This dataset was generated by the Remote Sensing Group of the TU Wien Department of Geodesy and Geoinformation, within projects funded by the European Structural and Investments Funds, the Austrian Space Applications Programme, and the EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (H SAF). Open use is granted under the CC BY 4.0 license. The provided dataset publication aims to support users of the ASCAT soil moisture data as provided by the EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (H SAF) to mask invalid retrievals due to subsurface scattering. This phenomenon has been discerned as the principal source of error in the current version of ASCAT soil moisture retrievals, as it contradicts the assumption that soil backscatter increases monotonically with soil moisture content. This happens because, in dry soil conditions, the presence of stones, rocks, or distinct soil layers can disrupt this expected relationship. At TU Wien, we have developed one statistical (P_ano) and two physically based indicators (P_sub, S_sub) that show the widespread occurrence of subsurface scattering not only in desert regions but also in more humid climates with distinct dry seasons. These indicators offer a means to identify subsurface scattering effects, enabling users of H SAF ASCAT soil moisture data to mask such effects. By selecting one of the provided indicators and setting a suitable threshold, users can tailor the masking process to their specific needs. As a baseline, we recommend using the monthly subsurface scattering mask included here in its dedicated file. We encourage the user community to leverage this data record to enhance the design of ASCAT soil moisture validation and application experiments. Moreover, this resource invites a reevaluation of conclusions drawn from earlier ASCAT studies and presents an opportunity to extend insights to other active microwave sensors operating at lower microwave frequencies (S-, L-, and P-band). Dataset Record ASCAT Mask Results This file contains the a mask recommended for ASCAT soil moisture retrievals, designed to address various conditions that can render ASCAT soil moisture data unreliable. These conditions encompass frozen soils, snow cover, wetlands, and areas with dense vegetation. Regarding the mitigation of subsurface scattering effects, the applied mask is derived from the monthly statistical indicator P_ano, with a threshold set at 0.1. This mask is applied to pixels that exceed this threshold for more than nine months. The results are gridded to the 12.5 km fixed Earth grid used for ASCAT (WARP5 grid). By implementing this mask, ASCAT pixels with potentially unreliable soil moisture measurements are excluded. Note that the subsurface scattering mask represents mean monthly conditions over the years 2007 to 2021, i.e. the behavior in single years may deviate from these conditions. The interpretation of values is as follows:

0: indicates no mask is applied, signifying that the soil moisture retrieval is considered reliable 1: indicates the application of the mask, implying that the soil moisture retrieval may not be reliable ERA5-Land Data Analysis For those seeking a more comprehensive exploration of subsurface scattering or desiring to generate customized masks utilizing the provided indicators, this file offers insights into ASCAT backscatter and surface soil moisture in combination with ERA5-Land soil moisture as well as ancillary information. The underlying data spans from 2007 to 2021 and is again gridded to the 12.5 km WARP5 grid. Within this file, users can access the following information: Subsurface scattering indicators:

P_ano - Probability of the occurrence of backscatter anomalies, depicts how frequently the ASCAT backscatter data exhibit anomalies P_ano_MM - P_ano calculated on a monthly basis P_sub - Probability of detecting subsurface scattering, derived from a physically based method that compares the goodness of fit of two backscatter models S_sub - Subsurface scattering signal strength, displays the signal range of the subsurface scatteringterm from completely dry to wet conditions Correlations before and after masking:

R_unmasked - Pearson correlation coefficient between ASCAT surface soil moisture and ERA5-Land soil moisture with no mask applied R_masked - Pearson correlation after applying the mask provided in the file above Selected specific masks:

cold_mask - Frozen soil and snow cover mask using ASCAT confidence flag (bit 1), ERA5-Land soil temperature (≤ 2 °C) and snow depth data (> 0 mm after averaging with a sliding window of 31 days) veg_mask - Dense vegetation mask based on ASCAT confidence flags (bits 4 and 5) and Copernicus Global Land Monitoring service (CGLS) leaf area index (LAI) data (LAI > 3) wet_mask - Wetland mask using Global Lakes and Wetlands Database (GLWD), land cover information, and ASCAT confidence flag subsurface_mask - Subsurface scattering mask based on monthly P_ano data (P_ano > 0.1) if threshold is exceeded for more than nine months Again, the values can be interpreted in the following way:

0: indicates no mask is applied 1: indicates the application of the corresponding mask Some pixels exhibit no values, even with none of the provided masks applied. This can be attributed to supplementary masking via ASCAT confidence flags as well as the exclusion of problematic CCI land cover classes not explicitly supplied. Ancillary dataset information:

cci_lc - ESA CCI land cover classification cfvo - 5 - 15 cm coarse fragment layer - International Soil Reference and Information Centre (ISRIC) SoilGrids™ 250m 2.0 dem - Terrain height - ETOPO 2022 global relief model glwd - Global Lakes and Wetlands Database (GLWD) classification isric - SoilGrids™ 250m 2.0 soil types by the ISRIC classification karst - World Karst Aquifer Map (WOKAM) classification kg - Köppen-Geiger climate classification lai - CGLS LAI sand - Sand content in the surface layer (0 - 5 cm) - ISRIC SoilGrids™ 250m 2.0 For more extensive details regarding ancillary data, please consult the associated references to the original datasets provided in the accompanying publication. ISMN Data Analysis For users interested in the comparison of ASCAT data with in-situ soil moisture stations, this file comprises the results for all ISMN stations with sufficient data from 2007 to 2021. All ancillary data are retrieved from the nearest ASCAT grid point index for better comparability to ERA5-Land data. The content of the file is exclusively comprised of the data points corresponding to ISMN stations, and does not encompass gridded data. The only additional parameter not used in the ERA5-Land analysis file is:

sensor - Identifier string of the ISMN station sensor the ISMN soil moisture is retrieved from For additional information on ISMN measurements, please refer to the corresponding documentation. Related Software

Software packages ascat, ecmwf_models and ismn facilitated the reading and conversion process of data from ASCAT on-board the series of Metop satellites, from the ECMWF reanalysis models, and from the International Soil Moisture Database (ISMN). EOmaps, an open-source Python package, has been used to visualise and analyse the provided datasets in the corresponding publication and is a suitable means for users seeking to create geographical maps. We also encourage users to take advantage of the grid point locator to quickly find locations through grid point indices (gpis) provided in the WARP5 grid: https://dgg.geo.tuwien.ac.at/ Acknowledgements This study was funded by the SustES project supported by the European Structural and Investments Funds (Adaptive Strategies for Sustainability of Ecosystem Services and Food Security in Harsh Natural Conditions, Reg. No. CZ.02.1.01/0.0/0.0/16 019/0000797), the ROSSIHNI project supported by the Austrian Space Applications Programme (FFG Pr.No. FO999892643), and the EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (H SAF). Reference Wagner, W., Lindorfer, R., Hahn, S., Kim, H., Vreugdenhil, M., Gruber, A., Fischer, M. & Trnka, M. (2023). Global Scale Mapping of Subsurface Scattering Signals Impacting ASCAT Soil Moisture Retrievals. TechRxiv, Preprint, https://doi.org/10.36227/techrxiv.24013890.v1