next up previous contents
Next: Further notes on the Up: Preparation and sampling of Previous: Organisation of input files

Running the SAMPLER to obtain nodes for a depth grid

TRIGRID has a menu system which is usually invoked by keying in 'trigrid' and pressing the 'Enter' key. Next, enter the number corresponding to the SAMPLER in the main menu. A large empty square window should appear after various preliminary messages, together with a smaller rectangular window on the right, identifying which version of the SAMPLER you are using. At the top of the screen, there is a menu of options which can now be selected with the mouse. The top menu in the SAMPLER offers the following options:

Some of these options lead to further sub-menus. The following notation is used throughout this manual for defining which program option is being referred to:

{TOP} indicates the top menu for the program in use
{TOP}DISPLAY denotes item DISPLAY in {TOP}
{TOP.DISPLAY}PAN denotes item PAN in submenu {DISPLAY} below {TOP}

and so on, if there are further levels of submenu. Where no confusion should result, {TOP.DISPLAY} will sometimes be written as {.DISPLAY}

Continuing with the example, pick option {TOP}BOUNDARY with the mouse. This brings up a number of options in the right hand window. A convention used in all the interactive TRIGRID programs is that items which appear in the colour cyan in the right hand panel are 'pickable', that is, some action will ensue if one of these is picked with the mouse. In the present case, picking with the mouse on NONE in the entry 'Boundary File : NONE' will produce a prompt asking the user to enter the name of the file containing digitised boundary data, in this case bnd.dig, via the keyboard. Subsampling rates then have to be set for the land and sea boundaries. Picking on the default value '1' under 'Select Kth land node' and entering 30, say, means that every 30th digitised point on land sections of the outer boundary and on island boundaries will be selected as a boundary node. Similarly, picking on the default value '1' under 'Select Lth open node' and entering 10 means that every 10th digitised point on stretches of sea boundary will be selected as a boundary node. Also pick on the default value 1.000 of minimum distance to be maintained between first and last points selected from each boundary (entry 'Min. dist. 1st-last') and enter 0.05, which is a suitable value for the demonstration case. Finally, pick ACCEPT, which will return control to the top menu.

At this point, picking {TOP}SAMPLE will result in sampling of the boundary data in file at the rates just set and display of the selected boundary nodes, superposed on the digitised boundaries. With the values K = 30 and L = 10 just entered, the land boundaries appear rather sparsely sampled compared to the sea boundaries. Pick {TOP}BOUNDARY again and this time set K = 15 and L = 20, then pick ACCEPT and {TOP}SAMPLE again to see the new selection of boundary nodes.

Now pick {TOP}CONTOUR, then pick NONE in the entry 'Contours file : NONE' and enter the name of the file (cont.dig)containing digitised contour data. Picking on the default value '1' under 'Select Mth contour node' and entering 8 means that every 8th point on each contour in the file will be selected as an interior node for the depth grid. The minimum distance is already set to the same value used for boundaries. Now picking ACCEPT and then {TOP}SAMPLE will show the interior nodes selected, superposed on the digitised contours. In addition, the boundary nodes corresponding to the current settings ( K = 15, L = 20 ) are also shown.

It remains to add the soundings in sound.nod for use as further interior nodes. Pick {TOP}SOUNDING and then 'NONE' in the entry 'Sounding file : NONE'. Enter the name sound.nod and pick ACCEPT, then {TOP}SAMPLE. Note that at present, all nodes supplied in the soundings file are used as internal nodes; there is no provision for subsampling. The set of nodes now displayed is sufficiently dense and well enough deployed to form a reasonably good basis for a depth grid. To save these nodes to file, pick {TOP}SAVE and choose a name for the output file. Since the output will be in NODE format, it is suggested that you use .nod for the file extension. The selected nodes are output by program SAMPLER in NODE format (see Part III).

With the subsampling parameters K = 15, L = 20 and M = 8, the file output from the SAMPLER should be identical to file dep.nod in directory trigrid/demodata (see Figure 2.2).

Figure: The boundary and interior nodes that were selected using program SAMPLER. These nodes are used as the basis for the depth grid. The same set of boundary nodes is used later for the model grid.

Use of the SAMPLER is now complete for the demonstration case and one can proceed with the triangulation of the depth grid in the next section. Though there should be no need for it in this demonstration case, there is a program in the package called OKNODE which can be used to verify NODE format files. In general, the depth node file should be checked with OKNODE at this stage.

next up previous contents
Next: Further notes on the Up: Preparation and sampling of Previous: Organisation of input files

Channel Consulting Ltd.