Click here to return to course outline.
Click here to return to W.R. Church's home page.
Click here to return to Earth Science course list.
Files: c: iomega(course):\aacrse\505\idr\ex5; Earthnt\public\505\ex5
In this exercise we are provided with two raster images, one showing the regional distribution of various types of granite in a batholith, the other the interpolated pattern of Thorium/Uranium ratio values in the batholith. On the assumption that potassic granites with low Th/U ratios (< 9) are most likely to be mineralized, create an image showing the distribution of granitic rocks with these attributes, and calculate their area.
Procedure:
1) Go to your folder in 'users on 'Earthnt' (H:)', where 'H' is the name of the mapped drive Earthnt\Users on your computer (it could be some letter other than 'H', or it may not exist and will have to be created), and make a folder named 'your initials5' KEEP READING! - (do not include the quotes, and 'your initials' really means your initials, e.g. wrc5, and not the string 'yourinitials'). If you have not previously used the computers in room 17 and are baffled by this instruction, do not panic, request the help of the instructor.
Fetch the window corresponding to 'Earthnt\Public\505\idr' or 'Earthnt\Public\Es300b'and copy the folder 'ex5' and its contents into the folder you have just created, by clicking and dragging the 'ex5' folder from the 'Public' window into the 'Users' window.
Load IDRISI (via Start and Programs) and set the Environment to the folder you have just created in your User area: to do this click (press the left mouse button) Environment on the Idrisi menu bar followed by ENVIRON; set the drive (e.g. 'Earthnt' (H:)) and directory ('your name/yourintials5/ex5') path to your folder, and the default reference units to metres. The 'Path of Working Data Directory' box should show something like "drive name\wrchurch\wrc5\ex5". If it doesn't, get help from the instructor.
The image KGRANEX5 shows the distribution of a set of granites: monzonite (cell value 1), high-K granite (cell value 2), granodiorite (cell value 3), tonalite (cell value 4), and diorite (cell value 5). Display KGRANEX5 (map of granite types) using the the QUAL16 palette: 1) with autoscaling, and 2) without autoscaling. Examine the QUAL16 palette and explain the difference between the two images. (Note that the autoscaled image takes every third colour in the palette (ie. 5 granite classes x 3 = 15 colours, whereas the non-autoscaled image takes the first first five colours of the palette.). Examine the cell values using the Query tool (the ? button on the tool bar.)
Display TUEX5 (a map of Thorium/U ratios) with the QUAL16 palette.
Use DESCRIBE (in Files) TUEX5 to examine the characteristics of the file; note the min and max Z values are 5 and 16, respectively, and that the values are REAL. Given the number of rows and columns and the min/max coordinates of the image, what is the cell resolution?
CREATING BOOLEAN IMAGES
Note: a 'BOOLEAN IMAGE' is composed of values of only '0' (false) or '1' (true). In the case of the image BSTGRAN you are about to create from the image KGRANEX5, it reflects the question 'Does the cell represent a high-K granite or not?, YES or NO?'; in the case of the image TURECEX5 to be created from the image TUEX5, it reflects the question 'Does the cell have an attribute value of less than 9? YES or NO?'. In both cases, if the answer is 'yes', the cell is given a value of '1'(TRUE), if the answer is 'no', it is given a value of '0' (FALSE).
To create a Boolean image of Th/U values < 9 using RECLASS to carry out a single attribute query
Run RECLASS (in Analysis-DataBase Query): Image, User defined, Input TUEX5, Output TURECEX5, Boolean value units. Give a value 1 to Th/U values of 0 to <9 (=class 1). Click the right side augment spin button (triangle) on the right side of the menu and give a value of 0 to Th/U values 9 to <999 (=class2). Click OK.
The image TURECEX5 will be generated automatically. Compare it with the image TUEX5, and examine the values of the same cell in both images using the query tool (the question mark button on the tool bar).
To create a Boolean image of high K-Feldspar Granite using ASSIGN to carry out a single attribute query
Use EDIT (in Data Entry) to write a values file GRANITES, with granite classes 1 (monzonite), 3 (granodiorite), 4 (tonalite), and 5 (diorite) assigned to 0 and granite class 2 (high-K granite) to 1.
In this case simply give all cells representing high-K granite with a value of 2 a value of 1 by typing in the instruction:
2 1 (i.e. 2 space 1)
Any unassigned value, that is the values 1,3, 4, and 5, representing all the other granite classes, will be given a default value of 0. SAVE the file, and exit Edit.
Run ASSIGN (in Data Entry) inputting KGRANEX5 (image) and GRANITES (values) and outputting BSTGRAN. The latter image will show the distribution of only the high-K granite (granite class 2), the granite body we, as geologists, have decided will most likely have the most mineralization potential.
The image BSTGRAN will be generated automatically. Again, compare BSTGRANwith KGRANEX5.
Next we need to create an image showing only those high-K granites with an Th/U ratio of less than 9. This is done by overlaying the images BSTGRAN and TUEX5 and comparing the values of the same cell in both images. If both cells have a values of 0, they will be assigned a value of 0; if either of the cells has a value of 0, again the cell will be assigned a value of 0; if both cells have a value of 1 (TRUE), then the cell will be assigned a value of 1. You will note that this operation is easily carried out by multiplying the same cell value in both images one with the other. In other words: 0 x 0 = 0 (FALSE, no potential mineralization); 0 x 1 = 0 (FALSE, no potential mineralization; 1 x 1 = 1 (TRUE; potential mineralization)!!
To create a Boolean image showing high K-Feldspar Granite with Th/U ratios <9, using OVERLAY to carry out a multiple attribute query.
Run OVERLAY (in Analysis-Database Query) and multiply TURECEX5 and BSTGRAN, and name the new image BSTMIN.
CALCULATING AREA
To calculate the total area of High K-feldspar granite with Th/U < 9, using the operation AREA.
Run AREA (in Analysis-Database Query) and input BSTMIN, tabular output; area in square kilometres.
To calculate the areas of non-contiguous high K-Feldspar granite with Th/U < 9, using GROUP and AREA.
Theory:
'Group' is a boolean strategy that gives areally non-contiguous sets of cells with the same attribute value, a different attribute value. As illustrated in the following example, the top leftmost cell in BSTMIN, with an attribute value of '1', is assigned a value of '0', as are all other '1' cells located immediately to the right, directly below or diagonally below the cell. The third cell in row 1 with a value of '0' is then given a value of '1', as, again, are all other '0' cells located immediately to the right, directly below or diagonally below the cell. The fourth cell in the first row, although a '1', is not contiguous with the other '1' cells, and is therefore assigned an attribute value of '2'. The cell directly below is also given a value of '2'. The next non -contiguous cell (the '1' cell at the bottom right) would be assigned a value of '3' - and so on for each group of non-contiguous cells in the matrix. The new matrix would be as shown in the image PLOTS1. The attribute value of each cell is then increased by a value of 1, and each cell mulitplied by it's original attribute value stored in BSTMIN.
BSTMIN PLOTS1 PLOTS2 PLOTS3 1 1 0 1 0 0 1 2 1 1 2 3 1 1 0 3 1 0 0 1 ------> 0 1 1 2 ------> 1 2 2 3 -------> 1 0 0 3 0 1 0 0 Group 1 0 1 1 SCALAR 2 1 2 2 PLOTS2 * 0 1 0 0 1 0 0 1 0 1 1 3 + 1 1 2 2 4 BSTMIN 1 0 0 4The 'Plot3' matrix is now composed of four groups of attribute values, 1, 0, 3, and 4. The original single set of cell values of '1' (TRUE) now have values of 1, 3, and 4, whereas the background '0 set' (FALSE) cells once again have a value of '0'. (Tricky, n'est-ce pas?)
Run GROUP (in Analysis, Context Operators) on BSTMIN, output PLOTS1; include diagonals.
Use OVERLAY and multiply PLOTS1and BSTMIN to create PLOTS2.
It is not necessary to carry out theSCALAR addition of '1' in this particular operation because the first top left cell has a value of '0', and therefore the value of '0' will be assigned to a cell set already with a value of '0'. Other cells with a value of '0' located within an area of '1''s will be assigned a value other than '0' ('2', '4', '5', '6', '7', and '9' in the case of BSTMIN). However, once the OVERLAY operation has been carried out, the cell value as shown in PLOTS2 will revert to '0'. (Make sure you understand why - examination question!!) If the top left cell were not '0' it would have been necessary to carry out the scalar operation as indicated above before running OVERLAY.
Run AREA
to determine the area occupied by each set of contiguous cells.
Structural Provinces of North America.
RETURN TO:
Click here to return to beginning.
Click here to return to course outline.
