Webbased Data Analysis Tools for CDAAC

1
Web-based Data Analysis Toolsfor CDAAC
Doug Hunt University Corporation for Atmospheric
Research
2
Introduction

• Overview of CDAAC processing strategy
• Database table layout
• Web interface features
• Demo

3
CDAAC data processing

• Convert input files
• Ground station GPS data
• LEO GPS data
• IGS orbits and pole files
• Gridded weather model data
• To Output files, including
• Atmospheric profiles
• Ionospheric profiles
• Ground station zenith tropospheric delays

4
CDAAC data processing, cont

• Processing consists of
• perl scripts which obey common input and output
  rules
• Currently about 20 scripts are used in CHAMP
  processing
• A table which tells which order to run the
  scripts in and what conditions must be met before
  a script can be run
• Each of these scripts may choose to write a
  record to the CDAAC database describing the data
  file or files it created

5
Database format

• Tables form a tree structure
• ltmissiongt_ltparentgt_ltchildgt, for example
• champ_occt_atmprf
• Mission CHAMP
• Parent occt, data relating to occultations
• Child atmprf, dry atmospheric profile data
• Entries in child tables are linked to entries in
  parent tables by a master key
• champ_occt_atmprf.parent champ_occt.id
• Tables are so linked for database normalization
  purposes
• Avoiding redundant data and data consistency
  problems

6
Database table structure
7
Web interface features

• Allows access to database and data files
• Selection of database records of all types based
  on
• Time
• Location
• Data quality
• Processing status
• Comparison of profile data with
• NCEP AVN model
• Radio sondes
• NCAR/NCEP analysis
• ECMWF analysis
• JPL or GFZ profiles

8
Features, (cont.)

• Display occultation map, including
• LEO sub point
• Occultation perigee point
• Ray path
• Compute statistics with comparison data. User
  can select
• Set of occultations to include
• Comparison data type
• Altitude levels
• Interpolation type
• An optional scalar parameter to plot statistics
  against

9
Features, (cont.)

• Scatter plots and maps, 1-, 2- or 3-D
• Any variable from any table vs. any others
• Several hundred available
• Data download
• Automatic generation of tar or zip files from any
  user-specified set of data files

10
Interface levels

• High level interface
• Easy to use
• Select occultations based on date range only
• Limited access to features
• Low level interface
• Complex
• Select occultations based on all available
  criteria
• All features available

11
Try it out

• Try it out at
• http//www.cosmic.ucar.edu/cdaac/DBif
• (send bugs to dhunt_at_ucar.edu)