GRAS
Satellite Application Facility

RO Climate Data Survey

The survey is now concluded and it is no longer possible to submit replies. We would like to thank those who participated. The results will appear on the Survey page when they have been processed.

BACKGROUND

Occultations of the radio signals from GPS satellites can provide detailed information about the atmosphere. The data consist of high-resolution vertical profiles of atmospheric quantities like temperature, pressure, specific humidity and refractivity, from (near) the surface up to the upper stratosphere. The vertical resolution will be in the range 150-300 m, and radio occultation (RO) profiles have been demonstrated to contain a very high information content in the upper troposphere and lower stratosphere. Such RO data will soon be available from EUMETSAT's MetOp satellite, which will give about 500 vertical atmospheric profiles per day evenly distributed across the globe, and from the COSMIC satellites giving about 3000 vertical atmospheric profiles per day.

RO METHOD & BENEFITS

The basic principle of the RO method is that a receiver onboard a low-orbiting satellite tracks GPS signals as the transmitting satellite sets or rises behind the Earth. Due to refraction in the ionosphere and the neutral atmosphere the signal is delayed and its path bent, enabling calculation of profiles of the index of refraction (or refractivity) and subsequently temperature and humidity as a function of height. Many of the characteristics of RO data suggest them as a near- ideal resource for climate studies, particularly the global coverage, the all-weather capability, and the self-calibrated nature of the RO data. The latter property - which distinguishes RO from most other satellite observational techniques - allows for relatively easy inter-comparison of data from different satellites and RO instruments, which is required to construct long time series covering many years and even decades. The EUMETSAT Polar System (EPS), with its planned series of three Metop satellites, now provides an opportunity to create RO occultation based climatologies of high quality on a longer term. This will help us meet the requirements of both the scientific community and a wide range of climate data users. For these purposes, we are currently undertaking studies on how to best exploit the GRAS data from a climate perspective.

THIS QUESTIONNAIRE

The GRAS SAF is part of EUMETSAT's network of Satellite Application Facilities (SAFs). The objective of the GRAS SAF is to deliver operational RO products from the GRAS instruments onboard the three Metop satellites. The GRAS SAF will enter into the operational phase and deliver products from around the beginning of 2007. In the GRAS SAF project we now need to determine the interest in climate data derived from RO profiles, and what form the data should have. Hence, this questionnaire tries to build a picture of the priorities of potential climate data users. Because the users of RO data will be from a wide range of fields, with different levels of interest and expertise, we divide the survey into three parts. You may skip the last part if you are not familiar with the RO technique.

If you are not familiar with the radio occultation principle, you may want to read this short introduction.

Name:	*
Email Address:
Affiliation	*
Current position:	*

*mandatory fields

PART 1 User characteristics

1. How would you best describe your interests in climate data:
instruments, engineering & operations
atmospheric sciences
other sciences, please specify:
both science & engineering/operations
not interested


2. Would you be interested in using climate data based on RO data? The RO climate data will necessarily will be of short duration at the start of operations, but will have global coverage.
yes, I could use RO climate data right from the start
yes, but only as a complement to other climate data

no, RO climate data will not be of any use to me until they cover a long time span.

no, RO climate data has no interest for me

PART 2: Interests and priorities

3. The RO technique gives vertical profiles of atmospheric refractivity. Temperature, density, and humidity are derived from such refractivity profiles. Would you be interested in using refractivity as a climate variable?
yes
no, only standard meteorological variables

4. There will be up to 3500 vertical atmospheric profiles per day, distributed across the globe at random locations, i.e. the profiles do not fall in the same spots. Climate data will be derived from the profiles through interpolation or averaging-and-gridding techniques. What do you prefer to use?
large sets of single profiles
averaged climate data
both

5. Climate data can be put on a coarse grid (say 10 by 10 degrees). Would you use such coarsely gridded climate data if provided?
yes
perhaps:
no

6. Which time resolutions of climate data are you most interested in?
month to season
season to year
other:

7. Climate data can be provided in the form of graphical images (i.e. data plotted on geographical maps, plots of graphs, etc.) rather than numerical data. Would you use such images if provided?
yes
perhaps:
no

8. Climate data can be provided alone, or be provided together with uncertainty estimates. Would you use uncertainty estimates if provided?
yes
perhaps:
no

9. Climate data can be presented with associated 'meta-data'. Such meta-data describe circumstances of the data collection process, such as the satellite orbit or the reduction technique (algorithm type, version number, etc.) used. Would you use meta-data if provided?
yes
perhaps:
no

10. Some data products will be made available via a website equipped with tools for producing graphs and maps, etc., while other data products are intended for user download. If simple data-analysis tools (for plotting, averaging and basic statistics) were available at a web site, would you use these tools?
yes
perhaps:
no

11. Which data format do you prefer for the climate data retrieval?
ASCII/text
netCDF
HDF
GRIB
other:

12. How important would a help-desk function be for your use of RO data? Assume you could ask a knowledgeable scientist data-related questions and get an answer within 72 hours.
important
useful, but not very important
unimportant

PART 3 Expert background and interest

Only for those familiar with the RO techniques. Otherwise, you may skip them and submit directly.

13. The RO technique produces data at various levels of proximity to 'standard meteorological variables'. From the purely obser- vational excess phase, refractivity and geophysical variables are derived with an increasing level of model-dependency. It is important for us to understand what sort of data the potential users wish to work with: geophysical variables only, or more fundamental RO observables as well.

Select the data types you could be interested in:

"Level 1a": Measured excess phase and amplitude.
Raw bending angles, purely observed
"Level 1b": Statistically optimized bending angles, somewhat dependent on an assumed climatology, particularly above 35 km altitude.
"Level 2": Refractivity, somewhat dependent on an assumed climatology.
"Level 2": 'Real meteorological observables like p, T and q' somewhat dependent on ECMWF background fields.
"Level 3": processed climate data products.

Something else, please specify:

14. To provide the data in a format that users actually will use, we need to know what the users require in terms of spatial and time resolution. Data can be provided in a wide range of ways - as single profiles, as binned data on a grid of some sort, and in various interpolated ways. The interpolation methods allow for a spatial resolution higher than that which is supported by the number of profiles observed.

Indicate as many choices as apply - even if the question somewhat duplicates previous questions in section 2:

I want to use single RO profiles
I want to use RO data averaged in bins on a grid at a time-resolution of: and a spatial resolution of:
I would like to use 'assimilated RO data' which have been sampled at a regular grid of this spatial resolution: and this time resolution: . The assimilated data would be data from a model of the atmosphere into which RO data were assimilated, similar to the NCEP and ERA40 reanalysis data.
I would like a representation based on a fitted model that I could use to perform my own interpolations at arbitrary places, e.g. I would like a spherical harmonics representation with coefficients to the harmonics provided along with a description of the assumptions about the base functions used.
I want something else: