CDAW-9.2 User's Guide (11/27/89) V2.0
[This is part 3 of 3, converted from the original mixed text-and-graphics
format to this plain ASCII text, January 21, 1994.]
[Page numbers have been inserted here to approximately match the original,
so that the original Table of Contents remains useful. The intervals between
page numbers are non-uniform due to the absence of graphics in this plain ASCII
version.]
--------
Page i
WORKING TABLE OF CONTENTS/PART #3
IV. GETTING STARTED: DATA MANIPULATION 51
IV.A. Options in Data Manipulation 52
IV.A.1. File and Variable CDAW-9 Naming Conventions. 55
IV.B. LIST/SUBSET/TIME WINDOW (Full Screen Interface) 56
IV.C. More About TAE, with CDF / Data Set List as an Example 61
IV.C.1. Menu options: 61
IV.C.2. Tutor Options: 63
IV.D. How to Access and Work with CDF Data in IDL. 65
V. A "HODGEPODGE" OF "INTERESTING" GRAPHICS FEATURES 66
V.A. A Definition of the "Select" Menu Options: 66
V.B. Details with Respect to XY and Panel Plots 68
V.B.1. Specify Multiple X-Axes: 68
V.B.2. Specify an "animation" parameter: 68
V.B.3. Specify a pseudo-color parameter (XY plots only): 69
V.B.4. Sort / Change data representation values / Set Fonts: 69
V.C. Histograms 72
V.D. An Introduction to Handling Imaging Data 72
V.E. Map Projections Available under the CDAW Software. 73
VI. ADVANCED GRAPHICS 74
VII. DETAILS IN DATA MANIPULATION 74
--------
Page ii
WORKING LIST OF FIGURES / PART#3
Figure IV.1. The "Datamanip" Menu. 52
Figure IV.2. Definition of Data Manipulation Options. 53
Figure IV.3. Initial LIST/SUBSET/TIME WINDOW Screen. 56
Figure IV.4. LSTW with CDF Selection Pop-up Menu 57
Figure IV.5. LSTW Summary Screen while Selecting Variables. 58
Figure IV.6. LSTW (Optional) Pop-up Prompt for Discrete
Variables. 59
Figure IV.7. LSTW PF1 "Action" Menu. 60
Figure IV.8. A Sample LIST Output to Terminal. 60
Figure IV.9. A Sample "Tutor" (CDFLIST in Data Manipulation) 63
Figure IV.10. A Sample "Tutor" (LIST under CDFLIST). 64
--------
Page 51
IV. GETTING STARTED: DATA MANIPULATION
As the alert reader has undoubtedly noticed, we noted the existence of
something we've termed "Data Manipulation" in the introductory paragraphs to
this tutorial, then promptly immersed ourselves in the "Graphics" subsystem.
A lot of the orientation of the CDAW system is indeed about graphics and data
visualization; i.e., the "display" of data. "Data Manipulation" is generally
about "doing things" with data; e.g.:
o averaging data
o constructing subsets or super-sets of data
o applying mathematics to data, for example constructing a value for
total B field strength by adding the vector components, and
o making lists of data set contents.
The underlying philosophy of the CDAW system as implemented to date is that
"graphics" graphically displays existing data . If you want to work with
("manipulate") the data before graphing it (which includes making listings of
the data) , you MUST make a new data set out of the "manipulated" data and then
feed that into the "graphics" subsystem. This approach may not always seem the
most natural or convenient (for which we apologize), but it's the best we have
for now.
--------
Page 52
IV.1) After typing "cdaw" etc., to enter "Data Manipulation" enter "2" and
[CR] at the first TAE menu prompt. You will then find yourself confronted with
the full array of "data manipulation" options under a TAE menu (the "Datamanip"
menu, analogous in many ways to the graphics "Select" menu) as shown below.
IV.A. Options in Data Manipulation
Figure IV.1. The "Datamanip"
_________________________________________________
Menu: "DATAMANIP", library "nacs$menu"
DATA MANIPULATION - Subsystem Menu
1) Copy a CDF ( CDFCOPY )
2) Create Parameters ( PARAMS )
3) List/Subset/Time Window a CDF ( CDFLOLA )
4) List a CDF (non VT-100 emulation) ( CDFLIST )
5) Subset a CDF (non VT-100 emulation) ( CDFSUB )
6) Apply a Time Window to a CDF (non VT-100 emulation) ( CDFFILT )
7) Stratify a CDF ( CDFSTRAT )
8) Run a Model ( MAGLINE )
9) Edit a CDF Header ( CDFEDIT )
10) Manage User CDF Library ( CDFLIB )
11) Display/Change Data Point Parameter Values ( DSPCHNG )
12) Terminate
Enter: selection number, HELP, BACK, TOP, MENU, COMMAND, or LOGOFF.
?
_________________________________________________
The underlying philosophy of the Data Manipulation subsystem design is to
provide a series of tools that can be applied selectively and in modular
fashion by the user.
Option #3 is "full screen" or "summary" interface that has the same
functionality as options #4-6 but more transparently invoked. Options #1-2 and
#4-7 above, when selected, will immediately prompt for the name of the input
data set (or CDF). For "Create Parameters" (option #2), up to 5 input CDFs may
be specified. As appropriate, an output CDF name will also be requested.
After the data set information is provided, additional screens ask for needed
(and optional) settings to be able to perform the requested action.
--------
Page 53
Figure IV.2. Definition of Data Manipulation Options.
__________________________________________________
1) Copy a CDF
Copies a CDF from e.g. the system library in preparation for modifying (in
place) that CDF's contents. Not necessary if a new CDF is to be created.
2) Create Parameters
Accesses the data manipulation and computational capabilities of the
Interactive Data Language (IDL). Using IDL, the user can:
o create new parameters using IDL functions and procedures, NACS-provided
algorithms, and user-defined algorithms
o filter data
o add new variables to existing CDFs (once copied into the user's directory)
o list variables, values of variables
o list / get help on algorithms and procedures
Note that when working with more than one CDF, the CDFs involved must have the
exact same time values for "cross-CDF" operations to function correctly (cf.
stratification).
3) List/Subset/Time Window (FullScreen)
Combination of options #4-6 below.
4) List a CDF
Allows listing of up to 8 variables from a CDF in time order.
The user is prompted for a time range and can select an increment (every nth
record to be listed). The user is then prompted for all variables to
include/exclude from listing from what effectively appears as a table but is in
fact constructed using a TAE tutor.
5) Subset a CDF
Build a new CDF containing a subset of the variables in the source CDF.
Appropriate subsetting will improve performance in many functions and will help
conserve directory disk space.
--------
Page 54
6) Apply a Time Window
Restrict the size of a data set by specifying a time window, either inclusive
or exclusive.
Appropriate windowing may also improve performance / help conserve disk space.
7) Stratify a CDF
Put data from one CDF onto the time basis of another specified CDF or by
supplying an arbitrary time interval and time resolution.
If stratifying from a baseline CDF, the user may choose either linear or cubic
spline interpolation. Time-weighted average binning can also be specified when
stratifying based on time interval and time resolution.
All times in a baseline CDF are used to provide time points, regardless of
whether they contain fill data or not. Stratification will not extend beyond
the time range of the baseline CDF. If the CDF to be stratified contains fill
values, these will be excluded from the interpolation computations.
8) Run a Model
Accesses David Stern's implementation of the Tsyganenko-Usmanov magnetic field
model.
The model may be entered directly through a field-line tracing program. The
result can be geographic coordinates of the field-line foot points or a file of
locations along the field line (which can be plotted on the Versatec plotter
but cannot be screen displayed).
9) Edit a CDF Header
Edit information that describes a CDF (metadata) and its contents.
Can only be applied to user CDFs.
10) Manage User Library
List / prompt for delete of CDFs in a user library.
11) Display/Change Data Point Parameter Values
Display (via listing of variables of interest) data points in a defined time
interval that meet or do not meet a specified condition on one of the CDF
variables. At the user's option, data points so selected may be changed to
fill. Can be useful (if sometimes tedious) in e.g. eliminating bad data.
__________________________________________________
To enter any option, just type its number and [CR].
--------
Page 55
The user should be generally aware that many Data Manipulation functions create
new CDFs and cannot execute unless sufficient unused VMS "quota" exists in the
user's account directory. It is therefore important to frequently purge files
and delete unneeded files / CDFs. Note that data sets implemented as CDFs
appear as multiple physical files in a directory listing (e.g., files of the
form *.cdf, *.cdh, *.v01, *.v02 ...). For that reason, we suggest users
utilize CDAW/NACS system functions to delete CDFs, although it is certainly
possible to do so from the DCL command line for users modestly familiar with
VMS systems.
Also note that Data Manipulation handles fill data (values to represent missing
or known bad data) where possible, to ensure that computations are performed
only with appropriate data. A NACS standard system fill value (-1.0 E+09) is
used for all newly created variables built from NACS-supplied algorithms. When
using standard IDL (see later sections), the user must take necessary
precautions individually to filter out unwanted data.
IV.A.1. File and Variable CDAW-9 Naming Conventions.
Files
EVENTX:SSEEX
SS is generally a two-character satellite designation or a one/two-character
ground station data set identifier.
EE is an experiment number (an NSSDC designation, with 00 = ephemeris data).
X is the event identifier (events A - E at the time of the CDAW-9.1
meeting).
Variables
SSEENNNN
SS and EE are defined as above, so that all variables in a given data set
have their first four characters to match the characters of the data set
identifier.
NNNN is a one to four character identifier for the variable in question.
When creating their own CDFs, users are free to adapt whatever naming
conventions they desire, although we suggest they follow the general database
and variable naming conventions where reasonable.
--------
Page 56
IV.B LIST/SUBSET/TIME WINDOW (Full Screen Interface)
The LIST/SUBSET/TIME WINDOW option (hence abbreviated as LSTW) is designed to
extract data from a data set within some window defined in time and parameter
space. LSTW is capable of generating:
o an alphanumeric listing to the screen
o an ASCII file containing a listing
o simultaneous screen and file listings
o a new data set in CDF windowed as above.
The interface is functionally very similar to the full-screen data selection
and graphics summary specification screens, with some features specific to
multi-dimensional CDFs. We will take as an example a listing from the CC01
data set in Event 9A.
IV.2) Select option 3 [CR] from the DATAMANIP menu. At this point, the LSTW
output specification screen below will be displayed.
Figure IV.3. Initial LIST/SUBSET/TIME WINDOW Screen.
[The figure did not transfer into this plain ASCII file.]
IV.3) The cursor is positioned to enter the directory (and CDF name). The
system does support a directory display (like the Uselect option of the
Graphics Data Selection screen). Since we want Event9a, enter event9a: and
[CR]. You will then see the screen shown in Figure IV.4 below.
--------
Page 57
Figure IV.4. LSTW with CDF Selection Pop-up Menu
[The figure did not transfer into this plain ASCII file.]
Using up-down arrows and u-d keys, move the highlight to the
desired data set (here CC01A) and select via [CR].
IV.4) At this point, some of the fields in the specification screen will be
filled out; e.g. with the data set dimensionality, number of variables, etc.
You are then presented with another prompt for output options (i.e.; Terminal
alone, Terminal + (ASCII) File, (ASCII) File alone, or output CDF. Note that
when the Terminal + File option is selected, interrupting the terminal output
will abort the full File output as well. For this example, we have chosen
"Terminal".
--------
Page 58
After selecting the output option, you are then positioned on the field to
select variables for listing (Figure IV.5 below). Specification of variables
works very much like that of Specification Page of the CDAW (Graphics)Summary
Screen (PF3, up-down arrows and [CR]). The key definitions applicable are
repeated from Section II for your convenience below the Figure.
Figure IV.5. LSTW Summary Screen while Selecting Variables.
[The figure did not transfer into this plain ASCII file.]
Key definitions with no pop-up menu displayed generally include:
[CR] move one field right, wrapping at end of line to next line
cntl-H move one field left, wrapping at start of line to last line
up-arrow move one field up
down-arrow move one field down
left-arrow move one character right (on entering min/max)
right-arrow move one character left (on entering min/max)
PF1 execute, edit and run/quit options
PF3 display "pop-up" selection menu for field (if applicable)
"delete" null selection of variable on line where cursor is currently
located (if applicable)
cntl-W redraw the screen (no change in options selected)
Key definitions with pop-up menu displayed generally include:
up-arrow move highlight one selection up
down-arrow move highlight one selection down
u page highlight up
d page highlight down
[CR] select highlighted value & move one field right
cntl-H no change in selected value & move one field left
IV.5) Select EPOCH (completely equivalent to TIME in the graphics context)
with [CR]. Move to the field for the 2nd independent variable and select
CC01ENRG. Note now that when you are positioned to the minimum / maximum range
fields, these values can either be directly typed or the user can call a pop-up
prompt via PF3 to indicate the range of values (Figure IV.6) below.
--------
Page 59
Figure IV.6. LSTW (Optional) Pop-up Prompt for Discrete Variables.
[The figure did not transfer into this plain ASCII file.]
For the listing in this example, we will choose CC01ENRG to be for a single
energy (0.067 MeV/e) (selected by [CR] above. We will also list all pitch
angles (so select CC01PANG as the 3rd listed independent variable) but enter
the inclusive range 0-90 degrees.
IV.6) At this point, we need to choose the dependent variable to be listed
(namely CC01FLUX). To do this, invoke the PF1 "action" menu (shown Figure IV.7
below) and move the highlight to select "Dependent Variables". Then use PF3 to
bring up a pop-up selection menu and select flux as the listed dependent
variable.
At this point, the definition of the listing is complete. Use PF1 to bring up
the "action" menu again and select "Generate Output". The resulting 1st page
of terminal output is shown in Figure IV.8. When completed (or interrupted
with any key + [CR], you may define another listing, return to the DATAMANIP
menu via PF1 and continue, return to graphics, or exit the CDAW system.
--------
Page 60
Figure IV.7. LSTW PF1 "Action" Menu (Before Selecting Dependent Variables for
Listing).
[The figure did not transfer into this plain ASCII file.]
Figure IV.8. A Sample LIST Output to Terminal.
_______________________________________________________
YYYYMMDDHHMMSS CC01ENRG CC01PANG CC01FLUX
19860401070224 0.06700 10.000 1873996.0
19860401070224 0.06700 20.000 1867685.8
19860401070224 0.06700 30.000 2170781.5
19860401070224 0.06700 40.000 2425756.3
19860401070224 0.06700 50.000 2813635.5
19860401070224 0.06700 60.000 3142426.0
19860401070224 0.06700 70.000 3464096.3
19860401070224 0.06700 80.000 3668155.8
19860401070224 0.06700 90.000 3828790.8
19860401070848 0.06700 10.000 1746982.5
19860401070848 0.06700 20.000 1786783.1
19860401070848 0.06700 30.000 1987588.6
19860401070848 0.06700 40.000 2315200.8
19860401070848 0.06700 50.000 2537206.8
19860401070848 0.06700 60.000 2819927.3
19860401070848 0.06700 70.000 3118492.8
19860401070848 0.06700 80.000 3448063.0
19860401070848 0.06700 90.000 3654568.0
19860401071512 0.06700 10.000 1680274.3
19860401071512 0.06700 20.000 1601858.8
19860401071512 0.06700 30.000 1950274.9
Hit <CR> to continue listing, any other key + <CR> to end
____________________________________________________________
--------
Page 61
IV.C More About TAE, with CDF / Data Set List as an Example
The original CDAW graphics system was built on the Transportable Applications
Executive (TAE), which was a pioneering attempt to create an interface control
system for applications that could be easily and consistently ported across
many different operating systems and computer hardware platforms. The price
for making the system sufficiently generic that this would be possible was a
fair degree of limitation on what that interface could easily present. This
limitation has been one motivation for investing in the Summary screen
interfaces of the CDAW-9 system.
But we have taken the effort to preserve a connection back into TAE, so you can
define e.g. an XY or Panel plot to the maximum degree possible in the Summary
screen, then go back to TAE to access those few advanced features of the
application logic for "Y versus X" type plots (including histograms and
pseudo-color or "z-variance" plotting ) still not integrated into the more
powerful Summary interface. The capability to display imaging data in the
CDAW system is also very powerful but can only be accessed via the TAE standard
interface at present. Data manipulation functions are currently only
accessible via TAE.
TAE defines two basic kinds of interfaces/screens, one the so-called "menu"
screen (of which the "Select" menu is an example) and the other the "tutor"
screen (of which the screen on which the post-processor file was specified is
an example).
TAE forces the design of the overall application program such as NACS to follow
a "tree" structure. With some degree of over-simplification, "menus" are the
means by which one can branch among the individual application program
"procedures" or program units of which the overall application is then
composed, "tutors" are the means by which parameters are defined that are
needed to run those procedures.
IV.C.1. Menu options:
As an example, we refer to the "menu" of Data Manipulation choices shown above
in Figure IV.1. All operations are performed at the TAE prompt line at the
bottom of the screen.
As has been noted in the examples, typing the appropriate number selects the
specified option from the menu screen. Typing "help" will bring up a general
help on the choices presented. Typing "help n" or "help option_name", where
"n" is the number to the left associated with a given option on the menu and
"option_name" is the name displayed to the right of the screen in association
with the number, will bring up a help text for that option. Any of the "help"
screens can be exited by typing "exit". A carriage return / [CR] moves to the
next page of the help and typing "page n" moves to page "n" of the help (if
that page exists).
--------
Page 62
The command "menu MENUNAME" or "m MENUNAME" entered on the prompt line will
jump to the specified menu (if it exists). An early example we have discussed
is "m select" entered at the NACS initialization menu. the command "back"
moves back along the tree structure to the next preceding menu, "top" move to
the so-called "root" menu (the initialization menu for NACS). Note that moving
around the displayed menus does NOT in and of itself do anything to change the
working environment. But jumping around menus, and then selecting menu options
without understanding how the overall system functions, can easily leave
critical application parameters undefined when executing a procedure (e.g.,
drawing a plot).
The command "command" or "c" entered on the prompt line puts the user into TAE
command mode and expecting further inputs in TAE Command Language or TCL.
Assuming you have a fairly standard terminal (i.e., not terminal type "OTHER"),
TAE command mode is generally of little use to the novice except for the "dcl",
"abort" and "exit" commands. In command mode (the "NACS>" prompt), any valid
VAX DCL command can be entered as
NACS>dcl VAX_command
without leaving the CDAW system. Both simple commands such as "dir" and more
complex commands (e.g., "mail" and "edit") can be entered.
It is NOT recommended to do a "set directory" from within NACS, because there
are internal system pointers set with the system path at the time you type
"cdaw" that will be made inconsistent with your current directory in that
event.
To get out of TAE command mode, type
NACS>m
to redraw the current menu. Another option from command mode is
NACS>ex
which will exit NACS without purging or deleting any files.
IV.7) Enter "1" and [CR]. The screen that results is shown in Figure IV.2.
--------
Page 63
Figure IV.9. A Sample "Tutor" (CDFLIST in Data Manipulation)
_______________________________________________
Tutor: proc "CDFLIST", library "ndms$menu" Pg 1.
List a CDF - Input
parm description value
---- ----------- -----
CDFNAME Name of input CDF " "
Enter: parm=value,HELP,PAGE,SELECT,SHOW,RUN,EXIT,SAVE, RESTORE; RETURN to page.
?
_________________________________________________
IV.C.2. Tutor Options:
You would use the up-arrow key to bring the prompt to enter the 1st CDF to the
TAE prompt line. To access successive prompts, use the down-arrow key. "page
n" will jump to the nth page of the tutor if it exists (another page exists
beyond the page displayed if the page number appears as "n+".
To enter a CDF name, backspace/rubout over the right quote mark, then type the
CDF name followed by the ". We're not in the summary screen, so you'd have to
enter the real name and a real path. The PF2 key can be used for some
character-string parameters (such as variable names when in those tutors) to
cycle FORWARD on the command line through the valid choices, PF3 to cycle
BACKWARD . Character strings need only be surrounded in quotes when they
contain spaces. A value is entered by [CR]. The sequence "esc esc" will clear
the command line.
As earlier noted,the mnemonics "EVENT9A" through "EVENT9E" have been defined
for the long form of the directory paths, hence "event9b:ij01b" will find the
event B data for the data set ij01 generally. To actually save the selection
you have made in a tutor and execute the tutor, you must type "run." Typing
"exit" or "back" will leave the tutor without saving any selections and
executing the associated procedure.
IV.8) Enter the IJ01B data set , then "run" and [CR].
CDFLIST now immediately passes into another tutor, in which the limits for the
listing (by time) are set. The same techniques are used as discussed above:
up-arrow to pull first field to prompt line (with highlight on main menu),
enter field, [CR], down-arrow to next field, etc and then "run", [CR] on the
tutor looking something like that below.
--------
Page 64
Figure IV.10. A Sample "Tutor" (LIST under CDFLIST).
___________________________________________________
Tutor: proc "list", library "/LOCAL/" Pg 1+
List a CDF - Input
parm description value
---- ----------- -----
TIMEMIN Minimum time for window "19860402180000"
Null = start time of CDF
(YYYYMMDDHHMMSS)
TIMEMAX Maximum time for window "19860403100000"
Null = stop time of CDF
(YYYYMMDDHHMMSS)
INCREM Sample Increment 1
OUTPUT Output Device "TERMINAL"
Enter: parm=value,HELP,PAGE,SELECT,SHOW,RUN,EXIT,SAVE,RESTORE; RETURN to page.
?
__________________________________________________
IV.9) At this point, you are based to a third tutor to specify the specific
variables to be listed. Values in the fields are all "yes/no", so use up-
arrow to pull the first field to the prompt line, PF2 to switch "no" to "yes"
(if you wanted to list that variable, then down-arrow to move to other desired
fields, and "run", [CR] to execute the tutor and finally bring the list to the
screen.
--------
Page 65
IV.D How to Access and Work with CDF Data in IDL.
Following is extracted from section 5.2 of CDAW-8 User's Guide.
--------
Page 66
V. A "HODGEPODGE" OF "INTERESTING" GRAPHICS FEATURES
In order to construct a tutorial-like presentation, a fairly large number of
details and features have been skipped. Some of those are really only likely
to be of interest to a user involved in advanced data visualization techniques
and are specific to specific kinds of graphics workstations. What discussion
there will be of these features is deferred to the summary of "Select" menu
options in section V.B and section VII (not available at the time of CDAW 9.1)
on Advanced Graphics.
Other features are of more general interest. In most cases, they are planned
to be integrated into the XY/Panel Summary screen display but have not yet been
so integrated due to resource constraints. They are currently accessible
through the TAE menus and tutors that underlay the CDAW system (refer back to
section IV.A). Some discussion of basic ways to handle imaging instrument data
(such as the DE-1 auroral imager) is also included .
V.A A Definition of the "Select" Menu Options:
Histogram Variable (X) versus frequency (number of occurrences)
X-Y Variable (Y) versus variable (X), overlaid on a single grid with
separate scalings. Up to 5 separate Y axis - X-axis data point
combinations may be defined. The capability to color code the
value of a third variable axis (Z) can be used on color
monitors.
Panel Variable (Y) versus variable (X) on 3 panels or grids. Up to 2
traces may be specified per panel, for a total of 6 variables.
Autoscaling is available, Z-axis color-coding is not.
--------
Page 67
Mapped XY XY plot of geographic coordinates (latitude and longitude) with
the option for a superimposed grid including continental
outlines and political boundaries. Z-axis color coding is
available. A wide variety of geographic projection techniques
are supported.
Contour Plots lines of equal value of a variable (Z) against a grid of
two other variables (X and Y).
Mapped Contour Specialized form of contour map where the X and Y grid are in
geographic coordinates, with map options for overlay outlines
and projections as in Mapped-XY option above.
Pseudo-Color Image The "Image" plot type allows color coding the value
of a variable (Z) against a grid of two other variables (X and
Y). Image plots are not possible on all color terminals.
Mapped Pseudo-Color Image Specialized/advanced form of pseudo-color image
where X and Y are geographic coordinates, with map options for
overlay outlines and projections as in Mapped-XY option above.
Surface Diagram Allows generation of a "perspective" plot; i.e., an
oblique view with the appearance of a 3-D image.
Mapped Surface Diagram Specialized form of "perspective" plot where X and Y
are geographic coordinates, with map options for overlay
outlines and projections as in Mapped-XY option above.
Merlin Plot The Megatek/Merlin 9200 series is an extremely powerful graphics
workstation capable of creating and manipulating images of great
complexity. Several Merlins reside in room 153, NSSDC facilities
in Building 26.
_________________________________________________
--------
Page 68
V.B Details with Respect to XY and Panel Plots
Some of the capabilities of the system for making XY and Panel plots are not
yet accessible from the Summary screens. These are briefly tabulated below,
with some discussion following.
PLOT OPTIONS ADDITIONAL CAPABILITIES (Beyond Features in Summary Screen)
-------------------------------------------------------------------------------
Plot Axes o Specify multiple X-axes (XY only)
o Specify animation parameter
o Specify pseudo-color parameter (XY only)
Plot options o Sort
Data representations o Adjust default scalings
o Change variable labels to appear on plots
Output Controls o Set font size and resolution/style
V.B.1. Specify Multiple X-Axes:
The current (XY/Panel Graphics) CDAW Summary screen interface is oriented
toward time plots at present. One consequence of this orientation is the
assumption that a single X-axis variable is appropriate.
In fact, the underlying software is capable of supporting 5 distinct X-axes
choices in the same way that it supports 5 Y-axes variable choices. The effect
then on the plot is to see the (up to) 5 Y-axes to the left of the plot, the
(up to) 5 X-axes at the bottom of the plot.
Multiple X-axes may be accessed through the "Define X-Y Plot Axes" tutor
(option #1) of the TAE "XY Plot Menu".
V.B.2. Specify an "animation" parameter:
Animation allows the user to generate a sequence of plots in a single call to
the "Draw Plot" procedure. The animation parameter (a variable in fact) should
be a discretely varying quantity; i.e., a variable dependent on only one
dimension. One plot will be pre-generated (before displaying any plots on the
screen) for each step in the animation parameter, within any global filter
currently defined on that parameter.
--------
Page 69
Time is a valid variable on which to do animation in multi-dimensional data
sets, although typically it should be filtered to avoid generating an excessive
number of plots. As one example, for a data set consisting of fluxes as a
function of time, look angle and energy, one might generate a series of spectra
at fixed time for each look angle.
Once the plots start to display on the screen, the keys "f" (forward) and "b"
(backward) can be used to move rapidly back and forth through the animation
sequence as desired. Some care is of course appropriate in that asking for a
long series of plots will necessarily involve a long delay time between
invoking DRAW and seeing the first plot in the series displayed on the screen.
Animation sequences are also specially handled in the post-processor when
displaying to a terminal, where "n" is used to move forward between whole
sequences when multiple sequences (multiple invocations of DRAW) have been
written to a single plot file, "f" and "b" are used to move within a given
animation sequence.
Animation may be accessed through the "Define X-Y Plot Axes" tutor (option #1)
of the TAE "XY Plot Menu".
V.B.3. Specify a pseudo-color parameter (XY plots only):
When a variable is assigned as the pseudo-color parameter (effectively the Z
variable on an X/Y/Z plot), each (X,Y) point to be plotted is assigned a color
based on the value of that Z-parameter. The color ranges over the currently
defined scale range in a pre-defined way, with finer color resolution displayed
on terminals supporting the display of more simultaneous colors. The system
presently ignores a specification of logarithmic scaling.
Pseudo-color may be accessed through the "Define X-Y Plot Axes" tutor (option
#1) of the TAE "XY Plot Menu".
V.B.4. Sort / Change data representation values / Set Fonts:
Writeups not completed in time for CDAW 9.2.
--------
Page 70
--------
Page 71
--------
Page 72
V.C. Histograms
Histograms are option #1 of the Graphics "Select" menu (Figure II.4).
Parameters of a histogram plot are defined via TAE tutors, each of which must
"run" to set the parameters for the subsequent plot. The parameter to be
plotted is defined under the "Plot Axes" tutor (animation is also possible to
generate multiple histograms in a single pass). Whether statistics and/or a
fitted curve to the distribution are to be put on the curve is set in "Plot
Options".
The user should note that the bin widths of the histogram are actually set as
attributes of the individual variables in the CDF: hence, the "Data
Representation" tutor must be accessed, then paged to the variable of interest.
Set/reset the parameter BIN1 to produce a sensible binning. Only linear
scaling is supported for histograms. Again, the tutor must "run" for any
changes to the binning to be made effective.
V.D. An Introduction to Handling Imaging Data
It is in the ability to handle image data that the functional virtues of
building the CDAW system on a larger graphics research system are most evident.
In the sense used here, image data is really any 2-dimensional or higher
dimension data plotted with color or contours to show variation of a dependent
variable with the independent variables. Of course, auroral imaging such as
from DE or VIKING are of particular interest in the context of CDAW-9.
Image data can be displayed under a variety of options, including pseudo-color
encoded XY plots, contour plots and pseudo-color images. In addition, the
"Mapped" options for each of these plot types allows the application of map
projections (see section V.E), as well as allowing geographic outlines to be
overlaid where appropriate.
The range of options is generally very large and beyond the scope of this
section to explore in full generality. Since the most common application of
the CDAW system is to generate gridded auroral images (quite commonly to a
uniform orthographic projection), the following outline indicates common
settings (with some remarks) needed to define and execute the mapped
pseudo-color image option. The tutors to be "run" to enter these settings are
shown in the MAPIMAG menu below.
_____________________________________________________________________________
Menu: "MAPIMAG" Mapped Image Plot Menu
1) Define Mapped Image PLOT AXES (REQUIRED) (AMAPIMAG)
2) Define Mapped Image PLOT OPTIONS (PMAPIMAG)
3) Define Map OPTIONS (MAPOPTS)
4) Define DATA REPRESENTATION OPTIONS (DATAOPTS)
5) Define FILTERS (FILTERS)
6) Define OUTPUT CONTROLS (CONTROL)
7) DRAW Mapped Image Plot (MAPIMAG)
8) GRAPHTERM (GRAPHTERM)
__________________________________________________________________
Key settings of parameters within these tutors are tabulated below to produce a
single DE mapped orthographic projection of the auroral oval (using CDAW-8 data
as a specific example).
__________________________________________________________________
Tutor: proc "AMAPIMAG" "MAPIMAG" Plot Axes
parm description value
---- ----------- -----
X Latitude Positional Parameter "D103GMLT"
Y Longitude Positional Parameter "D103GMLN"
Z Value Parameter "D103TCNT"
ANIMATE Animation parameter OPTIONAL -- (null value)
__________________________________________________________________
NOTE: This example is a plot in geomagnetic (not geographic) coordinates.
__________________________________________________________________
Tutor: proc "PMAPIMAG" "MAPIMAG" Plot Options
parm description value
---- ----------- ------
STATS Display statistics with plot "ON"
ALGORITM Map Color Image Algorithm "NEAREST1"
RESOLUTN Number of Grid points along Y Edge 50.0
__________________________________________________________________
NOTE: The options above are CRUCIAL in defining the speed with which the plot
will be produced. In general, use NEAREST1 (for best speed ande SDBWEIGHT for
highest plot quality). A grid resolution of 50 is generally required to make
out much detail in the image with 100+ for high quality, although lower
resolutions are very useful in initially setting up the plot and verifying
limits and color scales.
--------
Page 73
__________________________________________________________________
Tutor: proc "MAPOPTS" Map Options
parm description value
---- ----------- -----
PROJECT Map Projection "ORTHOGRAPHIC"
POLEPOS Pole Position Latitude -90.0 (1)
Longitude 180.0 (2).
CENTROID Viewing Centroid Latitude -- (null value) (1)
Longitude (2).
FIDUINC Fiducial line increment value 10.0
MAGFACT Magnification Factor 3.0
MAPDATA Select database to plot world map 10
POLITIC Display political boundary "OFF"
FIDUCIAL Plot fiducial lines on world map "ON"
__________________________________________________________________
NOTE: For a pole projection, the viewing centroid does not need to be
explictly defined. FIDUINC is the grid line spacing. A MAGFACT = 3.0
generally leaves the entire auroral oval within view (geomagnetic coordinates).
MAGFACT can be pushed to 3.5 if desired. Higher magnifications will require
adjusting POLEPOS and/or viewing CENTROID specification to focus on features of
interest. The MAPDATA value is set to a large number to inhibit plotting of a
world map (which would be misleading, given that the plot is in geomagnetic
coordinates).
__________________________________________________________________
Tutor: proc "DATAOPTS" (Pg 8+) DATA REPRESENTATION OPTS
parm description value
---- ----------- -----
D103TCNT Brightness Count -- (null value)
SCALE8 Scale minimum for D103TCNT 0.0 (1)
Scale maximum for D103TCNT 4000.0 (2).
STYPE8 Scale type for D103TCNT "LINEAR"
BIN8 Number of bins for D103TCNT 0
__________________________________________________________________
NOTE: The color scale on an image plot is changed by adjusting the scalemin /
scalemax values for that variable in the Data Representation menu. Only the
page where D103TCNT scale is set is shown above.
__________________________________________________________________
Tutor: proc "FILTERS" FILTERS
parm description value
---- ----------- -----
FILT1 Mnemonic for filter #1 "TIME"
RANGE1 Minimum for TIME YYYYMMDDHHMMSS "19860401080000" (1)
Maximum for TIME YYYYMMDDHHMMSS "19860401081500" (2).
INCL1 Inclusive / exclusive "INCLUSIVE"
status for filter #1
__________________________________________________________________
NOTE: A filter on TIME must be set to select data from a particular image. The
range should be set to capture a single image to be sensible, unless animation
on TIME has been specified as a MAPIMAG optionback.
V.E. Map Projections Available under the CDAW Software.
The following pages are extracted from sections 2 and 3 of the NSSDC
publication #88-05, "A Flexible, Template-Based Software Package for
Generating Maps (Ni, Gough and Treinish), released in March 1988.
[The mixed graphics-and-text extracted did not transfer into anything
meaningful in this plain ASCII file.]
--------
Page 74
END OF PART III.
____________________________________________________________________________
Curator: H. Kent Hills (hills@nssdca.gsfc.nasa.gov)
Author and NASA Official: Robert E. McGuire (mcguire@nssdca.gsfc.nasa.gov)
Head, Space Physics Data Facility (Code 632), NASA/GSFC, Greenbelt, MD 20771
Last Revised: 22 Nov 1996 [HKH]
____________________________________________________________________________