| 1 | = PDAF_local_weight = |
| 2 | |
| 3 | |
| 4 | This page documents the routine `PDAF_local_weight` of PDAF. |
| 5 | |
| 6 | The routine is used for localization in the analysis step of a filter and computes a weight according to the specified distance and the settings for the localizing function. |
| 7 | |
| 8 | The interface is the following: |
| 9 | {{{ |
| 10 | SUBROUTINE PDAF_local_weight(wtype, rtype, cradius, sradius, distance, & |
| 11 | nrows, ncols, A, var_obs, weight, verbose) |
| 12 | }}} |
| 13 | with the following arguments: |
| 14 | * `wtype` : `integer, intent(in)`[[BR]] Type of weight function:[[BR]] (0) unit weight (=1 up to radius `cradius`) [[BR]] (1) exponential decrease (1/e at distance=sradius; 0 for distance>cradius)[[BR]] (2) 5th order polynomial with support radius sradius (Gaspari&Cohn 1999; 0 for distance>sradius) |
| 15 | * `rtype` : `integer, intent(in)`[[BR]] Type of regulated weighting:[[BR]] (0) no regulation[[BR]] (1) regulation using mean variances[[BR]] (2) regulation using variance of single observation point |
| 16 | * `cradius` : `real, intent(in)`[[BR]] Cut-off radius (weight is always =0 for distance>cradius) |
| 17 | * `sradius` : `real, intent(in)`[[BR]] Support radius |
| 18 | * `distance` : `real, intent(in)`[[BR]] Distance to observation |
| 19 | * `nrows` : `integer, intent(in)`[[BR]] Number of rows in matrix A |
| 20 | * `ncols` : `integer, intent(in)`[[BR]] Number of columns in matrix A |
| 21 | * `A` : `real, intent(in), dimension(nrows,ncols)`[[BR]] Input matrix (only used for regulated localization) |
| 22 | * `var_obs` : `real, intent(in)`[[BR]] Observation variance (only used for regulated localization) |
| 23 | * `weight` : `real, intent(out)`[[BR]] computed localization weight |
| 24 | * `verbose` : `integer, intent(in)`[[BR]] Verbosity flag |
| 25 | |
| 26 | |
| 27 | |