Gap Filling of High-Resolution Soil Moisture for SMAP/Sentinel-1: A Two-Layer Machine Learning-Based Framework

Code for paper Gap Filling of High-Resolution Soil Moisture for SMAP/Sentinel-1: A Two-Layer Machine Learning-Based Framework . An open access version of the paper can be found here.

Data Preprocessing

Code for data preprocessing is contained in folder data_preprocessing. Functions are then called by generate_experiment_data.py to generate experiment data.

Data from various sources are first converted to a unified format netCDF4 with their original resolutions being kept. They are then rescaled to have the same resolution as the SMAP/Sentinel-1 3 km soil moisture product. More details can be found in the paper.

Machine Learning Modeling

1. Features for brightness temperature downnscaling and soil moisture prediction are defined in queries_single_day/queries_tb_v_disaggregated.txt and queries_single_day/queries_soil_moisture.txt separately. You can define different feature sets at the same time by giving each set a unique number. Predictions will be output to a subfolder named by the given number.

2. Experiments including regional learning ones (spatial, temporal, and spatiotemporal), temporal limitation exploration, real gap filling are called from regional_learning_experiments.py.

   Experiments for the spatial limitation exploration are called from single_day_experiments.py.

3. Code for machine learning models are contained in soil_moisture_downscaling/machine_learning. New machine learning models can be added here.

Cite this work

Mao, H., Kathuria, D., Duffield, N., & Mohanty, B. P. ( 2019). Gap filling of high‐resolution soil moisture for SMAP/Sentinel‐1: A two‐layer machine learning‐based framework. Water Resources Research, vol. 55, no. 8, pp. 6986–7009, 2019. https://doi.org/10.1029/2019WR024902

Biblatex entry:

@article{map2019gap,
    author = {Mao, Hanzi and Kathuria, Dhruva and Duffield, Nick and Mohanty, Binayak P.},
    title = {Gap Filling of High-Resolution Soil Moisture for SMAP/Sentinel-1: A Two-Layer Machine Learning-Based Framework},
    journal = {Water Resources Research},
    volume = {55},
    number = {8},
    pages = {6986--7009},
    keywords = {soil moisture, machine learning, multiresolution gap filling, SMAP satellite, SENTINEL-1 satellite, spatial/temporal machine learning},
    doi = {10.1029/2019WR024902},
    url = {https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2019WR024902},
    eprint = {https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2019WR024902},
    year = {2019}
}