Influence of GIA uncertainty on climate applications from satellite gravimetry

<p>Global coupled climate models are important for predicting future climate conditions. Due to sometimes large and often systematic model uncertainties, it is crucial to evaluate the outcome of model experiments against independent observations. Changes in the distribution and availability of terrestrial water storage (TWS), which can be measured by the satellite gravimetry missions GRACE and GRACE-FO, represent an important part of the climate system. However, the use of satellite gravity data for the evaluation of coupled climate models has only very recently become feasible. Challenges arise, e.g., from the still rather short time series of satellite data and from signal separation issues related to GRACE/-FO observing all mass change including non-water related variations such as glacial isostatic adjustment. Apart from climate model uncertainties, these challenges might be the reason for a disagreement between the direction of linear water storage trends of models and observations in several regions of the world, one of them located in Eastern Canada.</p> <p>This presentation will highlight the latest results achieved from our ongoing research on climate model evaluation based on the analysis of an ensemble of models from the Coupled Model Intercomparison Project Phase 6 (CMIP6). We will focus on long-term wetting and drying conditions in TWS. Using an ensemble of 52 GIA models that differ in the applied ice history, solid Earth rheology, and numerical code, this presentation will discuss how GIA modeling uncertainty does influence (i) the determination of water storage trends from GRACE/FO data, and (ii) the (dis-)agreement between drying/wetting trends in satellite gravimetry and CMIP6 climate models. We will show that the apparent disagreement between observations and models in highly GIA-affected regions in North America crucially depend on the particular model chosen for reducing the GIA effect from the GRACE satellite data.</p>