Assesing climatological forcing information provided in NetCDF format in ArcGIS environment

1
Assesing climatological forcing information
provided in NetCDF format in ArcGIS environment

• CEE 6440 GIS in Water Resources
• Fall 2004
• Term Paper Presentation
• ENRIQUE X ROSERO R

2
Motivation

• Interest in crossing boundaries between GIS and
  Atmospheric Science communities.
• First challenge in integrated analysis and
  display is data interoperability.

Objective

• To present a general example of data set
  conversion.

3
Atmospheric data into GIS

• Atmosphere has 3 spatial independent variables
  and changes in time
• Scientific data include gridded output from
  numerical models, image data from satillites,
  point data from stations and trajectory from
  soundings

• Layered data model conceptualization

4
Scientific Data Format NetCDF

• Machine-independent binary data format used for
  oceanographic and meteorological data exchange
• Stores a multidimensional array of variables and
  attributes (data and metadata)
• Interface through C, Fortran, C, Java and Pearl
  libraries

5
Data interoperatibility

• Scenarios to integrate and share information
• Development of NetCDF converters for ingesting
  data into GIS

Series of TimeSeries Tables
API generates a snapshot in GeoTIFF format
6
Case Rainfall events in Arkansas-Red Basin

• Radar-derived precipitation stored in NetCDF

7
Use of available API for file conversion

• ncBrowser allows no export
• GDAL export GeoTIFF and ASCII grid but no
  geoinformation

8
Development of custom import tool

• Fortran interface to retrieve data into array

9
Conversion Tools ASCII to Raster

• Straightforward
• Precipitation grid projected onto a HUC

10
Precipitation event 15 to 19 Nov 2004

• Ready for raster calculation, zonal statistics,
  integration to modeling, etc.

11
Zonal statistics

• Mean Median

12
Conclusions

• Tons of information available in netCDF (e.g.
  OpenDAP, THREDDS)
• There is no general nc2GIS conversion tool
• Custom import has to be defined on a particular
  file structure basis
• Seems unlikely that scientist replace particular
  tools (GrADS, Ferret, R, Matlab) with general
  purpose GIS tools
• Have to wait years until a domain-specific data
  model is developed, files become standarized and
  conversion tools are provided

13
References

• Betancourt, T.(2003) Some concepts in
  Atmospheric data.
• Yuan, M and Mcintosch, J. (2004). Data retrieval
  on spatiotemporal characteristics of weather
  events.
• Nativi, S. et al. (2003) Differences among the
  data models used by the Geographical Information
  Systems and Atmospheric Science Communities.
• Murray, D. et al. (2004) Integrating GIS data
  with Geosciences data in unidatas IDV
• Caron, J. and Nativi, S. (2004). NetCDF to GIS
  the Good, the Bad and the Ugly.
• Wilhelmi, O and Boehnert, J. (2004). Developing
  an Atmospheric Data Model for GIS.
• Zeiler, M. (1999). Modeling our world.
• Unidata (2002). NetCDF users guide for Fortran
  90.
• Unidata netCDF website
• http//my.unidata.ucar.edu/content/software/netcd
  f/index.html