>From: address@hidden (Chris Hennon) >Organization: Ohio State >Keywords: 200102121741.f1CHf4L28141 McIDAS mosaic IMGREMAP Chris, Sorry I couldn't get to this yesterday. >> The job of compositing in McIDAS is as simple as: >> >> o create an AREA that covers the region of interest >> o remap one image into the that AREA >> o remap the other image into the same AREA that the first was remapped >> into > >I have "mcidas terminology deficiency syndrome" (MTDS) - how do you >"create an AREA"? There are a couple of different ways to create an AREA file. Since you are wanting to do something very similar to what we do to composite GOES-East and West, I will describe the process that is instantiated in the GEWCOMP.MCB and GEWCOMP1.MCB example scripts. >Say my area of interest is 0,100W (lower left corner), 50N,2.5W (upper >right corner). There really is a different mind set in McIDAS and GEMPAK. In GEMPAK one always refers to a map area in terms of a lower left corner and an upper right corner. In McIDAS one always refers to a map area in terms of the lower right corner and upper left corner. Pretty weird, isn't it. In defence of McIDAS, all I can say is that it was doing this before GEMPAK :-) Before I begin, I should ask what this composite image will be used in/for. If you intend for the image to be usable in GEMPAK, then we may want to stay away from using a METEOSAT image as the base for remapping. This is because GEMPAK does not support the METEOSAT navigation (or it didn't the last time I looked). This is too bad since METEOSAT images are really nice for using as a remapping base. The reason for this is that the subsatellite point is a simple parameter in the navigation block. Changing it effectively rotates the image to whatever longitude you set the value to be. Remapping into that image is then straightforward. Let's first start with using a METEOSAT image as a remapping base ('cause this one is easy). In the Unidata McIDAS distribution is a number of topographic images. One of these is an image that covers the western hemisphere. This image is in AREA9012, and if you setup the ADDE datasets as per the online installation/configuration instructions, this image will be accessible to you as TOPO/WHEMI. If you didn't setup the TOPO dataset, we can still proceed by setting up the MYDATA dataset. <login as whoever you are running McIDAS as> <start a McIDAS session> BATCH MYDATA.BAT This will setup the MYDATA dataset. To see the elements of this dataset, run: DSSERVE LIST MYDATA You should get a listing that looks like: Group/Descriptor Type Format & Range RT Comment ------------------------ ----- ------------------ -- -------------------- MYDATA/GRIDS GRID GRID 1-9999 All gridded data in GRID fo rmat MYDATA/IMAGES IMAGE AREA 1-9999 All images in AREA format MYDATA/PTSRCS POINT MD 1-9999 All point source data files in MD file format MYDATA/TOPO IMAGE AREA 9000-9019 All topographic images in A REA format DSSERVE: done After this, you should be able to load the WHEMI image in two different ways: SF 1 IMGDISP MYDATA/TOPO.13 EU=TOPO - or - IMGDISP MYDATA/IMAGES.9012 EU=TOPO As you can see, the subsatellite point for this image lies at 0,100W: IMGLIST MYDATA/TOPO.13 FORM=EXP Image file directory listing for:MYDATA/TOPO Pos Satellite/ Date Time Center Res (km) Image_Size sensor Lat Lon Lat Lon --- ------------- ------------ -------- ---- ---- ----- ----- ------------ 13 TOPOGRAPHY 1 JAN 96001 00:00:00 0 100 Band: 1 No Information Available 9.0 9.0 1232 x 1252 proj: 0 created: 1996001 0 memo: --- 1-BYTE TOPOGRAPHY FILE --- type:PRD cal type:BRIT offsets: data= 1280 navigation= 256 calibration= 768 auxillary= 0 doc length: 0 cal length: 0 lev length: 0 PREFIX= 0 valcod: 0 zcor: 1 avg-smp: N pdl: 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 start yyddd: 1996019 start time:160000 start scan: 2500 lcor: 51 ecor: 1 bytes per pixel: 1 ss: 11 Image Center Point Res (derived) Lat: 9.04 (km) Lon: 8.98 (km) IMGLIST: done Now, let's make a copy of this image. This copy will be the AREA that we will use to remap into: IMGCOPY MYDATA/TOPO.13 MYDATA/IMAGES.1234 SIZE=ALL I chose the positon '1234' since the file it represents, AREA1234, is not one typically used by realtime data ingestion at Unidata sites. If you have an image stored in AREA1234 that you do not want to mess with, chose a different position number. For now, I will continue using positon 1234 as an example. Now, comes the part where one has to know something about the navigation block of McIDAS AREA files. This information is available online in the McIDAS Programmers Reference Manual: http://www.ssec.wisc.edu/mug/prog_man/prog_man.html Select Version 7.6 (it was not updated for McIDAS-X Version 7.7), and then go to Chapter 7 - Format of the Data Files. Click on AREAnnnn Area (image) files, and you will get to the section that describes the format of the AREA file. If you read through the description of the AREA file, you will eventually get to the 'Satellite-specific characteristics' page. When there, click on the 'Meteosat PDUS' link. In the link page, you will see a section called 'Meteosat PDUS navigation block' This defines the meaning of the words in the navigation block. The numbering of the words is 1-256, but they start in the AREA file at word 64 (0 based). The documentation tells us that the 'center longitude of rectification (west positive)' is listed in ddmmss (degrees, minutes, and seconds). You can list out the contents of the Meteosat navigation block for the image copy you just made using the McIDAS LWU application: LWU LIST AREA1234 64 64. 1297301844 96016 HEX: 4D534154 17710 ASCII: MSAT w 66. 93356 0 HEX: 16CAC 0 ASCII: l 68. 0 1250 HEX: 0 4E2 ASCII: 70. 1000000 0 HEX: F4240 0 ASCII: B@ 72. 0 96016 HEX: 0 17710 ASCII: w 74. 0 0 HEX: 0 0 ASCII: 76. 0 0 HEX: 0 0 ASCII: 78. 0 0 HEX: 0 0 ASCII: 80. 0 0 HEX: 0 0 ASCII: 82. 0 0 HEX: 0 0 ASCII: 84. 0 0 HEX: 0 0 ASCII: 86. 0 0 HEX: 0 0 ASCII: 88. 0 0 HEX: 0 0 ASCII: 90. 0 0 HEX: 0 0 ASCII: 92. 0 0 HEX: 0 0 ASCII: 94. 0 0 HEX: 0 0 ASCII: 96. 0 0 HEX: 0 0 ASCII: 98. 0 0 HEX: 0 0 ASCII: 100. 0 0 HEX: 0 0 ASCII: 102. 0 0 HEX: 0 0 ASCII: --END OF LISTING Word 70 in this list is the subsatellite longitude * 10000 (1000000). Since you are interested in longitudes that range from 2.5W to 100W, we should change the subsatellite point to 51.25 W. (NOTE that longitudes in McIDAS are specified as West positive (go figure)). 51.25 degrees is 51 degrees and 15 minutes. We use LWU again to change the subsatellite longitude LWU POKE AREA1234 511500 70 Now, the image has effectively been rotated so that the center of the image is at the center of your area of interest. Next display the image: SF 2 IMGDISP MYDATA/IMAGES.1234 LATLON=30 51:15 EU=TOPO MAP H Notice how the map does not conform to the topographic features. This is to be expected since we changed the subsatellite point, but we havn't (yet) done anything with the data. The next step is remapping the GOES-East and METEOSAT images into this base. I don't know what your datasets for these images will be, so I will show the steps used to remap RTIMAGES/GE-IR into the image and then display the results: IMGREMAP RTIMAGES/GE-IR MYDATA/IMAGES.1234 SMOOTH=YES IMGDISP MYDATA/IMAGES.1234 LATLON=30 51:15 EU=IMAGE REFRESH='EG;MAP H' So far, we have remapped the GOES-East IR image into the centralized projection we want. The next step would be to merge the METEOSAT image into the same thus creating a composite. Since I don't have a METEOSAT image to use, I will simply pretend that there is a dataset of METEOSAT IR images called MSAT/IR. IMGREMAP MSAT/IR MYDATA/IMAGES.1234 MERGE=YES SMOOTH=YES IMGDISP MYDATA/IMAGES.1234 LATLON=30 51:15 EU=IMAGE REFRESH='EG;MAP H' At this point, you will have a composite of METEOSAT and GOES-East IR, but you may say that this is not how you wanted to create the composite (since the values for the meteosat image will overlap/replace those of the GOES-East image everwhere the METEOSAT image values exist). We are not done at this point. We can do some other steps so that the compositing will be done along a particular longitude so that every pixel to the west will be from GOES-East, and every pixel to the East will be from METEOSAT. Before we get into how to do this, I think that you should give the above steps a try and see if you follow what is happening. As soon as you are past this, we can work through doing the compositing using a split method. For an example of composited GOES-East and GOES-West images, take a look at: DATALOC ADD RTIMAGES adde.ucar.edu IMGDISP RTIMAGES/GEW-IR LATLON=30 100 EU=IMAGE REFRESH='EG;MAP H' The compositing for these images is done stepwise. First the GEOS-West image is added and when the new GOES-East image is received it is added. If your display only contains the west image, then do: IMGDISP RTIMAGES/GEW-IR.-1 LATLON=30 100 EU=IMAGE REFRESH='EG;MAP H' to display the previously done composite. Let me know if you have any difficulties in the example above. As soon as you are comfortable with what is going on, we can proceed to more exotic ways of combining the images. Tom
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.