Backscatter overs ocean 30/09/2017 Radar altimetry backscattering signatures at Ka, Ku, C and S bands over the ocean Blarel F.*a, Frappart F.a,b; Blumstein D.a,c, Birol F. a, Morrow R. a, Nino F. a (a)- Laboratoire d’Etudes en Géophysique et Océanographie Spatiales (LEGOS-GRGS), UMR 5566, CNES/CNRS/IRD/UPS, Observatoire Midi-Pyrénées, 14 Avenue Edouard Belin, 31400 Toulouse, France, (b)- Géosciences Environnement Toulouse (GET-GRGS), UMR 5563, CNRS/IRD/UPS, Observatoire Midi-Pyrénées, 14 Avenue Edouard Belin, 31400 Toulouse, France, (c)- Centre National d’Etudes Spatiales (CNES), 18 Avenue Edouard Belin, 31400 Toulouse, France, Introduction Datasets Method In this study, we present global maps of along-track average backscattering coefficients (from ocean retracker) over ocean and associated standard deviation at : Ku band using ERS-2, ENVISAT, and Jason-2 data S band using ENVISAT data C band using Jason-2 data Ka band using SARAL data The resulting maps have been obtained from a original method developed to compute any statistics along the pass. It is well known that near-surface wind speed is a first-order geophysical parameter that can be retrieved from backscatter observations. But the sea surface roughness also often deviates from simple local wind forcing behavior. The final objective of this study is to gain some insight on the ocean processes other than wind and wave that may produce sea state perturbations observed by altimeter backscatter measurements (i.e. surface currents, eddies, ...). Global statistics will be used to identify regional study cases for field studies that will be performed in a second time. Along-track altimetry data were first sorted using a normalized pass [3] in order to merge the data measurements into cells regularly organized along the pathway. The mean pass is then defined by four parameters: the mean longitude and latitude of each cell composing the mean pass, and its size given by its dimensions along (δAlong-track) and cross track (δCross-track). If δCross-track is chosen arbitrarily (equals two kilometers in this study as this distance corresponds to the maximum cross-track variations along the orbit of ERS-2, ENVISAT, SARAL and Jason-2 missions), δAlong-track is given by : δAlong-track = Vsat δt where Vsat is the velocity of the satellite along the orbit equals to 7.45 km.s-1 and δt is chosen equals to 1 second, that mean that every cell contains, each cycle a maximum of 20 measurements. These parameters of normalized pass is integrated into the GDR products of the CTOH data base. ERS-2: The data product REAPER release []. Its data are available from May 1995 to June 2003 (cycle 1 to 84). ENVISAT v2.1 : The data product REAPER release [1]. Its orbit characteristics are the same as for ERS-2. Its data are available from May 2002 to October 2010 (cycle 7 to 93). JASON-2 mission: Its orbit is different of the above missions. Its data used are from August 2008 to June 2016 (cycle 0 to 288). SARAL: The data product CNES release [2]. Placed in the same orbit as the ENVISAT satellite. Its data are available from March 2013 to June 2016 (cycle 1 to 33). Mean pass Mean altimetry pass (black arrow)obtained as the average of all altimetry data acquired during the life time of an altimetry mission. Altimetry measurements are shown using dots, each color representing a cycle. Rectangles represent the cells along the altimetry pass defined by δAlong-track and δCross-track for a given location (λcell, φcell) (yellow and black star) 1 (λcell, φcell) Cell along average track. Table of mean and standard derivation of the backscatter coefficient over ocean (at 200 km from the coast and between 60 degrees latitude North and South) for each mission and band. Mission Cycles Band Mean (dB) Std (dB) ers2.r 1 to 84 ku 10.36 0.84 envisat.v21 7 to 93 11.09 1.80 7 to 64 s 11.26 1.57 s-ku -3.61 4.08 jason2d 0 to 288 13.46 0.66 c 15.29 0.64 c-ku 10.58 0.41 saral.s 0 to 33 ka 11.10 1.23 Altimetry data in one cell for several cycles δAzimuth=Vsat*δt δCross-track ENVISAT.v21 Ku band (13.575 GHz) S band (3.2 GHz) S-Ku Band difference ERS-2 reaper Ku Band (13.8 GHz) Fig. 1 Fig. 2 Fig. 3 Fig. 7 Bs (dB) Bs (dB) Bs (dB) Bs (dB) Fig. 4 Fig. 5 Fig. 6 Fig. 8 Std Bs (dB) Std Bs (dB) Std Bs (dB) Std Bs (dB) JASON 2 Ku band (13.575 GHz) C band (5.4 GHz) C-Ku Band difference Backscatter overs ocean Fig. 9 Fig. 10 Fig. 11 The altimetrer Backscatter responses at Ku- (ERS-2 ENVISAT, Jason2d) , S- (Envisat), Ka- (Saral) and C- bands (Jason2d) exhibit a wide range sea estate variations over the oceans. Whatever the band, the maps of mean backscatter (Fig. 1, 2, 3, 4, 9, 10, 11 and 15) shows the same ocean's pattern. Differences appear only on the dynamic of the backscatter signature for Saral (Fig. 15). In order to compare the dynamic of the mean backscatter parameter of each band, the map scale have been centered on the mean backscatter (see above table) using a range amplitude of 4 dB. Saral's map (Fig.15) shows the mean backscatter signature at Ka-band the most contrasted with large difference between the maximum and the minimum and shows also wider areas with high backscatter value (>13.0 dB). This effect should be due to the higher sensitivity of the Ka-band to several atmospheric factors like liquid water and water vapor []. All the mean maps (Fig. 1, 2, 3, 4, 9, 10, 11 and 15) show some well known areas having high mean backscatter: a close to the equator in the East and West Pacific ocean where el nino phenomenon takes place and also in the Indian see. Other areas having the lower mean backscatter: in the Austral Ocean and the North Atlantic. Between these areas, there are areas with medium value located on the main oceanic gyre place. The spatial pattern of mean backscatter in the maps appears strongly related to the wind stress (Fig.17) and its effect on the sea surface. The areas having high mean value of backscatter are areas where the the wind stress is low. While the areas having low mean values are areas where the the wind stress is stronger. The maps of standard derivation of the backscatter (above the maps of the mean backscatter: Fig. 4, 5, 6, 8, 12, 13, 14 and 16) show roughly the same pattern areas as the mean. The main oceanic gyre are also observed form each side of the equator, within which the StD values are lower. Bs (dB) Bs (dB) Bs (dB) Fig. 12 Fig. 13 Fig. 14 Std Bs (dB) Std Bs (dB) Std Bs (dB) Saral Ka Band (35.75 GHz) Fig. 17: Annual mean wind stress on the ocean. A contour of 1 represents a wind-stress of magnitude 0.1 Nm−2. Stresses reach values of 0.1 to 0.2 Nm−2 under the middle-latitude westerlies, and are particularly strong in the southern hemisphere. The arrow is a vector of length 0.1 Nm−2. Marschall et al [5]. Fig. 17 Fig. 15 Bs (dB) Fig. 16 Results & Conclusion This first result enable to validated the normalized pass method at 1Hz. Checks that the observation of the backscattered energy over ocean enable to find again the spatial distribution of the wind stress like the first order signal. For the further, we plan: To look regional areas like Mediterranean sea and used normalized pass at higher frequency (20 Hz, 40Hz). To observe the second order signal of backscatter and to find the related geophysical parameter. Std Bs (dB) References Mean Backscatter (dB) over Mediterranean Sea for Jason-2 [1]-ESA.2011.https://earth.esa.int/documents/10174/1912962/ra2-mwr.ProductHandbook.2_2.pdf [2]- CNES. 2013.http://www.aviso.altimetry.fr/fileadmin/documents/data/tools/SARAL_Altika_products_handbook.pdf [3]- Blarel et al. 2015. CTOH Technical report: ”Normalized Altimetric Pass”. http://ctoh.legos.obs-mip.fr [4]- Saleh Abdalla, Calibration of SARAL/AltiKa Wind Speed, IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, VOL. 11, NO. 6, JUNE 2014 [5]- Marschall, J., Plumb, R.A.: Atmosphere, Ocean, and Climate Dynamics. Elsevier Academic Press, USA (2008) Ocean Surface Topography Science Team, November 2016, La Rochelle, France