Inland Water Altimetry

Measuring inland water heights using satellite radar altimetry is an established application developed over many years, with several sites offering access to river and lake data from prior satellites. The new generation of SRAL satellite radar altimeters presents new opportunities and challenges, particularly for monitoring of river targets. This paper presents a global analysis of Cryosat2 SAR and LRM mode waveforms over river systems, and combines this with outcomes from height retrieval from prior missions to give an estimate of the Sentinel3 capability to retrieve river system heights. For Cryosat2 data in both SAR and LRM modes, 50-80% of the echoes are found to be multi-target responses, a far higher percentage than from previous missions. However, a small number of well-behaved single-target responses are also found, well distributed throughout the river systems. Prior mission global analysis outcomes demonstrate that river height timeseries can successfully be retrieved even where only one or two ‘clean’ echoes are retracked at each satellite overpass. The conclusion from this part of the study is therefore that heights should be retrieved from Sentinel3 SAR mode waveforms from the majority of the river systems examined, if the SRAL tracker is at least as good as that on Cryosat2. Estimates of the enhanced retrieval from Sentinel3 SRAL FBR echoes are made using Envisat burst echoes and Cryosat2 SAR FBR data. One conclusion is that a target must be at least ‘80Hz’ wide in order to be successfully monitored, due to the discontinuous sampling of the SRAL; however, whilst the noise is found to be higher on 80Hz waveforms, small numbers of ‘clean’ echoes are still retrieved.

Successfully retracking these echoes to retrieve river heights is a complex problem. The widely used OCOG retracking is not found to give optimal results for SAR mode river waveforms, where multi-target responses complicate the echo shapes both at 20Hz and, using stacked SAR FBR echoes, at 80Hz. A new SAR waveform retracker has therefore been developed, and the results from first analysis and assessment of the retracking performance are presented here using Cryosat2 SAR mode waveforms. Improvement in along-track RMS ranges from 10% to 50% compared to OCOG retracking. An analysis of the stability of this retracker to the increased waveform noise at 80Hz is also made: the results are promising. Combining all the analysis results, a first look towards Sentinel3 SAR mode echo characterization and height retrieval is presented over example river basins.

Remote sensing provides valuable data for parameterization and updating of hydrological models, for example water level measurements of inland water bodies from satellite radar altimeters. Many studies have used satellite altimetry data from repeat-orbit missions such as Envisat, ERS or Jason, or synthetic wide-swath altimetry data as expected from the SWOT mission. This study is one of the first hydrologic applications of altimetry data from a drifting orbit satellite mission, namely CryoSat-2. CryoSat-2 is equipped with the SIRAL instrument, a new type of radar altimeter similar to SRAL on Sentinel-3.

CryoSat-2 SARIn level 2 data is used to improve a 1D hydrodynamic model of the Brahmaputra river basin in South Asia set up in the DHI MIKE 11 software. CryoSat-2 water levels were extracted over river masks derived from Landsat imagery. After discharge calibration, simulated water levels were fitted to the CryoSat-2 data: In a first step, the average simulated water levels along the river were calibrated to the CryoSat-2 data by adapting the hydrodynamic cross section datums. Subsequently the simulated water level amplitudes were fitted to those obtained from Envisat virtual station time series by adapting the cross section shapes.

The water level was only calibrated for the Brahmaputra in the Assam valley due to a lack of Envisat data further upstream. Despite the steep and rugged terrain in the upstream part of the Brahmaputra, the CryoSat-2 data was found usable after a Landsat river mask was applied. After calibration a hydrodynamic model with accurate spatio-temporal representation of water levels is obtained.

This is a prerequisite for real-time model updating by assimilation of CryoSat-2 altimetry or multi-mission data in general. For this task, a data assimilation framework has been developed and linked with the MIKE 11 model, enabling the integration of any kind of water level measurements. It is a flexible framework that can assimilate water level data which are arbitrarily distributed in time and space. Different types of error models and data assimilation methods can easily be used and tested. Furthermore, it is not only possible to update the water level of the hydrodynamic model, but also the states of the rainfall-runoff models providing the forcing of the hydrodynamic model. The setup has been used to assimilate CryoSat-2 observations over the Assam valley for the years 2010 to 2013, testing different data assimilation methods and model error representations.  Performance improvement in terms of discharge forecast due to the assimilation of satellite altimetry data was then evaluated.

Here we present a new system for retracking of SAR waveforms over rivers and lakes. Satellite altimetry offers frequent and global sampling across borders, which can be used to validate and calibrate hydrological models in remote areas where in situ measurements are scarce.

The system was developed using CryoSat-2 20Hz SAR data, but due to the similarities between the Sentinel-3 SRAL altimeter and the SIRAL altimeter on-board CryoSat-2 an adaption of the method will be straightforward.

The SAMOSA-3 retracker has previously demonstrated to perform better than other existing SAR retrackers over water surfaces and has therefore been chosen as the primary retracker whenever applicable. To find the waveforms for which the SAMOSA-3 retracker is appropriate, a classification is performed using first the k-means for clustering sample waveforms into classes and then the Naïve Bayes classifiers supported by training classes from the k-means for determining classes of all waveforms. The waveforms that are found suitable for the SAMOSA-3 retracker are isolated; these are the ocean-like and the highly specular waveforms. If the SAMOSA-3 retracker is found unsuitable or if the obtained result from this retracker is unsatisfactory, the waveform is retracked using the newly developed Multiple Waveform Persistent Peak (MWaPP) sub-waveform retracker, which studies adjacent waveforms to reveal the sub-waveform most likely to represent the echo from nadir.

The method has been validated against lake and river gauges around the world to ensure that the system works for multiple types of inland water, and reveals that the combined method performs very well, but does not obtain significantly higher precisions compared to the MWaPP retracker. The results also show the MWaPP retracker provides height estimates with precisions very similar to those of the SAMOSA-3 retracker.

The key drivers for improving the inland water height level monitoring capability from space observations are two-fold. The first one consists in increasing the temporal/spatial coverage of the radar altimeter measurements by enlargement of the altimeter constellation or densification of the observation points (between virtual stations over rivers for missions on repetitive tracks). The second one requires improvements in the water level estimation availability and precision.

The new generation of radar altimeters featuring a delay/doppler mode (currently on board the Cryosat-2 satellite and in the upcoming Sentinel-3A, -3B and future Jason-CS missions) will clearly fulfill this second point offering improved performances with respect to conventional altimeters. The high along-track spatial resolution of the delay/doppler altimeter may enable the measurement of high-resolution water level transects across rivers and other water basins where conventional altimeters echoes are impacted by land contamination due to their larger footprint size. Such dataset will also permit to extract spatial and temporal derivatives of water level, which are characteristics of major interest in hydrology.

Based on Cryosat-2 measurement analyses obtained with innovative methods (at level 1 and level 2), delay/doppler mode (Sentinel-3 will be first mission to cover the entire hydrological surfaces with a delay/doppler mode) has already demonstrated its great performances providing new insights into the hydrological processes and phenomena occurring in hydrological basins (rivers and lakes). As well, observation in Ka band provides highly valuable height measurements.

With the continuous increase of the altimeter constellation (that will be complemented by the future exploitation of the SWOT mission around 2020), space observations are regularly providing a significant increase of the continental water body monitoring capabilities that can be exploited by hydrologists and model developers.

In this context, we propose to give an illustrated overview of the altimeter constellation potentiality and performances over hydrological conditions, taking benefit of the synergy and complementarity between the different altimeter missions.

This work is the continuation of the work we presented in 2015 [Fabry et Bercher, Venise 2015] [Fabry et Bercher, Frascati 2015] on the understanding of SAR mode altimetry in hydrology. It is part of the SEOM Sentinel-3 Hydrologic Altimetry Processor prototypE (SHAPE) study which aims at boosting the use of “SAR” mode altimetry data in hydrology. The main reason why Space Hydrology is still not operational at global scale is the variety of inland water scenes and scenarios which cannot properly be taken into account via a fixed processing chain. The complexity coming from the spatial diversity is emphasized by the strong temporal variability related to seasonal trends, extreme events and human action. The radar backscatter properties of water depend on wind conditions, surface current and trophic phenomenons.

The key aspect of the SHAPE project that is addressed here is the qualitative and quantitative characterization of the existing SAR Mode Altimetry data (CryoSat-2) over the inland water domain.

We defined and implemented a new conceptual framework that eases and partially automates the analysis of CryoSat-2 measurements over inland waters through the use of water masks and the altimeter footprints. This serves to analyse the altimeter data together with the information on the water fraction within the instrument footprint. This new framework is inherently adaptive to the existing water masks.

The following physical and statistical indicators rare computed:

the Mean Individual Echoes, Stacks and Waveforms (SAR and RDSAR) as reference signals,

the Kurtosis (peakiness), Skewness (asymmetry), Standard Deviation for Stacks and Waveforms (SAR and RDSAR),

Specifically for the MLD: Stack centre (mispointing and/or mean surface slope), Range Integrated Power (RIP) statistics, MLD static & animated plots with a crossing over the non-water area, transition area and water area categories.

Several improvements have been made in both the computations and the method to analyse the results :

- tracker range used to properly locate the footprint on ground is now tropospheric corrected,

- the real satellite attitude (pointing angle) is used to properly locate the footprint on ground,

- not only Beam-Doppler limited footprint is used but also Pulse-Doppler limited footprint to beter locate water content,

- more recent masks are used (Sentinel-1 SAR images derived water masks from our segmentation tool)

- we randomly reject some of the members of the over populated water fraction ; especially in the 0% (non water) class.

The outcome of this work may impact future retrackers and ease the mapping of water bodies through water detection criteria. Results are provided over several areas of interest.