2026
Language
EN
Document type
Dataset