Available online at www.sciencedirect.com Advances in Space Research 51 (2013) 1542–1550 www.elsevier.com/locate/asr From TOPEX/Poseidon to Jason-2/OSTM in the Amazon basin Frédérique Seyler a,⇑, Stéphane Calmant b, Joecila Santos da Silva c, Daniel Medeiros Moreira d, Franck Mercier e, C.K. Shum f,g a IRD/ESPACE-DEV, 500 Rue Jean Francßois Breton, 34093 Montpellier, France b IRD/LEGOS, 14 Av. Edouard Belin, 31400 Toulouse, France c UEA/CESTU, Av. Djalma Batista 3578, 69058-807 Manaus, Brazil d UFRJ/CPRM, Av. Pasteur 404, 22290-040 Rio de Janeiro, Brazil e CLS, Collecte Localisation Satellites, 8–10, rue Hermès, Parc Technologique du Canal, 31520 Ramonville Saint-Agne, France f Division of Geodetic Sciences, School of Earth Sciences, Ohio State University, 125 South Oval Mall, 43210 Columbus, OH 43210, United States g Institute of Geodesy and Geophysics, Chinese Academy of Sciences, Wuhan, China Available online 12 November 2012 Abstract A major interest of radar altimetry over rivers is to monitor water resources and associated risk in basins where there is little or no conventional in situ data. The objective of the present study is to calibrate altimetry data in a place where conventional data are available, and use the results to estimate the potential error committed in the estimation of water levels in an ungauged or poorly gauged basin. The virtual stations extracted with Jason-2 in this study concern a very broad sample of river channel width and complexity. Minimum channel width has been estimated at 400 m. Unlike TOPEX/Poseidon (T/P), Jason-2 seems to have the capability to distinguish the river bed from its floodplain. The quality of the results obtained with Jason-2 is incomparably better than that obtained with T/P. Despite the fact that no absolute calibration has been assessed for river in this study, the bias calculated converge around 0, 35 m, which could be then the error estimated on the water stage derived from Jason-2 ranges, when no other validation is available. ICE3 algorithm seems to be performing as well as ICE1, and further research is needed to design retracking algorithm specifically for continental water. Ó 2012 COSPAR. Published by Elsevier Ltd. All rights reserved. Keywords: Altimeter calibration; Jason-2; Amazon basin; Hydrology 1. Introduction The family of TOPEX/Poseidon (T/P) satellites (Fu et al., 1991) extends over 20 years of altimetry history, since T/P was launched August 10, 1992 from Kourou in French Guiana. It is only in 1996 that the retracking of T/P archive by the Science Working Team on the T/P project, achieves 2–3 cm error in estimating the ocean surface. This accuracy has been reached because of reduced radial orbit errors (Bertiger et al., 1994), reaching the sub-centimetric accu⇑ Corresponding author. E-mail addresses: frederique.seyler@ird.fr (F. Seyler), stephane.calmant@ird.fr (S. Calmant), jsdsilva@uea.edu.br (Joecila Santos da Silva), daniel.moreira@cprm.gov.br (D.M. Moreira), franck.mercier@cls.fr (F. Mercier), ckshum@osu.edu (C.K. Shum). racy for the Jason-2 mission (Bertiger et al., 2010). It is also about that time that are emerging the first applications of altimetry for inland waters (Morris and Gill, 1994; Birkett, 1995a,b, 1998; Ponchaut and Cazenave, 1998). It is only 10 years later that are published early works on multi altimeter mission for inland waters (Berry et al., 2005; Frappart et al., 2006). Jason-1 that was highly anticipated as a following of T/P, was for inland waters a “gap” in data since very few data of Jason-1 are useful for monitoring inland waters. It was not until 2008 and the launch of Jason-2 to continue the Poseidon family on inland water bodies. In the studies of the ocean, it is possible to combine data from different missions as long as the relative biases are estimated. This is due to the spatial and temporal continuity of the ocean environment. The case of rivers and lakes is different. Most lakes of small area will no longer be 0273-1177/$36.00 Ó 2012 COSPAR. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.asr.2012.11.002 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 monitored if the trace of the orbit changes during a mission or between two different missions. In the case of rivers, there exists a spatial continuity but limited to the hydrographic network. The water level in the river system results from the combination of hydrodynamic factors (slope, flow velocity, roughness of the bed), morphological factors (shape of the section that is variable throughout the hydrological cycle), the contribution of the tributaries and the various exchanges with the watershed and the floodplain (direct rainfall, evaporation, diffuse runoff, groundwater contribution and subtraction). It is easily understandable that this combination of factors is highly and non linearly variable in space and time. In addition, the use of satellite altimetry for monitoring river stage allowed defining the notion of virtual station (Frappart et al., 2006; Leon et al., 2006; Roux et al., 2008). A virtual station is constituted by the measurement points from the ground track portion located at the intersection with the river. This concept was designed partly to be close to terrestrial water level monitoring networks that consist of fixed stations. The interest and value of a water level measurement station is linked to its lifespan in a specific place. This persistence allows following changes in the hydrological regime in the long term and predicting extreme events. Monitoring and forecasting is of course even more significant in the context of climate change. It is why the concept of lineage of the missions is an important concept for hydrology, as at a family of satellite corresponds a common orbit, which ensures continuity in the river or lake monitoring. It is the first objective of this study to compare the results obtained by T/P and by Jason-2 in estimating the water stage at virtual stations. In hydrology, the need for temporal continuity of the missions is accompanied by a need for a spatial distribution as dense as possible. This is one of the great advantages of altimetry over conventional hydrological measurement networks to be globally distributed in a dense network constituted by the different satellite ground tracks. For example, the study of Frappart et al. (2005) calculated the storage of the inundation plain of the Rio Negro sub-basin. This sub-basin of about 700 000 km2 counts 25 in situ stations and T/P altimetric virtual stations added 88 monitoring points of the water stage. Using multi-mission sources allows densifying the monitoring network of virtual stations. In this respect the relatively loose mesh of the Jason-2 orbit is complementary to that of Envisat, much denser. The reverse is true for the temporal resolution. The revisit period greater than one month prevents a number of hydrological applications to be contemplated with Envisat data. In this respect, Jason-2 is better suited for some applications needing frequent observations. However for using multi-mission data, it is necessary to determine possible bias between sensors. If there are a number of studies on absolute bias of Jason-1 and 2 in the oceanic domain (Dettmering and Bosch, 2010; Bonnefond et al., 2010; Haines et al., 2010; Mertikas et al., 2010, 2011; Arnault et al., 2011; Watson et al., 2011; Washburn et al., 2011, among others), a few studies 1543 of calibration over lakes (Birkett and Beckley, 2010; Cheng et al., 2010; Cretaux et al., 2009, 2011), there is currently no estimate of absolute bias for Jason-2 on rivers. This is the second objective of this paper to determine if absolute calibration for Jason-2 is possible over river. The advantage of using data processed by tracker designed specifically for inland waters has been repeatedly shown (Frappart et al., 2006; Birkett and Beckley, 2010). Cited studies established that the algorithm ICE1 was more robust than the other retracking algorithms (OCEAN, SEAICE and ICE2) for both Envisat and Jason-2 over inland waters. When estimating the water height, ICE1 do not always give the less noisy result but it gives nearly the better result in most cases. In addition comparing with other retracking algorithm, ICE1 has the lowest rate of data loss. As soon as 2004, considerable effort has been made by the CASH project (Contribution of Satellite Altimetry to Hydrology, funded by the Réseau Terre & Espace,1 on behalf of the French Ministry of Research) for processing the whole T/P archive for the nominal orbit, with the trackers used for the Envisat mission (namely ICE1, ICE2 and SeaIce). During the PISTACH project, Mercier et al. (2007) have developed for CNES a tracker specific for inland waters called ICE3, that has been applied to the Jason-2 data. No studies comparing the two trackers ICE3 and ICE1 have been published and this is the third objective of the present study. 2. Data and methods Data used in this study were from two radar altimeter satellite mission, T/P and Jason-2/OSTM. Were also used data from GPS acquisition campaigns and in situ limnimetric data. T/P satellite was launched on August 10th, 1992 by NASA, U.S. space agency and CNES, the French space agency. It is given a sea level overall accuracy over one month better than 2 cm. In September 2002, T/P was moved to a new orbit, now used by Jason-1, liberating the orbit for Jason-2 after a tandem phase. The tandem formation was maintained (both ground track within 1 km of each other) from the launch of Jason-2 in June 2008 until January 26th, 2009, when Jason-1 was moved from that orbit into the T/P one. This phase has been used for calibrating the entire altimetric system (Quartly, 2010), mainly over ocean. It could not have been used over land surface as very few continental Jason-1 data is available. The T/P data used in this study encompass then the period 1992–2002 (http://www.aviso.oceanobs.com/en/missions/past-missions/topexposeidon.html). In the scope of the CASH project, the whole archive of T/P data on the nominal orbit has been retracked with the four algorithms used for the Envisat mission, i.e. ocean, ICE1, ICE2, and SEAICE. In this study we used the T/P data of the CASH 1 In French – means Earth and space network. 1544 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 archive retracked by ICE1, in order to be coherent with the processing used for Jason-2. Jason-2 mission was launched on June 20th, 2008. It has on-board the Poseidon 3 altimeter, coupled with the real time tracking system DIODE of DORIS, which should allow better and more frequent measurements in coastal areas, inland waters and ice caps. Poseidon 3 altimeter is a two-frequency solid-state altimeter, measuring range with accurate ionospheric corrections, operating at 13.575 GHz (Ku-band) and 5.3 GHz (C-band). Expected precision is about 2 cm over ocean. Raw data are processed by SSALTO (Segment Sol multimissions d’ALTimétrie, d’Orbitographie.2 In this study were used data from the geophysical data record (GDR-C), a fully validated product (Dumont et al., 2008), processed in delayed-mode processing (AVISO), distributed by CTOH (http://ctoh.legos.obs-mip.fr/). What is new with Jason-2 and allows retrieving a good proportion of data over land, unlike Jason-1, is an on-board tracking technique called “open loop tracking”. The internal memory of the altimeter includes a Digital Terrain Model (DTM) including elevations of areas flown along track. These DTM, coupled with the DIODE system, are used to anticipate the time position of the reception window of the radar in order to improve the data collection over the continental domain (http://www.altimetry.info/ html/alti/principle/waveform/onboard_tracking_en.html). T/P and Jason-2 shared the same orbit at successive times with a gap between 2002 and 2008. The orbit is 1336 km high, with a repeat time of 10 days, the distance between ground tracks being 315 km at the equator, and the along track resolution is 700 m for land applications. In order to compare the two retracking algorithm ICE1 and ICE3, two different datasets were used in this study. First, the altimetric ranges were extracted from CTOH database (http://ctoh.legos.obs-mip.fr/), using the geophysical and propagation corrections recommended for continental water: Wet troposphere correction calculated over the continents using NCEP (National Centers for Environmental Prediction) data (Mercier, 2003); ionospheric, tropospheric, pole tide and solid earth tide corrections provided in the GDRs have been applied. The range chosen was that retracked by the algorithm ICE1 (Bamber, 1994). ICE1 has been determined as the more robust for hydrological studies comparing with the other three algorithms available in the GDRs (OCEAN, ICE2 and SEAICE) (Frappart et al., 2006; Silva et al., 2010). Second, we used altimetric ranges retracked with the ICE3 algorithm, made available in the scope of the PISTACH project (Mercier et al., 2007). ICE3 has been developed to process altimetric products specifically for coastal areas and continental water bodies. Better geophysical corrections are applied to IGDR with a 20 Hz sampling rate, and 2 In French when in italic: Orbitography and altimetry multimission ground segment. pre-classification of the waveforms following GLOBCOVER land cover allows better discriminating the water bodies. For the three datasets (T/P retracked with ICE1, Jason2 retracked by ICE1 and ICE3), we used the VALS software (http://www.hybam.org) to build the temporal series (series of water stage varying with time). VALS software was designed by our team specifically for creating virtual stations. The method developed in VALS consists in extracting the data comprised within a polygon defined at the crossing of the tracks and the water body. A second step consists in performing a projection of the portion of the track onto a plane perpendicular to the river flow. The display in this vertical plane across the river bed allows the extraction of the part of each cycle comprised within the two margins of the river at all period of the hydrological cycle. Generally the selected points for the cycles acquired during the low flow period will be fewer than the points selected at high flow, as the river bed is narrower. Last, the median of the measurements for each cycle within the bed of the river is computed. The median is less sensitive than the average to possible outliers. The method used is best described in Silva et al. (2010). Most of in situ gauging stations used in this study are maintained by ANA (National Agency for Water of Brazil) jointly with CPRM (Geological Survey of Brazil). All data are available on hidroweb website (http://hidroweb.ana.gov.br/). Some of them are maintained in the scope of calibration projects funded by CNES/TOSCA program (program of the French Space Agency), IRD (French Institute for Development), CPRM, CNPq (Brazilian National Council for Scientific and Technological Development) and FINEP (Brazilian Financier of Studies and Projects). Readings of the water levels are made twice a day by an observer at the gauging station, at 7:00 and 17:00, and then the daily average value is calculated and dated at 12:00 local time. This is the reference daily water level for the given station. Details on the operation of the ANA/CPRM management of the Brazilian limnimetric network is given in hidroweb website. In order to level the in situ gauging stations to a common reference level, GPS observations have been conducted using dual-frequency receivers and micro-centered antennas and collected during various field campaigns. Position and ellipsoid height of each gauging station were computed using the GINS-PC software developed by CNES (Marty, 2009; Loyer et al., 2012) from GPS recorded observations lasting from a couple of hours to several days. All the GPS solutions were relative to ITRF2005 (Altamimi et al., 2007). To convert these GPS ellipsoidal heights of the gauge onto orthometric heights, we use the EGM2008 geoid model (Pavlis et al., 2012). Absolute daily river levels are then computed from the river stage at a given date corrected from the orthometric height of the zero of the gauge station. Detailed description of the leveling of the in situ gauging stations is given (Calmant et al., 2013). F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 Table 1 Direct comparison between a track and a gauge nearby (distance <10 km). Coari (241) Santa Luzia (063) Urucurituba (152) ICE 1 Jason-2 absolute bias (m) ICE 3 Jason-2 absolute bias (m) Dbias 0.582 ± 0.047 0.503 ± 0.026 0.506 ± 0.018 0.578 ± 0.042 0.459 ± 0.024 0.467 ± 0.018 0.004 0.044 0.039 1545 have been retained from the daily records of the in situ gauging stations. Second, for three tracks located at the vicinity of an in situ gauging station, it was performed a direct comparison between water stage estimated at the virtual station and the water level recorded at the in situ station (the three tracks are listed Table 1). 3. Results and discussion Biases for Jason-2 were computed for the two different datasets (retracked by ICE1 and ICE3). First biases were computed by difference between the altimetric time series and that of the two closest leveled gauging stations, one upstream and the other downstream of the virtual station (altimetric station). A constant linear slope model has been used for best-fitting the bias between the altimetric station and the two gauging stations. Only the concordant dates 3.1. From T/P to Jason-2 A number of 72 virtual stations (blue dots on Fig. 1) have been extracted from the Jason-2 mission within the Amazon basin. They have been chosen to be well distributed within the basin, from the largest stem of the Amazon near the mouth until the farthest reaches into the Andean piedmont, minimum channel width being estimated at Fig. 1. Location map; Blue dots represent the Jason-2 virtual station in the Amazon basin extracted for this study. Green symbols are the T/P virtual stations. Magenta large dot marks the location of case A Fig. 2, red large dot marks the location of case B Fig. 2, yellow large dot mars the location of the case C Fig. 2. The background is from SRTM (Shuttle Radar Terrain Model). 1546 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 Fig. 2. Three cases of comparison between T/P, Jason-2 and in situ gauge station; grey line/black triangles represent time series of ranges calculated for T/ P mission from 1992 to 2002, grey lines black dots represent time series for Jason-2 from 2008 on. Red line represents the time series of leveled river stage at the closest in situ gauge station. The graph above the time series represents the differences in meter between the altimetry time series and the conventional one. (For interpretation of the references to colours in this figure legend, the reader is referred to the web version of this article.) 1547 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 400 m. For each one of these virtual stations, T/P data (T/ P) have been processed in the same way (described here above) in order to extract virtual stations. From these 72 virtual stations, only 41 could be extracted for T/P. From these 41, when the time series is compared (for the concordant dates) with the time series of the closest in situ gauge station, only 18 have a RMS between the two stations less than ±1.50 m. These 18 cases are reported Fig. 1 (green symbols) and Fig. 2 gathers some of these favorable cases. In the three parts of the figure, the beginning of the grey time series (From 1992 to 2002) is made of T/P data and the end of Jason-2 data, from 2008. The red line represents the daily time series of the nearest gauging station. The graph above the time series illustrates the differences between virtual and in situ series. Within these 18 favorable cases, it is possible to distinguish three different situations. The first case is the most frequent and is illustrated by the Purus case (Fig. 2A, magenta large dot on Fig. 1). Only the upper part of the stage variations is captured by T/P. It could be due to the loss of data during low water when the river narrows. But the constant level of the low water enables interpreting this lack of data as the measurement by T/P on the dried floodplain. Few cases (four cases) show a phase concordance with the in situ time series, but the differences between the two series is of ±2–3 m with a high number of outliers (Fig. 2B, red large dot on Fig. 1). In this situation, the floodplain do not completely dries up, but the measurements at low flow are highly contaminated by the margins of the river. In some very rare cases (two cases), the T/P time series matches quite well the in situ time series. This is the case of the track 63 near Manaus (Fig. 2C, yellow large dot on Fig. 1); another one concerns the Madeira River. These two cases concern very large river without drying floodplain during the low flow season. The typical difference is more or less the same than that of the Jason2 time series, only the number of outliers is higher for T/ P than for Jason-2. T/P is lacking of resolution in comparison with Jason-2, and even using the same process for the two missions, T/P is not able to distinguish between the river and the floodplain. It can perform good measurements at high water when the major bed of the river is flooded but cannot lock onto the minor bed of the river at low water, and the measured range is either that of the dried floodplain or contaminated by the margins. It is noteworthy that these results are much less optimistic than previous studies conducted in the Amazon basin: For example Birkett et al. (2002), who found for T/P mission a mean RMS of 1.1 m when comparing with about 50 gauge stations in the Amazon basin. In the present study, in situ gauges are referenced to sea level. In Birkett et al. (2002), the water stage were referenced to the gauge zero. Therefore the comparison between the altimetry series and the in situ series were performed by matching the mean of both series. This relative leveling considerably reduced the deviation. Particularly, the cases shown Fig. 2A (the most frequent one) should demonstrate lesser RMS if relatively leveled. It is interesting also to recall that the errors were reported in various precursor studies to be higher for low water than for high water, which is not surprising when in most cases, T/P is measuring the whole dried floodplain instead of the river stage. In comparison with these poor results of T/P, we can enhance the improvements made with the Jason-2 mission. It has been possible to extract time series for the 72 virtual stations (blue dots Fig. 1) with Jason-2, including upstream situations on the Amazonas of Peru, or narrow Napo in Ecuador, or upper Rio Negro in Brazil, also for minor tributaries of the Amazon like Rio Icßa. Mean RMS is 0.31 m. If a large part of that improvement can be expressed in terms of a better resolution of the altimeter, these are very promising results for the next generation of nadir altimeter like SARAL/AltiKa. Table 2 Absolute bias calculated for Jason-2 mission with the two retracking algorithms ICE1 and ICE3. Jason-2 track Sub-basin Upstream in situ gauge Downstream in situ gauge Distance upstream (km) 139 152 228 063 Amazonas Amazonas Amazonas Madeira Itacoatiara Jatuarana Parintins Manicore 249 72 32 102 152 254 076 241 254 063 076 089 165 178 254 Madeira Madeira Negro Negro Negro Solimoes Solimoes Solimoes Solimoes Solimoes Solimoes Borba Porto Velho Barcelos Barcelos Curicuriari Codajas Codajas Sao Paulo de Olivencßa Fonte Boa Sao Paulo de Olivencßa Fonte Boa Parintins Iracema Obidos Fazenda Vista Alegre Urucurituba Humaita Moura Moura Tapuraquara Anama Anama Fonte Boa Tefe Fonte Boa Tefe 132 163 28 103 128 53 53 28 95 169 237 Total distance (km) Ice1 bias (m) StDev (m) Ice3 bias (m) StDev (m) 18 30 143 98 267 102 175 200 0.369 0.669 0.236 0.475 0.036 0.020 0.023 0.036 0.176 0.716 0.203 0.330 0.071 0.076 0.021 0.044 6 80 134 61 100 47 47 423 163 288 17 138 243 162 164 228 100 100 451 258 457 254 0.615 0.122 0.430 0.646 0.081 0.463 0.240 0.512 0.410 0.456 0.278 0.099 0.030 0.045 0.038 0.019 0.043 0.042 0.056 0.034 0.057 0.047 0.539 0.148 0.406 0.649 0.133 0.306 0.268 0.748 0.398 0.404 0.309 0.088 0.042 0.047 0.036 0.029 0.042 0.041 0.068 0.041 0.044 0.044 Distance downstream (km) 1548 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 Nevertheless, the first objective of the study, which was to determine if a continuity of monitoring could be obtained by the T/P to Jason-2 family, is not reached, or at least only partially. Only 18 virtual stations can be reasonably used for a long term monitoring, with suspicious results at low water for T/P. 3.2. Bias calculation Tables 1 and 2 report the bias calculated from the difference between Jason-2 stations and in situ stations leveled by GPS. In Table 1, results are obtained by direct comparison, as the track was located in the immediate vicinity of the in situ gauge for three cases. In Table 2, the results were obtained following the method described in chapter 2. This method is based on a strong hypothesis: It supposes that the slope between the three stations (two in situ and one virtual) may vary during the hydrological cycle but that it remains the same all along the river reach considered. Bias of Jason-2 when calculated by comparison with one close by in situ station is in average of 0.53 ± 0.03 m for ICE1, and 0.498 m ± 0.028 cm for ICE3. When compared with two in situ gauges leveled by GPS, mean values of bias are 0.244 ± 0.041 m for ICE1 and 0.191 ± 0.048 m for ICE3. These values could seem closer to the absolute bias calculated by other authors in different environments (lakes and ocean) than the values presented above. But these bias of Jason-2 are positive in 11 of the 15 cases and negative in 4 cases. They are comprised between 0.51 and +0.69 m. As negative bias occurs only in cases where there is a large distance between stations, it appears that the hypothesis of constant slope could not be satisfied when the distance is great. When removing the cases for a total distance greater than 200 km, mean bias for ICE1 is 0.356 ± 0.047 cm and 0.352 m ± 0.044 cm for ICE3, with no negative values. The two sets of results appear to be far from the results found by other authors for ocean and lakes: 0.162 m for lake Issyk Kul (Crétaux et al., 2011), 0.173–0.171 m found by Mertikas et al. (2011) for descending and ascending passes over ocean, 0.148 m over the great lakes (Cheng et al., 2010), and 0.15 m over Corsica (Bonnefond et al., 2010). As there is no evident reason why the absolute bias should be significantly different over rivers, it appears from these results that there does not exist any way of determining the potential error of altimetric stage when compared with in situ gauged water level. Even when the gauging station is located nearby the altimetric track (less than 10 km), the hypothesis of constant slope seems to be irrelevant. 3.3. Comparison between ICE1 and ICE3 Table 3 reports the relative bias between the two retracking algorithms, the sampling rate for each one (percent of calculated ranges in reference to the total number of measurements), and the RMS for the two trackers, for the Table 3 Relative bias (Db in m) between Ice1 and Ice3 water level and sampling rate (Tx) by direct comparison of retracking algorithm for each track. 1–228 Amazonas 2–139 Amazonas 3–050 Amazonas 4–152 Amazonas 5–063 Amazonas 6–076 Solimoes 7–241 Solimoes 8–254 Solimoes 9–165 Solimoes 10- Solimoes 11- Solimoes 12- Amazonas 14–191 Maranon 15–204 Maranon 16–102 Javari 18–026 Javari 19–178 Jutai 20–178 Jutai 21–089 Jutai 22–102 Jutai 23–178 Jurua 24–102 Jurua 25–089 Jurua 26–026 Jurua 27–076 Purus 29–165 Purus 30–102 Purus 31–026 Purus 34–152 Madeira 35–063 Madeira 37–063 Madeira 38–076 Madeira 39–254 Madeira 40–241 Madeira 41–241 Madeira 42–241 Madeira 43–050 Xingu 44–050 Xingu 45–050 Xingu 46–050 Iriri 47–089 Negro 48–089 Negro 50–254 Negro 51–165 Negro 52–013 Negro 53–178 Negro 54–063 Negro 55–076 Negro 56–241 Negro 57 178 Uaupes 58 102 Ica 59 013 Ica _ 60 191 Ica 61 026 Ica 62 115 Ica 63 089 Ica 64 026 Napo 65 165 Japura 66 089 Japura 67 178 Japura 69 102 Japura 70 191 Japura 71 026 Japura 72 115 Japura Db (ice1–ice3) m tx (ice1) tx (ice3) 0.116 0.190 0.257 0.122 0.008 0.065 0.050 0.037 0.058 0.116 0.036 0.083 0.131 0.145 0.350 0.007 0.049 0.005 0.061 0.096 0.292 0.059 0.254 0.134 0.122 0.011 0.004 0.154 0.106 0.153 0.075 0.106 0.108 0.036 0.002 0.009 0.226 0.208 0.169 0.078 0.067 0.042 0.055 0.110 0.047 0.016 0.118 0.079 0.047 0.055 0.46 0.061 0.029 0.041 0.056 0.090 0.047 0.065 0.072 0.089 0.046 0.124 0.097 0.082 98 96 98 99 96 100 97 97 97 96 97 95 94 94 97 96 100 100 100 100 100 99 100 100 99 100 95 96 97 98 100 100 98 98 95 91 93 94 79 93 98 100 92 94 93 100 94 97 96 99 94 99 91 92 91 95 95 100 100 96 98 79 96 92 96 92 97 96 88 96 91 87 81 86 89 89 93 93 94 92 98 98 94 96 98 97 96 98 97 94 92 94 94 98 95 98 100 96 95 92 94 98 77 92 92 94 92 93 93 97 88 96 94 98 89 95 88 82 82 91 93 97 96 88 97 78 94 94 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 72 Jason-2 virtual stations. Relative bias is varying from 0.192 to +0.257 m, with a mean value of +0.051 m. As the mean RMS calculated from the comparison between the two trackers is 0.29 m (min = 0.082, max = 1.408), we cannot assume a systematic bias between the two tracking algorithm. By comparing the bias in relation to the in situ stations (Table 2), there is no relevant difference between the two trackers either. The bias calculated for each tracker is always of same sign, and very close from each other. Mean difference is + 0.054 m (minimum 0.052, maximum difference 0.193). As for the sampling rate, it is better for ICE3 than for ICE1 in three cases, equal in four cases, and better for ICE1 than for ICE3 in the 65 other cases. 4. Conclusion Jason-2 altimeter gives far better results when estimating the river water level than T/P. A total of 72 virtual stations have been extracted within the Amazon basin for Jason-2, with a mean RMS of 0.31 m, for rivers varying from several kilometers wide to 400 m. With the same methodology of extraction, same geophysical corrections, and the same tracker (ICE1), only 18 virtual stations have a RMS less than ±1.50 m for T/P. Analyzing from various examples the reasons for such poor results for T/P, it seems that the results are comparable with those of Jason-2 in the case of large rivers without floodplain. There is only a greater number of outliers with T/P than with Jason-2 in these cases. For all the other situations, T/P captures only the ranges for high flow. For low flow the values of the range are either contaminated by the surroundings of the river, or lost. To our knowledge, our study is the first one estimating an error for T/P by comparing with water stage obtained at leveled in situ stations. This could explain that better results for T/P have been found previously when comparing the series only relatively. These results show that unfortunately, there is no continuity to be expected for T/P to Jason-2 missions for continental rivers, with some rare exceptions. Nevertheless, the good results of Jason-2 that can be attributed to a better resolution of the altimeter seem promising for the next generation of nadir altimeter like SARAL/AltiKa. As far as we know, absolute bias has never been calculated for Jason-2 altimetry missions applied to river level. In this study, comparing 15 virtual stations obtained with Jason-2 with time series at the in situ gauges closest upstream and downstream from the virtual station, we found a mean bias of 0.244 ± 0.041 m for ICE1 and 0.191 ± 0.048 m for ICE3 algorithm. In both cases, the distribution of biases is characterized by a large spread around the mean value. Such a result was also observed by Calmant et al. (2013) for the Envisat ranges also retracked with the ICE1 algorithm. Noteworthy, these spreads around the mean values of biases are far above the potential error sources usually invoked in altimetry calibration over oceans or even over lakes, such as orbit or pulse propagation errors. We propose that these great dis- 1549 persions reflect the fact that the slope between the virtual and in situ stations is not only varying with time but also along the course of the river. Therefore, it is not only impossible to estimate an absolute bias for Jason-2 altimeter in the case of rivers, but also impossible to estimate an error of the altimeter range by comparing with one or several in situ gauging stations, even in the closest cases (less than 10 km apart). Nevertheless, it is worth noting that these results give a fair estimation of the range of error that can be expected when measuring river stage by altimetry. There is a remarkable convergence around the value of 0.35 m when estimating bias for Jason-2. It seems then reasonable to apply a bias of 0.15 m (calculated either for ocean or lakes) to Jason-2 data in future studies, estimating a potential error of ±0.35 m, when no other validation is available. And this could be always the case, even when in situ data is available, unless permanent dispositive of validation can be designed, installed and maintained at the exact location of the satellite tracks. ICE3 algorithm seems to have no better results than ICE1 when estimating the range for Jason-2. Further research has to be devoted to find tracker better adapted to continental water. Despite the complexity of continental waters, altimetry reveals being an invaluable tool for monitoring water resources in ungauged or poorly gauged basins. Great expectations is placed in SARAL/AltiKa mission and in the continuation of Jason-2 mission, in order to maintain the continuity of the great number of virtual station already functioning in the great tropical basins and to ensure the needed complementarities between the different missions. Acknowledgments CASAM and IHESA Projects supported this research. GPS field work was funded by CPRM and IRD (Dinâmica Fluvial do Sistema Amazonas-Solimões) and by CNES (TOSCA project FOAM). The authors thank the students involved in the Research Initiation Program of RHASA, Research Team at Amazonas State University, Brazil, who produced the Jason 2 time series. The Ohio State University (OSU) component of the research is partially supported by NASA, and by the OSU Climate, Water, and Carbon (CWC) Program. References Altamimi, Z., Collilieux, X., Legrand, J., et al. ITRF2005: a new release of the international terrestrial reference frame based on time series of 454 station position and Earth orientation parameters. J. Geophys. Res. 112, http://dx.doi.org/10.1029/2007JB004949, 2007. Arnault, S., Pujol, I., Melice, J.L. In situ validation of Jason-1 and Jason-2 altimetry missions in the Tropical Atlantic Ocean. Marine Geodesy 34 (3–4), 319–339, http://dx.doi.org/10.1080/01490419.2011.584833, 2011. Bamber, J.L. Ice sheet altimeter processing scheme. Int. J. Remote Sens. 15 (4), 925–938, http://dx.doi.org/10.1080/01431169408954125, 1994. Berry, P., Galick, J.D., Freeman, J.A., Mathers, E.L. Global inland water monitoring from multi-mission altimetry. Geophys. Res. Lett. 32, L16401, http://dx.doi.org/10.1029/2005GL022814, 2005. 1550 F. Seyler et al. / Advances in Space Research 51 (2013) 1542–1550 Bertiger, W.I., Bar-Sever, Y.E., Christensen, E.J., et al. GPS precise tracking of TOPEX/POSEIDON: results and implications. J. Geophys. Res. 99 (C12), 24449, http://dx.doi.org/10.1029/94JC01171, 1994. Bertiger, W.I., Desai, S.D., Haines, B., et al. Single receiver phase ambiguity resolution with GPS data. J. Geod. 84 (5), 327–337, http:// dx.doi.org/10.1007/s00190-010-0371-9, 2010. Birkett, C.M. The global remote sensing of lakes, wetlands and rivers for hydrological and climate research. In: Geoscience and Remote Sensing Symposium, 1995 (IGARSS ‘95), Quantitative Remote Sensing for Science and Applications, vol. 3, pp. 1979–1981, 1995. Birkett, C.M. The contribution of T/P to the global monitoring of climatically sensitive lakes. J. Geophys. Res. 100 (C12), 25179–25204, 1995b. Birkett, C.M. Contribution of the TOPEX NASA radar altimeter to the global monitoring of large rivers and wetlands. Water Resour. Res. 34, 1223–1239, 1998. Birkett, C.M., Mertes, L.A.K., Dunne, T., et al. Surface water dynamics in the Amazon basin: application of satellite radar altimetry. J. Geophys. Res. 107 (D20), 8059, http://dx.doi.org/10.1029/2001JD000609, 2002. Birkett, C.M., Beckley, B. Investigating the performance of the Jason-2/ OSTM radar altimeter over lakes and reservoirs. Marine Geodesy 33 (S1), 204–238, http://dx.doi.org/10.1080/01490419.2010.488983, 2010. Bonnefond, P., Exertier, P., Laurain, O., Jan, G. Absolute calibration of Jason-1 and Jason-2 altimeters in corsica during the formation flight phase. Marine Geodesy 33, 80–90, http://dx.doi.org/10.1080/ 01490419.2010.487790, 2010. Calmant, S., Santos da Silva, J., Moreira, D., et al. Detection of Envisat RA2/ICE1 retracked altimetry bias over the Amazon basin rivers using GPS, 51 (8), 1551–1564, 2013. Cheng, K-H., Kuo, C.K., Tseng, H-Z., et al. Lake surface height calibration of Jason-1 and Jason-2 over the great lakes. Marine Geodesy 33, 186–203, http://dx.doi.org/10.1080/01490419.2010. 487802, 2010. Cretaux, J.-F., Calmant, S., Romanovski, V., et al. An absolute calibration site for radar altimers in the continental domain: lake Issyk-kul in Central Asia. J. Geod. 83, 723–735, 10.100/ s00190.008.0289.7, 2009. Crétaux, J.-F., Calmant, S., Romanovski, V., et al. Absolute calibration of Jason radar altimeters from GPS kinematic campaigns over lake Issykkul. Marine Geodesy. 34 (3–4), 291–318, http://dx.doi.org/ 10.1080/01490419.2011.585110, 2011. Dettmering, D., Bosch, W. Global calibration of Jason-2 by multi-mission crossover analysis. Marine Geodesy 33, 150–161, http://dx.doi.org/ 10.1080/01490419.2010.490419, 2010. Dumont, J.P., Rosmorduc, V., Picot, N., et al., OSTM/Jason-2 Products Handbook, 2008. Frappart, F., Seyler, F., Martinez, J.-M., et al. Floodplain water storage in the Negro river basin estimated from microwave remote sensing of inundation area and water levels. Remote Sens. Environ. 99, 387–399, 2005. Frappart, F., Calmant, S., Cauhopé, M., et al. Preliminary results of Envisat RA-2-derived water levels validation over the Amazon basin. Remote Sens. Environ. 100, 252–264, http://dx.doi.org/10.1016/ j.rse.2005.10.027, 2006. Fu, L., Christensen, E.J., Lefebvre, M. TOPEX /POSEIDON: the ocean topography experiment. Eos. Trans. Am. Geophys. Union. 72 (35), 1991. Haines, B., Desai, S.D., Born, G.H. The harvest experiment: calibration of the climate data record from TOPEX/Poséidon, Jason-1 and the OSTM. Marine Geodesy 33, 91–113, http://dx.doi.org/10.1080/ 01490419.2010.491028, 2010. Leon, J.G., Calmant, S., Seyler, F., et al. Rating curves and average water depth estimation at the upper Rio Negro from altimetric spatial data and calculated remote discharges. J. Hydrol. 328, 481–496, 2006. Loyer, S., Perosanz, F., Mercier, F., et al. Zero-difference GPS ambiguity resolution at CNES–CLS IGS analysis center. J. Geod., http:// dx.doi.org/10.1007/s00190-012-0559-2, 2012. Marty, J.C. Documentation algorithmique du programme GINS, version 5 juillet 2009, 563 available at <http://www.igsac.cnes.fr/documents/ gins/GINS_Doc_Algo.html> (in French), 564, 2009. Mercier, F. Satellite altimetry over non ocean areas: an improved wet tropospheric correction from meteorological models. In: Joint EGS– AGU meeting, Nice, 2003. Mercier, F., Picot, N., Lombard, A., et al. Improved Jason-2 altimetry products for coastal zones and continental waters (PISTACH Project). Eos Trans. AGU, 89 (53), Fall Meet. Suppl., Abstract G31B–0662, 2007. Mertikas, S.P., Ioannides, R., Tziavos, I., et al. Statistical model and latest results in the determination of the absolute bias for the radar altimeters of Jason satellites using the gavdos facility. Marine Geodesy 33, 114– 119, http://dx.doi.org/10.1080/01490419.2010.488973, 2010. Mertikas, S.P., Daskalakis, A., Tziavos, I.N., et al. Ascending and descending passes for the determination of the altimeter bias of Jason satellites using the Gavdos facility. Marine Geodesy 34 (3–4), 261–276, 2011. Morris, C.S., Gill, S.K. Evaluation of the T/P altimeter system over the great lakes. J. Geophys. Res. 99 (C12), 24527–24539, 1994. Pavlis, N.K., Holmes, S.A., Kenyon, S.C., Factor, J.K. The development and evaluation of the Earth gravitational model 2008 (EGM2008). J. Geophys. Res. 117, B04406, http://dx.doi.org/10.1029/2011JB008916, 2012. Ponchaut, F., Cazenave, A. Continental lake level variations from T/P (1993–1996). C. R. Acad. Sci. Paris, Ser. IIa 326, 13–20, 1998. Quartly, G.D. Jason-1/Jason-2 metocean comparisons and monitoring. Marine Geodesy 33 (Special issue: OSTM/Jason-2 calibration/validation), Supplement 1, 2010. Roux, E., Cauhopé, M., Bonnet, M-P., et al. Daily water stage estimated from satellite altimetric data for large river basin monitoring. Hydrol. Sci. J. 53 (1), 2008. Silva, J.S., Calmant, S., Rotuno Filho, O.C., et al. Water levels in the Amazon basin derived from the ERS-2 and Envisat radar altimetry missions. Remote Sens. Environ. 114, 2160–2181, http://dx.doi.org/ 10.1016/j.rse.2010.04.020, 2010. Washburn, S.A., Haines, B.J., Born, G.H., Fowler, C. The harvest experiment LIDAR system: water level measurement device comparison for Jason-1 and Jason-2/OSTM calibration. Marine Geodesy 34 (3–4), 277–290, 2011. Watson, C., White, N., Church, J., et al. Absolute calibration in bass strait, Australia: TOPEX, Jason-1 and OSTM/Jason-2. Marine Geodesy 34, 242–260, 2011.