NASA QUIKSCAT Project

Description

This dataset consists of the version 2 Level 2B science-quality ocean surface wind vector retrievals from the Oceansat-2 scatterometer (OSCAT), which was designed and launched by the Indian Space Research Organization (ISRO) 23 September 2009. This Level 2B dataset is produced by the Jet Propulsion Laboratory (JPL) QuikSCAT Project in cooperation with ISRO. The retrievals are provided on a non-uniform grid within the swath at 12.5 km pixel resolution. This resolution is achieved through a slice composite technique in which high resolution slice measurements from L1B data are composited into a 12.5 km wind vector cell. This newest version contains an improved geophysical model function (GMF), known as QSCAT12, consistent with the Remote Sensing Systems Ku2011 GMF, and an improved rain detection and flagging algorithm; these algorithms are consistent with the latest reprocessed version 3 QuikSCAT L2B dataset. Each L2B file corresponds to a specific orbital revolution (rev) number, which begins at the southernmost point of the ascending orbit. As a Ku-band dual pencil-beam rotating scatterometer, OSCAT design specs bear a strong resemblence of the Ku-band SeaWinds scatterometers on QuikSCAT and Midori-II (ADEOS-II). The primary difference between OSCAT and SeaWinds is the ~4 degree increase in the OSCAT incidence angle, which acts as an offset to the relatively lower altitude of Oceansat-2 to help provide a nearly identical swath width to SeaWinds. In the early phase of cal/val, the JPL QuikSCAT Project identified several problems, most of which have been corrected in this latest L2B version. This dataset release is expected to have an accuracy similar to that of the version 3 L2B QuikSCAT product, with minor caveats, all of which are described by Jaruwatanadilok et al. (2014) and summarized in the User Guide document which is made available in PO.DAAC Drive at https://podaac-tools.jpl.nasa.gov/drive/files/allData/oceansat2/L2B/oscat/jpl/docs/ . Read software is made available in MATLAB, Python, R, and IDL and is accessible in PO.DAAC Drive at https://podaac-tools.jpl.nasa.gov/drive/files/allData/oceansat2/L2B/oscat/jpl/sw/ .

QSCAT_L1C_NONSPINNING_SIGMA0_WINDS_V2

This dataset is Version 2 of the geo-located and averaged Level 1B Sigma-0 measurements and wind retrievals from the SeaWinds on QuikSCAT platform, initiated in the months following the failure of the rotating antenna motor on 22 November 2009, using the various incidence angles at which QuikSCAT was pointed during the time period from November 2009 until present. Incidence angles were varied in order to cross-calibrate the Oceansat-2 and RapidScat scatterometers and to extend the known Ku-band geophysical model function. The averaging of the L1B input data combined with the wind vector processing results are both contained in this product referred to hereafter as Level 1C (L1C). The fixed and repointed beam processing is relative to either the one corresponding to the vertically polarized "outer" beam or the other corresponding to the horizontally polarized "inner" beam. The Sigma-0 values from the fixed operating beam for each frame are averaged to a single value representing approximately 50 samples. The data points are land flagged, collocated with ECMWF surface winds, and have climatological nadir attenuations provided for the location and time of the data (not applied to the sigma0). The following enhancements have been applied in the Version 2 re-processing: 1) the GMF has been updated (QNS2016a) to make use of ECMWF nowcast 1x1 degree resolution wind direction information for the entire historical data record; 2) the new QNS2016a GMF leverages a calibration adjustment from Remote Sensing Systems (RSS) resulting in a consistently lower Normalized Radar Cross Section (NRCS or Sigma-0) measurements that establishes a Sigma-0 bias of -0.25 dB (-5.9% linear scale) compared to the L1C Version 1 data; 3) the new QNS2016a GMF also applies an azimuthal modulation that is decreased by several tenths of a dB (for Sigma-0) in variation with wind speed; this results in a more consistent wind speed retrieval comparison between "non-spinning" and "spinning" modes of the QuikSCAT instrument; 4) spacecraft attitude was re-estimated using slice data over multiple orbits as a replacement for lost echo-tracking capability during the "non-spinning" mode of the instrument; this new attitude estimation follows an unpublished manual technique that leverages the echo power of individual slice observations; since only a small subset of slice observations are analyzed, rapid variations in attitude are not captured; 5) continues data production beyond October 2016 through the end of mission on 30 August 2018. Retrieved wind directions are only slightly different from ECMWF values and should not be considered an independent measurement of wind direction. Retrieved wind speeds do not depend significantly on ECMWF speeds as evidenced by the fact that they agree closely with WindSAT polarimetric radiometer speeds whenever WindSAT and ECMWF disagree. The Sigma0 values have also been corrected for scan loss (due to the fact that the antenna does not scan) and for X-factor changes due to repointing.

QSCAT_LEVEL_2B_OWV_COMP_12

This dataset contains the latest reprocessed version 3 of the Level 2B science-quality ocean surface wind vector retrievals from the QuikSCAT scatterometer. The retrievals are provided on a non-uniform grid within the swath at 12.5 km pixel resolution. Higher resolution is achieved through a slice composite technique in which high resolution slice measurements from L1B data are composited into a 12.5 km wind vector cell. Version 3 processing begins with the same L1B (time-ordered backscatter) data as used in the previous processing. Version 3 has several improvements over the previous JPL processing of the QuikSCAT L2B winds: 1) changes to measurement binning, which was done in order to decrease noise and reduce gaps in the 12.5 km L2B wind retrievals, 2) an improved geophysical model function (GMF) to model the effect of wind on backscatter, 3) a neural network approach to correct rain contaminated winds speeds, 4) cross-track dependent wind speed biases were estimated and removed from the wind retrievals. The 12.5 km binning resolution enables users to obtain wind vector retrievals 10 km closer to shore when compared to the 25 km L2B dataset (only available in versions 1 and 2). More details to the processing changes and improvements are noted by Fore et al. (2014): PO.DAAC Drive at https://podaac-tools.jpl.nasa.gov/drive/files/allData/quikscat/L2B12/docs/fore_et_al_ieee_2014.pdf . Each L2B file corresponds to a specific orbital revolution (rev) number, which begins at the southernmost point of the ascending orbit. This is the official dataset produced by the QuikSCAT Project through the SeaWinds Processing and Analysis Center (SeaPAC). The Version 3 User Guide document is accessible from https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-docs/quikscat/open/L2B12/docs/qscat_l2b_v3_ug_v1_0.pdf.

QSCAT_LEVEL_2B_OWV_COMP_12_LCR_3.1

This dataset contains the latest reprocessed version 3.1 of the Level 2B science-quality ocean surface wind vector retrievals from the QuikSCAT scatterometer. The retrievals are provided on a non-uniform grid within the swath at 12.5 km pixel resolution. Higher resolution is achieved through a slice composite technique in which high resolution slice measurements from L1B data are composited into a 12.5 km wind vector cell. Version 3.1 processing begins with the same L1B (time-ordered backscatter) data as used in the previous Version 3.0 processing. Version 3.1 improves upon the previous Version 3.0 processing by incorporating enhanced coastal processing using a Land Contamination Ratio (LCR) method with a fixed threshold. The 12.5 km binning resolution combined with the LCR processing enables this dataset to provide wind vector retrievals with approximately half the coastal gap as compared to the Version 3.0 12.5 km L2B dataset. The geophysical model function used to produce the wind vector cell retrievals remains unchanged between Version 3.0 and 3.1. Each L2B file corresponds to a specific orbital revolution (rev) number, which begins at the southernmost point of the ascending orbit. This is the official dataset produced by the NASA QuikSCAT Project through the SeaWinds Processing and Analysis Center (SeaPAC). More details to the processing changes and improvements are to be published in the near future, but for now can be referenced by the following presentation: https://mdc.coaps.fsu.edu/scatterometry/meeting/docs/2016/Thu_AM/coastal-poster.pdf .

QSCAT_LEVEL_2B_OWV_COMP_12_KUSST_LCRES_4.1

This dataset contains the latest reprocessed version 4.1 of the Level 2B science-quality ocean surface wind vector retrievals from the QuikSCAT scatterometer. The retrievals are provided on a non-uniform grid within the swath at 12.5 km pixel resolution. Higher resolution is achieved through a slice composite technique in which high resolution slice measurements from L1B data are composited into a 12.5 km wind vector cell. Each L2B file corresponds to a specific orbital revolution (rev) number, which begins at the southernmost point of the ascending orbit. This is an official dataset produced by the NASA QuikSCAT Project through the SeaWinds Processing and Analysis Center (SeaPAC). Version 4.1 processing begins with the same L1B (time-ordered backscatter) data as used in the previous Version 4.0 processing. This new version has a number of key improvements and changes over the previous version 4.0, including: 1) winds are now retrieved to within 5-km and 10-km of the coast within oceans/seas and lakes respectively; 2) coastal winds are now flagged as poor coastal quality and likely corrupted in orbits with estimated spacecraft pitch error greater than 0.04 degrees, which affects 150 orbits of data where coastal winds are severely contaminated by land due to poor attitude knowledge (note: attitude error tracking can identify pitch error but not yaw error, so when estimated pitch error is far from zero, it implies yaw error is large and uncorrected); 3) coastal winds are flagged based upon the long term mean wind speed and standard deviation of wind speed for each place on the ground; 4) four quantities, means and standard deviations computed with and without the land contamination correction algorithm applied (note: higher mean and smaller standard deviation are evidence of land contamination), are used to estimate the expected wind speed bias with respect to neighboring wind vector cells over open water; 5) wind vector cells with estimated speed bias greater than 0.4 m/s are flagged as poor coastal quality and likely corrupted; 6) winds within 5-km of the coast of an ocean/sea and 10-km of the coast of a lake are flagged as poor coastal quality and likely corrupted; the larger distance threshold for lakes is due to higher variability in lake water levels.

QSCAT_LEVEL_1B_V2

The SeaWinds on QuikSCAT Level 1B dataset contains the geo-located Sigma-0 measurements and antenna pulse "egg" and "slice" geometries as derived from ephemeris and the Level 1A dataset. The pulse "egg" represents the complete footprint of the pulse, which has a spatial geometry of approximately 25 km by 35 km. There are 8 slices that constitute the range-binned components of a pulse each of which has a spatial geometry of approximately 25 km by 7 km. The orientation of the long dimension of the slices varies with the rotation of the antenna and thus does not align with the along/across track orientation of the wind vector grid in the L2B/L2A products. This dataset represents the second reprocessed version of the Level 1B release. Special note: QuikSCAT went into a "non-spinning" mode on 22 November 2009. The final rev number in the nominal Operational "spinning" mode is 54296; the "non-spinning" mode of the instrument continued predominantly until the end of the time series. There were some brief periods of "spinning" in between, which include the following days and rev numbers (identified in parenthesis): 1) 29 January 2013 to 5 February 2013 (7909-71011), 2) 14 March 2013 (71536-71549), 3) 18 March 2013 to 21 March 2013 (71590-71634), and 4) 28 March 2013 to 31 March 2013 (71735-71769). Data during the "non-spinning" mode is not consistently calibrated with data from the "spinning" mode. Furthermore, incidence angles change periodically during the "non-spinning" mode. It is therefore advised that only "expert" users attempt using the data during the "non-spinning" mode. For standard L1B data users who wish to access consistently calibrated L1B data during the "non-spinning" mode, please consider using the L1B Averaged Sigma-0 dataset as alternative, which may be accessed by contacting podaac@podaac.jpl.nasa.gov

Data Discovery

Explore this data using NASA's Earthdata Search, a comprehensive tool for discovering and visualizing Earth science datasets.

Data Access

Access requires an Earthdata Login account. Read our guide on obtaining AWS credentials to retrieve this data from AWS.

Update Frequency

Varies by dataset

License

Creative Commons BY 4.0

Documentation

https://earthdata.nasa.gov/

Managed By

See all datasets managed by NASA.

Contact

https://earthdata.nasa.gov/contact

How to Cite

NASA QUIKSCAT Project was accessed on DATE from https://registry.opendata.aws/nasa-quikscat.