Enhancing Discharge Interpolation for SWOT
Abstract
The highly intermittent nature of the SWOT river observations in both space and time poses a major challenge to users for most applications. Therefore, it is critical to assemble the discrete SWOT river observations into a spatially complete and temporally continuous data record. Different spatial/temporal interpolation methods have been proposed and many involve implementing a river routing model and assimilating SWOT data into the model. We focus on a fixed-window Kalman Smoothing technique based on the so-called Inverse Streamflow Routing (ISR), which applies a two-sweep procedure to propagate observed information exhaustively up and down the river channel network in time/space. One goal of this approach is to minimize the use of model background estimates and to make the data record as purely "SWOT" as possible, so it is designed to work with or without background estimates at all. However, ongoing research shows that the SWOT data gaps can dictate the efficacy of such interpolation and larger gaps may not be well reconstructed if available neighboring data are slightly too far away in space or time. In other words, the observed information fails to propagate far enough. To enhance the propagation of observed information from SWOT, which is already exhaustive across the river network, we introduce spatial and temporal error correlation in runoff fields (i.e. "discharge" at the furthest possible upstream) into the smoothing algorithm. The basic premise is that if one SWOT overpass on the river picks up a storm event from its contributing basin, then it is likely that the neighboring basins may also be affected by the same storm and that should be reconstructed even if the neighboring basins are not observed. We expect the introduction of spatial and temporal coherence in storm runoff to enable (1) stronger gap-filling power of the discharge interpolation procedure and (2) that the SWOT observed information can be propagated to the maximum possible range in time and space.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMOS53C1342P
- Keywords:
-
- 1845 Limnology;
- HYDROLOGYDE: 1856 River channels;
- HYDROLOGYDE: 4520 Eddies and mesoscale processes;
- OCEANOGRAPHY: PHYSICALDE: 4544 Internal and inertial waves;
- OCEANOGRAPHY: PHYSICAL