Consistency Analysis of Global Terrestorial Water Cycle: Comparison of observations with Today's Earth simulations
Abstract
The terrestrial water cycle is a key element of the climate system representing all water exchange processes at the surface. To investigate those, Authors have developed the global terrestrial hydrological simulation system "Today's Earth (TE)". The TE system aims to evaluate & provide long-term hydrological dataset utilizing satellite derived information and to contribute the society as a part of the climate services by calculating risk indices of water hazards, particularly floods.
The TE system simulates energy and water balances over global land surface using MATSIRO5 (Takata et al., 2003) as land surface model, and then calculate river discharge or flood area fraction by the CaMa-Flood, global hydrodynamic model (Yamazaki et al., 2010). The system is currently operated as global model with 0.5-degree latitude/longitude grid for MATSIRO5 and 0.25-degree grid for CaMa-Flood. As the forcing data of the system, we utilize JMA's 55-year reanalysis (JRA-55; Kobayashi et al., 2005) as baseline experiment (TE-JRA-55). Another experiment utilizing Global Satellite Mapping of Precipitation (GSMaP) is also evaluated (TE-GSMaP) to see how those two different experiments describe the terrestrial water cycle. First, we compared the global terrestrial water cycle (P: precipitation, E: evapotranspiration and dS/dt: storage change) estimated by the TE system with observations. Here we utilized the GPCC rain gauge dataset, GLEAM satellite based evapotranspiration, and GRACE derived TWS, respectively, and found that the GSMaP results were generally consistent with observations in the tropics, but dS/dt was overestimated in the mid-latitudes (30-60 degrees), where overestimation of GSMaP was reported, indicating differences in accuracy among regions. Next, basin-scale water balances (P, E, dS/dt, and R: river discharge) were analyzed in a river basin with a catchment area of more than 300,000 km2 for which in-situ observation data are available. The monthly mean residual in the water balance using independent observation data was limited to about ±2 mm/day. The sign of the trend of each of the 10 targeted river basin averages for each of the variables was consistent with observations and simulations, suggesting an acceleration of the water cycle over the 14 years covered.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMH082...10Y
- Keywords:
-
- 1836 Hydrological cycles and budgets;
- HYDROLOGY;
- 1855 Remote sensing;
- HYDROLOGY;
- 1878 Water/energy interactions;
- HYDROLOGY;
- 1910 Data assimilation;
- integration and fusion;
- INFORMATICS