You are here: Home Naiad Project What is Naiad system? Virtual variables

Virtual variables

Virtual variables are variables which are not contained natively within the source products served through Naiad system. They are added or computed on the fly, in order to enrich, complement or simplify the content of the source dataset.

 Enriching content

Virtual variables intended for native content enrichment provide new information that is not present in the source data but complement it and help interpretation or usage of the data. This new information may be computed completely from native content, such as the sun zenital angle computed from original pixel latitudon/longitude and time, or use an external source of information (such as climatology to add the climatological value directly on top of each pixel or to compute anomaly with the actual observed value).

Virtual-variable-zenital-solar-angle largeVirtual variable for the sun zenital angle, computed from the latitude, longitude and time variables within a source product. This variable can be computed for any dataset containg space and time information at pixel level.

Simplifying content

Interpretation and understanding of some product fields may be very confusing for users with poor knowledge of the product or of a particular type of remote-sensing. Getting a correct and accurate value of some geophysical parameter will often involve usage of several ancillary fields in order to select the correct value or remove suspect or bad values. For instance, using scatterometer data often involves dealing with with several wind solutions for a given pixel as the wind retrieval algorithm is ambiguous, and then selecting the correct one among all the solutions (often using an additional field providing the rank of the most likely value) : this can be highly simplified by providing to user just the "best wind" field by creating a virtual variable performing the selection job in a transparent way for the user. The product becomes then more understandable and straightforward to use.

Virtual-variable-wind-vector-ambiguity-1 largeFirst solution for wind vector retrieval from ASCAT scatterometer. The image exhibits clearly mismatches in data since close wind vector have opposite direction (like between the swath right edge and the inner swath).
virtual-variable-wind-ambiguity-2 largeSecond solution for wind vector retrieval from ASCAT scatterometer. The image exhibits clearly mismatches in data since close wind vector have opposite direction (like between the swath right edge and the inner swath).
Virtual-variable-best-wind-solution largeVirtual variable providing the best wind selection from ASCAT scatterometer. This variable is computed from the two ambiguities provided in the source data, and the field providing the prefered solution (1 or 2), but combining all that together in a single and easier to use variable giving simply the most accurate value.

 

Other examples of such virtual variables include :

  • "best sea surface temperature" : high quality sea surface temperature cleaned from any bad, suspect, cloud or dust contaminated values
  • ...