Temporal interpolation for "one-point" gaps in hourly temperature

1. Locate a single missing value
2. Check if data available immediately before and after the interval
3. If available generates first estimate from linear interpolation
4. Checks for the Max and Min temperature for the missing value from the measurements registered in the hour afterwards
5. Provide an estimate T1 for the missing value as the average of 0.5*(MaxT + MinT)
6. Update the correcte dvalue following the rules below:

If an existing corrected value exists and it was generated by HQC:

• do not update, leave correction from HQC

If an existing corrected value exists and not from HQC:

• check if existing value: MaxT >= existing corrected >= MinT
• if FALSE update corrected with T1

If no corrected value exists:

• update corrected value with T1

• Non-real time (Qc2)

It shall be possible to run the algorithm for

• Any arbitrary time range.
• In automatic mode for ann arbitrary sliding time window (e.g. the last N days from today).
• The algorithm is specific to hour temperature measurements.

CONTROLINFO AND/OR USEINFO

• The trigger for the method is the identification of a missing original value (i.e. -32767).

TYPEID, PARAMIDS, OTHER METADATA

• Paramid=211

Note: The MinT and MaxT data are located in the paramids 213 and 215 respectively, as the regular kvalobs variables TAN and TAX.

SPECIFICATION OF CONTROLFLAGS TO BE SET FOR EXPECTED RESULTS OF THE ALGORITHM, I.E. CONSIDER BOTH POSITIVE AND NEGATIVE RESULTS

Follow the standard defined in the original kvalobs specification.

controlinfo(7), ftime

and add to cfailed that the method appiled was TAN, TAX average.

NB if the TAN TAX method is used (i.e. the correct information is available) and no HQC is overwritten, the result will always be ftime=1.

EXPECTIONS OF THE USEINFO FLAGS CORRESPONDING TO CONTROLINFO FLAG SETTINGS

• TBD

• …
• …
• …