Advancing AI-based pan-European groundwater monitoringdoi:10.26165/JUELICH-DATA/ZBLDIRJülich DATA2021-12-142Ma, Yueling; Montzka, Carsten; Naz, Bibi; Kollet, Stefan, 2021, "Advancing AI-based pan-European groundwater monitoring", https://doi.org/10.26165/JUELICH-DATA/ZBLDIR, Jülich DATA, V2Advancing AI-based pan-European groundwater monitoringdoi:10.26165/JUELICH-DATA/ZBLDIRMa, YuelingMontzka, CarstenNaz, BibiKollet, StefanJülich DATAMa, YuelingMa, Yueling2021-12-13Earth and Environmental SciencesGroundwaterAnomaliesLSTM-TLLong Short-Term Memory (LSTM) networksTransfer learning (TL)EuropeThis study proposes an AI-based methodology combining Long Short-Term Memory (LSTM) networks and transfer learning (TL) to estimate water table depth anomalies (wtd_a) at the European scale in the absence of consistent water table depth (wtd) observational data sets, which is named LSTM-TL. The data repository provides: i) data utilized for evaluating LSTM-TL performance, i.e., input averaged monthly precipitation and soil moisture anomalies from common observational data sets (pr_a,o and θ_a,o), wtd_a estimates obtained from LSTM-TL (wtd_a,lstm-tl), wtd_a estimates obtained from LSTM networks with modeling data as input (wtd_a,lstm(m)), and wtd_a estimates obtained from LSTM networks trained on observations (wtd_a,lstm(o)); ii) reconstructed European monthly wtd_a,lstm-tl data RD1-6 from the early 1980s to the near present and their input pr_a,o and θ_a,o data; and iii) a Jupyter Notebook showing an example about the implementation of LSTM-TL. All the data sets in the repository have a spatial resolution of 0.11 degrees (~12.5 km).CC0 WaiverMa, Y., Montzka, C., Naz, B. and Kollet, S.: Advancing AI-based pan-European groundwater monitoring, in preparation.COSMO_REA6_monthly_precipitation_anomaly_FZJ-IBG3_COSMO_0.1_degree_v1.1995_082019.ncMonthly precipitation anomalies (pr_a) derived from the COSMO-REA6 dataset (referred to: Bollmeyer, C., Keller, J. D., Ohlwein, C., Wahl, S., Crewell, S., Friederichs, P., Hense, A., Keune, J., Kneifel, S., Pscheidt, I., Redl, S. and Steinke, S.: Towards a high‐resolution regional reanalysis for the European CORDEX domain, Q. J. R. Meteorol. Soc., 141(686), 1–15, doi:10.1002/qj.2486, 2015.)application/x-netcdfERA5_bias_corrected_monthly_precipitation_anomaly_FZJ-IBG3_WFDE5_0.1_degree_v1..ncMonthly precipitation anomalies (pr_a) derived from the ERA5 bias corrected dataset (referred to: Muñoz Sabater, J.: Near surface meteorological variables from 1979 to 2019 derived from bias-corrected reanalysis, Copernicus Clim. Chang. Serv. Clim. Data Store, doi:10.24381/cds.20d54e34, 2021.)application/x-netcdfERA5_Land_monthly_precipitation_anomaly_FZJ-IBG3_ERA5_Land_0.1_degree_v1.1981_052021.ncMonthly precipitation anomalies (pr_a) derived from the ERA5 Land dataset (referred to: Muñoz Sabater, J.: ERA5-Land hourly data from 1981 to present, Copernicus Clim. Chang. Serv. Clim. Data Store, doi:10.24381/cds.e2161bac, 2021.)application/x-netcdfERA5_Land_monthly_soil_moisture_anomaly_FZJ-IBG3_ERA5_Land_0.1_degree_v1.1981_052021.ncMonthly soil moisture anomalies (θ_a) derived from the ERA5 Land dataset (referred to: Muñoz Sabater, J.: ERA5-Land hourly data from 1981 to present, Copernicus Clim. Chang. Serv. Clim. Data Store, doi:10.24381/cds.e2161bac, 2021.)application/x-netcdfExample_script_for_the_implementation_of_LSTM_TL.ipynbA Jupyter Notebook showing an example about the implementation of LSTM-TL application/x-ipynb+jsonGLEAM_monthly_soil_moisture_anomaly_FZJ-IBG3_GLEAM_0.1_degree_v1.1980_2020.ncMonthly soil moisture anomalies (θ_a) derived from the GLEAM dataset (referred to: Martens, B., Miralles, D. G., Lievens, H., van der Schalie, R., de Jeu, R. A. M., Fernández-Prieto, D., Beck, H. E., Dorigo, W. A. and Verhoest, N. E. C.: GLEAM v3: satellite-based land evaporation and root-zone soil moisture, Geosci. Model Dev., 10(5), 1903–1925, doi:10.5194/gmd-10-1903-2017, 2017. Miralles, D. G., Holmes, T. R. H., De Jeu, R. A. M., Gash, J. H., Meesters, A. G. C. A. and Dolman, A. J.: Global land-surface evaporation estimated from satellite-based observations, Hydrol. Earth Syst. Sci., 15(2), 453–469, doi:10.5194/hess-15-453-2011, 2011.)application/x-netcdfLSTM_m_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_0.1_degree_v1.1996_2016.ncMonthly water table depth anomalies (wtd_a) estimates at pixels with water table depth (wtd) observations over Europe obtained by LSTM networks with modeling results as inputapplication/x-netcdfLSTM_obs_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_0.1_degree_v1.1996_2016.ncMonthly water table depth anomalies (wtd_a) estimates at pixels with water table depth (wtd) observations over Europe obtained by LSTM networks trained on observationsapplication/x-netcdfLSTM_TL_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1996_2016.ncMonthly water table depth anomalies (wtd_a) estimates at pixels with water table depth (wtd) observations over Europe obtained by LSTM-TLapplication/x-netcdfLSTM_TL_RD1_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1981_2019.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD1, obtained by LSTM-TL with ERA5 bias corrected precipitation anomalies (pr_a) and ERA5 Land soil moisture anomalies (θ_a) as inputapplication/x-netcdfLSTM_TL_RD2_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1980_2019.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD2, obtained by LSTM-TL with ERA5 bias corrected precipitation anomalies (pr_a) and GLEAM soil moisture anomalies (θ_a) as inputapplication/x-netcdfLSTM_TL_RD3_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1995_082019.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD3, obtained by LSTM-TL with COSMO-REA6 precipitation anomalies (pr_a) and ERA5 Land soil moisture anomalies (θ_a) as inputapplication/x-netcdfLSTM_TL_RD4_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1995_082019.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD4, obtained by LSTM-TL with COSMO-REA6 precipitation anomalies (pr_a) and GLEAM soil moisture anomalies (θ_a) as inputapplication/x-netcdfLSTM_TL_RD5_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1981_052021.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD5, obtained by LSTM-TL with ERA5 Land precipitation anomalies (pr_a) and ERA5 Land soil moisture anomalies (θ_a) as inputapplication/x-netcdfLSTM_TL_RD6_monthly_groundwater_table_depth_anomaly_FZJ-IBG3_LSTM_TL_0.1_degree_v1.1981_2020.ncReconstructed European monthly water table depth anomaly (wtd_a) data RD6, obtained by LSTM-TL with ERA5 Land precipitation anomalies (pr_a) and GLEAM soil moisture anomalies (θ_a) as inputapplication/x-netcdfobs_averaged_monthly_precipitation_anomaly_FZJ-IBG3_0.1_degree_v1.1996_2016.ncInput averaged monthly precipitation anomalies (pr_a) from observational datasets for the period 1996-2016application/x-netcdfobs_averaged_monthly_soil_moisture_anomaly_FZJ-IBG3_0.1_degree_v1.1996_2016.ncInput averaged monthly soil moisture anomalies (θ_a) from observational datasets for the period 1996-2016application/x-netcdf