Evidence suggests that catchment state variables such as groundwater can exhibit multiyear trends. This means that their state may reflect not only recent climatic conditions but also climatic conditions in past years or even decades. Here we demonstrate that five commonly used conceptual “bucket” rainfall‐runoff models are unable to replicate multiyear trends exhibited by natural systems during the “Millennium Drought” in south‐east Australia. This causes an inability to extrapolate to different climatic conditions, leading to poor performance in split sample tests. Simulations are examined from five models applied in 38 catchments, then compared with groundwater data from 19 bores and Gravity Recovery and Climate Experiment data for two geographic regions. Whereas the groundwater and Gravity Recovery and Climate Experiment data decrease from high to low values gradually over the duration of the 13‐year drought, the model storages go from high to low values in a typical seasonal cycle. This is particularly the case in the drier, flatter catchments. Once the drought begins, there is little room for decline in the simulated storage, because the model “buckets” are already “emptying” on a seasonal basis. Since the effects of sustained dry conditions cannot accumulate within these models, we argue that they should not be used for runoff projections in a drying climate. Further research is required to (a) improve conceptual rainfall‐runoff models, (b) better understand circumstances in which multiyear trends in state variables occur, and (c) investigate links between these multiyear trends and changes in rainfall‐runoff relationships in the context of a changing climate.

Satellite altimeter sea surface heights, in combination with Argo ocean temperature and salinity observations, provide an independent measure of global mean ocean mass (GMOM) change. Over the period January 2005 to April 2020, GMOM rates observed by the Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow‐On (GFO) range from 1.88 ± 0.10 to 2.63 ± 0.10 mm/year, compared to 2.85 ± 0.37 mm/year from Altimeter‐Argo. Over much of the GRACE era, GRACE GMOM estimates agree well with Altimeter‐Argo over a broadband of frequencies. However, during the late stage of the GRACE mission (after August 2016) and into the GFO era, differences between GRACE/GFO and Altimeter‐Argo GMOM estimates become substantially larger and systematic, which may be related to the single accelerometer mode of operation during late‐stage GRACE and GFO missions and/or errors of Argo and altimeter data.

Rising atmospheric and ocean temperatures have caused substantial changes in terrestrial water circulation and land surface water fluxes, such as precipitation and evapotranspiration, potentially leading to abrupt shifts in terrestrial water storage. The European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) soil moisture (SM) product reveals a sharp depletion during the early 21st century. During the period 2000 to 2002, soil moisture declined by approximately 1614 gigatonnes, much larger than Greenland’s ice loss of about 900 gigatonnes (2002–2006). From 2003 to 2016, SM depletion continued, with an additional 1009-gigatonne loss. This depletion is supported by two independent observations of global mean sea level rise (~4.4 millimeters) and Earth’s pole shift (~45 centimeters). Precipitation deficits and stable evapotranspiration likely caused this decline, and SM has not recovered as of 2021, with future recovery unlikely under present climate conditions.

We carry out a comprehensive error assessment of Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow‐On (GFO) Release‐6 (RL06) solutions from the Center for Space Research (CSR) at the University of Texas at Austin, NASA Jet Propulsion Laboratory (JPL), and Geoforschungszentrum (GFZ). The study covers the period April 2002 to August 2020 and uses two different methods, one based upon open ocean residuals (OOR) and the other a Three‐Cornered Hat (TCH) calculation. General results from the two methods are similar. With 300 km Gaussian smoothing OOR RMS errors for CSR, JPL, and GFZ solutions are ∼2.01, 3.19, and 3.67 cm, respectively. With additional decorrelation filtering OOR RMS values are reduced to ∼1.24, 1.53, and 1.69 cm, respectively. TCH analysis also shows that CSR has the lowest noise levels with similar RMS values, and additional decorrelation filtering reduces error levels. TCH may underestimate errors if there are common errors among geophysical background models. Errors in GFO's first two years (25 solutions for 2018.06 to 2020.08) are comparable to those of GRACE when zonal degree 2 and 3 coefficients are replaced by Satellite Laser Ranging estimates. The OOR method reveals mismodeled intra‐seasonal dynamic ocean signals associated with the Argentine Gyre during de‐aliasing, while the TCH method shows differences between ocean tide models near Australia and Antarctica. Both OOR and TCH RMS analysis offer a means to assess the noise level of GRACE/GFO estimated mass change. The actual uncertainty of GRACE/GFO estimate averaged (or totaled) over a given region is also affected by other error sources.

Global mean ocean mass change derived from the Gravity Recovery and Climate Experiment (GRACE) gravity solutions generally agrees well with ocean mass change inferred from satellite altimeter sea surface height and Argo floats observations during the period January 2005 to December 2015. However, there is a systematic annual phase lag (~10°) between GRACE and Altimeter‐Argo estimates. This phase lag is attributed to the enforced mass conservation in GRACE gravity solutions, in which the ΔC₀₀ coefficients (representing changes in total Earth mass) are set to zero. After a correct implementation of global mass conservation by removing global mean atmospheric mass from the GRACE solutions using atmospheric model predictions, the annual phase lag is nearly completely gone, yielding significantly improved agreement between GRACE and Altimeter‐Argo estimates. In addition, retaining GRACE ΔJ₂ coefficients provides better GRACE ocean mass estimates at both seasonal and long‐term time scales.