Description: We compute displacements of global SLR station coordinates by atmospheric loading based on surface pressure data from European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-interim data. Inhouse we generate two branches: firstly straightforward following Farrel’s theory but using updated load Love numbers, secondly from utilizing localized Green’s functions instead of global ones. Externally provided displacements are available f.i. from the International Mass Loading Service (IMLS) based on different input data and modeling. We compare these displacements and apply them to Satellite Laser Ranging (SLR) data processing of a recent six years period of the LAGEOS, LARES, AJISAI, STARLETTE and STELLA geodetic missions. We assess the impact of the loading models on precise orbit determination and Earth parameters of interest.

Description: We apply synthetic aperture radar data and geophysical modeling to assess ground deformation changes at the 284 km2 large Toktogul Reservoir in Kyrgyzstan, Central Asia, which is used for hydropower generation and irrigation. The reservoir’s water level is prone to significant changes during the year, but also shows inter-annual variations due to overall water recession or accumulation. We use Envisat ASAR data to analyse the ground deformation during a time of exaggerated use of water between 2004 – 2009 (net water level drop of 60 m / 13.5 km3 ) and Sentinel-1 data to derive the ground deformation during a time of overall water level increase between 2014 – 2016 (net water level plus of 51 m / 11.2 km3 ). The deformation pattern was measured by generating an interferometric time-series using the Small BAseline Subset (SBAS) approach. After removing heavily impacting atmospheric effects by applying the elevation dependent powerlaw approach, results show that both sensors are able to image related uplift and subsidence signals in the order of approximately 1 mm per 1 m water level change for the investigated time periods. Moreover, time-series results from Sentinel-1 also resolve intra-annual changes induced by 40 m periodical water level changes. Reasons for this superior behaviour of Sentinel-1 data are a short temporal baseline of 12 days and a small orbital tube, which both lead to a higher temporal sampling compared to the Envisat setting and at the same time to a better correlation of points within the interferograms. The derived spatial pattern of land-deformation rate is validated against modeling of the elastic deformation, based on a Love-number approach. The load forcing due to lake-level changes is derived from satellite-based radar altimetry.

Description: Short-term forecasts of atmospheric, oceanic, and terrestrial hydrospheric effective angular momentum functions (EAM) of Earth rotation excitation are combined with least-squares extrapolation and auto-regressive modelling to routinely predict polar motion (PM) and ∆UT1 for up to 90 days into the future. Based on several experiments with more than 500 individual hindcasts from 2016 and 2017, a best-performing parametrization for the method was identified. At forecast horizons of 10 days, the prediction accuracy is 3.02 mas and 5.39 mas for PM and ∆UT1, respectively, corresponding to improvements of 34% and 44% with respect to Bulletin A. At forecast horizons of 60 days, prediction accuracies are 12.52 mas and 107.96 mas for PM and ∆UT1, corresponding to improvements of 34% and 8% over Bulletin A. The 90 days-long EAM forecasts leading to those improved EOP predictions are routinely published once per day at www.gfz-potsdam.de/en/esmdata and are thus ready-for-use for operational EOP prediction efforts.

Description: The influence of the elastic Earth properties on seasonal or shorter periodic surface deformations due to atmospheric surface pressure and terrestrial water storage variations is usually modeled by applying a local half-space model or an one dimensional spherical Earth model like PREM from which a unique set of elastic load Love numbers, or alternatively, elastic Green's functions are derived. The first model is valid only if load and observer almost coincide, the second model considers only the response of an average Earth structure. However, for surface loads with horizontal scales less than 2500 km2, as for instance, for strong localized hydrological signals associated with heavy precipitation events and river floods, the Earth elastic response becomes very sensitive to inhomogeneities in the Earth crustal structure. We derive a set of local Green's functions defined globally on a 1° × 1° grid for the 3-layer crustal structure TEA12. Local Green's functions show standard deviations of ±12% in the vertical and ±21% in the horizontal directions for distances in the range from 0.1° to 0.5°. By means of Green's function scatter plots, we analyze the dependence of the load response to various crustal rocks and layer thicknesses. The application of local Green's functions instead of a mean global Green's function introduces a variability of 0.5 − 1.0 mm into the hydrological loading displacements, both in vertical and in horizontal directions. Maximum changes due to the local crustal structures are from −25% to +26% in the vertical and −91% to +55% in the horizontal displacements. In addition, the horizontal displacement can change its direction significantly. The lateral deviations in surface deformation due to local crustal elastic properties are found to be much larger than the differences between various commonly used one-dimensional Earth models.

Description: GRACE/GRACE-FO Level-3 product based on GFZ RL06 Level-2B products (Dahle & Murböck, 2019) representing Ocean Bottom Pressure (OBP) variations provided at 1° latitude-longitude grids as defined over ocean areas. The OBP grids are provided in NetCDF format divided into yearly batches. The files each contain seven different variables: 1) 'barslv': gravity-based barystatic sea-level pressure 2) 'std_barslv': gravity-based barystatic sea-level pressure uncertainties 3) 'resobp': gravity-based residual ocean circulation pressure resobp 4) 'std_resobp': gravity-based residual ocean circulation pressure uncertainties 5) 'leakage': apparent gravity-based bottom pressure due to continental leakage 6) 'model_ocean': background-model ocean circulation pressure 7) 'model_atmosphere': background-model atmospheric surface pressure These Level-3 products are visualized at GFZ's web portal GravIS (http://gravis.gfz-potsdam.de). Link to data products: ftp://isdcftp.gfz-potsdam.de/grace/GravIS/GFZ/Level-3/OBP

Description: The Earth System Modelling Group of GeoForschungsZentrum Potsdam (ESMGFZ) routinely operates global Earth system models to provide real-time information about gravity variations, Earth rotation excitations, and surface load deformations related to mass redistributions in the atmosphere, ocean, and terrestrial water storage. The general ocean circulation model MPIOM and the hydrological model LSDM are both consistently forced with 3-hourly atmospheric data from ECMWF. Gravity variations are available as GRACE AOD1B release 06 dealiasing product (spherical harmonic coefficients d/o 180, 3h); Earth rotation excitations are available as effective angular momentum time series AAM (3h), OAM (3h), HAM (24h), and SLAM(24h); Non-tidal deformations due to surface mass loads are available as gridded data sets NTAL (0.5◦ , 3h), NTOL (0.5◦ , 3h), and HYDL(0.5◦ , 24h). All data sets start in the year 1976 and are updated daily at about 11 UTC for the last day. The updates include also 6-day forecasts. ESMGFZ products are accessible via the GFZ webpage www.gfz-potsdam.de/en/esmdata.

Description: Since its launch in 2002, the Gravity Recovery and Climate Experiment (GRACE) mission provides a unique way to monitor the terrestrial water storage (TWS) variations at large spatial scale (〉300km) by measuring month-tomonth changes of the Earth’s gravity field. We apply TWS variations estimated from GRACE to assess the ability of four hydrological and land surface models to simulate the continental branch of the global water cycle. Based on four different validation metrics that focus on variability on sub-seasonal to inter-annual time scales, we demonstrate that for the 31 largest discharge basins worldwide all model runs agree with the observations to a very limited degree only, together with large spreads among the models themselves. In particular, we focus on selected basins with very different climatic conditions and discuss time series of individual water storage components such as surface water, soil moisture, and snow depth. Since we are applying a common atmospheric forcing data-set to all models considered, we conclude that the discrepancies found are not due to differences in the forcing, but are mainly related to the model structure and parametrization. By investigating the relative performance of these different models, we attempt to give directions for further development of global numerical models in the areas of large-scale hydrology and land-atmosphere interactions.