High-resolution soil moisture forecasts for drought early warning in Germany: IFS–mHM ensemble forecast dataset (Najafi et al.) 0 17404

Project
POF3 - T53 - From models to predictions - Sub-seasonal Hydroclimatic Forecasting System - HS2S
Description
This dataset contains daily, high-resolution soil moisture index (SMI) data for Germany for the period 1 January–29 September 2025. It represents the reference soil moisture conditions derived from total soil column moisture simulated by the mesoscale Hydrological Model (mHM). The initializtion period 1 April–31 August 2025 is for evaluating the HS2S forecasts in the associated study.

For the 2025 dataset, the SMI calculation uses a trailing 7-day mean of simulated total soil column moisture. The value assigned to each date therefore represents the average soil moisture for that date and the preceding six days. This temporal averaging is performed internally by the Fortran-based Computational Hydrosystems This dataset contains daily, high-resolution ensemble soil moisture index (SMI) forecasts for Germany, initialized between 1 April and 31 August 2025. The forecasts cover lead times of up to 14 days and were produced using the IFS–mHM modelling chain.

The forecast-production chain is:

**ECMWF IFS ensemble forecasts at 0.1° resolution → interpolation to the 0.01° mHM grid → bias correction against DWD gridded meteorological reference data using the corresponding ECMWF reforecasts → ensemble mHM simulations → total soil column moisture → trailing 7-day mean → ensemble SMI forecasts.**

The ECMWF Integrated Forecasting System (IFS) ensemble meteorological forecasts were first spatially interpolated from 0.1° to the 0.01° mHM model grid. The interpolated forecasts were then bias-corrected against gridded meteorological reference data from the German Weather Service (DWD). The bias correction was derived with respect to the corresponding ECMWF reforecasts, thereby accounting for systematic forecast biases as a function of forecast lead time.

The resulting bias-corrected meteorological ensemble forecasts were used to force the mesoscale Hydrological Model (mHM). Each IFS ensemble member was propagated separately through mHM to generate an ensemble of daily total soil column moisture forecasts.

The SMI calculation uses a trailing 7-day mean of simulated total soil column moisture. For each forecast date and ensemble member, the soil moisture value represents the mean for that date and the preceding six days. This temporal averaging is performed internally by the Fortran-based Computational Hydrosystems (CHS) SMI software as part of the SMI calculation and is not applied as a separate preprocessing step.

The SMI expresses the forecast 7-day mean soil moisture at each model grid cell as a quantile relative to the historical soil moisture distribution for the corresponding time of year. The historical reference distribution is based on mHM simulations for 1975–2024. The SMI is dimensionless and ranges from 0 to 1, with low values indicating unusually dry conditions and high values indicating unusually wet conditions. For example, an SMI value of 0.2 indicates that the forecast soil moisture is at approximately the 20th percentile of the historical distribution for that time of year.

The SMI was calculated using the Fortran-based CHS SMI software and methodology available at https://git.ufz.de/chs/progs/SMI.

This dataset is a derived data product generated from spatially interpolated and reforecast-based bias-corrected IFS ensemble meteorological forecasts, mHM-simulated total soil column moisture, and the historical SMI reference distribution. It supports the forecast verification, probabilistic drought analyses, figures, and tables presented in the associated manuscript, “High-resolution soil moisture forecasts for drought early warning in Germany.”


DOI
https://doi.org/10.48758/ufz.17404
Cite as (APA)
Najafi, H. (2026). High-resolution soil moisture forecasts for drought early warning in Germany: IFS–mHM ensemble forecast dataset (Najafi et al.) [Dataset]. Helmholtz-Centre for Environmental Research. https://doi.org/10.48758/UFZ.17404
Size
275,2 GB
License and Terms of Use
LICENSE and TERMS OF USE
Data quality
quality assured data
(required)
(optional)
17 − 8 + 3  =
 
If you wish to contact us regarding the displayed data set, please fill out the following form completely.
Your message will be sent automatically to the correct contact persons.
(maximum 100 characters)
(maximum 4000 characters)
75 − 3 + 8  =
 
Metadata Catalogues
Dublin Core
URI
https://www.ufz.de/record/dmp/archive/17404
DOI
10.48758/ufz.17404
Subject
soil moisture index (SMI)
Publication Date
24.07.2026
Type of data
dataset (structured data in a defined format)
Data quality
quality assured data
Status of data
processed data
Data format
netcdf
Creators
Najafi, Husain
Contributors
Samaniego-Eguiguren, Luis
Publisher
Helmholtz-Centre for Environmental Research
Rights holders
Helmholtz-Centre for Environmental Research
Citation note
data are managed by Husain Najafi
Contact
husain.najafi@ufz.de
Time or Period of creation
22.07.2026
Language of the content
English
Target group
Anyone interested in ensemble soil mositure forecasts of 2025 German drought
Access permissions
free access