Coastal Altimetry Data Handbook Issue 2.0, 02 September 2014

Transcription

1 Coastal Altimetry Data Handbook Issue 2.0, 02 September 2014 Edited by the Coastal & Marine Research Centre, University College Cork

2 DOCUMENT DETAILS Author Paolo Cipollini, Rory Scarrott, Helen Snaith Document Title Product Data Handbook: Coastal Altimetry Document Reference D180G_HB_SL1 Product Reference SL1, NRT9 Issue 2.0 Date of Issue 02 September 2014 CHANGE RECORD Version Date Change Description Author First draft HS, PC Second Draft PC, RS Feedback sought from ESA RS Final edits PC, RS Version 1.0 released RS Version 1.0 reviewed. Version 1.1 compiled for editing Partial update PC Version 1.2 reviewed. Version 1.3 compiled for editing Final update PC Version 2.0 released PC, RS RS RS Front cover image credit: ESA, NASA, ABC Action News, COASTALT 2

3 CONTENTS Summary... 6 Abbreviations and Acronyms THE COASTAL ALTIMETRY DATA PRODUCT Introduction Satellite Altimetry - the principles behind the data Issues Concerning Altimetry in the Coastal Zone Need for improved re-tracking Need for optimised correction Processing satellite altimetry observations into coastal altimetry data Altimeter Missions included in esurge Models used for re-tracking and corrections esurge specific models esurge retrackers Product Validation Product Quality Flags The Near Real Time Coastal Altimetry product Input data to the NRT Coastal Altimetry processor Processing details for NRT Coastal Altimetry Availability of NRT Coastal Altimetry on the esurge server Potential Uses of Altimetry Data in Storm Surge Applications PRODUCT DETAILS Technical Description Accessing the Product Viewing the product in the esurge website Download the product Accessing the product outside esurge Using the Product Overview Organisation of the Product Typical computation from altimetry data Corrected Altimeter Range Wet Troposphere Correction Dry Troposphere Correction

4 2.3.7 Ionosphere Correction Sea State Bias Correction Sea Surface Height and Sea Level Anomaly Tide Effects Total Water Level Envelope (TWLE) Surface Air Pressure Effects Geophysical Surface - Mean Sea Surface or Geoid Flags and other quality control variables Mean Sea Surface and Adjustment of the Cross Track Gradient Smoothing Ionosphere Correction Generation of 1 Hz Averages Constraints on Use FURTHER INFORMATION AND CONTACTS REFERENCES AND FURTHER READING Applicable Technical Documentation Relevant esurge Project Documents ANNEX A: SATELLITE ALTIMETRY FACTS & CONVENTIONS A.1 Distance Conventions A.2 Orbits, revolutions, passes and repeat cycles A.3 Reference ellipsoid A.4 Correction conventions A.5 Time convention A.6 Unit convention A.7 Flagging and editing ANNEX B: FULL NetCDF DESCRIPTION OF VARIABLES IN THE CGDRs B.1 Envisat CGDR B.2 CryoSat-2 LRM CGDR B.3 CryoSat-2 SAR CGDR ANNEX C: THE ESURGE PROJECT C.1 About esurge C.2 The esurge Consortium

5 List of Tables and Figures Table 0-1 Table 1-1 Table 1-2 Table 1-3 Table 1-4 Table 1-5 Table 2-1 Table 2-2 Table 2-3 Table 2-4 Table 2-5 Table 2-6 Table 2-7 Table 2-8 Table A-0-1 Figure 1-1 Figure 1-2 Figure 1-3 Figure 1-4 Figure 1-5 Figure 1.6 Figure 1-7 Figure 2-1 Figure 2-2 Figure 2-3 Figure 2-4 Figure 2.5 Figure 2-6 Figure 2-7 Figure A-0-1 Acronyms and abbreviations Models and standards for Envisat Models and standards for CryoSat-2 LRM Models and standards for CryoSat-2 SAR Results of the internal validation for range Results of the internal validation for SWH Coastal altimetry product technical specifications NRT Coastal altimetry product technical specifications Available range values in the CGDR. Available wet tropospheric correction values in all CGDR products. Additional wet tropospheric correction values in a limited number of enhanced CGDR products. Available ionospheric correction values in the CGDR Available ocean tide correction values in the CGDR Available SSH, SSHA and TWLE variables in the CGDR Envisat 35-day phase orbit parameters. Altimetric distances relationship between altitude, range and height. An Illustration of waveform corruption when the altimeter approaches the coast. Schematic of the coastal altimetry processor for the ESA esurge Project Delay/Doppler (SAR) Altimeter footprint Envisat passes over the North Indian Ocean when the area was affected by cyclone SIDR Example and validation of SSH for ENVISAT Example and validation of SWH for ENVISAT Accessing the data on the esurge website The data access page of the esurge website The Cyclone Sidr dataset inventory on the esurge web-service Previewing the Cyclone Sidr altimetry dataset using the Preview link Viewing dataset metadata and information using OPeNDAP Exploring Cyclone Sidr datasets in the esurge GIS web-viewer Downloading data using the OPeNDAP dataset information page Example of inclined orbit and impact of the inclination drift 5

6 Summary Satellite altimetry is a powerful earth observation technique that, through the analysis of radar echoes from the sea surface, allows the estimation of sea surface height, significant wave height and surface wind speed. It is arguably the most successful technique to explore ocean dynamics and is proving to be invaluable in measuring and monitoring global sea levels. Until recently, however, altimeter data near the coast were usually discarded as being inaccurate or difficult to interpret. Through reprocessing the radar echoes and improving some of the corrections that need to be applied to the altimetric measurements, meaningful measurements can be retrieved in the coastal strip (approximately 0-50 km from the coast with extent varying dependant on local conditions). This need for reprocessing underpins the rationale for considering this as a new branch of altimetry simply referred to as coastal altimetry. Coastal altimetry has a prominent role to play in storm surge research, as it directly measures the Total Water Level Envelope (TWLE), i.e. the sea level taking into account tides, waves, effects of a storm surge and impact of river discharges, and one of the key quantities required by storm surge applications and services. However, coastal altimetry can also provide important information on the wave field in the coastal strip, aiding the development of more realistic wave models which in turn, can be used to improve the forecast of wave setup and overtopping processes. This handbook describes the delayed-time coastal altimetry products (Coastal Geophysical Data Records or CGDR containing, amongst other quantities, TWLE and significant wave height) developed for the ESA Data User Element esurge Project. It covers data from different altimetry missions (Envisat, Jason-1, Jason-2, and CryoSat). This guide describes the fundamentals of how sealevel height and TWLE measurements are extracted from satellite altimetry echoes, and how such processing must be adapted to provide coastal altimetry data pertaining to sea level heights and TWLE in the coastal zone. The guide also outlines how the multi-mission coastal altimetry processor is integrated into the esurge data access and delivery system, clearly outlines information such as product technical and theoretical specifications, essential background information, and guidance for further reading. The theoretical knowledge outlined here is illustrated in select cases with examples of how coastal altimetry data has provided further information on selected significant surge events. Please note that the esurge also provides access to Near-Real-Time (NRT) coastal altimetry products, which are also described in this handbook These products and associated handbook were compiled for the esurge Project, funded by the European Space Agency under the ESA Data User Element (DUE) Programme.. For more information on the esurge project go to 6

7 Abbreviations and Acronyms Table 0-1 lists the acronyms and abbreviations used within this document. Table 0-1: Acronyms and abbreviations. Acronym AOI BGP BOR CGDR CMRC COASTALT DMI DUE ECMWF EO ENVISAT ESA GDR GIS GNSS GPD KNMI MERIS MODIS NETCDF NOAA NOC OSTST Meaning Area of Interest Brown plus Gaussian Peak Brown Ocean Retracker Coastal Geophysical Data Record Coastal and Marine Research Centre ESA Development of COASTal ALTimetry for Envisat Danmarks Meteorologiske Institut (Danish Meteorological Institute). Data User Element European Centre for Mid-Range Weather Forecast Earth Observation Environmental Satellite European Space Agency Geophysical Data Record Geographical Information Systems Global Navigation Satellite System GNSS-derived Path Delay Koninklijk Nederlands Meterologisch Instituut (Royal Netherlands Meteorological Institute) Medium Resolution Imaging Spectrometer Moderate Resolution Imaging Spectroradiometer Network Common Data Format National Oceanographic and Atmospheric Administration National Oceanography Centre Ocean Surface Topography Science Team 7

8 RADS RMS SAR SEARS SEV SGDR SPOT SSB SSH SSHA SWH TWLE UCC USO Radar Altimeter Database System Root Mean Square Synthetic Aperture Radar Surge Event Analysis and Repository Service Storm surge Event Sensor Geophysical Data Record Satellite Pour l'observation de la Terre Sea State Bias Sea Surface Height Sea Surface Height Anomaly Significant Wave Height Total Water Level Envelope University College Cork Ultra-Stable Oscillator WGS1984 World Geodetic System

9 1. THE COASTAL ALTIMETRY DATA PRODUCT 1.1 Introduction Altimeters allow us to measure the height of the sea surface. To an observer looking from shore, the ocean will look flat, except of course for two very familiar phenomena: the surface roughness due to wind (i.e. the short wavelets with horizontal scales of a few cm across), and the wave field (i.e. the troughs and crests with horizontal scales of meters to tens of meters). However, if the same observer could look at the ocean over distances of tens or hundreds of km from the vantage point of satellites, he or she would see variations ( bumps and troughs ) in height of several meters these being the variations which a satellite altimeter measures. A number of factors shape the nature of these bumps and troughs. If we assume the shape of the Earth is a smooth ellipsoid (i.e. the solid derived from the rotation of an ellipse about one if its axes, fuller at the equator and flattened at the poles), the sea surface deviates from that shape, firstly due to local perturbations in the Earth s gravitational field (in turn due to the nonhomogeneous distribution of mass inside our planet and in its crust). As a first approximation, the sea surface follows a surface of equal (constant) gravity, called the geoid, which would be the shape of the sea surface if there were no currents, winds and tides, and which varies up to 100 meters above or below the reference ellipsoid. Additional to the effects of gravity, there are the effects of moving water, i.e. currents and tides. Major oceanic currents correspond to variations in sea surface of up to a couple of metres over 100 km, while tides can be up to a few metres. Oceanographers are extremely interested in monitoring currents, and this can be done by looking at the altimeter signal, if the geoids signal can be removed (which is becoming increasingly possible with more recent geoid models). And for many applications we don't even need a geoid: it is sufficient to look at the variation of the sea surface over time (i.e. the height anomalies), because the geoid has not changed (it only does over very long time scales). One such application is to monitor the rise of sea level, regionally or globally. Additionally, in most cases the contribution to sea surface height due to tides can also be estimated. Given that we know exactly (from centuries of tidal observations) which periodical signals to seek and model for in the acquired data, repeated altimetric measurements over long time spans can be analysed to achieve this. Satellite altimeters have the additional advantage of potentially capturing tidal information over entire oceans, in contrast to in-situ coastal tide gauges, which are situated along continental or island coastlines. Furthermore, in addition to information on range, which is then converted into sea height and used as explained, altimetric pulses can also convey information on the waves and surface winds field. These phenomena also change the shape and strength of the reflected echoes allowing wave and wind fields to be estimated, studied and monitored. Taking these applications into account, it can be seen that the primary measurement derived from altimeters (sea surface height) is useful in a variety of applications - contributing to geoid characterisation, mapping sea bottom topography, observing oceanic currents and tides, studying and monitoring sea-level rise, whilst also allowing us to study and monitor wind and wave fields. 9

10 1.2 Satellite Altimetry - the principles behind the data Altimetry satellites essentially determine the distance from the satellite to a target surface by sending a radar pulse towards the surface and measuring the time it takes to come back (figure 1-1). If the satellites orbital position is known with a sufficient degree of precision (i.e. its orbital altitude with respect to reference surface, such as the ellipsoid), we can compute the height of the sea surface through difference analysis (for a detailed description of terms such as range, orbit, altitude, see figure 1-1). Orbit Range Altitude Height Ocean Surface Reference Ellipsoid Ocean Bottom Figure 1-1: Altimetric distances relationship between altitude, range and height. In addition to surface height, through looking at the shape and amplitude of the returned waveform, it is also possible to measure wave height and surface wind speed over the oceans surface. While the working principle of altimetry is simple, what makes the measurement complicated is the required precision, a few cm (the typical amplitude of the signal due to mesoscale features in the ocean) from a satellite orbiting at ~1000 km over the surface - a precision greater than one part in ten million. Envisioned as an everyday comparison, it is as if you went to the bakery to buy a standard loaf of bread (about 1 kg) and wanted to know its weight to an accuracy of less than 0.1 mg. This one part in ten million precision needs to be achieved and maintained, whilst the entire altimeter system (i.e. the instrument, but also the processing chain with the various corrections) is required to be stable in time, if we want to estimate long-term sea level rise. Currently, we can measure changes in global sea level with an accuracy 1 of 0.5 mm/y. To obtain the desired accuracy we first need precise knowledge of the satellite's orbital position, achieved through the use of several positioning systems on board. Moreover, we need to correct accurately for the slowing down of the radar pulse due to electrons in the ionosphere ( ionospheric 1 See relevant research carried out in the ESA Sea Level Climate Change initiative

11 correction) and to gases and water vapour in the troposphere ( dry tropospheric and wet tropospheric correction). All these can be corrected either through the use of ancillary measurements or the use of models. Finally, waves are not perfectly sinusoidal, with returns from troughs stronger than those from crests, which can result in range overestimation due to sea state ( sea state bias ). This also needs to be corrected for, normally through the use of empirical models. Altimetry thus requires a lot of factors to be taken into account (usually in the form of corrections, as we have seen, which are summed to the raw range estimate), plus the use of specialised processing, before the data can be deemed usable. A final complication may arise depending on what component of the sea surface height we are interested in for our application: for example, if the signal due to currents is required, then we need to remove the contribution due to tides and to high-frequency atmospheric signals. This can be achieved through the use of models. However, here it is worth noting that for some applications, such as the study of storm surges, correcting for tidal and high-frequency atmospheric signals may not be required, as those signals are an integral part of the Total Water Level Envelope, which is often the main quantity of interest for monitoring and assessing impacts. 1.3 Issues Concerning Altimetry in the Coastal Zone Satellite altimeters have been monitoring the Earth s oceans for more than 20 years, with an excellent degree of accuracy. However, in the coastal strip (approximately 0-50km from the coast), data are generally flagged as being of poor quality for a number of technical reasons, and are therefore rejected. Prompted by the tantalizing prospect of recovering 20 years of data over the coastal ocean, and encouraged by the improved suitability for coastal applications of new and future altimeters (such as those on CryoSat-2, AltiKa and Sentinel-3), an active community of researchers in coastal altimetry has formed, engaged in developing techniques to recover useful measurements of sea level and significant wave height in coastal waters, as well as implementing and promoting new applications (for a detailed account of the entire discipline see Vignudelli et al, 2011, while a summary of the applications is in Cipollini et al, 2010). The larger space agencies are strongly supporting research and development in this newly emerged field through initiatives such as ESA s COASTALT (for Envisat) and CNES PISTACH (for Jason-2) 2, and now through the European Space Agency s DUE esurge Project 3. NASA and CNES are also supporting coastal altimetry through several Ocean Surface Topography Science Team (OSTST) projects. Two classes of problems are encountered when trying to recover meaningful estimates of geophysical parameters (sea level, significant wave height and wind speed) from altimetry data in the coastal zone. Firstly, there are identified issues which arise from the modification of altimetric echoes (waveforms), which occur when land enters the footprint of the instrument (illustrated in figure 1-2). Secondly, there are problems due to the unavailability of, or inaccuracies in, some corrections that need to be applied to raw altimetric measurements to account for instrumental, atmospheric, or other geophysical effects (such as tides). Both these classes of problems are described in sections 1.3.1, and See for further information 3 See for further information 11

12 corrupted waveform Figure 1-2: An illustration of waveform corruption when the altimeter approaches the coast Need for improved re-tracking The first class of problems mentioned in Section 1.3 comes from the waveform modification, which occurs when land enters the footprint of the instrument. In normal satellite altimetry, parameter retrieval is carried out by fitting a waveform model to the waveforms, a process known as retracking, and then extracting the parameters from it. While waveforms over the open ocean are well fitted by the Brown model (Brown, 1977), in close proximity to the coast (0 10km) more sophisticated re-tracking approaches are needed (Gommenginger et al., 2011). The impact of land on waveforms not only depends upon crude distance from coast but also on the coastal topography. In the coastal zone, waveforms will normally suffer some attenuation due to missing ocean surface elements in the altimetric footprint (as land returns are normally weaker than those from the ocean). However, their shape is also modified by returns from land elements close to the coastline and with variable surface elevation. Features in the waveforms due to these and other effects (for instance the peaks due to quasi-specular reflections from calm waters described by Gómez-Enri et al., 2010) will migrate along the waveform, from one waveform to the next. The simple Brown model is often unable to produce a good fit to the observed waveforms in such conditions, resulting in inaccurate derived parameters. To overcome this, several re-tracking schemes optimized for the coastal environment have been proposed in recent years, within the framework of the COASTALT and PISTACH projects as well as by other members of the international coastal altimetry community. Techniques to ensure that each waveform is not treated in isolation, but in contrast using the information from the previous and following waveforms, have also been developed. This remains is an area of very active on-going research that will hopefully pave the way for the next generation of altimeter retrackers. A significant milestone has already been achieved with a substantial contribution by the esurge project, i.e. the development and validation of a novel subwaveform retracker for pulse-limited 12

13 altimetry (Adaptive Leading Edge Subwaveform, or ALES, Passaro et al., 2014). ALES has two very appealing characteristics it shows improved results in the retrieval of range and significant wave height in the coastal zone across different missions, and its performance over open ocean is virtually the same as the standard Brown retracker therefore in 2014 it was selected as the retracker of choice for pulse-limited altimetry in esurge and the Coastal Altimetry products which are being reprocessed 4. SAR altimetry, by virtue of the high resolution (~ 300m) in the along-track direction, presents intrinsic potential for the monitoring of coastal sea level and significant wave height, at least in all those cases where the satellite tracks are favourably oriented (i.e. near-orthogonal) to the coastline. SAR altimetry uses simultaneous Fourier processing of several echoes received by the instrument to improve the along-track resolution, and is superior to conventional pulse-limited altimetry in terms of signal to noise ratio; this potential is being demonstrated with the CryoSat-2 altimeter, launched in 2010, which is the first one of this kind to ever have been flown, although it has been operating in SAR mode only over a few dedicated oceanic regions, reverting to pulse-limited (or Low-Resolution Mode, LRM) over large parts of the Earth s oceans. Altimeters on boards the Sentinel-3 A and B missions will also be of the SAR variety, and the latest plan 5 is to operate them in SAR mode over the global oceanic area. The technical issues arising when applying SAR altimetry to oceanography, and the generation of ocean products have been studied in detail within the ESA-funded SAMOSA 6 and CryoSat Plus for Oceans (CP4O 7 ) projects. The esurge processor adopts the SAMOSA3 waveform model (Ray et al., 2015) derived within SAMOSA for the retracking of SAR altimetry waveforms, which after the Fourier processing, have shapes distinctly different from the conventional pulselimited waveform. For conventional waveforms in LRM mode we use ALES Need for optimised correction The second set of issues mentioned in Section 1.3 concern the corrections implemented upon the raw altimetric data during processing to obtain measured Sea height parameters. In coastal areas, the unavailability of, or inaccuracies in, some corrections that need to be applied to raw altimetric measurements (to account for instrumental, atmospheric or other geophysical effects such as tides) can severely impact upon the usability of data. In the coastal zone, the most critical corrections concern ocean tides, water vapour, and the sea state bias (SSB). Currently, the SSB (i.e. the bias on range estimates dependent on the sea state, caused by the shape of the sea surface not being perfectly sinusoidal) is in need of further investigation, to successfully migrate from the actual empirical correction models towards physically based (or more complex and natural system based) ones. This is a difficult task concerning open ocean data, let alone data concerning the coastal zone. However the largest uncertainty associated with coastal altimetry data arises from inaccurate removal of tide-associated corrections. Global tidal models improve with every new release, yet still do not perform well in many coastal and shelf locations. Inaccurate correction of the path delay due to water vapour, known as wet tropospheric correction, is also a pressing concern and an issue that must be addressed to make coastal zone data usable. Insofar as tides are concerned, an obvious 4 This reprocessing is happening gradually over 2014 and into early Contact the esurge project for further information. 5 Correct as of this handbooks release date

14 solution is to develop accurate local tidal models, which then need to be merged with the appropriate global models. Notably, some applications do not require the application of this tidal correction a case in point being the study of storm surges where the contribution due to tides must be included in the measured water level (TWLE), in order to fully describe the point at which, and the extent to which coastal flooding occurs. Concerning the wet tropospheric correction, this is normally estimated from a multi-channel passive microwave radiometer on the same platform as the altimeter. However, this estimate quickly becomes corrupted as soon as land enters the radiometer footprint, i.e Km from the coast. Alternative corrections have been devised and appear to be successful at least in some conditions (Desportes et al., 2007; Brown, 2010). One quite promising scheme (reported on in Fernandes et al., 2010) is attempting to estimate the wet tropospheric path delay from GPS measurements, known as GPD (GNSS-derived Path Delay), and developed within the framework of the COASTALT project. Some of the techniques described above have been implemented in software processors used to routinely reprocess coastal altimetry data, such as the Jason-2 data generated by PISTACH and made available through AVISO 8. The GPD wet tropospheric correction, recommended as the correction of choice for the coastal zone by the ESA-funded Sea Level Climate Change Initiative (SL-CCI) Project, has in principle been chosen also for the esurge products, including the coastal altimetry product outlined here, but will only be added in future releases of the esurge products as it is not yet readily accessible via the native Sensor Geophysical Data Records (SGDRs, i.e. the original data records coming from the Space Agencies which are ingested and reprocessed by the esurge processor) or via other archives such as the Radar Altimeter Database System (RADS) 9. At the moment the wet tropospheric corrections available in the esurge products are the ECMWF model-derived Wet Tropo for both Envisat and CryoSat-2, and the MWR (Microwave Radiometer Derived) Wet Tropo for Envisat; the latter is normally noisy and unreliable in the coastal environment so we recommend using the ECMWF model-derived one. 1.4 Processing satellite altimetry observations into coastal altimetry data The coastal altimetry data product associated with this handbook is provided through the esurge service. The dataset products consist of along-track records of geophysical parameters ( Coastal Geophysical Data Records or CGDR a terminology introduced by the COASTALT project for products specifically reprocessed for the coastal zone, and maintained here even if the products are not solely coastal but may extend well into the open ocean). Essentially, the output data available is comprised of satellite altimetry data which has been run through an esurge coastal altimetry processor where waveform data are reprocessed, and ancillary data are used to update corrections specific to coastal areas. The data are at processing level 2 (therefore are typically referred to as level 2 data ) and have been derived by reprocessing altimetric waveforms (level 1b data - which have already been processed using satellite altimetry techniques), with the additional inclusion of a number of corrections for propagation and surface effects. 8 See for further information

15 The high-level schematic of the satellite altimetry to coastal altimetry data process is shown in figure 1-3. Its functionalities, highlighted in red in the figure are: 1. ingest the SGDR products (i.e. the original data records coming from the Space Agencies) from the various missions 2. ingest ancillary data (meteo data and external correction fields, like ionospheric models. These include all corrections at 1 Hz from the RADS archive). reprocess the waveform data, using a number of alternative retrackers to generate high resolution data which may be more useful in coastal areas (examples are, as discussed in the ALES retracker for conventional pulse-limited altimetry and the SAMOSA3 retracker for SAR altimetry). 3. generate new geophysical corrections from these new data where possible, or copy across the corrections from the SGDR and from RADS. 4. generate higher data rate geophysical correction data, by interpolation, as necessary for correcting the higher rate range data. 5. output all the relevant original and new fields into a single file per pass, in a self-describing format (NetCDF), which also allows easily addition of extra fields (for instance, a user model output) The basic coastal product includes fields that can be determined for any coastal region, using the data from the altimeter itself, or instruments mounted on the same platform, and global models, listed below. The output product has been designed to allow use of the new, retracked, range, significant wave height and backscatter values, together with some of the geophysical corrections that rely on them (such as ionospheric correction and sea-state bias corrections 10 ). They also contain the comparable original data, to enable users to readily compare the SGDR and CGDR values. One enhancement of the source data is achieved through providing geophysical correction fields at the higher (18 Hz or 20 Hz, depending on the mission) data rate (the corresponding variable names start with hi_ or hz18_ ). This involves interpolation of the existing 1 Hz values. All geophysical corrections provided on the SGDR have been interpolated, using a simple linear interpolation, to provide high-rate correction values in the CGDR products. 10 as of 2 September 2014 the ionospheric and SSB corrections for Envisat are not yet recomputed using the dedicated retracker estimates, and default to the values provided in the SGDR; this is in consideration of the fact that more research is needed on the behavior of these corrections (especially the SSB) in the coastal zone. For updates please do not hesitate to contact Dr. Paolo Cipollini (see Section 3.0 of this handbook). 15

16 Figure 1-3: schematic of the coastal altimetry processor for the ESA esurge Project, The numbers in red correspond to the various processor functionalities described in the text Note that the esurge processor is designed to be able to be used by experienced investigators to generate their own products, specific to their area of interest. To this end, the processor has been designed to be flexible in its use. Many of the parameters are defined by configuration files accessed at run-time, which can be tailored to specific requirements. The output product can be restricted to include only data falling within a pre-defined region, specified by its latitude and longitude limits. For further information, or to discuss the opportunity and feasibility of specific runs of the processor, see section 3.0 of this handbook Altimeter Missions included in esurge As of August 2014, the altimeter data reprocessed with the esurge processor and distributed via the esurge archive include CGDRs for the following instruments over the esurge SEVs (Surge EVents) and Areas of Interest (AoIs) 11 : ESA Envisat RA-2 CryoSat-2 over LRM regions (also in Near-Real-Time) CryoSat-2 over SAR regions (Near-Real-Time capability for CryoSat SAR has been demonstrated for a small number of specific events but is not currently available) The extension of the processor to Jason-2 and AltiKa is in progress (as of August 2014). Every altimeter has its own set of peculiarities (i.e. single- vs dual- frequency band, presence/absence of a Microwave radiometer) that require some dedicated adjustments to the processing. The following section details those adjustments. 11 The number of events and number and extent of Areas of interest are continuously growing within esurge as new (and past) SEVs and AoIs are added to the archive see 16

17 1.4.2 Models used for re-tracking and corrections The CGDR products are based on the SGDR products for each mission. The input SGDR values have not been altered during translation to the CGDR product, except in the case of some flags (detailed below in table 1-1) and the generation of interpolated values. All algorithm specifications for variables determined directly from the SGDR values can be found in the relevant documentation for specific missions (i.e. Envisat, CryoSat-2, etc). It is advisable that users familiarise themselves with the algorithms applied in deriving mission-specific SGDR values, and remain aware of these as they also consider the algorithms required to reprocess SGDR values into CDGR values. The tables below summarises the models and standards that are adopted for esurge CGDR value derivation for Envisat and CryoSat-2, and also lists the corresponding variables in the CGDR netcdf files. The following Section provides more details on models added in the esurge processor, whilst more general information on included models is provided in [AD 19]. It should be noted that, while the initial ( ) esurge coastal altimetry products for Envisat were based on the legacy of the COASTALT processor, and therefore included the output of a specular peak retracker and a mixed (Brown + specular peak) retracker, with the advent and validation of the ALES sub-waveform retracker (Passaro et al., 2014) it has been decided to use the latter in preference, as it clearly outperforms the specular and mixed ones. The latest esurge products for pulse-limited altimetry (not only Envisat, but any other pulse-limited mission) therefore are based on the output of ALES. For SAR altimetry, as previously stated, the SAMOSA3 retracker (Ray et al., 2015) has been used, which has also been adopted by ESA for the Sentinel-3 Detailed Processing Model. 17

18 Table 1-1: Models and Standard for Envisat Parameter Model and [source] Variable name in CGDR Orbit Based on DORIS and laser tracking data. [SGDR v2.1] alt_cog_ellip hz18_alt_cog_ellip Altimeter Re-tracking (both for range and SWH) Dry Troposphere Range Correction Wet Troposphere Range Correction (from Model) Wet Tropospheric Correction (from onboard microwave radiometer) Ionospheric Correction Sea State Bias Correction Inverse Barometer Correction Non-tidal Highfrequency Dealiasing Correction Tide Solution 1 Tide Solution 2 Solid Earth Tide Model Ocean re-tracking: Brown model as defined in (Brown, 1977)[AD 1] [SGDR v2.1] esurge Brown re-tracking: Brown Ocean Retracker (BOR) (Halimi et al., 2012) a simplified version of (Brown, 1977) [computed] esurge specialized re-tracking: ALES (Passaro et al., 2014) [computed] From ECMWF atmospheric pressures and model for S1 and S2 atmospheric tides [SGDR v2.1] From ECMWF model [SGDR v2.1] Obtained with a neural algorithm from the 23.8 GHz and 36.5 GHz brightness temperatures (in K) interpolated to RA-2 time tag, and the ocean backscatter coefficient for Ku-band (db), not corrected for atmospheric attenuation [AD 2][AD 7] [SGDR v2.1] from the DORIS daily maps of Total Electron Content [SGDR v2.1] from the GIM model for products processed with CMA v7.1 or higher [SGDR v2.1] Empirical function of Ku-band's significant wave height and the RA-2 wind speed from Ocean re-tracking algorithm [AD 1]. The function has been derived from one year of Envisat data (cycles 25 to 35), using crossover SSH differences and applying the non-parametric estimation technique (Gaspar and Florens, 1998) [SGDR v2.1] Computed from ECMWF atmospheric pressures after removing S1 and S2 atmospheric tides [SGDR v2.1] Mog2D High Resolution ocean model. Ocean model forced by ECMWF winds and atmospheric pressures after removing S1 and S2 atmospheric tides difference of this value and the inverse barometer correction is reported (Carrère, 2003; Carrère and Lyard, 2003) [SGDR v2.1] GOT00.2b total geocentric tide (Cartwright et al 1991; Ray, 1999) including the load tide and the long-period equilibrium tide contributions [SGDR v2.1] FES2004 total geocentric tide (Lefèvre et al, 2002) including the load tide (Francis and Mazzega, 1990) and the long-period equilibrium tide contributions (the long-period tide is the same as for solution 1) [SGDR v2.1] From Cartwright and Taylor (1971) tidal potential [SGDR v2.1] ku_band_ocean hz18_ku_band_ocean ku_sig_wv_ht hz18_ku_sig_wv_ht esurge_brown_range_ku hz18_esurge_brown_range_ku esurge_brown_swh_ku hz18_esurge_brown_swh_ku hz18_ales_range_ku hz18_ales_swh_ku mod_dry_tropo_corr hz18_mod_dry_tropo_corr mod_wet_tropo_corr hz18_mod_wet_tropo_corr mwr_wet_tropo_corr hz18_mwr_wet_tropo_corr ion_corr_doris_ku hz18_ion_corr_doris_ku ion_corr_mod_ku hz18_ion_corr_mod_ku sea_bias_ku hz18_sea_bias_ku inv_barom_corr hz18_inv_barom_corr dib_hf hz18_dib_hf tot_geocen_ocn_tide_ht_sol1hz 18_tot_geocen_ocn_tide_ht_sol1 tidal_load_ht_sol1 hz18_tidal_load_ht_sol1 long_period_ocn_tide_ht hz18_long_period_ocn_tide_ht tot_geocen_ocn_tide_ht_sol2 hz18_tot_geocen_ocn_tide_ht_sol2 tidal_load_ht_sol2 hz18_tidal_load_ht_sol2 solid_earth_tide_ht hz18_solid_earth_tide_ht Pole Tide Model Equilibrium model (Wahr, 1985) [SGDR v2.1] geocen_pole_tide_ht hz18_geocen_pole_tide_ht Mean Sea Surface CLS01 (Hernandez and Schaeffer, 2000; 2001) [SGDR v2.1] m_sea_surf_ht hz18_m_sea_surf_ht 18

19 Table 1-2: Models and Standard for CryoSat-2 LRM 12 Parameter Model and [source] Variable name in CGDR Orbit Altimeter Re-tracking (both for range and SWH) Dry Troposphere Range Correction Wet Troposphere Range Correction (from Model) 13 Ionospheric Correction Sea State Bias Correction Inverse Barometer Correction Non-tidal Highfrequency Dealiasing Correction Tide Solution Solid Earth Tide Model Based on DORIS Navigator Level-0 Data for NRT products and DORIS precise orbits for delayed-time products. [Level1b] esurge Brown re-tracking: Brown Ocean Retracker (BOR) (Halimi et al., 2012) a simplified version of (Brown, 1977) [computed] NOTE: range is already converted into TWLE by applying appropriate corrections esurge specialized re-tracking: ALES (Passaro et al., 2014) [computed] NOTE: range is already converted into TWLE by applying appropriate corrections From ECMWF analysed grids supplied by Meteo France via the CNES SSALTO system [Level1b] From ECMWF atmospheric pressures and model supplied by Meteo France via the CNES SSALTO system [Level1b] From either the Global Ionospheric Map (GIM), which uses GPS measurements, or the Bent model, selected at processing time. The GIM is the nominal choice with the Bent model available as a alternative solution if GIM data is not available [Level1b] not yet provided at this time (August 2014) --- Computed from ECMWF atmospheric pressures supplied by Meteo France via the CNES SSALTO system [Level1b] Mog2D High Resolution ocean model via the CNES SSALTO system, difference of this value and the inverse barometer correction is reported [Level1b] FES2004 total geocentric tide including the load tide and the long-period equilibrium tide contributions [Level1b] From Cartwright and Taylor (1971) tidal potential [Level1b] alt_cog hi_alt_cog hi_h_twle_brown swh_ocean hi_swh_brown hi_h_twle_ales hi_swh_ales corr_trop_dry_mod hi_corr_trop_dry_mod corr_trop_wet_mod hi_corr_trop_wet_mod corr_iono_mod hi_corr_iono_mod corr_ib hi_corr_ib corr_dib_hf hi_corr_dib_hf h_tide_ocean_tot_geocen hi_h_tide_ocean_tot_geocen h_tide_load hi_h_tide_load h_tide_ocean_long_period hi_ h_tide_ocean_long_period h_tide_solid_earth hi_h_tide_solid_earth Pole Tide Model Equilibrium model (Wahr, 1985) [Level1b] h_tide_pole_geocen hi_h_tide_pole_geocen Mean Sea Surface UCL04, a hybrid global model compiled from the Arctic Gravity Project Geoid (see and PIPS Mean Dynamic Topography (from the US Naval Postgraduate School) above 81.5 degrees north, the ERS MSS (see between 60 and 81.5 degrees north and CLS01 MSS (see over the rest of the globe, to provide better accuracy in the arctic ocean h_mss hi_h_mss 12 in CryoSat-2 nomenclature, the Level-1 input products functionally equivalent to the Envisat SGDR (i.e. inclusive of averaged waveforms that can be retracked) are known as Level1b 13 Note that thereis no on-board microwave radiometer on CryoSat-2 19

20 Table 1-3: Models and Standard for CryoSat-2 SAR Parameter Model and [source] Variable name in CGDR Orbit Altimeter Re-tracking (both for range and SWH) Dry Troposphere Range Correction Wet Troposphere Range Correction (from Model) 14 Based on DORIS Navigator Level-0 Data for NRT products and DORIS precise orbits for delayed-time products. [Level1b] SAMOSA3 waveform model (Ray et al., 2015) [computed] NOTE: additional range-derived variables are also provided: 1) corrected range, 2) corrected SSH and 3) TWLE From ECMWF analysed grids supplied by Meteo France via the CNES SSALTO system [Level1b] From ECMWF atmospheric pressures and model supplied by Meteo France via the CNES SSALTO system [Level1b] Ionospheric Correction GPS-based: from Global Ionospheric Map (GIM), which uses GPS measurements[level1b] Model: Bent model [Level1b] Sea State Bias Correction Inverse Barometer Correction Non-tidal Highfrequency Dealiasing Correction Tide Solution not yet provided at the moment --- Computed from ECMWF atmospheric pressures supplied by Meteo France via the CNES SSALTO system [Level1b] not provided at the moment --- FES2004 total geocentric tide including the load tide and the long-period equilibrium tide contributions [Level1b] hi_alt_cog hi_range_samosa hi_range_corrected_samosa hi_h_corrected_samosa hi_h_twle_samosa3 hi_swh_samosa hi_corr_trop_dry_mod hi_corr_trop_wet_mod hi_corr_iono_gps hi_corr_iono_mod hi_corr_ib hi_h_tide_ocean_tot_geocen_s ol1 hi_h_tide_load_sol1 hi_h_tide_ocean_long_period Solid Earth Tide Model From Cartwright and Taylor (1971) tidal potential [Level1b] hi_h_tide_solid_earth Pole Tide Model Equilibrium model (Wahr, 1985) [Level1b] hi_h_tide_pole_geocen Mean Sea Surface not provided at the moment (UCL04 is used in computations of SSH and TWLE) Note that there is no on-board microwave radiometer on CryoSat-2 20

21 1.4.3 esurge specific models esurge retrackers As introduced in 1.4.2, several physically based waveform retrackers are implemented and run in parallel within the esurge processor in its standard configuration. They are: 1. a conventional Brown theoretical Ocean Retracker, in the simplified form that coincides with the one used in Halimi et al. (2012). 2. the novel Adaptive Leading-Edge Subwaveform retracker or ALES (Passaro et al., 2014) 3. for SAR waveforms, the SAMOSA model (Ray et al., 2015) Retracker 1: Brown theoretical ocean retracker (BOR) This is an implementation of the well-known Brown ocean waveform retracker, known to perform well for altimeter waveforms over the open ocean and (typically) up to ~10 km from the coast. The complete expression of the power P r (t) at time t is given by: ( ) { ( and ) } ( ( ) ) { ( ( ) ) ( ( ) ) ( ( ) )} Eq. 1 ( ) { } { ( ( ) ) ( ( ) ) ( ( ) )} where: sigma0 = the normalised radar backscatter cross-section at nadir h = the satellite height = the mispointing angle (from nadir) c = the velocity of light in vacuo t 0 = the time origin (=2h/c) corresponding to the mid-point on the waveform leading edge I 0 (t) = the modified Bessel function of the first kind [0] Tn = the thermal noise erfc(x) = the complementary error function [0] = the ocean skewness SWH = the Significant Wave Height He 2 = the Hermite polynomial of order 2 Eq. 2 21

22 and const1= G 2 0 l 2 R chp T s P p 4( 4p ) 2 l p 2 Eq. 3 Eq. 4 Eq. 5 where: G 0 = the antenna gain parameter R = the radar wavelength = the pulse compression ratio P T = the transmitted power p = 0.53*(compressed pulse width) l p = the two-way propagation loss over and above the free-space loss Y B = the antenna beam in radians s = the standard deviation of the sea surface elevation, related to the significant wave height by s ~ SWH/4. These are the equations implemented in the NOC BOR retracker, which has been used successfully to retrack Envisat RA-2 Ku-band waveforms in the COASTALT processor [AD 6][AD 11], although with the following simplifications: four parameters retrieval: range (t 0 ), SWH ( s ), Sigma 0, Thermal Noise (as opposed to 5 parameter retrieval i.e. the fifth parameter - the off-nadir mispointing angle ξ - is either set to zero or provided from independent observations such as for the on-board star-trackers) Linear wave statistics (skewness and cross-skewness set to zero) Least-square fitting (with provision for weighted least square fitting in future) It is worth noting that with the simplifications above the COASTALT Brown retracker coincides with the simplified Brown component described by Halimi et al. (2012) when the instrument thermal noise is estimated separately (for instance using the portion of the waveform preceding the leading edge); this is the form implemented in the esurge coastal altimetry processor. Retracker 2: Adaptive Leading-Edge Subwaveform (ALES) ALES adopts the Brown-Hayne theoretical ocean model (Brown, 1977; Hayne, 1980), the same as in use for the open-ocean retrackers, which describes the average return power of a rough scattering surface (i.e. what we simply call waveform). The return power V m is 22

23 ( ) [ ( )] ( ) Eq. 6 Where ( ) ( ) ( ) ( ) ( ) a ( ) ( ) ( ) and c = is the speed of light h = the satellite altitude R e = the Earth radius ξ = the off-nadir mispointing angle θ 0 = the antenna beam width τ = the Epoch with respect to the nominal tracking reference point σ c = the rise time of the leading edge (depending on a term σ s linked to SWH and on the width of the radar point target response u p ) σ p = radar point target response P u = the amplitude of the signal T n = the thermal noise level In practice, the model in equation 6 is a raised sigmoid [1+erf(u)]/2 describing the increasing power in the waveform leading edge and the subsequent plateau, multiplied by a negative exponential which models the reduction of power in the waveform tail (decay), plus thermal (additive) noise T n. The amplitude of the signal P u is attenuated by a term a ξ dependant on mispointing ξ. P u can be converted into a measurement of the backscatter coefficient σ 0 on the basis of the instrument calibration. Note that the significant wave height SWH, in addition to affecting the rise time of the waveform, also has a small effect on the sigmoid location (variable u) and on the waveform decay (variable v), via the term c ξ σ c 2. More details and a full validation of ALES for Envisat, Jason-1 and Jason-2 are in Passaro et al.,

24 Time delay Retracker 3: SAMOSA model for SAR (delay-doppler) altimetry waveforms SAR altimetry exploits the Doppler information contained in the received radar echoes to achieve improved performance in terms of along-track resolution and precision of the range and SWH retrievals (Raney, 1998). While the hardware is largely similar to that of a conventional altimeter, the received signal undergoes delay-doppler processing that allows to resolve narrow along-track cells (as shown in Figure 1) rather than a large circular footprint. Multiple looks over the same along-track resolution cell are then accumulated and averaged. The resulting multilooked waveform has a complicated analytical expression, whose detailed model (the SAMOSA3 model used by the esurge processor) is described in Ray et al. (2015). Along-track Figure 1-4: delay/doppler (SAR) altimeter footprint from Cotton et al., (2008) 1.5 Product Validation [AD 1] The main coastal altimetry products (height and significant wave heights) from the esurge processor for Envisat have undergone an internal validation (or verification) by comparison with respect to the following references: Retracked values in the Envisat SGDRs Values from the RADS archive by NOAA and Technical University of Delft This verification has been carried out by means of simple diagnostics (RMS of the 18-Hz samples in each data block, and bias), for the passes in the SEARS archive for two SEVs over different AOIs, namely Cyclone SIDR (AOI-90, see fig 1-4) N Indian Ocean 15 November 2007 Major North Sea storm surge (AOI-10) 8/9 Nov

25 Figure 1-5: Envisat passes over the North Indian Ocean (AOI-90) during 8-19 Nov 2007, when the area was affected by cyclone SIDR The validation was initially done by visually checking that the reprocessed quantities were realistic in comparison with the reference, as shown in figure 1-5 for a pass over the SIDR affected region for the particular example of the sea surface height. Figure 1.6 shows the plots of significant wave height and its 20-Hz RMS value. Figure 1-6: Example of validation of sea surface height (SSH) for Envisat pass 0438, cycle 63 over the Bay of Bengal (14 Nov 2007); (left) ground track; (centre) uncorrected SSH for esurge compared with that in SGDR and RADS the CLS01 Mean Sea Surface (MSS) is also plotted for reference; (right) zoom in on a shorter track segment. 25

26 Figure 1-7: Example of validation of significant wave height (SWH) for Envisat pass 0438, cycle 63 over the Bay of Bengal (14 Nov 2007); (left) ground track; (centre) comparison of 1-Hz SWH for esurge with that from SGDR and RADS; (right) standard deviation of the 18-Hz data over 1.1 s blocks. Subsequently, cumulative statistics (median of the 18-Hz values for range and SWH, and mean range and SWH biases) have been computed for all the profiles over the two selected SEVs. The results, shown in tables 1-2 and 13, confirm that esurge Brown retracker is internally validated with respect to RADS, which constitutes a widespread accepted standard, as both range and SWH RMS and biases are very close to reference ones, so this is a solid foundation on which to add specialized retracking (such as with the ALES algorithm) for coastal scenarios (and wind estimation) Table 1-4: Results of the internal validation for range 26

27 Table 1-5: Results of the internal validation for SWH Validation of coastal altimetry data against in-situ observations for Envisat and for other missions over specific surge events will be carried out in collaboration with the esurge user community and its results will be included in future updates to this document as appropriate Product Quality Flags Flags are used to convey quality information or operating modes. They are usually set to zero to mean OK and 1 for not OK. Any spare flags are set to zero. There may be exceptions, in which case a particular description of the flag's use is provided. For example, flags may be used to provide information on the operation mode of the instrument. The flag values and flag meanings of simple flags are defined in their attributes in the CGDR products. Additional editing is usually required to determine the data quality. In some cases, flags have been pre-determined, using quality criteria e.g. data exceeding given thresholds. However, in other cases, users will also need to apply their own data quality editing criteria 1.6 The Near Real Time Coastal Altimetry product The esurge system has been set up to handle near real time surge events, which includes near real time data, to service the storm surge modelling community. This is the esurge Live service as foreseen and requested in the esurge Statement of Work [SOW]. A significant component of esurge Live is the provision of reprocessed NRT Coastal Altimetry, on which this section provides some details Input data to the NRT Coastal Altimetry processor esurge-live has been demonstrated on a small number of NRT data for CryoSat-2 SAR, over the Indian coast SAR mode polygon that is included in Geographical mode mask 3.4 active since October A small amount of NRT SAR data was also made available by ESA during the Xaver/Sinta Claas storm over the North Sea in December 2013, following a specific request by the esurge project; this allowed the project to observe the resulting surge within few hours from acquisition of the data 15. Other data available for NRT processing, but not yet tested in the esurge processor, are the Jason-2 Operational Geophysical Data Records (OGDR), which are routinely generated and distributed by Eumetsat with a short delay of three to five hours from acquisition. These will be ingested in a future extension of the processor. 15 See 27

28 1.6.2 Processing details for NRT Coastal Altimetry The modifications to the esurge coastal altimetry software needer in order to be able to process NRT data are minimal, and all deal with ensuring a proper ingestion of the various fields in the Level 1b NRT data stream. Taking CryoSat-2 as an example, the NRT products whose name starts with CS_NRT_ on the ESA Kiruna server have the same format than the offline products whose name starts with CS_OFFL. The main differences in the data will be the availability and accuracy of some of the fields, mainly the orbital height (delayed-time, consolidated orbits are normally significantly more accurate than the NRT ones) which is used in the computation of SSH and TWLE. Then the processor applies exactly the same retracking algorithms and correction interpolation as those described in section 1.4. Estimates of SWH and wind in the NRT products are not affected by the orbit field and are expected to be of similar precision in NRT and delayed-time products Availability of NRT Coastal Altimetry on the esurge server On the esurge server NRT products and delayed time products are kept separate and have different product codes: NRT9 for the NRT data and SL1 for the delayed-time data. This makes a comparison of the two products over the same pass segment possible, once the delayed-time products for that segment are produced. In most cases the differences are only due to the improvement in the consolidated orbits, therefore at scales of little relevance to storm surge observations. 1.7 Potential Uses of Altimetry Data in Storm Surge Applications One of the most promising applications of coastal altimetry is their potential use in the study of storm surges. Understanding and realistic modelling of surges supports both preparation and mitigation activities, should therefore bring enormous societal benefits, particularly to some of the world s poorest countries (like Bangladesh). Coastal altimetry data have a prominent role to play here, as it provides direct measurements of the total water level envelope (TWLE), i.e. one of the key quantities required by storm surge applications and services. As mentioned, such storm surge applications of coastal altimetry data do not require tidal correction, given the importance of deriving measurements for the TWLE, which includes the water mass component due to tides. Moreover, coastal altimetry data can also provide important information on the wave field in the coastal strip, helping to develop more realistic wave models that in turn can be used to improve the forecast of wave setup and overtopping processes. 28

29 2. PRODUCT DETAILS 2.1 Technical Description The coastal altimetry products provided through the esurge system are essentially comprised of satellite altimetry data which has been run through the esurge coastal altimetry processor, effectively reprocessing the waveform data and using ancillary data to update corrections specific to coastal areas. The products consist of along-track records of geophysical parameters at processing level 2 (and therefore normally referred to as level 2 data ), having been derived by reprocessing altimetric waveforms (level 1b data) which have already been processed using satellite altimetry techniques. Moreover, in deriving coastal altimetry data from the satellite altimetry inputs there are the additional inclusions of a number of corrections for propagation and surface effects. As already mentioned in section 1, the elementary data record is called Coastal Geophysical Data Record (CGDR a terminology been introduced by the COASTALT project for the development of Envisat Altimetry for the Coastal Zone). esurge CGDRs are available in NetCDF format, with each separate file corresponding to a single overpass pass of the satellite over a specific esurge Area of Interest (AOI). A list of all esurge AOIs is available in [RB]. Time series of parameters over specific locations can be built by extracting the parameter from multiple CGDRs and combining them. The important feature in the esurge CGDR is that they have been generated with specialized retracking algorithms optimized for the coastal environment, and also that they contain as an explicit variable an estimate of the Total Water Level Envelope (TWLE), which is a crucial quantity for surge studies. The esurge CGDRs are NetCDF files, one for each separate altimetric overpass over an AOI, for the different altimetric missions considered in esurge (Envisat, CryoSat-2, Jason-2 and any future additions). They have been computed starting from each mission s Level 1B data (waveforms), which are usually available in the form of SGDRs (Sensor Geophysical Data Records), through processing with the esurge coastal Altimetry processor. The basic coastal product includes fields that can be determined for any coastal region, using the data from the altimeter itself, or instruments mounted on the same platform, and global models (many of these fields are available in the SGDRs so are just copied through and interpolated to the CGDRs, see below). This product does not include fields that would require specific auxiliary information, such as a region-specific tidal model, or in-situ observational data. The output product has been designed to allow use of the new, retracked, range, significant wave height and backscatter values, together with the geophysical corrections that rely on them (such as ionospheric correction and sea-state bias corrections). It also contains the comparable original data, to enable users to independently compare the SGDR and CGDR values. One enhancement of the source data is achieved through providing geophysical correction fields at the high (18 or 20 Hz) data rate. This involves interpolation of the existing 1 Hz values. All geophysical corrections provided on the SGDR have been interpolated, using a simple linear interpolation, to provide high rate correction values in the esurge CGDR products. 29

30 A summary of the technical specifications of the coastal altimetry data product (SL1) can be found in Table 2-1. A specific summary for the NRT coastal altimetry data (product NRT9) is in Table 2-2. Note that the CGDRs for SL1 and NRT9 are in the same format: what changes is their availability for different mission and different areas. Product Provider Areas of interest covered Description Source Data Source Data Format Reprocessing Source of Error Information Accuracy (estimated) Resolution Table 2-1: Coastal altimetry product technical specifications. NOC SL1 - Coastal altimetry esurge areas: all AoIs See for further information on AOI locations Altimeter data reprocessed by coastal altimetry processor, with specialized retracking and corrections, to obtain the TWLE. Missions and time spans: Envisat RA-2: May Apr 2012 CryoSat: Jul 2010 onwards Jason-2 (being added): Jul 2008 onwards Sensor Geophysical Data Records (SGDRs) for all missions Envisat/CryoSat: from ESA J-1/J-2: from AVISO Envisat, CryoSat: ESA custom (binary) Jason-1/2: NetCDF SGDR hi-rate waveforms are retracked with specialized retrackers. For conventional pulse-limited missions (including CryoSat-2 over LRM mode areas) the ALES retracker is used. For CryoSat-2 over SAR mode areas the SAMOSA retracker is used. Corrections to range are recomputed with techniques optimized for the coastal strip and/or retrieved from updated archives such as the RADS archive. The corrected heights (i.e. orbital heights corrected ranges) correspond to the TWLE Goodness of Fit shall be generated by coastal altimetry processor during waveform fitting 16 5cm at 10 km from the coast, 6cm at 5km from the coast, 8cm at 3km from the coast (single 20Hz measurements) 350m along-track for the high-rate (20Hz) data 16 not yet implemented as of 2 September

31 Product Provider Table 2-2: NRT Coastal altimetry product technical specifications. NOC NRT9 Near Real Time Coastal altimetry Areas of interest covered Description Source Data Source Data Format Reprocessing Source of Error Information Accuracy (estimated) Resolution esurge areas: in principle all AoIs ; in practice for CryoSat only where NRT data are available (in SAR mode, the Indian Coastal SAR polygon) See for further information on AOI locations Altimeter data reprocessed by coastal altimetry processor, with specialized retracking and corrections, to obtain the TWLE. Missions and time spans: CryoSat: Jul 2010 onwards (only for specific event in NRT) Jason-2 (being added): Jul 2008 onwards (only for specific event in NRT) NRT Sensor Geophysical Data Records (SGDRs) for all missions CryoSat: CS_NRT from ESA J-2: OGDR from EUMETSAT CryoSat: ESA custom (binary) Jason-2: NetCDF SGDR hi-rate waveforms are retracked with specialized retrackers. For conventional pulse-limited missions (including CryoSat-2 over LRM mode areas) the ALES retracker is used. For CryoSat-2 over SAR mode areas the SAMOSA retracker is used. Corrections to range are recomputed with techniques optimized for the coastal strip and/or retrieved from updated archives such as the RADS archive. The corrected heights (i.e. orbital heights corrected ranges) correspond to the TWLE Goodness of Fit shall be generated by coastal altimetry processor during waveform fitting 17 9cm at 10km from the coast, 10cm at 5km from the coast, 12cm at 3km from the coast (single 20Hz measurements) 350m along-track for the high-rate (20Hz) data 17 not yet implemented as of 2 September

32 2.2 Accessing the Product For many storm cases (events) where storm-surges are reported to have arisen, coastal altimetry data have been collected and made available through the esurge web-service (see shown in figure 2-1 and figure 2-2). We would encourage users to look up these events and the altimetry and various ocean data contained on the facility database. Figure 2-1: Accessing the data on the esurge website (tabs encircled). Figure 2-2: The data access page of the esurge website. 32

33 2.2.1 Viewing the product in the esurge website The esurge web-service provides users with a number of ways to explore and assess the utility of data they are interested in. Upon navigating to a storm surge event of interest (e.g. Cyclone Sidr shown in figure 2-3), the individual datasets can be previewed (figure 2-4) using the preview tab alongside the dataset name as shown encircled in figure 2-3. Alongside the dataset name also lies the link to view global attributes and dataset information through the OPeNDAP viewer (figure 2-5). Alternatively, users can quick link to the esurge GIS web-viewer, where they can visualise and explore all the datasets available concerning the event (figure 2-6) Figure 2-3: The Cyclone Sidr dataset inventory on the esurge web-service. Preview option links are encircled. 33

34 Figure 2-4: Previewing the Cyclone Sidr altimetry dataset using the Preview link. Figure 2-5: Viewing dataset metadata and information using OPeNDAP. 34

35 Figure 2-6: Exploring Cyclone Sidr datasets in the esurge GIS web-viewer Download the product If users are satisfied that the product suits their purposes, there is an option to download the data using the OPenDAP dataset information window (figure 2-7). Simply click on the Get ASCII or Get Binary tab to download the data. Note that the ability to download requires registration. Figure 2-7: downloading data using the OPeNDAP dataset information page (tabs encircled). 35

36 2.2.3 Accessing the product outside esurge Outside of the esurge-listed events, the source data from which these inundation datasets were derived are also publically available (also frequently subject to registration). The originators of the data are clearly outlined in each datasets metadata should users need to consult the data sources. If data sources were unclear, the source data were not used to derive the available inundation datasets. Note that the inundation datasets specifically provided through the esurge system are not available through third parties, but only through the esurge web-service. 2.3 Using the Product Overview This section will give the reader a guide to the use of the CGDR data from the esurge coastal altimetry processor, using Envisat products as an example. These products are experimental, and provided primarily for research purposes. They have not yet been fully validated, and users should proceed with caution, particularly in regard to the coastal specific fields. While this handbook tries to be correct and complete, note that nothing can replace the information to be gained at conferences and other meetings from those using these data. The user must proceed with caution and explore the use of the data with due diligence. While Envisat is used as an example, the illustrated procedure can be easily replicated for products from other satellites (for instance Jason-2 and CryoSat-2 in LRM or SAR mode) once the differences in variable naming detailed in Table 1-1, Table 1-2, andtable 1-3 are taken into account. The instruments on Envisat made direct observations of the following quantities of interest to satellite altimetry users: altimeter range, ocean significant wave height, ocean radar backscatter cross-section (primarily a function of surface wind speed), ionospheric electron content in the nadir direction and tropospheric water content. Ground based laser station and DORIS station measurements of the satellite location and speeds are used in precision orbit determination (POD). The DORIS stations also measure the ionospheric electron content along the line of sight to the satellite. All of these measurements are useful in themselves, but they are made primarily to derive the sea surface height with the highest possible accuracy. Such a computation also needs external data (not collected aboard Envisat), e.g., atmospheric pressure. In addition, instrument health and calibration data are collected onboard and used to make corrections to the main measurements and to monitor the instrument stability in the long term. The esurge CGDRs contains all relevant corrections needed to calculate the sea surface height. For the other geophysical variables in the CGDR: ocean significant wave height, tropospheric water content, ionospheric electron content (derived by a simple formula) and wind speed, the necessary instrument and atmospheric corrections have already been applied. The following sections explain the organisation of the products, and the rationale for how the corrections should be applied. 36

37 2.3.2 Organisation of the Product The organisation of the product follows the specifications already defined for the COASTALT project [AD 7]. The CGDR products are arranged as one file per segment per pass over an Area of Interest (so the same for the source SGDR products). The product names are based on the convention used for the Envisat Level 2 Products, with the addition of the esurge Area of Interest (AOI) identifier at the end (i.e. before the.nc extension), namely: Filename = <product_id><processing_stage_flag><originator_id> <start_day>_<start_time>_<duration> <phase><cycle>_<relative_orbit>_<absolute_orbit>_<counter>_<aoi identifier>.nc With: <product_id> <processing_stage_flag> <originator_id> RA2_MWS_2P R F-P Processing stage flag (N-V, where N is NRT and letter closer to V are higher levels of consolidation. R are the data from reprocessing v 2.1 A 3 character ID code for the originator of the Level 2 products (F-P is the French PAF, AVISO) <start_day> yyyymmdd UTC date of first record in file in year month day order <start_time> Hhmmss UTC time of first record in file in hour minute second format <duration> Ssssssss duration of product in seconds <phase> A mission phase single character A or B, or 3 for the extended phase <cycle> CCC the cycle, a three digit number, eg 021 = cycle 21 <relative_orbit> <absolute_orbit> <counter> <AOI indentifier> XXXXX XXXXX XXXX AOI_XXX the relative orbit number within the cycle, a five digit number from to 00501, eg is relative orbit 42 the absolute orbit number since the start of the mission, a five digit number, eg incremental counter for product, from 0000 to 9999 then wraps to 0000 XXX is the AOI number, see for further information on AOI locations 37

38 For example product: RA2_MWS_2PRF-P _203914_ _00259_51565_3453_AOI_010.nc is based on the SGDR product: RA2_MWS_2PRF-P _203914_ _00259_51565_3453.N1 Which is a level 2 RA2 / MWR product, at processing stage R, originating from the French PAF, with the first record in the file from 8 th January 2012, 20:39:14, including 3006 seconds of data from phase 3 (extended Envisat phase), cycle 110, relative orbit 259, absolute orbit 51565and was product 3453 generated (in base 10000). Important note on orbit and pass numbering for Envisat 18 : the data for a single orbit will be contained in two separate files, one for the ascending (south to north) pass (a pass is half an orbit) and the other for the descending (north to south) pass. To make things slightly more complicated, the numbering of <absolute_orbit> and <relative_orbit> vary in a staggered pattern, i.e. an absolute orbit goes from the northernmost point in the orbit to the next northernmost point in the orbit, where a relative orbit goes from the southernmost point in the orbit to the next southernmost point in the orbit. Pass number (rather than orbit number) is normally used by altimeter specialists to refer to passes (half-orbits) and avoid any confusion, for instance it is extensively used in the RADS archive. Of the two SGDR files with the same <relative_orbit> number, the first one will be the ascending pass (with a odd pass number, equal to 2*<relative_orbit> 1), while the second one is the descending pass (with an even pass number, equal to 2*<relative_orbit>). The CGDR product is arranged with related values clustered together in the product. Hence, the first section of the product contains the dimension variables and coordinate variables then the orbit information (including the positional information). The complete list of variables included in the products is given in ANNEX B: FULL NetCDF DESCRIPTION OF VARIABLES IN THE CGDRs Within each section, there will be variables reported at 1 Hz and variables reported at 18 Hz. The 1 Hz variables (actually, 0.9 Hz) represent an average of 20 of the 18 Hz values. The 18 Hz variables are represented as 2 dimensional variables, with the primary dimension equal to the dimension of the 1 Hz data, and the second dimension being 20 the number of samples in the 1 Hz average. The 1 Hz average is calculated as occurring at the centre-point of the 18 Hz values, ie between the location of the 10 th an 11 th 18 Hz values. In general the 1 Hz and 18 Hz variables will have the same variable name, with the 18 Hz versions having the variable name prefixed by hz18_, eg the primary coordinate variables are lat and lon, whilst the coordinate variables associated with the 18 Hz values are hz18_lat and hz18_lon. 19 In addition to the Ku band, Envisat (and other altimeters such as Jason-1 and Jason-2) had a second, lower frequency band available (S-band for Envisat, C-band for Jason-1/2) whose purpose is the derivation of an accurate ionospheric correction. S-band waveforms can be retracked to derive range and swh estimates (it is indeed the difference between Ku-band and S-band estimated ranges 18 Note that versions of Envisat SGDRs prior to v2.1 did not have an unequivocal convention on orbit numbering; the convention described in this paragraph only applies to v2.1, and to the esurge Envisat products 19 For CryoSat-2 high-rate data the variable name is prefixed by hi_ 38

39 that allows the computation of the ionospheric correction). S-band-derived values for some quantities are present in the SGDRs but they are not copied across to the esurge CGDRs, as the Kuband-derived ones are the most accurate ones Typical computation from altimetry data In this section references are made to specific CGDR parameters by name using the name of the variable as described in the netcdf data sets. These names were initially defined by the parameter descriptions from the Envisat Level-2 product definition documents, for ease of referencing of the source algorithms. Some minor modification has occurred for internal consistency within the product. WARNING: Default values, provided in the _FillValue attribute, are given to data when computed values are not available, the user must screen parameters to avoid using those with default values. Also the user must check flag values. The related flags are given in the description of each variable (see [AD 7] although some discussion of flags appears in this section.) In all cases the user must ensure that all values are available at the same frequency. 1 Hz version of 18 Hz values can be calculated by averaging over the 20 samples. 18 Hz data can be generated by interpolating the 1 Hz values, using time as the interpolation base variable and remembering that the 1 Hz average represents the centre of the associated 18 Hz measurements (between the 10th and 11th points). All SGDR corrections available on the CGDR, as well and any 1-Hz updated corrections form the RADS archive, have also been linearly interpolated to 18 Hz for ease of use of the data Corrected Altimeter Range The main data of the CGDR are the altimeter ranges. The CGDR provides ranges measured at Kuband (see Table 1-1, and Table 2-3 below), i.e. the one used for most applications. The given ranges are corrected for instrumental effects. The given ranges must be corrected by the user for path delay in the atmosphere through which the radar pulse passes and for the nature of the reflecting sea surface. All range corrections are defined and they should be ADDED to the range. A corrected (Ku-band) range is given by: Corrected Range = Range + Wet Troposphere Correction + Dry Troposphere Correction + Ionosphere Correction + Sea State Bias Correction For the specific case of Envisat, in the past (for SGDR prior to v2.1) a further correction needed to be applied to ranges: the USO Correction compensating for the drift of the Ultra-Stable Oscillator on board. This correction has now been accounted for in the reprocessing leading to Envisat v2.1 SGDRs, released since January 2012, so it is already applied in the esurge Envisat CGDRs derived from v2.1 SGDRs. 39

40 Range: This may be one of the SGDR range values, or one of the esurge retracked ranges, as given in Table 2-3 below. Table 2-3: Available range values in the CGDR 1 Hz range value 18 Hz range value source ku_band_ocean hz18_ku_band_ocean SGDR Ocean retracker esurge_brown_range_ku hz18_esurge_brown_range_ku hz18_esurge_bgp_range_[ku/s] esurge Brown retracker (BOR see ) ALES retracker (see ) The esurge Brown retracker output gives ranges very similar to the SGDR Ocean retracker, as discussed above (section 1.4.3) The ALES Retracker has been validated in the coastal zone in a number of cases (Passaro et al., 2014) and should normally be used in preference in the coastal strip. It is not recommended that users change range source within a single analysis, as there may be unknown biases between the retracker ranges. The ALES retracker usually offers a good performance also over the open ocean so can be used for analysis that extend from the coast to offshore waters Wet Troposphere Correction Table 2-4: Available wet tropospheric correction values in all CGDR products 1 Hz correction 18 Hz correction source mod_wet_tropo_corr hz18_mod_wet_tropo_corr ECMWF Model mwr_wet_tropo_corr hz18_mwr_wet_tropo_corr MWR The model wet tropospheric correction will exist for all 1 Hz measurements, but the MWR measurement is expected to more accurately capture short-scale tropospheric processes affecting the water vapour concentration. Enhanced CGDR products may be available in the uture, that contain an additional wet tropospheric correction, at 18 hz, generated using the novel GPD algorithm (Fernandes et al., 2010). The GPD correction has been generated at 1 Hz by University of Porto, then interpolated to 18 Hz using simple linear interpolation, before addition to the CGDR products. 40

41 Table 2-5: Additional wet tropospheric correction values in a limited number of enhanced CGDR products. 1 Hz correction 18 Hz correction source hz18_gpd_wet_tropo_corr GPD algorithm (Fernandes et al., 2010). Note: the GPD wet tropospheric correction, and its associated flag, error and variance (see [AD 7]), are not yet part of the esurge baseline CGDRs but will be added in the near future. The user should check for presence of these variables in their products Dry Troposphere Correction Only one dry tropospheric correction is provided: the ECMWF model correction at 1 Hz (mod_dry_tropo_corr), together with the correction interpolated to 18 Hz (hz18_mod_dry_top_corr) Ionosphere Correction There are two ionospheric corrections available on the CGDR, given in Table 2-6. These are the DORIS and model corrections. We have discontinued the dual-frequency derived correction for Envisat as the S-band ceased operation on 18 January 2008 (cycle 65), so for all data after that date one needs to use an alternative correction, and models have got very close in terms of accuracy to dual-frequency correction anyway. In future releases of the products other corrections could be computed from the Ku and S-band ranges given by a particular retracker and then they should be applied only to the Ku-band range for that particular retracker. Table 2-6: Available ionospheric correction values in the CGDR. 1 Hz correction 18 Hz correction source ion_corr_doris_ku hz18_ion_corr_doris_ku DORIS ionospheric measurements GIM ionospheric model (from CMA 7.1 ion_corr_mod_ku hz18_ion_corr_mod_ku onwards) IMPORTANT: See Section "Smoothing the Ionosphere Correction". 41

42 2.3.8 Sea State Bias Correction The sea state bias correction is determined from the SGDR Ocean retracker ranges, wave heights and wind speeds, and available for both Ku and S-band in the SGDR, but for the reasons discussed above only the ku-band correction (sea_bias_ku and hz18_sea_bias_ku) is provided in the CGDR. As this correction is determined empirically, it is probable that the table used to determine the relationship of wave height and wind speed to sea state bias correction will be different for the different retrackers. However, this has not yet been determined and should be subject of further research, made possible by the availability of the retracked quantities from the various esurge retrackers Sea Surface Height and Sea Level Anomaly Sea surface height (SSH) is the height of the sea surface above the reference ellipsoid. It is calculated by subtracting the corrected range from the orbital altitude: Sea Surface Height = Altitude Corrected Range For most oceanographic purposes, the effects of non-geostrophic processes on the sea surface height will also be removed using geophysical corrections. Note: all the geophysical parameters here are heights, as defined in section 1.2, and as such are positive upwards. Corrected Sea Surface Height = Altitude Corrected Range Solid Earth Tide Height Geocentric Ocean Tide Height Pole Tide Height Inverted Barometer Height Correction HF fluctuations of the Sea Surface Note: for coastal applications, where comparison with tide gauges or other measurement systems is to be undertaken, not all the geophysical corrections may need to be applied. In particular, the inverted barometer correction may be omitted. A case in point for storm surge applications is the computation of the Total Water Level Envelope (TWLE), as discussed in the specific section below. The sea level anomaly (SLA), also referred to as Sea Surface Height Anomaly (SSHA), is defined here as the sea surface height minus the mean sea surface. It is given by: Sea Level Anomaly = Corrected Sea Surface Height Mean Sea Surface 42

43 The Mean Sea Surface used is the CLS01 model from the SGDR.The SLA contains information about: Real changes in ocean topography related to ocean currents Differences between the true dynamic response to atmospheric pressure and the applied inverse barometer and model high frequency response. Differences between tides and the tide models Differences between the mean sea surface model and the true mean sea surface at the altimeter location Un-modelled or mis-modelled measurement effects (skewness, sea state bias, altimeter errors, tropospheric corrections, ionospheric correction, etc.) Orbit errors There is naturally also random measurement noise. The available SSH, SSHA and TWLE in the esurge products are listed in Table 2-8 below. Altitude This is the orbital altitude (see parameter altitude in 1.2). This is available at 1 Hz (alt_cog_ellip) and 18 Hz (hz18_alt_cog_ellip). Corrected Range See section above Tide Effects The total tide effect on the sea surface height is the sum of three values from the CGDR: Tide Effect = Geocentric Ocean Tide + Solid Earth Tide + Pole Tide Of these, the largest component (and the most familiar to oceanographers) is the geocentric ocean tide, described in more detail below Geocentric Ocean Tide The geocentric ocean tide provided on the CGDR is the sum total of the ocean tide, with respect to the ocean bottom, and the loading tide height of the ocean bottom. Geocentric Ocean Tide = Ocean Tide + Load Tide The CGDR provides a choice of two geocentric ocean tide values, each also available interpolated to 18 Hz (see Table 2-7). Each uses a different model for the sum total of the ocean tide and loading tide heights from the diurnal and semidiurnal tides, but both include an equilibrium representation of the long-period ocean tides at all periods except for the zero frequency (permanent tide) term. Note: the CGDR also explicitly provides the loading tide height from each of the two models that are used to determine the two geocentric ocean tide values, at 1 Hz (tidal_load_ht_sol1, tidal_load_ht_sol2) and interpolated to 18 Hz (hz18_tidal_load_ht_sol1, hz18_tidal_load_ht_sol2). 43

44 Obviously, the geocentric ocean tide values and loading tide values should not be used simultaneously, since the loading tide height would be included twice. Table 2-7: Available ocean tide correction values in the CGDR 1 Hz correction 18 Hz correction source tot_geocen_ocn_tide_ht_sol1 hz18_tot_geocen_ocn_tide_ht_sol1 GOT00.2b tot_geocen_ocn_tide_ht_sol2 hz18_tot_geocen_ocn_tide_ht_sol2 FES2004 Solid Earth Tide As the Earth s liquid oceans respond to the gravitational pull of the Moon and the Sun, so too does the solid material that makes up the bulk of the planet, responding to lunisolar gravitational attraction. See for a comprehensive overview on Solid Earth Tides. Within the esurge Altimetry product, a single solution, at 1 Hz (solid_earth_tide_ht) and interpolated to 18 Hz (hz18_solid_earth_tide_ht), is provided. NOTE: Zero frequency (permanent tide) term is also not included in this parameter. Pole Tide Variations in the geocentric position of the earth's rotation axis (polar motion) cause deformation within the earth (Wahr, 1985). The ocean pole tide is the ocean response to the variation of both the solid Earth and the oceans to the centrifugal potential that is generated by small perturbations to the Earth's rotation axis. Modelling the pole tide requires knowledge of proportionality constants, the so-called Love numbers, and a time series of perturbations to the Earth's rotation axis, a quantity that is now measured routinely with space techniques. Within the esurge Altimetry product, a single solution, at 1 Hz (geocen_pole_tide_ht) and interpolated to 18 Hz (hz18_geocen_pole_tide_ht), is provided Total Water Level Envelope (TWLE) The TWLE is the quantity of choice in storm surge studies, as it represents the actual level that the water reaches for the combined effects of ocean dynamics, tidal effects and atmospheric forcing. In practise, the altimetric TWLE is computed as the Sea Surface Height Anomaly, i.e. from the SSH corrected for ionospheric delay, wet/dry tropospheric delays, sea state bias, loading tides and solid earth tides. Note that in contrast to SSHA, oceanic tides (ocean, pole, and long period tide) and the local response to atmospheric forcing (inverse barometer and high-frequency) are left in (not corrected for). The TWLE is therefore a measure of the actual water level experienced by an observer at the coast. 44

45 Table 2-8: Available SSH, SSHA and TWLE variables in the CGDR Quantity 1 Hz variable 18 Hz variable source SSH - uncorrected SSHA - uncorrected hz18_ssh_uncorrected hz18_ales_ssh_uncorrected hz18_ssha_uncorrected hz18_ales_ssha_uncorrected SGDR Ocean retracker ALES Retracker SGDR Ocean retracker ALES Retracker TWLE twle hz18_twle SGDR Ocean retracker hz18_ales_twle ALES Retracker Surface Air Pressure Effects The effect of the surface air pressure is split into two the inverse barometer, or the direct response of the ocean to atmospheric pressure, and the high frequency response of the ocean to changes in the atmospheric pressure. Both terms are included on the CGDR. Inverted Barometer Height Correction A single solution, at 1 Hz (inv_barom_corr) and interpolated to 18 Hz (hz18_inv_barom_corr), is provided. HF Fluctuations of the Sea Surface A single solution, at 1 Hz (dib_hf) and interpolated to 18 Hz (hz18_dib_hf), is provided Geophysical Surface - Mean Sea Surface or Geoid The geophysical fields Geoid (geoid_ht, hz18_geoid_ht, not currently provided in the CGDR) and Mean Sea Surface (m_sea_surf_ht, hz18_m_sea_surf_ht) are heights (ie distances above the reference ellipsoid), as is the Sea Surface Height. The quantity normally required for oceanography is the dynamic topography, or the height of the sea surface above the geoid. However, the properties of the geoid at high frequency are only known through inference from the mean sea surface, which approximates to the geoid. Hence, use of the geoid for determining dynamic topography is only appropriate for long-wavelength oceanographic features. Instead, it is more normal to reference to the mean sea surface, to determine the sea level anomaly (or the total water level envelope, which is also an anomaly). Fro this reason only the MSS is currently provided in the CGDR. 45

46 Flags and other quality control variables Some versions of the CGDRs may contain also a number of flags and some quality control variables such as the RMS or goodness of fit from the various retrackers. Flags and quality control variables are aimed at expert users, who can find full details in [AD 1][AD 7][AD 19] Mean Sea Surface and Adjustment of the Cross Track Gradient To study sea level changes between two dates, it is necessary to take the difference between sea surface heights from different cycles at the exact same latitude-longitude, to remove errors in the time-invariant geoid, which is not well known at short-wavelengths. However, the satellite ground track is allowed to drift by +1 km from its nominal position and so each repeat cycle of the satellite samples a different geoid profile. Differencing these profiles will introduce an error due to the poorly known cross-track geoid gradient. This error was estimated by [0] as about 2 cm km -1 over most of the ocean. However, this is effect is much larger where the expected geoid slopes are greater, eg over continental shelf slopes or coastal region. In addition, measurements are provided approximately every 1.1 s along the pass (about 7 km) for the 1 Hz data, or every s for the 18 Hz data (about 0.35 km). The position of these measurements along-track will be at different latitude-longitude locations on different cycles. Hence, even if the passes repeated exactly, it would be necessary to interpolate along the pass. The Mean Sea Surface height is known to much greater accuracy, courtesy of previous altimetry missions, than the geoid height. Hence the use of SSHA (or TWLE for the study of surges) is of benefit in looking at repeated data. Interpolation of SSHA to a repeated track location generates smaller errors than interpolating the Sea Surface Height. Over the open ocean it is common to interpolate data to a common set of along-track points, or "reference" track. Whilst this approach may be beneficial in the open ocean, the smaller spatial scales of the coastal zone might make this less beneficial, and alternative approaches, including the use of a locally generated mean sea surface might be more appropriate. An example of such use is in the X-TRACK processor (Roblou et al., 2007) Smoothing Ionosphere Correction Model and DORIS derived ionospheric corrections currently provided in the CGDRs do not need any smoothing. However, in future issues of the products, we may provide ionospheric range corrections from dual-frequency range measurements, either for Envisat (prior to the failure of the S-band, see 2.3.7) or for other missions such as Jason-1 and Jason-2. These dual-frequency ionospheric correction values are typically due to instrument noise effects. To reduce the noise, it is recommended that these parameters are averaged over 100 km or more (Imel, 1994). In order to provide a reversible correction, no averaging will be performed on the dual frequency ionospheric corrections when provided on the CGDR. Users may smooth the dual-frequency ionospheric correction along-track before applying, although care should be taken close to land. Typical and maximum smoothing scales are km for local times between 06 and 24 hours and km for local times between 00 and 06 hours. The shorter (longer) smoothing time is also more appropriate during times of high (low) solar activity 46

47 Generation of 1 Hz Averages Each 1 Hz value from the altimeter measurements (range, wave height and sigma-0) is derived from the linear regression of the valid 18 Hz parameters determined from the retracking algorithms. For coastal regions, users may well wish to generate averaged values from the 18 Hz measurements from the specific coastal retrackers. 2.4 Constraints on Use The esurge CGDR products are experimental, and provided primarily for research purposes to demonstrate their suitability for storm surge monitoring and forecasting. They have not yet been formally validated, and users should proceed with caution, particularly with regard to the coastal specific fields. While this handbook tries to be correct and complete, note that nothing can replace the information to be gained at conferences and other meetings from those using these data. The user must proceed with caution and explore the data with due diligence. 47

48 3. FURTHER INFORMATION AND CONTACTS For queries regarding Coastal Altimetry data, please do not hesitate to contact: Product Originator Contact: Dr. Paolo Cipollini For queries regarding the esurge Project, please do not hesitate to contact: General esurge contact: Website address: Phillip Harwood For queries regarding the European Space Agency (ESA) Data User Elements (DUE) Programme, see Note that the coastal altimetry community holds regular workshops where the science and techniques of coastal altimetry are reviewed and the various applications are showcased and discussed. See for more detail. 48

49 4. REFERENCES AND FURTHER READING Abdalla, S (2006). A wind retrieval algorithm for satellite radar altimeters. ECMWF. Technical Memorandum. Abramowitz, M and I M Stegun (1968). Handbook of mathematical functions with formulas, graphs, and mathematical tables. Dover, N.Y., 1046pp. Brenner, A C, C J Koblinsky and B D Beckley (1990). A Preliminary Estimate of Geoid-Induced Variations in Repeat Orbit Satellite Altimeter Observations. Journal of Geophysical Research, 95(C3): Brown, G. S. (1977). The average impulse response of a rough surface and its applications. IEEE J. Oceanic Eng., OE-2, Brown, S. (2010). A Novel Near-Land Radiometer Wet Path-Delay Retrieval Algorithm: Application to the Jason-2/OSTM Advanced Microwave Radiometer. IEEE Trans. On Geoscience and Remote Sensing, 48, , doi: /TGRS Caballero, I, J Gómez-Enri, G Navarro and P Villares (2011). Towards a validation of Envisat RA-2 high rate significant wave height in coastal systems: case study of the Gulf of Cadiz. 5 th EARSeL Workshop on Remote Sensing of the Coastal Zone, Prague, Czech Republic, 1 st 3 rd June, 2011 Callahan, P S (1984). Ionospheric Variations affecting Altimeter Measurements: A brief synopsis. Marine Geodesy, 8: Carrère, L (2003). Etude et modélisation de la réponse haute fréquence de l océan global aux forçage météorologiques. PhD. Carrère, L and F Lyard (2003). Modelling the barotropic response of the global ocean to atmospheric wind and pressure forcing comparisons with observations. Geophysical Research Letters, 30(6): Cartwright, D E and A C Edden (1973). Corrected tables of tidal harmonics. Geophysical Journal of the Royal Astronomical Society. 33: Cartwright, D E and R J Taylor (1971). New computations of the tide-generating potential. Geophysical Journal of the Royal Astronomical Society, 23: Cartwright, D E, R D Ray and B V Sanchez (1991). Oceanic tide maps and spherical harmonic coefficients from Geosat altimetry. Goddard Space Flight Center. NASA Tech. Memorandum : 74. Chambers, D P, B D Tapley and R H Stewart (1998). Reduction of geoid gradient error in ocean variability from satellite altimetry. Marine Geodesy, 21: Chandler, S.C. (1891) On the variation of latitude, I, Astronomical Journal, 11,

50 Chelton, D B, J C Ries, B J Haines, L-L Fu and P S Callahan (2001). Satellite Altimetry. In: Satellite Altimetry and Earth Sciences. L-L Fu and A Cazenave (Eds): Cipollini, P, J Benveniste et al. (2010) The Role of Altimetry in Coastal Observing Systems, in Proceedings of OceanObs 09: Sustained Ocean Observations and Information for Society (Vol. 2), Venice, Italy, September 2009, Hall, J., Harrison, D.E. & Stammer, D., Eds., ESA Publication WPP-306. Cotton et al., (2008) Development of SAR Altimetry Mode Studies and Applications over Ocean, Coastal Zones, and Inland Water SAMOSA State of the Art Assessment, 59 pp., available from CLS (2006). Design and assessment of a method to correct the Envisat RA-2 USO anomaly, Contract rep. re contract ESA/Esrin 19049/05/I-OL. Defrenne, D and J Benveniste (2004). A global land elevation and ocean bathymetry model from radar altimetry. QWG meeting minutes. Deng, X and W E Featherstone (2006). A coastal re-tracking system for satellite radar altimeter waveforms: Application to ERS-2 around Australia. Journal of Geophysical Research, 111(C06012), doi: /2005jc Desportes, C., E.Obligis, and L.Eymard (2007). On the wet tropospheric correction for altimetry in coastal regions, IEEE Trans. Geosci. Remote Sens., vol. 45, no. 7, pp Dumont, J P, V Rosmorduc, N Picot, S Desai, H Bonekamp, J Figa, J Lillibridge and R Scharroo (2009). OSTM/Jason-2 Products Handbook; Issue: 1 rev 4. CNES Rep No. SALP-MU-M-OP CN, August 3, 2009 Fernandes, M. J., C. Lázaro, A. L. Nunes, N. Pires, L. Bastos, V. B. Mendes (2010). GNSS-derived Path Delay: an approach to compute the wet tropospheric correction for coastal altimetry, IEEE Geosci. Rem. Sens Lett., vol. 7, no. 3, pp Francis, O and P Mazzega (1990). Global charts of ocean tide loading effects. Journal of Geophysical Research, 95: 11,411-11,424. García Lafuente J and J Ruiz (2007). The Gulf of Cádiz pelagic ecosystem: A review. Progress in Oceanography, 74(2-3): Gaspar, P and J P Florens (1998). Estimation of the sea state bias in radar altimeter measurements of sea level: Results from a new non parametric method. Journal of Geophysical Research, 103: Gaspar, P, S Labroue, F Ogor, G Lafitte, L Marchal and M Rafanel (2002). Improving non parametric estimates of the sea state bias in radar altimeter measurements of sea level. Journal of Atmospheric and Oceanic Technology, 19: Gille S T & C W Hughes (2001). Aliasing of high-frequency variability by altimetry: Evaluation from bottom pressure recorders. Geophysical Research Letters, 28 (9):

51 Gómez-Enri, J, C P Gommenginger, M A Srokosz, P G Challenor and J Benveniste (2007). Measuring global ocean wave skewness by re-tracking RA-2 ENVISAT waveforms. Journal of Atmospheric and Oceanic Technology, 24: Gómez-Enri, J, S Vignudelli, G D Quartly, C P Gommenginger, P Cipollini, P G Challenor and J Benveniste (2010), Modeling Envisat RA-2 waveforms in the coastal zone: Case-study of calm water contamination, IEEE Geosci. Rem. Sensing Lett., vol. 7, no. 3, pp Gommenginger, C., P. Thibaut, L. Fenoglio-Marc, G. Quartly, X. Deng, J. Gómez-Enri, P. Challenor, Y. Gao (2011). Re-tracking altimeter waveforms near the coasts - A review of re-tracking methods and some applications to coastal waveforms, in Coastal Altimetry, Eds S. Vignudelli, A. Kostianoy. P. Cipollini, J. Benveniste, Springer. Gross, R.S. The excitation of the Chandler wobble, Geophysical Research Letters, 27(15), Halimi, A, C Mailhes, Y-H Tourneret, P Thibaut, F Boy, (2012) Parameter Estimation for Peaky Altimetric Waveforms, IEEE Transactions on Ge.oscience and Remote Sensing. Hayne, O S (1980). Radar Altimeter Mean Return Waveforms from Near-Normal-Incidence Ocean Surface Scattering. IEEE Transactions: Antennae and Propagation, AP-28(5): Hernandez, F and P Schaeffer (2000). Altimetric Mean Sea Surfaces and Gravity Anomaly maps intercomparisons. CLS. AVI-NT CLS: 48. Hernandez, F and P Schaeffer (2001). The CLS01 Mean Sea Surface: A validation with the GSFC00.1 surface. Secondary The CLS01 Mean Sea Surface: A validation with the GSFC00.1 surface. Technical Report, CLS Ramonville St Agnes: 14pp. Imel, D (1994). Evaluation of the Topex/Poseidon dual-frequency ionosphere correction, Journal of Geophysical Research, 99(24): , 1994 Labroue, S (2005). RA-2 Ocean and MWR measurement long term monitoring report for WP3, Task2 SSB estimate for RA-2 altimeter. CLS_DOS-NT Labroue, S. and E. Obligis (2003). Neural network retrieval algorithm for the Envisat/MWR. CLS. ESA contract report (contract n 13681/99/NL/GD). CLS/DOS/NT/ Le Provost, C (2001). Ocean Tides. Satellite Altimetry and Earth Sciences. In: Satellite Altimetry and Earth Sciences. L-L Fu and A Cazenave (Eds): Le Provost, C, M. Genco, F Lyard, P Vincent and P Canceil (1995). Spectroscopy of the world ocean tides from a finite element hydrodynamic model. Journal of Geophysical Research, 99: Lefèvre, F (2002). Modélisation de la marée océanique à l'échelle globale par la méthode des éléments finis avec assimilation de données altimétriques. CLS. SALP-RP-MA-E CLS: 87. Lefèvre, F, F H Lyard, C Le Provost and E J O Schrama (2002). FES99: a global tide finite element solution assimilating tide gauge and altimetric information. Journal of Atmospheric and Oceanic Technology, 19:

52 Lemoine, F G, S C Kenyon, K Factor, R G Trimmer, N K Pavlis, D S Chinn, C M Cox, S M Klosko, S B Luthcke, M H Torrence, Y M Wang, R G Williamson, R H Rapp and T R Olson (1998). The Development of the joint NASA GSFC and NIMA Geopotential Model EGM96. NASA Goddard Space Flight Center. NASA/TP Letellier, T, F Lyard and F Lefèvre (2004). The new global tidal solution: FES2004. Ocean Surface Topography Science Team Meeting, St. Petersburg, Florida. Madsen, K. S., Høyer, J. L. and Tscherning, C. C. (2007). Near-coastal satellite altimetry: Sea surface height variability in the North Sea Baltic Sea area. Geophysical Research Letters, 34, L14601, doi: /2007gl Martini, A, P Feminias, G Alberti and M P Milagro-Perez (2005). RA-2 S-Band Anomaly: Detection and waveform reconstruction. Proc. of 2004 Envisat & ERS Symposium, Salzburg, Austria September 2004 (ESA SP-572). Passaro, M., Cipollini, P., Vignudelli, S., Quartly, G., Snaith, H. (2013). ALES: a multi-mission adaptive sub-waveform retracker for coastal and open ocean altimetry, Remote Sensing of Environment, Volume 145, 5 April 2014, Pages , Pavlis, N and R H Rapp (1990). The development of an isostatic gravitational model to degree 360 and its use in global gravity modeling. Geophysical Journal International, 100: Raney, R. K. (1998). The delay Doppler radar altimeter. IEEE Transactions on Geoscience and Remote Sensing, 36(5), Rapp, R H, R S Nerem, C K Shum, S M Klosko and R G Williamson (1991). Consideration of Permanent Tidal Deformation in the Orbit Determination and Data Analysis for the TOPEX/POSEIDON Mission. Goddard Space Flight Center. NASA Tech. Memorandum Rapp, R H, Y M Wang and N K Pavlis (1991). The Ohio State 1991 Geopotential and Sea Surface Topography Harmonic Coefficient Models. Dept. of Geodetic Science and Surveying, The Ohio State University Ray, R D (1999). A global ocean tide model from TOPEX/POSEIDON altimetry: GOT99.2. Goddard Space Flight Center. NASA Tech. Memorandum Ray, R D and B V Sanchez (1989). Radial deformation of the Earth by oceanic tidal loading. Goddard Space Flight Center. NASA Tech. Memorandum Ray, C., Martin-Puig, C., Clarizia, M.P., Ruffini, G., Dinardo, S., Gommenginger, C., Benveniste, J. "SAR Altimeter Backscattered Waveform Model," IEEE Transactions on Geoscience and Remote Sensing, vol.53, no.2, pp.911,919, Feb. 2015doi: /TGRS Rio, M-H and F. Hernandez (2004). A mean dynamic topography computed over the world ocean from altimetry, in-situ measurements, and a geoid model. Journal of Geophysical Research, 109(C12032). 52

53 Rio, M-H, P Schaeffer, J-M Lemoine, and F Hernandez (2005). Estimation of the ocean Mean Dynamic Topography through the combination of altimetric data, in-situ measurements and GRACE geoid: From global to regional studies. GOCINA international workshop, Luxembourg. Roblou L, F Lyard, M Le Hénaff and C Maraldi (2007): X-TRACK, A new processing tool for altimetry in coastal oceans. Proc. Envisat Symposium, Montreux, Switzerland. Rodriguez, E, Y Kim and J M Martin (1992). The effect of small-wave modulation on the electromagnetic bias. Journal of Geophysical Research, 97(C2): Smith, W H F and D T Sandwell (1994). Bathymetric prediction from dense satellite altimetry and spare shipboard bathymetry. Journal of Geophysical Research, 99: Stacey, F D (1977). Physics of the Earth. J. Wiley. Stammer, D, C Wunsch and R m Ponte (2000). De-aliasing of global high frequency barotropic motions in altimeter observations. Geophysical Research Letters, 27: Tierney, C, J Wahr, F Bryan and V Zlotnicki (2000). Short-period oceanic circulation: implications for satellite altimetry. Geophysical Research Letters, 27: Tournadre, J and J C Morland (1998). The effects of rain on TOPEX/POSEIDON Altimeter data. IEEE Trans. Geosci. Remote Sensing, 35: Vignudelli, S, A G Kostianoy, P Cipollini and J Benveniste (Editors) (2011), Coastal Altimetry, Springer- Verlag, Berlin Heidelberg, doi: / , 578 pp. Wahr, J M (1985). Deformation Induced by Polar Motion. Journal of Geophysical Research-Solid Earth and Planets, 90(B11): Wessel, P and W H F Smith (1996). A Global Self-consistent, Hierarchical, High-resolution Shoreline Database, Journal of Geophysical Research, 101(B4): Witter, D L and D B Chelton (1991). A Geosat altimeter wind speed algorithm and a method for altimeter wind speed algorithm development. Journal of Geophysical Research, 96: Wunsch, C (1972). Bermuda sea level in relation to tides, weather and baroclinic fluctuations. Reviews Geophysics Space Physics, 10: Yi, Y (1995). Determination of Gridded Mean Sea Surface from TOPEX, ERS-1 and GEOSAT Altimeter Data. Dept. of Geodetic Science and Surveying, The Ohio State University. 434:

54 4.1 Applicable Technical Documentation [AD 1] Envisat RA-2/MWR Product Handbook, Issue 2.2, 27 Feb 2007: [AD 2] Envisat RA-2/MWR Level 2 User Manual, v1 rev.2, 20/06/2006 [AD 3] Envisat-1 Product Specifications, ANNEX A: PRODUCT DATA CONVENTIONS PO-RS- MDA-GS-2009, Is.: 3, Rev.: D, Date: 05/05/2004 [AD 4] Envisat-1 Product Specifications, Volume 5: RA-2 Product Structure PO-RS-MDA-GS- 2009, Is.: 3, Rev.: D, Date: 22/11/2007 [AD 5] Envisat-1 Product Specifications, Volume 14: RA-2 Product Specifications PO-RS- MDA-GS-2009, Is.: 4, Rev.: C, Date: 30/01/2009 [AD 6] COASTALT Waveform Retracker Software Technical Specifications. COASTALT STS001 v1.2, 28 July [AD 7] COASTALT Product Specification v2.0 rev 3 12 July [AD 8] Envisat RA2/MWR ocean data validation and cross-calibration activities. Yearly report CLS.DOS/NT/ Issue 1 rev 1, June [AD 9] Envisat Altimetry Data Set Version 2.0 Level 1B and Level 2 processing upgrades. IDEAS-VEG-IPF-TSP-0543, Issue Feb [AD 10] COASTALT WP2 Technical Note Improvement of Corrections in Coastal Areas, NOCS report, 11 September 2008, 34pp. [AD 11] COASTALT WP3.1 Technical Note Coastal waveform retracking: definition development and prototyping, U. Cádiz, 27pp. [AD 12] Impact of the Envisat Mission Extension on SAR data, ESA Technical Note, revision 1.0, 12 October 2010 ( n_on_sar_data_-_1_01.pdf) [AD 13] Technical Note on Wet Tropospheric Corrections in Coastal Areas, COASTALT Deliverable 2.1b, v 1.2, 30 Jun 2009 [AD 14] Global assessment of GNSS-derived tropospheric corrections, COASTALT2 Deliverable 2.1a, COASTALT2-D21a-11, v 1.1, 26/07/2010 [AD 15] GPD output for CGDR for European coasts, COASTALT2 Deliverable 2.1b, COASTALT2-D21b-11, v 1.2, 08/02/2011. [AD 16] COASTALT EWP1 Deliverable D1.2a. Processor Improvements: Technical Note,, Version 1 25 July [AD 17] COASTALT EWP1 Deliverable D1.2b. Processor: Plug and Play User Guide COASTALT Processor version 2.0 revision 3, Version 1 25 July [AD 18] COASTALT Coastal Mask Tool User Manual Version 1, ESA/ESRIN Contract No /08/I-LG Contract report [AD 19] Envisat Coastal Altimetry Product Handbook, COASTALT2 Deliverable 4.1b, COASTALT2-D41b-201, (ESA ref: ENVI-DTEX-EOPS-TN ), v2.0.1, 16/09/

55 4.2 Relevant esurge Project Documents [Contract] ESA contract for esurge Ref: /11/I-LG - e-surge [SOW] Statement of Work, ref. EOP/SM2135, Issue 2, Revision 1, dated 21 October 2010 [TP] [CLA-1] [CLA-2] [CLA-3] [KO-MOM] [RB] [ICD] [TS] [DARD] [TN1] [PMP] esurge Technical Proposal, Ref. UK_672_eSurge, Issue 1.1, dated 8 June 2011 Three sets of clarifications, ref. UK_672/Clarifications/01, UK_672/Clarifications /02 and UK_672/Clarifications /03 Minutes of the negotiation meeting held on 8 June 2011, ref. esurge_mom_ko, Issue 1.0, dated 8 June 2011 esurge Requirement Baseline Document Ref esurge_d50 v2.1 esurge Interface Control Document Ref esurge_d120 v2.1 esurge Technical Specification Ref esurge_d110 v3.1 Data Access Requirements Document Ref esurge_d130 v1.0 Technical Note on Reanalysis Experiments Ref esurge_d150 v2.1 esurge Project Management Plan Ref esurge_d440 v2.0 55

56 ANNEX A: SATELLITE ALTIMETRY FACTS & CONVENTIONS A.1 Distance Conventions In order to reduce confusion in discussing altimeter measurements and corrections, the following terms are used in this document as defined below: Distance and Length are general terms with no special meaning in this document. Height is the distance above a reference surface. The reference surface used is the reference ellipsoid (see below). Positive is upwards (away from the centre of the earth). Range is the distance from the satellite to the surface of the Earth, as measured by the altimeter. Thus, the altimeter measurement is referred to as "range" or "altimeter range," not height. Altitude is the height of the satellite or altimeter, usually given as the height of the centre of mass of the satellite. This distance is computed from the satellite orbit (ephemeris) data. Sea Surface Height is the height of the sea surface. The sea surface height is the difference of the altimeter range from the satellite altitude. Strictly, it is the difference of the altimeter range, corrected for atmospheric delays, from the satellite altitude. The Geoid is an equipotential surface i.e. a surface along which the gravity potential remains constant. The shape of this surface is controlled by the earth s gravity field. Sea Surface Topography, or dynamic topography, is that part of the sea surface height caused by ocean currents. It is equal to the departure of the sea surface height from the geoid height, once geophysical effects, such as tides, atmospheric pressure effects and sea state bias have been removed. A.2 Orbits, revolutions, passes and repeat cycles An Orbit is one circuit of the earth by the satellite as measured from one ascending node crossing to the next. Every satellite mission has a characteristic orbit inclination (i.e. the inclination of the orbital plane with respect to the earth s equatorial plane), which sets the maximum (in absolute value) latitude, i.e. how close to the pole the orbit reaches. For instance, if the orbit inclination is 82, the satellite will cover from 82 S to 82 N. An ascending node occurs when the sub satellite point crosses the earth's equator (at an angle equal to the orbit inclination) going from south to north, i.e. during an ascending pass. A Revolution (REV) is synonymous with orbit. A Pass is half a revolution of the earth by the satellite from extreme latitude to the opposite extreme latitude. For altimetry data, it is normal to organize data by pass to avoid having data boundaries in the middle of the oceans. Ascending passes are odd numbered and descending passes are even numbered. For Envisat, an Ascending Pass begins at latitude 82 and ends at +82. A Descending Pass is the opposite (+82 to 82 ). And orbit with an inclination greater than 90 (retrograde orbits), say 90 +θ with θ>0, has the same latitudinal coverage of the corresponding orbit with an inclination of 90 θ (prograde orbit), only the orientation of the ascending and descending pass is reversed. For 56

57 instance, the inclination of the TOPEX/Poseidon and Jason-1/2 missions is 66 (i.e. prograde), which gives a coverage between 66 S and 66 N, with the ascending passes oriented SSW to NNE, while the inclination of Envisat is 98 (i.e. retrograde) which corresponds to a coverage between 82 S and 82 N but the ascending passes have an orientation SSE to NNW. Still taking Envisat as an example, from launch until 22 October 2010, Envisat maintained a 35 day exact repeat cycle. During this phase, after each repeat cycle of 1002 passes, taking 35 days, Envisat revisited the same ground-track within a margin of ±1 km. That means that every location along each pass of the Envisat ground-track was measured every 35 days. On 2 November 2010 (and until the end of the mission on 8 April 2012) Envisat began a new phase, with a 30 day near-repeat cycle. During this extension orbit phase, there were 862 passes per cycle. The inclination of the orbit was not maintained and hence the ground tracks are not repeated exactly, but with slight changes in inclination. The inclination drift induces a rotation of the orbital plane around 38 (North for descending tracks, South for ascending tracks), hence at these latitudes there was no longitudinal drift of the tracks and they will repeat almost exactly. For other latitudes, the passes would drift over time, as shown in figure A-1. The resultant drift at the equator was expected to be approximately 15 km over 3 years. Figure A-0-1: Example of 98.5 inclined orbit (Envisat - the passes shown are the descending passes) and impact of the inclination drift for the same track of two successive cycles (exaggerated inclination change) for the latest Envisat Phase. The passes are numbered 1 to 1002 for the 35-day phase and 1 to 862 for the 30-day phase, with pass number 1 defined as the pass that has its ascending node closest to zero longitude (actually E for the 35-day phase, 0.6 E, drifting west with time for the 30-day phase) and the cycles are defined so that this is the first pass within a cycle. 57

58 Table A-0-1: Envisat 35-day phase orbit parameters. Orbit parameter Value Orbits per Day 14 11/35 Repeat Cycle (days) 35 Orbits in Cycle 501 Orbit Period (min) Mean Local Solar Time at descending node 10:00 Inclination (deg) Semi-Major Axis [Orbit Radius] (km) Orbit Velocity (km/s) 7.45 Mean Altitude (km) Orbital Altitude Range (km) A.3 Reference ellipsoid The Reference Ellipsoid is a first-order definition of the non-spherical shape of the Earth as an ellipsoid of revolution. The reference ellipsoid used for the Envisat mission is the WGS84 ellipsoid, defined as having an equatorial radius of km and a flattening coefficient of 1/ A.4 Correction conventions All environmental and instrument corrections are computed so that they should be added to the quantity which they correct. That is, a correction is applied to a measured value by Corrected Quantity = Measured Value + Correction This means that a correction to the altimeter range for an effect that lengthens the apparent signal path (e.g., wet troposphere correction) is computed as a negative number. Adding this negative number to the uncorrected (measured) range reduces the range from its original value toward the correct value. A.5 Time convention Example: Corrected Range = Measured Range + Range Correction. Times are UTC and referenced to January 1, :00: (this is valid for Jason-2 too) 58

59 A.6 Unit convention Wherever possible, values are reported in SI units, as defined in the UD-units package (where the definitions exists), so as to be compliant with the CF-conventions. All distances and distance corrections are reported in meters, although this may require application of a scale factor, as reported in the product, with the exception of the distance from coast, which is reported in km. A.7 Flagging and editing Flags are used to convey quality information or operating modes. They are usually set to zero to mean OK and 1 for not OK. Any spare flags are set to zero. There may be exceptions, in which case a particular description of the flag's use is provided. For example, flags may be used to provide information on the operation mode of the instrument. The flag values and flag meanings of simple flags are defined in their attributes in the CGDR products. Additional editing is usually required to determine the data quality. In some cases, flags have been pre-determined, using quality criteria e.g. data exceeding given thresholds. However, in other cases, users will also need to apply their own data quality editing criteria 59

60 ANNEX B: FULL NetCDF DESCRIPTION OF VARIABLES IN THE CGDRs B.1 Envisat CGDR Example file:ra2_mws_2prf-p _203914_ _00259_51565_3453_aoi_010 {dimensions: time = 233 ; samples = 20 ; band_0 = 128 ; band_1 = 64 ; variables: double time(time) ; time:long_name = "time in sec since " ; time:standard_name = "Lo rate time" ; time:units = "seconds since :00:00" ; time:comment = "Determined from mdsr_time" ; int lat(time) ; lat:_fillvalue = ; lat:scale_factor = 1.e-06 ; lat:long_name = "Lo rate geodetic latitude" ; lat:standard_name = "latitude" ; lat:units = "degress north" ; lat:comment = "Lo rate latitude value, defined as the latitude of the source packet centre (i.e. average of blocks 9 and 10).It is not corrected for surface slope and so represents the orbit track position" ; lat:source = "SGDR MDSR field 4 + MDSR field 63" ; int lon(time) ; lon:_fillvalue = ; lon:scale_factor = 1.e-06 ; lon:long_name = "Lo rate geodetic longitude" ; lon:standard_name = "longitude" ; lon:units = "degrees east" ; lon:comment = "Lo rate longitude value, defined as the longitude of the source packet centre (i.e. average of blocks 9 and 10). It is not corrected for surface slope and so represents the orbit trackposition" ; lon:source = "SGDR MDSR field 5" ; int alt_cog_ellip(time) ; alt_cog_ellip:_fillvalue = ; alt_cog_ellip:scale_factor = ; alt_cog_ellip:long_name = "Altitude of CoG above reference ellipsoid" ; alt_cog_ellip:units = "m" ; alt_cog_ellip:comment = "Obtained by interpolating the Orbit State Vectors in the DORIS precise orbit files" ; alt_cog_ellip:source = "SGDR MDSR field 9" ; alt_cog_ellip:coordinates = "lon lat" ; short ku_band_ocean(time) ; ku_band_ocean:_fillvalue = 32767s ; ku_band_ocean:scale_factor = ; ku_band_ocean:long_name = "esurge Ku-band on board retracker range" ; ku_band_ocean:standard_name = "altimeter ranges" ; ku_band_ocean:units = "m" ; ku_band_ocean:coordinates = "lon lat" ; short ku_sig_wv_ht(time) ; ku_sig_wv_ht:_fillvalue = 32767s ; ku_sig_wv_ht:scale_factor = ; ku_sig_wv_ht:long_name = "Ku-band Significant wave height" ; ku_sig_wv_ht:standard_name ="sea_surface_wave_significant_height" ; ku_sig_wv_ht:units = "m" ; ku_sig_wv_ht:comment = "The 1Hz estimate from the 18Hz output ocean retracking" ; ku_sig_wv_ht:source = "SGDR MDSR field 55" ; ku_sig_wv_ht:coordinates = "lon lat" ; short sd_18hz_ku_swh(time) ; sd_18hz_ku_swh:_fillvalue = 32767s ; sd_18hz_ku_swh:scale_factor = ; sd_18hz_ku_swh:long_name = "Ku-band Significant wave height standard deviation" ; sd_18hz_ku_swh:standard_name = "sea_surface_wave_significant_height" ; sd_18hz_ku_swh:units = "m" ; sd_18hz_ku_swh:comment = "The 1Hz standard deviation of the 18Hz significatn wave height output" ; sd_18hz_ku_swh:coordinates = "lon lat" ; short inv_barom_corr(time) ; inv_barom_corr:_fillvalue = 32767s ; inv_barom_corr:scale_factor = ; inv_barom_corr:long_name = "Inverted barometer correction" ; inv_barom_corr:standard_name = "sea_surface_height_correction_due_to_air_pressure_at_low_frequency" ; 60

61 ; inv_barom_corr:units = "m" ; inv_barom_corr:source = "SGDR MDSR field 40" ; inv_barom_corr:coordinates = "lon lat" ; short mod_wet_tropo_corr(time) ; mod_wet_tropo_corr:_fillvalue = 32767s ; mod_wet_tropo_corr:scale_factor = ; mod_wet_tropo_corr:long_name = "Model wet tropospheric correction" ; mod_wet_tropo_corr:standard_name = "altimeter_range_correction_due_to_wet_troposphere" ; mod_wet_tropo_corr:units = "m" ; mod_wet_tropo_corr:comment = "From the computational grid of the ECMWF run" ; mod_wet_tropo_corr:source = "ECMWF model. SGDR MDSR field 41" ; mod_wet_tropo_corr:coordinates = "lon lat" ; short mod_dry_tropo_corr(time) ; mod_dry_tropo_corr:_fillvalue = 32767s ; mod_dry_tropo_corr:scale_factor = ; mod_dry_tropo_corr:long_name = "Model dry tropospheric correction" ; mod_dry_tropo_corr:standard_name = "altimeter_range_correction_due_to_dry_troposphere" ; mod_dry_tropo_corr:units = "m" ; mod_dry_tropo_corr:comment = "From the computational grid of the ECMWF run" ; mod_dry_tropo_corr:source = "ECMWF model, SGDR MDSR field 39" ; mod_dry_tropo_corr:coordinates = "lon lat" ; short mwr_wet_tropo_corr(time) ; mwr_wet_tropo_corr:_fillvalue = 32767s ; mwr_wet_tropo_corr:scale_factor = ; mwr_wet_tropo_corr:long_name = "MWR derived wet tropospheric correction" ; mwr_wet_tropo_corr:standard_name = "altimeter_range_correction_due_to_wet_troposphere" ; mwr_wet_tropo_corr:units = "m" ; mwr_wet_tropo_corr:comment = "Obtained with a neural algorithm from the 23.8GHz and 36.5GHz brightness temperatures (in K) interpolated to RA-2 time tag, and the ocean backscatter coefficient for Ku-band(db), not corrected for atmospheric attenuation" ; mwr_wet_tropo_corr:source = "Microwave radiometer, SGDR MDSR field 42" ; mwr_wet_tropo_corr:coordinates = "lon lat" ; short ion_corr_doris_ku(time) ; ion_corr_doris_ku:_fillvalue = 32767s ; ion_corr_doris_ku:scale_factor = ; ion_corr_doris_ku:long_name = "Ionospheric correction from DORIS on Ku-band" ; ion_corr_doris_ku:standard_name = "altimeter_range_correction_due_to_ionosphere" ; ion_corr_doris_ku:units = "m" ; ion_corr_doris_ku:comment = "Obtained form the DORIS daily maps of Total Electron Content" ; ion_corr_doris_ku:source = "DORIS TEC maps, SGDR MDSRfield 45" ; ion_corr_doris_ku:coordinates = "lon lat" ; short ion_corr_mod_ku(time) ; ion_corr_mod_ku:_fillvalue = 32767s ; ion_corr_mod_ku:scale_factor = ; ion_corr_mod_ku:long_name = "Ionospheric correction from model on Ku-band" ; ion_corr_mod_ku:standard_name = "altimeter_range_correction_due_to_ionosphere" ; ion_corr_mod_ku:units = "m" ; ion_corr_mod_ku:comment = "Obtained form the GIM model for products processed with CMA v7.1 or higher" ; ion_corr_mod_ku:source = "SGDR MDSR field 47" ; ion_corr_mod_ku:coordinates = "lon lat" ; short dib_hf(time) ; dib_hf:_fillvalue = 32767s ; dib_hf:scale_factor = ; dib_hf:long_name = "MOG2D HF contribution" ; dib_hf:standard_name = "sea_surface_height_correction_due_to_air_pressure_and_wind_at_high_frequency" dib_hf:units = "m" ; dib_hf:comment = "Difference between the MOG2D estimate and the inverse barometer" ; dib_hf:source = "SGDR MDSR field 51" ; dib_hf:coordinates = "lon lat" ; short sea_bias_ku(time) ; sea_bias_ku:_fillvalue = 32767s ; sea_bias_ku:scale_factor = ; sea_bias_ku:long_name = "Sea state bias corection on Ku-band" ; sea_bias_ku:units = "m" ; sea_bias_ku:comment = "Computed by biniliear interpolation" ; sea_bias_ku:source = "SGDR MDSR field 49" ; sea_bias_ku:coordinates = "lon lat" ; short tot_geocen_ocn_tide_ht_sol1(time) ; tot_geocen_ocn_tide_ht_sol1:_fillvalue = 32767s ; tot_geocen_ocn_tide_ht_sol1:scale_factor = ; tot_geocen_ocn_tide_ht_sol1:long_name = "Total geocentric ocean tide height (solution 1)" ; tot_geocen_ocn_tide_ht_sol1:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; tot_geocen_ocn_tide_ht_sol1:units = "m" ; tot_geocen_ocn_tide_ht_sol1:comment = "GOT00.2b ocean tide model which consists of independent near-global estimates of eight components (Q1,O1,P1,K1,N2,M2,S2 and K2)" ; tot_geocen_ocn_tide_ht_sol1:source = "GOT00.2b. SGDR MDSR field 101" ; tot_geocen_ocn_tide_ht_sol1:coordinates = "lon lat" ; short tot_geocen_ocn_tide_ht_sol2(time) ; tot_geocen_ocn_tide_ht_sol2:_fillvalue = 32767s ; tot_geocen_ocn_tide_ht_sol2:scale_factor = ; tot_geocen_ocn_tide_ht_sol2:long_name = "Total geocentric ocean tide height (solution 2)" ; 61

62 tot_geocen_ocn_tide_ht_sol2:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; tot_geocen_ocn_tide_ht_sol2:units = "m" ; tot_geocen_ocn_tide_ht_sol2:comment = "FES 2004 ocean tide model generated at LEGOS" ; tot_geocen_ocn_tide_ht_sol2:source = "FES SGDR MDSR field 102" ; tot_geocen_ocn_tide_ht_sol2:coordinates = "lon lat" ; short solid_earth_tide_ht(time) ; solid_earth_tide_ht:_fillvalue = 32767s ; solid_earth_tide_ht:scale_factor = ; solid_earth_tide_ht:long_name = "Solid earth tide height" ; solid_earth_tide_ht:standard_name = "sea_surface_amplitude_due_to_earth_tide" ; solid_earth_tide_ht:units = "m" ; solid_earth_tide_ht:source = "Cartwright and Taylor tidal potential. SGDR MDSR field 105" ; solid_earth_tide_ht:coordinates = "lon lat" ; short geocen_pole_tide_ht(time) ; geocen_pole_tide_ht:_fillvalue = 32767s ; geocen_pole_tide_ht:scale_factor = ; geocen_pole_tide_ht:long_name = "Geocentric pole tide height" ; geocen_pole_tide_ht:standard_name = "sea_surface_height_amplitude_due_to_pole_tide" ; geocen_pole_tide_ht:units = "m" ; geocen_pole_tide_ht:comment = "Geocentric tide height due to polar motion interpolated to 18Hz" ; geocen_pole_tide_ht:source = "Wahr [1985]. SGDR MDSR field 106" ; geocen_pole_tide_ht:coordinates = "lon lat" ; int m_sea_surf_ht(time) ; m_sea_surf_ht:_fillvalue = ; m_sea_surf_ht:scale_factor = ; m_sea_surf_ht:long_name = "Mean sea-surface height" ; m_sea_surf_ht:standard_name = "sea_level" ; m_sea_surf_ht:units = "m" ; m_sea_surf_ht:comment = "CLS01 mean sea surface estimated on 1/30 degree grid using a local inverse method" ; m_sea_surf_ht:source = "CLS01 mean sea surfqace. SGDR MDSR field 98" ; short ave_chirp_ku(time) ; ave_chirp_ku:_fillvalue = 32767s ; ave_chirp_ku:long_name = "Ku band chirp value" ; ave_chirp_ku:standard_name = "chirp_band" ; ave_chirp_ku:comment = "0 -> at least 1 record at 320MHz, 1 -> at least one record at 80MHz, 2 -> all input records are at 20MHz" ; ave_chirp_ku:source = "SGDR MDSR field 121" ; ave_chirp_ku:units = "" ; short long_period_ocn_tide_ht(time) ; long_period_ocn_tide_ht:_fillvalue = 32767s ; long_period_ocn_tide_ht:scale_factor = ; long_period_ocn_tide_ht:long_name = "Long period ocean Tide height" ; long_period_ocn_tide_ht:standard_name = "sea_surface_height_amplitude_due_to_equilibrium_ocean_tide" ; long_period_ocn_tide_ht:units = "m" ; long_period_ocn_tide_ht:comment = "This equilibrium tide is added to the total geocentric ocean tide" ; long_period_ocn_tide_ht:source = "Cartwright and Taylor tidal potential. SGDR MDSR field 103" ; long_period_ocn_tide_ht:coordinates = "lon lat" ; short tidal_load_ht_sol1(time) ; tidal_load_ht_sol1:_fillvalue = 32767s ; tidal_load_ht_sol1:scale_factor = ; tidal_load_ht_sol1:long_name = "Tidal loading height (solution 1)" ; tidal_load_ht_sol1:units = "m" ; tidal_load_ht_sol1:comment = "Tidal loading height induced by the ocean tide calculated from the GOT00.2 model" ; tidal_load_ht_sol1:source = "GOT00.2b. SGDR MDSR field 114" ; tidal_load_ht_sol1:coordinates = "lon lat" ; short tidal_load_ht_sol2(time) ; tidal_load_ht_sol2:_fillvalue = 32767s ; tidal_load_ht_sol2:scale_factor = ; tidal_load_ht_sol2:long_name = "Tidal loading height (solution 2)" ; tidal_load_ht_sol2:units = "m" ; tidal_load_ht_sol2:comment = "Tidal loading height induced by the ocean tide calculated from the FES2002 model" ; tidal_load_ht_sol2:source = "FES2002. SGDR MDSR field 104" ; tidal_load_ht_sol2:coordinates = "lon lat" ; int esurge_brown_range_ku(time) ; esurge_brown_range_ku:_fillvalue = ; esurge_brown_range_ku:scale_factor = ; esurge_brown_range_ku:long_name = "esurge 1Hz Ku band ocean ranges" ; esurge_brown_range_ku:standard_name = "altimeter_ranges" ; esurge_brown_range_ku:units = "m" ; esurge_brown_range_ku:comment = "From an ocean retracking algorithm applied to the 18Hz Ku band waveform and averaged to 1Hz" ; esurge_brown_range_ku:source = "Ocean retracker (Ku band) computed according to esurge retracker" ; esurge_brown_range_ku:coordinates = "lon lat" ; short esurge_brown_swh_ku(time) ; esurge_brown_swh_ku:_fillvalue = 32767s ; esurge_brown_swh_ku:scale_factor = ; esurge_brown_swh_ku:long_name = "esurge Ku-band Significant wave height" ; esurge_brown_swh_ku:standard_name = "sea_surface_wave_significant_height" ; 62

63 esurge_brown_swh_ku:units = "m" ; esurge_brown_swh_ku:comment = "The 1Hz estimate from the 18Hz output ocean esurge retracker" ; esurge_brown_swh_ku:source = "Computed according to esurge retracker" ; esurge_brown_swh_ku:coordinates = "lon lat" ; short twle(time) ; twle:_fillvalue = 32767s ; twle:scale_factor = ; twle:long_name = "Total water level envelope" ; twle:standard_name = "total_water_level_envelope" ; twle:units = "m" ; twle:comment = "Computed by esurge processor, averaged onto 1Hz" ; twle:source = "The altimetric TWLE is the Sea Level Anomaly corrected for ionospheric delay, wet/dry tropospheric delays, sea state bias, loading tides and solid earth tides. It still includes oceanic tides (ocean, pole, and long period tide) and the local response to atmospheric forcing (inverse barometer and high-frequency). The TWLE is therefore a measure of the actual water level experienced by an observer at the coast" ; twle:coordinates = "lon lat" ; double hz18_time(time, samples) ; hz18_time:long_name = "time in sec since " ; hz18_time:standard_name = "Hi rate time" ; hz18_time:units = "seconds since :00:00" ; hz18_time:comment = "Determined from 1Hz averaged time and 18Hz time differences from 1Hz time" ; int hz18_lat(time, samples) ; hz18_lat:_fillvalue = ; hz18_lat:scale_factor = 1.e-06 ; hz18_lat:long_name = "Hi rate geodetic latitude" ; hz18_lat:standard_name = "latitude" ; hz18_lat:units = "degrees north" ; hz18_lat:comment = "Reconstructed by adding the 18Hz slope corrected latitude difference to the 1Hz latitude value" ; hz18_lat:source = "SGDR MDSR field 4 + MDSR field 63" ; int hz18_lon(time, samples) ; hz18_lon:_fillvalue = ; hz18_lon:scale_factor = 1.e-06 ; hz18_lon:long_name = "Hi rate geodetic longitude" ; hz18_lon:standard_name = "longitude" ; hz18_lon:units = "degrees east" ; hz18_lon:comment = "Reconstructed by adding the 18Hz slope corrected longitude difference to the 1Hz longitude value" ; hz18_lon:source = "SGDR MDSR field 5 + MDSR field 64" ; int hz18_alt_cog_ellip(time, samples) ; hz18_alt_cog_ellip:_fillvalue = ; hz18_alt_cog_ellip:scale_factor = ; hz18_alt_cog_ellip:long_name = "18Hz altitude of CoG above reference ellipsoid" ; hz18_alt_cog_ellip:units = "m" ; hz18_alt_cog_ellip:comment = "Obtained by summing alt_cog_ellip and 18Hz differences from alt_cog_ellip (extracted from the input L1B records) provided on SGDR" ; hz18_alt_cog_ellip:source = "SGDR MDSR field 9 + SGDR MDSR field 10" ; hz18_alt_cog_ellip:coordinates = "hz18_lon hz18_lat" ; int hz18_ku_band_ocean(time, samples) ; hz18_ku_band_ocean:_fillvalue = ; hz18_ku_band_ocean:scale_factor = ; hz18_ku_band_ocean:long_name = "18Hz Ku band ocean ranges" ; hz18_ku_band_ocean:standard_name = "altimeter_ranges" ; hz18_ku_band_ocean:units = "m" ; hz18_ku_band_ocean:comment = "From an ocean retracking algorithm applied to the 18Hz Ku band waveform" ; hz18_ku_band_ocean:source = "Ocean retracker (Ku band). SGDR MDSR field 19" ; hz18_ku_band_ocean:coordinates = "hz18lon hz18lat" ; short hz18_ku_sig_wv_ht(time, samples) ; hz18_ku_sig_wv_ht:_fillvalue = 32767s ; hz18_ku_sig_wv_ht:scale_factor = ; hz18_ku_sig_wv_ht:long_name = "18Hz Ku-band Significant wave height" ; hz18_ku_sig_wv_ht:standard_name = "sea_surface_wave_significant_height" ; hz18_ku_sig_wv_ht:units = "m" ; hz18_ku_sig_wv_ht:comment = "Interpolated 1Hz estimate from the 18Hz output ocean retracking" ; hz18_ku_sig_wv_ht:source = "SGDR MDSR field 55" ; hz18_ku_sig_wv_ht:coordinates = "hz18lon hz18lat" ; short hz18_inv_barom_corr(time, samples) ; hz18_inv_barom_corr:_fillvalue = 32767s ; hz18_inv_barom_corr:scale_factor = ; hz18_inv_barom_corr:long_name = "Inverted barometer correction interpolated to 18Hz" ; hz18_inv_barom_corr:standard_name = "sea_surface_height_correction_due_to_air_pressure_at_low_frequency" ; hz18_inv_barom_corr:units = "m" ; hz18_inv_barom_corr:source = "SGDR MDSR field 40" ; hz18_inv_barom_corr:coordinates = "hz18lon hz18lat" ; short hz18_mod_wet_tropo_corr(time, samples) ; hz18_mod_wet_tropo_corr:_fillvalue = 32767s ; hz18_mod_wet_tropo_corr:scale_factor = ; hz18_mod_wet_tropo_corr:long_name = "Model wet tropospheric correction" ; hz18_mod_wet_tropo_corr:standard_name = "altimeter_range_correction_due_to_wet_troposphere" ; hz18_mod_wet_tropo_corr:units = "m" ; 63

64 hz18_mod_wet_tropo_corr:comment = "From the computational grid of the ECMWF run interpolated to 18Hz" ; hz18_mod_wet_tropo_corr:source = "ECMWF model. SGDR MDSR field 41" ; hz18_mod_wet_tropo_corr:coordinates = "hz18lon hz18lat" ; short hz18_mod_dry_tropo_corr(time, samples) ; hz18_mod_dry_tropo_corr:_fillvalue = 32767s ; hz18_mod_dry_tropo_corr:scale_factor = ; hz18_mod_dry_tropo_corr:long_name = "Model dry tropospheric correction" ; hz18_mod_dry_tropo_corr:standard_name = "altimeter_range_correction_due_to_dry_troposphere" ; hz18_mod_dry_tropo_corr:units = "m" ; hz18_mod_dry_tropo_corr:comment = "From the computational grid of the ECMWF run interpolated to 18Hz" ; hz18_mod_dry_tropo_corr:source = "ECMWF model, SGDR MDSR field 39" ; hz18_mod_dry_tropo_corr:coordinates = "hz18lon hz18lat"; short hz18_mwr_wet_tropo_corr(time, samples) ; hz18_mwr_wet_tropo_corr:_fillvalue = 32767s ; hz18_mwr_wet_tropo_corr:scale_factor = ; hz18_mwr_wet_tropo_corr:long_name = "MWR derived wet tropospheric correction" ; hz18_mwr_wet_tropo_corr:standard_name = "altimeter_range_correction_due_to_wet_troposphere"; hz18_mwr_wet_tropo_corr:units = "m" ; hz18_mwr_wet_tropo_corr:comment = "Obtained with a neural algorithm from the 23.8GHz and 36.5GHz brightness temperatures (in K) interpolated to RA-2 time tag, and the ocean backscatter coefficient for Ku-band(db), not corrected for atmospheric attenuation interpolated to 18Hz" ; hz18_mwr_wet_tropo_corr:source = "Microwave radiometer, SGDR MDSR field 42" ; hz18_mwr_wet_tropo_corr:coordinates = "hz18lon hz18lat"; short hz18_ion_corr_doris_ku(time, samples) ; hz18_ion_corr_doris_ku:_fillvalue = 32767s ; hz18_ion_corr_doris_ku:scale_factor = ; hz18_ion_corr_doris_ku:long_name = "Ionospheric correction from DORIS on Ku-band" ; hz18_ion_corr_doris_ku:standard_name = "altimeter_range_correction_due_to_ionosphere" ; hz18_ion_corr_doris_ku:units = "m" ; hz18_ion_corr_doris_ku:comment = "Obtained from the DORIS daily maps of Total Electron Content interpolated to 18Hz" ; hz18_ion_corr_doris_ku:source = "DORIS TEC maps, SGDR MDSR field 45" ; hz18_ion_corr_doris_ku:coordinates = "hz18lon hz18lat" ; short hz18_ion_corr_mod_ku(time, samples) ; hz18_ion_corr_mod_ku:_fillvalue = 32767s ; hz18_ion_corr_mod_ku:scale_factor = ; hz18_ion_corr_mod_ku:long_name = "Ionospheric correction from model on Ku-band" ; hz18_ion_corr_mod_ku:standard_name = "altimeter_range_correction_due_to_ionosphere" ; hz18_ion_corr_mod_ku:units = "m" ; hz18_ion_corr_mod_ku:comment = "Obtained form the GIM model for products processed with CMA v7.1 or higher interpolated to 18Hz" ; hz18_ion_corr_mod_ku:source = "SGDR MDSR field 47" ; hz18_ion_corr_mod_ku:coordinates = "hz18lon hz18lat" ; short hz18_dib_hf(time, samples) ; hz18_dib_hf:_fillvalue = 32767s ; hz18_dib_hf:scale_factor = ; hz18_dib_hf:long_name = "MOG2D HF contribution" ; hz18_dib_hf:standard_name = "sea_surface_height_correction_due_to_air_pressure_and_wind_at_high_frequency"; hz18_dib_hf:units = "m" ; hz18_dib_hf:comment = "Difference between the MOG2D estimate and the inverse barometer interpolated to 18Hz" ; hz18_dib_hf:source = "SGDR MDSR field 51" ; hz18_dib_hf:coordinates = "hz18lon hz18lat" ; short hz18_sea_bias_ku(time, samples) ; hz18_sea_bias_ku:_fillvalue = 32767s ; hz18_sea_bias_ku:scale_factor = ; hz18_sea_bias_ku:long_name = "Sea state bias correction on Ku-band" ; hz18_sea_bias_ku:units = "m" ; hz18_sea_bias_ku:comment = "Computed by biniliear interpolation interpolated to 18Hz" ; hz18_sea_bias_ku:source = "SGDR MDSR field 49" ; hz18_sea_bias_ku:coordinates = "hz18lon hz18lat" ; short hz18_tot_geocen_ocn_tide_ht_sol1(time, samples) ; hz18_tot_geocen_ocn_tide_ht_sol1:_fillvalue = 32767s ; hz18_tot_geocen_ocn_tide_ht_sol1:scale_factor = ; hz18_tot_geocen_ocn_tide_ht_sol1:long_name = "Total geocentric ocean tide height (solution 1)" ; hz18_tot_geocen_ocn_tide_ht_sol1:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; hz18_tot_geocen_ocn_tide_ht_sol1:units = "m" ; hz18_tot_geocen_ocn_tide_ht_sol1:comment = "GOT00.2b ocean tide model which consists of independent near-global estimates of eight components (Q1,O1,P1,K1,N2,M2,S2 and K2) interpolated to 18Hz" ; hz18_tot_geocen_ocn_tide_ht_sol1:source = "GOT00.2b. SGDR MDSR field 101" ; hz18_tot_geocen_ocn_tide_ht_sol1:coordinates = "hz18lon hz18lat" ; short hz18_tot_geocen_ocn_tide_ht_sol2(time, samples) ; hz18_tot_geocen_ocn_tide_ht_sol2:_fillvalue = 32767s ; hz18_tot_geocen_ocn_tide_ht_sol2:scale_factor = ; hz18_tot_geocen_ocn_tide_ht_sol2:long_name = "Total geocentric ocean tide height (solution 2)" ; hz18_tot_geocen_ocn_tide_ht_sol2:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; hz18_tot_geocen_ocn_tide_ht_sol2:units = "m" ; 64

65 hz18_tot_geocen_ocn_tide_ht_sol2:comment = "FES 2004 ocean tide model generated at LEGOS interpolated to 18Hz" ; hz18_tot_geocen_ocn_tide_ht_sol2:source = "FES SGDR MDSR field 102" ; hz18_tot_geocen_ocn_tide_ht_sol2:coordinates = "hz18lon hz18lat" ; short hz18_solid_earth_tide_ht(time, samples) ; hz18_solid_earth_tide_ht:_fillvalue = 32767s ; hz18_solid_earth_tide_ht:scale_factor = ; hz18_solid_earth_tide_ht:long_name = "Solid earth tide height" ; hz18_solid_earth_tide_ht:standard_name = "sea_surface_amplitude_due_to_earth_tide" ; hz18_solid_earth_tide_ht:units = "m" ; hz18_solid_earth_tide_ht:comment = "Solid earth tide height interpolated to 18Hz" ; hz18_solid_earth_tide_ht:source = "Cartwright and Taylortidal potential. SGDR MDSR field 105" ; hz18_solid_earth_tide_ht:coordinates = "hz18lon hz18lat"; short hz18_geocen_pole_tide_ht(time, samples) ; hz18_geocen_pole_tide_ht:_fillvalue = 32767s ; hz18_geocen_pole_tide_ht:scale_factor = ; hz18_geocen_pole_tide_ht:long_name = "Geocentric pole tide height" ; hz18_geocen_pole_tide_ht:standard_name = "sea_surface_height_amplitude_due_to_pole_tide" ; hz18_geocen_pole_tide_ht:units = "m" ; hz18_geocen_pole_tide_ht:comment = "Geocentric tideheight due to polar motion interpolated to 18Hz" ; hz18_geocen_pole_tide_ht:source = "Wahr [1985]. SGDR MDSR field 106" ; hz18_geocen_pole_tide_ht:coordinates = "hz18lon hz18lat"; int hz18_m_sea_surf_ht(time, samples) ; hz18_m_sea_surf_ht:_fillvalue = ; hz18_m_sea_surf_ht:scale_factor = ; hz18_m_sea_surf_ht:long_name = "Mean sea-surface height"; hz18_m_sea_surf_ht:standard_name = "sea_level" ; hz18_m_sea_surf_ht:units = "m" ; hz18_m_sea_surf_ht:comment = "CLS01 mean sea surface estimated on 1/30 degree grid using a local inverse method interpolated to 18Hz" ; hz18_m_sea_surf_ht:source = "CLS01 mean sea surface. SGDR MDSR field 98" ; short hz18_long_period_ocn_tide_ht(time, samples) ; hz18_long_period_ocn_tide_ht:_fillvalue = 32767s ; hz18_long_period_ocn_tide_ht:scale_factor = ; hz18_long_period_ocn_tide_ht:long_name = "Long period ocean Tide height" ; hz18_long_period_ocn_tide_ht:standard_name ="sea_surface_height_amplitude_due_to_equilibrium_ocean_tide" ; hz18_long_period_ocn_tide_ht:units = "m" ; hz18_long_period_ocn_tide_ht:comment = "This equilibrium tide is added to the total geocentric ocean tide interpolated to 18Hz" ; hz18_long_period_ocn_tide_ht:source = "Cartwright and Taylor tidal potential. SGDR MDSR field 103" ; hz18_long_period_ocn_tide_ht:coordinates = "hz18lon hz18lat" ; short hz18_tidal_load_ht_sol1(time, samples) ; hz18_tidal_load_ht_sol1:_fillvalue = 32767s ; hz18_tidal_load_ht_sol1:scale_factor = ; hz18_tidal_load_ht_sol1:long_name = "Tidal loading height (solution 1)" ; hz18_tidal_load_ht_sol1:units = "m" ; hz18_tidal_load_ht_sol1:comment = "Tidal loading height induced by the ocean tide calculated from the GOT00.2 model interpolated to 18Hz" ; hz18_tidal_load_ht_sol1:source = "GOT00.2b. SGDR MDSR field 114" ; hz18_tidal_load_ht_sol1:coordinates = "hz18lon hz18lat" ; short hz18_tidal_load_ht_sol2(time, samples) ; hz18_tidal_load_ht_sol2:_fillvalue = 32767s ; hz18_tidal_load_ht_sol2:scale_factor = ; hz18_tidal_load_ht_sol2:long_name = "Tidal loading height (solution 2)" ; hz18_tidal_load_ht_sol2:units = "m" ; hz18_tidal_load_ht_sol2:comment = "Tidal loading height induced by the ocean tide calculated from the FES2002 model interpolated to 18Hz" ; hz18_tidal_load_ht_sol2:source = "FES2002. SGDR MDSR field 104" ; hz18_tidal_load_ht_sol2:coordinates = "hz18lon hz18lat"; int hz18_esurge_brown_range_ku(time, samples) ; hz18_esurge_brown_range_ku:_fillvalue = ; hz18_esurge_brown_range_ku:scale_factor = ; hz18_esurge_brown_range_ku:long_name = "esurge 18Hz Ku band ocean ranges" ; hz18_esurge_brown_range_ku:standard_name = "altimeter_ranges" ; hz18_esurge_brown_range_ku:units = "m" ; hz18_esurge_brown_range_ku:comment = "From an ocean retracking algorithm applied to the 18Hz Ku band waveform" ; hz18_esurge_brown_range_ku:source = "Ocean retracker (Ku band) computed according to esurge retracker" ; hz18_esurge_brown_range_ku:coordinates = "hz18lon hz18lat" ; short hz18_esurge_brown_swh_ku(time, samples) ; hz18_esurge_brown_swh_ku:_fillvalue = 32767s ; hz18_esurge_brown_swh_ku:scale_factor = ; hz18_esurge_brown_swh_ku:long_name = "esurge Ku-band Significant wave height" ; hz18_esurge_brown_swh_ku:standard_name = "sea_surface_wave_significant_height" ; hz18_esurge_brown_swh_ku:units = "m" ; hz18_esurge_brown_swh_ku:comment = "18Hz output ocean esurge retracker" ; hz18_esurge_brown_swh_ku:source = "Computed according to esurge retracker" ; hz18_esurge_brown_swh_ku:coordinates = "lon lat" ; short hz18_twle(time, samples) ; hz18_twle:_fillvalue = 32767s ; hz18_twle:scale_factor = ; hz18_twle:long_name = "Total water level envelope" ; hz18_twle:standard_name = "total_water_level_envelope" ; 65

66 hz18_twle:units = "m" ; hz18_twle:comment = "Computed by esurge processor" ; hz18_twle:source = "The altimetric TWLE is the Sea Level Anomaly corrected for ionospheric delay, wet/dry tropospheric delays, sea state bias, loading tides and solid earth tides. It still includes oceanic tides (ocean, pole, and long period tide) and the local response to atmospheric forcing (inverse barometer and high-frequency). The TWLE is therefore a measure of the actual water level experienced by an observer at the coast" ; hz18_twle:coordinates = "hz18lon hz18lat" ; short hz18_ssh_uncorrected(time, samples) ; hz18_ssh_uncorrected:_fillvalue = 32767s ; hz18_ssh_uncorrected:scale_factor = ; hz18_ssh_uncorrected:units = "m" ; hz18_ssh_uncorrected:comment = "Computed by esurge processor" ; hz18_ssh_uncorrected:coordinates = "hz18lon hz18lat" ; short hz18_ssha_uncorrected(time, samples) ; hz18_ssha_uncorrected:_fillvalue = 32767s ; hz18_ssha_uncorrected:scale_factor = ; hz18_ssha_uncorrected:units = "m" ; hz18_ssha_uncorrected:comment = "Computed by esurge processor" ; hz18_ssha_uncorrected:coordinates = "hz18lon hz18lat" ; int hz18_ales_range_ku(time, samples) ; hz18_ales_range_ku:_fillvalue = ; hz18_ales_range_ku:scale_factor = ; hz18_ales_range_ku:units = "m" ; hz18_ales_range_ku:comment = "Computed by esurge processor" ; hz18_ales_range_ku:coordinates = "hz18lon hz18lat" ; short hz18_ales_swh_ku(time, samples) ; hz18_ales_swh_ku:_fillvalue = 32767s ; hz18_ales_swh_ku:scale_factor = ; hz18_ales_swh_ku:units = "m" ; hz18_ales_swh_ku:comment = "Computed by esurge processor" ; hz18_ales_swh_ku:coordinates = "hz18lon hz18lat" ; short hz18_ales_twle(time, samples) ; hz18_ales_twle:_fillvalue = 32767s ; hz18_ales_twle:scale_factor = ; hz18_ales_twle:long_name = "Total water level envelope" ; hz18_ales_twle:standard_name = "total_water_level_envelope" ; hz18_ales_twle:units = "m" ; hz18_ales_twle:comment = "Computed by esurge processor" ; hz18_ales_twle:source = "The altimetric TWLE is the Sea Level Anomaly corrected for ionospheric delay, wet/dry tropospheric delays, sea state bias, loading tides and solid earth tides. It still includes oceanic tides (ocean, pole, and long period tide) and the local response to atmospheric forcing (inverse barometer and high-frequency). The TWLE is therefore a measure of the actual water level experienced by an observer at the coast" ; hz18_ales_twle:coordinates = "hz18lon hz18lat" ; short hz18_ales_ssh_uncorrected(time, samples) ; hz18_ales_ssh_uncorrected:_fillvalue = 32767s ; hz18_ales_ssh_uncorrected:scale_factor = ; hz18_ales_ssh_uncorrected:units = "m" ; hz18_ales_ssh_uncorrected:comment = "Computed by esurge processor" ; hz18_ales_ssh_uncorrected:coordinates = "hz18lon hz18lat" ; short hz18_ales_ssha_uncorrected(time, samples) ; hz18_ales_ssha_uncorrected:_fillvalue = 32767s ; hz18_ales_ssha_uncorrected:scale_factor = ; hz18_ales_ssha_uncorrected:units = "m" ; hz18_ales_ssha_uncorrected:comment = "Computed by esurge processor" ; hz18_ales_ssha_uncorrected:coordinates = "hz18lon hz18lat" ; double waveform_band0(time, samples, band_0) ; waveform_band0:units = "" ; double halimi_fit_ku(time, samples, band_0) ; halimi_fit_ku:units = "" ; double hi_wfm_ales(time, samples, band_0) ; hi_wfm_ales:units = "" ; // global attributes: :band_0_name = "Ku" ; :band_1_name = "S" ; :sampling_lo = "1 Hz" ; :sampling_hi = "18 Hz" ; } 66

67 B.2 CryoSat-2 LRM CGDR Example file: CS_OFFL_SIR_LRM_ T160657_ T162105_B001_AOI_130 {dimensions: time = 573 ; samples = 20 ; band0 = 128 ; variables: int hi_alt_cog(time, samples) ; hi_alt_cog:_fillvalue = ; hi_alt_cog:scale_factor = ; hi_alt_cog:add_offset = 0. ; hi_alt_cog:long_name = "high rate altitude of CoG above reference ellipsoid" ; hi_alt_cog:units = "m" ; int hi_alt_rate(time, samples) ; hi_alt_rate:_fillvalue = ; hi_alt_rate:scale_factor = ; hi_alt_rate:add_offset = 0. ; hi_alt_rate:long_name = "high rate instantaneous altitude rate" ; hi_alt_rate:units = "m/2" ; short hi_corr_dib_hf(time, samples) ; hi_corr_dib_hf:_fillvalue = s ; hi_corr_dib_hf:scale_factor = ; hi_corr_dib_hf:add_offset = 0. ; hi_corr_dib_hf:long_name = "high rate MOG2D HF contribution" ; hi_corr_dib_hf:standard_name = "sea_surface_height_correction_due_to_air_pressure_and_wind_at_high_frequency"; hi_corr_dib_hf:units = "m" ; short hi_corr_ib(time, samples) ; hi_corr_ib:_fillvalue = s ; hi_corr_ib:scale_factor = ; hi_corr_ib:add_offset = 0. ; hi_corr_ib:long_name = "high rate inverted barometer correction" ; hi_corr_ib:standard_name ="sea_surface_height_correction_due_to_air_pressure_at_low_frequency" ; hi_corr_ib:units = "m" ; short hi_corr_iono_mod(time, samples) ; hi_corr_iono_mod:_fillvalue = s ; hi_corr_iono_mod:scale_factor = ; hi_corr_iono_mod:add_offset = 0. ; hi_corr_iono_mod:long_name = "high rate ionospheric correction from model" ; hi_corr_iono_mod:standard_name ="altimeter_range_correction_due_to_ionosphere" ; hi_corr_iono_mod:units = "m" ; short hi_corr_trop_dry_mod(time, samples) ; hi_corr_trop_dry_mod:_fillvalue = s ; hi_corr_trop_dry_mod:scale_factor = ; hi_corr_trop_dry_mod:add_offset = 0. ; hi_corr_trop_dry_mod:long_name = "high rate model dry tropospheric correction" ; hi_corr_trop_dry_mod:standard_name ="altimeter_range_correction_due_to_dry_troposphere" ; hi_corr_trop_dry_mod:units = "m" ; short hi_corr_trop_wet_mod(time, samples) ; hi_corr_trop_wet_mod:_fillvalue = s ; hi_corr_trop_wet_mod:scale_factor = ; hi_corr_trop_wet_mod:add_offset = 0. ; hi_corr_trop_wet_mod:long_name = "high rate model wet tropospheric correction" ; hi_corr_trop_wet_mod:standard_name = "altimeter_range_correction_due_to_wet_troposphere" ; hi_corr_trop_wet_mod:units = "m" ; int hi_h_mss(time, samples) ; hi_h_mss:_fillvalue = ; hi_h_mss:scale_factor = ; hi_h_mss:add_offset = 0. ; hi_h_mss:long_name = "high rate mean sea-surface height" ; hi_h_mss:units = "m" ; short hi_h_surge_ales(time, samples) ; hi_h_surge_ales:_fillvalue = s ; hi_h_surge_ales:scale_factor = ; hi_h_surge_ales:add_offset = 0. ; hi_h_surge_ales:long_name = "storm surge height" ; hi_h_surge_ales:units = "m" ; short hi_h_tide_load(time, samples) ; hi_h_tide_load:_fillvalue = s ; hi_h_tide_load:scale_factor = ; hi_h_tide_load:add_offset = 0. ; hi_h_tide_load:long_name = "high rate tidal loading height" ; hi_h_tide_load:units = "m" ; short hi_h_tide_ocean_long_period(time, samples) ; hi_h_tide_ocean_long_period:_fillvalue = s ; hi_h_tide_ocean_long_period:scale_factor = ; hi_h_tide_ocean_long_period:add_offset = 0. ; hi_h_tide_ocean_long_period:long_name = "high rate long period tide height" ; hi_h_tide_ocean_long_period:standard_name = "sea_surface_height_amplitude_due_to_equilibrium_ocean_tide" ; hi_h_tide_ocean_long_period:units = "m" ; 67

68 short hi_h_tide_ocean_tot_geocen(time, samples) ; hi_h_tide_ocean_tot_geocen:_fillvalue = s ; hi_h_tide_ocean_tot_geocen:scale_factor = ; hi_h_tide_ocean_tot_geocen:add_offset = 0. ; hi_h_tide_ocean_tot_geocen:long_name = "hi rate total geocentric ocean tide height" ; hi_h_tide_ocean_tot_geocen:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; hi_h_tide_ocean_tot_geocen:units = "m" ; short hi_h_tide_solid_earth(time, samples) ; hi_h_tide_solid_earth:_fillvalue = s ; hi_h_tide_solid_earth:scale_factor = ; hi_h_tide_solid_earth:add_offset = 0. ; hi_h_tide_solid_earth:long_name = "high rate solid earth tide height" ; hi_h_tide_solid_earth:standard_name = "sea_surface_height_amplitude_due_to_earth_tide" ; hi_h_tide_solid_earth:units = "m" ; short hi_h_twle_ales(time, samples) ; hi_h_twle_ales:_fillvalue = s ; hi_h_twle_ales:scale_factor = ; hi_h_twle_ales:add_offset = 0. ; hi_h_twle_ales:long_name = "high rate total water-level envelope from ales processor" ; hi_h_twle_ales:standard_name = "total_water_level_envelope" ; hi_h_twle_ales:units = "m" ; short hi_h_twle_brown(time, samples) ; hi_h_twle_brown:_fillvalue = s ; hi_h_twle_brown:scale_factor = ; hi_h_twle_brown:add_offset = 0. ; hi_h_twle_brown:long_name = "high rate total water-level envelope from Brown processor" ; hi_h_twle_brown:standard_name = "total_water_level_envelope" ; hi_h_twle_brown:units = "m" ; float hi_lat(time, samples) ; hi_lat:_fillvalue = e+36f ; hi_lat:long_name = "high rate geodetic latitude" ; hi_lat:standard_name = "latitude" ; hi_lat:units = "degrees_north" ; float hi_lon(time, samples) ; hi_lon:_fillvalue = e+36f ; hi_lon:long_name = "high rate longitude" ; hi_lon:standard_name = "longitude" ; hi_lon:units = "degrees_east" ; double hi_sat_pitch(time, samples) ; hi_sat_pitch:_fillvalue = e+36 ; hi_sat_pitch:long_name = "high rate satellite pitch" ; hi_sat_pitch:units = "degrees" ; double hi_sat_roll(time, samples) ; hi_sat_roll:_fillvalue = e+36 ; hi_sat_roll:long_name = "high rate satellite roll angle"; hi_sat_roll:units = "degrees" ; double hi_sat_vel(time, samples) ; hi_sat_vel:_fillvalue = e+36 ; hi_sat_vel:long_name = "high rate satellite velocity" ; hi_sat_vel:units = "m/s" ; short hi_swh_ales(time, samples) ; hi_swh_ales:_fillvalue = s ; hi_swh_ales:scale_factor = ; hi_swh_ales:add_offset = 0. ; hi_swh_ales:long_name = "high rate significant wave height from ALES retracker" ; hi_swh_ales:standard_name = "sea_surface_wave_significant_height" ; hi_swh_ales:units = "m" ; short hi_swh_brown(time, samples) ; hi_swh_brown:_fillvalue = s ; hi_swh_brown:scale_factor = ; hi_swh_brown:add_offset = 0. ; hi_swh_brown:long_name = "high rate significant wave height from Brown retracker" ; hi_swh_brown:standard_name = "sea_surface_wave_significant_height" ; hi_swh_brown:units = "m" ; double hi_time(time, samples) ; hi_time:_fillvalue = e+36 ; hi_time:long_name = "high rate time in sec since " ; hi_time:standard_name = "time" ; hi_time:units = "sec since " ; int alt_cog(time) ; alt_cog:_fillvalue = ; alt_cog:scale_factor = ; alt_cog:add_offset = 0. ; alt_cog:long_name = "altitude of CoG above reference ellipsoid" ; alt_cog:units = "m" ; alt_cog:source = "C2 MDSR field 9" ; short corr_dib_hf(time) ; corr_dib_hf:_fillvalue = s ; corr_dib_hf:scale_factor = ; corr_dib_hf:add_offset = 0. ; corr_dib_hf:long_name = "MOG2D high frequency contribution" ; corr_dib_hf:standard_name = "sea_surface_height_correction_due_to_air_pressure_and_wind_at_high_frequency"; corr_dib_hf:units = "m" ; corr_dib_hf:source = "MDSR field 38" ; corr_dib_hf:ancillary_variables = "flag_corr flag_corr_error" ; short corr_ib(time) ; 68

69 corr_ib:_fillvalue = s ; corr_ib:scale_factor = ; corr_ib:add_offset = 0. ; corr_ib:long_name = "inverted barometer correction" ; corr_ib:standard_name = "sea_surface_height_correction_due_to_air_pressure_at_low_frequency" ; corr_ib:units = "m" ; corr_ib:source = "MDSR field 37" ; corr_ib:ancillary_variables = "flag_corr flag_corr_error" ; short corr_iono_mod(time) ; corr_iono_mod:_fillvalue = s ; corr_iono_mod:scale_factor = ; corr_iono_mod:add_offset = 0. ; corr_iono_mod:long_name = "ionospheric correction from model" ; corr_iono_mod:standard_name = "altimeter_range_correction_due_to_ionosphere" ; corr_iono_mod:units = "m" ; corr_iono_mod:source = "MDSR field 40" ; corr_iono_mod:ancillary_variables = "flag_corr flag_corr_error" ; short corr_trop_dry_mod(time) ; corr_trop_dry_mod:_fillvalue = s ; corr_trop_dry_mod:scale_factor = ; corr_trop_dry_mod:add_offset = 0. ; corr_trop_dry_mod:long_name = "model dry tropospheric correction" ; corr_trop_dry_mod:standard_name = "altimeter_range_correction_due_to_dry_troposphere" ; corr_trop_dry_mod:units = "m" ; corr_trop_dry_mod:source = "MDSR field 35" ; corr_trop_dry_mod:ancillary_variables = "flag_corr flag_corr_error" ; short corr_trop_wet_mod(time) ; corr_trop_wet_mod:_fillvalue = s ; corr_trop_wet_mod:scale_factor = ; corr_trop_wet_mod:add_offset = 0. ; corr_trop_wet_mod:long_name = "model wet tropospheric correction" ; corr_trop_wet_mod:standard_name = "altimeter_range_correction_due_to_wet_troposphere" ; corr_trop_wet_mod:units = "m" ; corr_trop_wet_mod:source = "MDSR field 36" ; corr_trop_wet_mod:ancillary_variables = "flag_corr flag_corr_error" ; int h_mss(time) ; h_mss:_fillvalue = ; h_mss:scale_factor = ; h_mss:add_offset = 0. ; h_mss:long_name = "mean sea-surface height" ; h_mss:units = "m" ; short h_tide_load(time) ; h_tide_load:_fillvalue = s ; h_tide_load:scale_factor = ; h_tide_load:add_offset = 0. ; h_tide_load:long_name = "tidal loading height" ; h_tide_load:units = "m" ; short h_tide_ocean_long_period(time) ; h_tide_ocean_long_period:_fillvalue = s ; h_tide_ocean_long_period:scale_factor = ; h_tide_ocean_long_period:add_offset = 0. ; h_tide_ocean_long_period:long_name = "long period Tide height" ; h_tide_ocean_long_period:standard_name = "sea_surface_height_amplitude_due_to_equilibrium_ocean_tide" ; h_tide_ocean_long_period:units = "m" ; short h_tide_ocean_tot_geocen(time) ; h_tide_ocean_tot_geocen:_fillvalue = s ; h_tide_ocean_tot_geocen:scale_factor = ; h_tide_ocean_tot_geocen:add_offset = 0. ; h_tide_ocean_tot_geocen:long_name = "total geocentric ocean tide height" ; h_tide_ocean_tot_geocen:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; h_tide_ocean_tot_geocen:units = "m" ; short h_tide_pole_geocen(time) ; h_tide_pole_geocen:_fillvalue = s ; h_tide_pole_geocen:scale_factor = ; h_tide_pole_geocen:add_offset = 0. ; h_tide_pole_geocen:long_name = "geocentric pole tide height" ; h_tide_pole_geocen:standard_name = "sea_surface_height_amplitude_due_to_pole_tide" ; h_tide_pole_geocen:units = "m" ; short h_tide_solid_earth(time) ; h_tide_solid_earth:_fillvalue = s ; h_tide_solid_earth:scale_factor = ; h_tide_solid_earth:add_offset = 0. ; h_tide_solid_earth:long_name = "solid earth tide height" ; h_tide_solid_earth:standard_name = "sea_surface_height_amplitude_due_to_earth_tide" ; h_tide_solid_earth:units = "m" ; float lat(time) ; lat:_fillvalue = e+36f ; lat:long_name = "geodetic latitude" ; lat:standard_name = "latitude" ; lat:units = "degrees_north" ; lat:source = "C2 MDSR field 7" ; float lon(time) ; lon:_fillvalue = e+36f ; lon:long_name = "longitude" ; lon:standard_name = "longitude" ; 69

70 } lon:units = "degrees_east" ; lon:source = "C2 MDSR field 8" ; short swh_ocean(time) ; swh_ocean:_fillvalue = s ; swh_ocean:scale_factor = ; swh_ocean:add_offset = 0. ; swh_ocean:long_name = "significant wave height fromocean retracker" ; swh_ocean:standard_name = "sea_surface_wave_significant_height" ; swh_ocean:units = "m" ; double time(time) ; time:_fillvalue = e+36 ; time:long_name = "time in sec since " ; time:standard_name = "time" ; time:units = "sec since " ; time:source = "C2 MDSR field 1" ; 70

71 B.3 CryoSat-2 SAR CGDR Example file: CS_OFFL_SIR_SAR_1B_ T230535_ T230712_B001_AOI_010 {dimensions: time = 110 ; samples = 20 ; variables: double hi_alt_cog(time, samples) ; hi_alt_cog:_fillvalue = e+36 ; hi_alt_cog:long_name = "altitude of CoG above reference ellipsoid" ; hi_alt_cog:units = "m" ; hi_alt_cog:source = "MDSR field 9" ; double hi_alt_rateg(time, samples) ; hi_alt_rateg:_fillvalue = e+36 ; hi_alt_rateg:long_name = "instantaneous altitude rate" ; hi_alt_rateg:units = "m/s" ; hi_alt_rateg:source = "MDSR field 10" ; short hi_corr_ib(time, samples) ; hi_corr_ib:_fillvalue = s ; hi_corr_ib:scale_factor = ; hi_corr_ib:long_name = "inverted barometric correction" ; hi_corr_ib:standard_name = "sea_surface_height_correction_due_to_air_pressure_at_low_frequency" ; hi_corr_ib:units = "m" ; hi_corr_ib:source = "MDSR field 37" ; short hi_corr_iono_gps(time, samples) ; hi_corr_iono_gps:_fillvalue = s ; hi_corr_iono_gps:scale_factor = ; hi_corr_iono_gps:long_name = "ionospheric correction from gps" ; hi_corr_iono_gps:standard_name = "altimeter_range_correction_due_to_ionosphere" ; hi_corr_iono_gps:units = "m" ; hi_corr_iono_gps:source = "MDSR field 39" ; short hi_corr_iono_mod(time, samples) ; hi_corr_iono_mod:_fillvalue = s ; hi_corr_iono_mod:scale_factor = ; hi_corr_iono_mod:long_name = "ionospheric correction from model" ; hi_corr_iono_mod:standard_name = "altimeter_range_correction_due_to_ionosphere" ; hi_corr_iono_mod:units = "m" ; hi_corr_iono_mod:source = "MDSR field 40" ; short hi_corr_trop_dry_mod(time, samples) ; hi_corr_trop_dry_mod:_fillvalue = s ; hi_corr_trop_dry_mod:scale_factor = ; hi_corr_trop_dry_mod:long_name = "model dry tropospheric correction" ; hi_corr_trop_dry_mod:standard_name = "altimeter_range_correction_due_to_dry_troposphere"; hi_corr_trop_dry_mod:units = "m" ; hi_corr_trop_dry_mod:source = "MDSR field 35" ; short hi_corr_trop_wet_mod(time, samples) ; hi_corr_trop_wet_mod:_fillvalue = s ; hi_corr_trop_wet_mod:scale_factor = ; hi_corr_trop_wet_mod:long_name = "model wet tropospheric correction" ; hi_corr_trop_wet_mod:standard_name = "altimeter_range_correction_due_to_wet_troposphere"; hi_corr_trop_wet_mod:units = "m" ; hi_corr_trop_wet_mod:source = "MDSR field 36" ; double hi_h_corrected_samosa(time, samples) ; hi_h_corrected_samosa:_fillvalue = e+36 ; hi_h_corrected_samosa:long_name = "Corrected Sea Surface height" ; hi_h_corrected_samosa:units = "m" ; hi_h_corrected_samosa:source = "NOC CryoSat processor;" ; short hi_h_tide_load_sol1(time, samples) ; hi_h_tide_load_sol1:_fillvalue = s ; hi_h_tide_load_sol1:scale_factor = ; hi_h_tide_load_sol1:long_name = "tidal loading height (solution 1)" ; hi_h_tide_load_sol1:units = "m" ; hi_h_tide_load_sol1:source = "MDSR field 43" ; short hi_h_tide_ocean_long_period(time, samples) ; hi_h_tide_ocean_long_period:_fillvalue = s ; hi_h_tide_ocean_long_period:scale_factor = ; hi_h_tide_ocean_long_period:long_name = "long period tide height" ; hi_h_tide_ocean_long_period:standard_name = "sea_surface_height_amplitude_due_to_equilibrium_ocean_tide" ; hi_h_tide_ocean_long_period:units = "m" ; hi_h_tide_ocean_long_period:source = "MDSR field 42" ; short hi_h_tide_ocean_tot_geocen_sol1(time, samples) ; hi_h_tide_ocean_tot_geocen_sol1:_fillvalue = s ; hi_h_tide_ocean_tot_geocen_sol1:scale_factor = ; hi_h_tide_ocean_tot_geocen_sol1:long_name = "total geocentric ocean tide height (solution 1)" ; hi_h_tide_ocean_tot_geocen_sol1:standard_name = "sea_surface_height_amplitude_due_to_geocentric_ocean_tide" ; hi_h_tide_ocean_tot_geocen_sol1:units = "m" ; hi_h_tide_ocean_tot_geocen_sol1:source = "MDSR field 41" ; short hi_h_tide_pole_geocen(time, samples) ; hi_h_tide_pole_geocen:_fillvalue = s ; hi_h_tide_pole_geocen:scale_factor = ; 71

72 hi_h_tide_pole_geocen:long_name = "geocentric pole tide height" ; hi_h_tide_pole_geocen:standard_name = "sea_surface_height_amplitude_due_to_pole_tide" ; hi_h_tide_pole_geocen:units = "m" ; hi_h_tide_pole_geocen:source = "MDSR field 45" ; short hi_h_tide_solid_earth(time, samples) ; hi_h_tide_solid_earth:_fillvalue = s ; hi_h_tide_solid_earth:scale_factor = ; hi_h_tide_solid_earth:long_name = "solid earth tide height" ; hi_h_tide_solid_earth:standard_name = "sea_surface_height_amplitude_due_to_earth_tide" ; hi_h_tide_solid_earth:units = "m" ; hi_h_tide_solid_earth:source = "MDSR field 44" ; double hi_h_twle_samosa3(time, samples) ; hi_h_twle_samosa3:_fillvalue = e+36 ; hi_h_twle_samosa3:long_name = "total water level envelope" ; hi_h_twle_samosa3:units = "m" ; hi_h_twle_samosa3:source = "NOC CryoSat processor; hi_h_corrected_samosa3 hi_mss" ; double hi_lat(time, samples) ; hi_lat:_fillvalue = e+36 ; hi_lat:long_name = "geodetic latitude" ; hi_lat:standard_name = "latitude" ; hi_lat:units = "degrees_north" ; hi_lat:source = "MDSR field 7" ; hi_lat:_coordinateaxistype = "Lat" ; double hi_lon(time, samples) ; hi_lon:_fillvalue = e+36 ; hi_lon:long_name = "geodetic longitude" ; hi_lon:standard_name = "longitude" ; hi_lon:units = "degrees_east" ; hi_lon:source = "MDSR field 8" ; hi_lon:_coordinateaxistype = "Lon" ; double hi_range_corrected_samosa(time, samples) ; hi_range_corrected_samosa:_fillvalue = e+36 ; hi_range_corrected_samosa:long_name = "Corrected Range" ; hi_range_corrected_samosa:units = "m" ; hi_range_corrected_samosa:source = "NOC CryoSat Processor: range corrected for ocean_loading_tide solidearth_tide wet_trop dry_trop gim_ion" ; double hi_range_samosa(time, samples) ; hi_range_samosa:_fillvalue = e+36 ; hi_range_samosa:long_name = "uncorrected range" ; hi_range_samosa:units = "m" ; hi_range_samosa:source = "NOC CryoSat Processor" ; double hi_sat_pitch(time, samples) ; hi_sat_pitch:_fillvalue = e+36 ; hi_sat_pitch:long_name = "satellite pitch" ; hi_sat_pitch:units = "degrees" ; double hi_sat_roll(time, samples) ; hi_sat_roll:_fillvalue = e+36 ; hi_sat_roll:long_name = "satellite roll angle" ; hi_sat_roll:units = "degrees" ; double hi_sat_vel(time, samples) ; hi_sat_vel:_fillvalue = e+36 ; hi_sat_vel:long_name = "satellite velocity" ; hi_sat_vel:units = "m/s" ; hi_sat_vel:source = "MDSR field 11" ; double hi_swh_samosa(time, samples) ; hi_swh_samosa:_fillvalue = e+36 ; hi_swh_samosa:long_name = "significant wave height from CryoSat processor" ; hi_swh_samosa:units = "m" ; hi_swh_samosa:source = "NOC CryoSat Processor" ; double hi_time(time, samples) ; hi_time:_fillvalue = e+36 ; hi_time:long_name = "time in seconds since " ; hi_time:standard_name = "time" ; hi_time:units = "sec since " ; hi_time:source = "MDSR field 1" ; double time(time) ; time:_fillvalue = e+36 ; time:long_name = "time in seconds since " ; time:standard_name = "time" ; time:units = "seconds since " ; time:source = "MDSR field 1" ; // global attributes: :Conventions = "CF-1.6" ; } 72

73 ANNEX C: THE ESURGE PROJECT C.1 About esurge Despite the potential utility of satellite data, the storm surge community has not made as much use of it as they could. Largely this is due to the lack of easy data access. Different datasets are stored in different locations, in different data formats and with different access requirements. esurge aims to change this, bringing relevant datasets together in an east to use, web-accessible database of data products, downloadable in a standardised format. The esurge project is being run in two phases. During the initial Development Phase (Phase 1) we have built the database, known as SEARS (Surge Event Analysis and Repository Service), and populated it with initial data for a selection of historical surge events. This will give a useful library that can be used for assessing and improving the performance of numerical models. Whilst most of the datasets are already available, and just need to be imported into the database, others are being created during the project. Following the launch of the SEARS database, esurge will move into a Service Demonstration phase (Phase 2). During this phase we will continue to add more historical data, but will also look at making data available for surge events as they occur. The aim is to show that it is feasible to provide satellite data in near real time, so that it could potentially be used in forecasting and warning systems. It is important to note that esurge is not itself a forecasting and warning system, it is a system to make data available to forecasters. There are dedicated agencies (such as the UK Environment Agency) whose role it is to warn of impending flooding. Making the data available is just part of the process of getting people to use it; we must also show the value of the data. To this end our partners at DMI and NOC will perform a series of experiments, focussing on the North Sea and North Indian Ocean. These will take existing models, such as DMI s HBM model and NOC s operational CS3X surge model, and will look at how incorporating satellite data could improve the models hindcast accuracy. These experiments will also investigate the best way to incorporate satellite data into models. This is a complex subject, and we do not expect to be able to resolve it in this project, but we aim to pave the way for future work 73

74 C.2 The esurge Consortium The esurge consortium consists of CGI (UK), the National Oceanography Centre (UK), the Danish Meteorological Institute (DK), University College Cork s Coastal and Marine Research Centre (IRL) and the Royal Dutch Meteorological Institute (NL). Consultants to Government and Industry (CGI) was founded in 1976 and is a global IT and business process services provider delivering business consulting, systems integration and outsourcing services. With 72,000 professionals operating in 400 offices in 40 countries, CGI fosters local accountability for client success while bringing global delivery capabilities to clients front doors. CGI applies a disciplined and creative approach to achieve an industry-leading track record of on-time, on-budget projects and to help clients leverage current investments while adopting new technology and business strategies. As a result of this approach, our average client satisfaction score for the past 10 years has measured consistently higher than 9 out of 10. We have a dedicated international Space and Satcoms business with over 300 specialists and a long track record in delivering mission critical software systems across the Space sector, and in particular for Navigation and GNSS systems. We have worked on many ESA Earth Observation projects, including GlobWave, CCI, GECA, PALSAR and many others. The National Oceanography Centre (NOC) is a wholly owned centre of the Natural Environment Research Council (NERC). The NOC was formed by bringing together the NERC-managed activity at Liverpool s Proudman Oceanographic Laboratory and the National Oceanography Centre, Southampton, creating the UK s leading institution for sea level science, coastal and deep ocean research and technology development. The NOC hosts both the National Tidal and Sea Level Facility, and the Permanent Service for Mean Sea Level (since 1933), and contributes to the Storm Tide Forecasting Service (STFS), developing operational tide-surge models that provide UK coastal flood warning (in partnership with the Met Office and the Environment Agency). It has been at the forefront in developing interfaces to data sources and information. NOC have been involved in ESA funded projects such as COASTALT, GlobColour and GlobWave. The Danish Meteorological Institute (DMI) is a public institute, providing meteorological, oceanographic and related services for the people of the Kingdom of Denmark (Denmark, the Faroe Islands and Greenland). DMI s area of activity comprises forecasting and warning services as well as continuous monitoring of weather, sea state, climate, and related environmental conditions in the 74

75 atmosphere, over land and in the sea. As such, it has national responsibility for carrying out storm surge model forecasts and issuing warnings for Danish areas to the Danish coastal authorities and the public in general. DMI is part of the Baltic Sea Operational Oceanographic System (BOOS) and North West Shelf Operational Oceanographic System (NOOS). DMI play the role as the real-time insitu sea level centre for the BOOS and NOOS communities. In the MyOcean project DMI leads the Baltic Model Forecasting Centre providing real time ocean forecasting for the Baltic Sea. DMI is part of the High Resolution Local Area Modelling (HIRLAM) developing consortium within numerical weather predictions. DMI is operationally running a number of numerical forecast models for European and Arctic regions, alongside regional and large scale ocean models (HBM and HYCOM). DMI is part of a collaboration developing a coupled atmosphere, ocean and sea ice climate model (EC-Earth), whilst a high resolution coupled ocean and ice forecast model (HYCOM/CICE) is currently being developed at the institute. The Coastal and Marine Research Centre (CMRC) in University College Cork was established in 1994 to undertake research into coastal and marine resource management. It is part of the Environmental Research Institute (ERI) and the Irish Maritime and Energy Resource Cluster (IMERC).Research and consultancy in the CMRC is undertaken by staff with a range of specialist backgrounds, all of whom work collaboratively in a project orientated environment. The Centre s expertise and skill sets are highly regarded both nationally and internationally. Fundamental and applied research in the CMRC is organised according to five specialist areas of interest: marine geomatics; applied remote sensing and GIS; marine and coastal governance; coastal processes and seabed mapping and marine ecology. The CMRC works with data from a wide range of satellite EO instruments including MERIS, MODIS, SAR and higher resolution optical datasets (e.g. Landsat, IRS, SPOT, and IKONOS) for land, coastal and marine applications. It lies at the forefront of geomatics research with Europe and internationally, with an ability to work with a variety of data in projects such as FP7 NETMAR, FP6 InterRisk and FP5 DISMAR. It has a track record of engaging end users and stakeholders in projects, organising the CoastColour users workshop in 2008 and, was part of the organising committee for the European Space Agency s Space innovation Powering Blue Growth Conference held in Cork in April The CMRC is currently being merged with sister centres in University College Cork under the banner of Beaufort Research. For further information see The Koninklijk Nederlands Meteorologisch Instituut, KNMI, (Royal Netherlands Meteorological Institute) is a government agency operating under the responsibility of the Dutch Ministry of Transport. It provides weather observations, weather forecasts and vital weather information, whilst carrying out applied and fundamental research in support of its operational tasks and as a global change research centre. Skilled and experienced groups, specialising in diverse topics such as instrument development and electronic read-out, automation, computing, operations control and quality control are employed within the institute, providing quality controlled, and cost effective data acquisition and data processing services. As an operational meteorological data centre and research institute in one, KNMI combines its international networks and collaborative projects in a practical way. It is an active member of the World Meteorological Organisation (Geneva, CH), the European Centre for Medium-range Weather Forecasts (Reading, UK) and the European Organisation for the Exploitation of Meteorological Satellites (Darmstadt, G), and Eumetsat's Ocean and Sea Ice Satellite Application Facility (SAF). 75

76 For more information on esurge please contact Phillip Harwood, esurge Project Manager, at 76