| 480 | | Each field variable yields a corresponding variable with the same name in the projected field. in addition integral quantities (circulation, flux...) can be calculated. The result of projection depends on the object type, the nature of the coordinates, the ''Role'' of field variables and on the projection mode !ProjMode: |
| 481 | | |
| 482 | | * ''' !ProjMode = 'projection': ''' this is a normal projection of the field data in a range of action around the object, as defined by the parameters 'RangeX', 'RangeY', "RangeZ'. The projection of an input variable defined on unstructured coordinates therefore remains unstructured. By contrast, an input variable defined on a regular grid always yields a projected variable on a regular grid (for instance on a line or plane). Error flags ? |
| 483 | | * 'points': each field variable is averaged in a sphere of radius RangeX (a single value) around each projection point and attributed to this point position. An ancillary variable U_nbval(i) indicates the number of (non-false) data found around each point. Ancillary data and warning flags are not projected on points. |
| 484 | | * 'line': for scattered coordinates, each initial data point within a range ''RangeY'' on each side of the line is normally projected on the line, keeping its field values. For grid lin interpolation and averaging. Vector quantities are furthermore projected on the line as longitudinal (X) and normal (Y) components. The line length and mean value of each variable along the line is also calculated (giving access to circulation and flux). Ancillary data and warning flags are not projected on points. |
| | 480 | Each field variable yields a corresponding variable with the same name in the projected field. Integral quantities (circulation, flux...) can be also calculated. The result of projection depends on the object type, the nature of the coordinates, the ''Role'' of field variables and on the projection mode '''!ProjMode''': |
| | 481 | |
| | 482 | * ''' !ProjMode = 'projection': ''' this is a normal projection of the field data in a range of action around the object, as defined by the parameters '''[RangeX], [RangeY], [RangeZ]'''. The projection of an input variable defined on unstructured coordinates therefore remains unstructured. By contrast, an input variable defined on a regular grid always yields a projected variable on a regular grid (for instance on a line or plane). Error flags ? |
| | 483 | * 'points': each field variable is averaged in a sphere of radius '''[RangeY]''' ('''Max''') around each projection point and attributed to this point position. The number of (non-false) data found around each point #i is recorded in the vqriable U_nbval(i). Ancillary data and warning flags are not projected on points. |
| | 484 | * 'line': for scattered coordinates, each initial data point within a range '''[RangeY]''' on each side of the line is normally projected on the line, keeping its field values. For grid lin interpolation and averaging. Vector quantities are furthermore projected on the line as longitudinal (X) and normal (Y) components. The line length and mean value of each variable along the line is also calculated (giving access to circulation and flux). Ancillary data and warning flags are not projected on points. |
| 489 | | * ''' !ProjMode |
| | 489 | * ''' !ProjMode = 'interp_lin': ''' Linear interpolation of scalar and vector field variables, after exclusion of false data (marqued by error flag). Ancillary data and warning flags are not projected in this mode. Gridded data are interpolated by ..., while fields with scattered coordinates are projected with the Matlab function .... Note that this function provides interpolation only within the convex hull of the initial data set, attributing 'NaN' (undefined) field values out of this domain. To avoid problems with further data processing, UVMAT transforms NaN values into zeros, but mark them with an error flag FF=1. |
| 490 | 490 | * 'points': linear interpolation on each point of the object. |
| 491 | 491 | * 'line','polyline', 'rectangle', 'polygon', 'ellipse': linear interpolation on points regularly spaced on the line, with mesh DX. The X coordinate is the distance following the line, with an origin at the starting point(the first point in 'line','polyline','polygon',the lower left corner for rectangle, the point along the main axis for an ellipse). The line length and mean value of each variable along the line is also calculated (giving access to circulation and flux). |
| 492 | 492 | * 'plane': linear interpolation on a regular grid with meshes DX, DY and ortigin at (X,Y)=(0,0). This grid is bounded by the two values of RangeX and RangeY along X and Y respectively. |
| 493 | 493 | |
| 494 | | * ''' !ProjMode |
| 495 | | |
| 496 | | * ''' !ProjMode |
| 497 | | |
| 498 | | * ''' !ProjMode |
| | 494 | * ''' !ProjMode = 'interp_tps': ''' This behaves with different objects line 'interp_lin', but using the more precise thin spline shell method. This is particularly usefull to calculate spâtial field derivatives. Furthermore this method provides data exrtrapolation outside the initial convex hull (although it is not reliable at large distances). This mode does require a previous calculation of tps weights, see [#a5.1Gridingofdata section 5.1], so it does not act on the initial field cells with scattered coordinates. This is done by UVMAT if tps projection is requested. Gridded data are linearly interpolated (to clarify...). |
| | 495 | |
| | 496 | * ''' !ProjMode = 'inside' and 'ouside '''': defined only for closed lines: rectangle, polygon, ellipse. For each field U, its probability distribution function Uhist inside, or respectively outside, the line is calculated, as well as the mean Umean. other statistics... |
| | 497 | |
| | 498 | * ''' !ProjMode = 'none', 'mask_inside', 'mask_outside': ''' no projection operation. The object is used solely for plotting purpose, to show a boundary or to prepare a mask, inside or outside a closed line, see [#a9-Masksandgrids section 9]). |
