LEROI_TEM

The file LEROI_TEM.F90 is adapted from the source file Leroi_FM_TEM_subs.f90, so that the LEROI_TEM can be run via Matlab's MEX interface.

Compiling

The MEX interface can be compiled with:

mex LEROI_TEM.F90

Compiler optimisation options can be specified with FOPTIMFLAGS, e.g.:

mex -v LEROI_TEM.F90 FOPTIMFLAGS='-O2'

The Fortran .mod files can be removed after compilation.

Input

Various input values are hard coded in the Fortran source, and are equivalent to the values in the provided Leroi.cfl file:

Offset Loop with 1 receiver - Model: 2 layers over basement, 2 plates 1 0 0 1 0 ! TDFD, DO3D, ISYS, PRFL, ISTOP 0 4 1 ! STEP, NSX, KRXW 1 ! SURVEY_TYPE 1 1 1 1 4 1 ! NLINES, MRXL, NTX, SOURCE_TYPE, MVRTX, NTRN 4 0 ! NVRTX TxZ 1 1 1 1 ! LINE IDTX, RX_TYPE, NRX 3 0 0 1 0 1 ! CMP SV_AZM, KNORM, IPLT, IDH, RXMNT

These values are used by the Leroi subroutine READ_SYSTEM_AND_SURVEY_DATA.

Nineteen additional input values are required from Matlab for the variables NCHNL, REFTYM, OFFTYM, TXON, TXAMP, TOPN, TCLS, SXE, SXN, RXE, RXN, RXZ, NLYR, NPLT, NLITH, LYTH, LITHL, THK and UNITS.

In the Leroi.cfl these would be specified as:

``` 20 1.05 1.05 ! NCHNL, REFTYM, OFFTYM 0.0 0.0 0.001 1.0 1.0492 1.0 1.05 0.0 ! TXON, TXAMP(NSX) 0.006000 0.007625 0.007625 0.009750 0.009750 0.012500 0.012500 0.015880 0.015880 0.020250 0.020250 0.025880 0.025880 0.033000 0.033000 0.042130 0.042130 0.053750 0.053750 0.068500 0.068500 0.087380 0.087380 0.111400 0.111400 0.151700 0.151700 0.181100 0.181100 0.231000 0.231000 0.294600 0.294600 0.375900 0.375900 0.479500 0.479500 0.611600 0.611600 0.780100 ! TOPN, TCLS(NCHNL) (in ms) 5 5 -5 5 -5 -5 5 -5 ! SXE, SXN(MVRTX) 4 ! UNITS 15 0 0 ! RXE, RXN, RXZ

3 0 3 !! NLYR, NPLT, NLITH 9.9773 -1 1 1 0 0 1 2035.84 -1 1 1 0 0 1 887.959 -1 1 1 0 0 1 !! LYTH(NLITH, 7) 1 94.2462 2 34.1019 !! LITHL(NLYR - 1), THK(NLYR - 1) 3 !! LITH(NLYR) ```

In Matlab, these would be specified as:

``` NCHNL = 20; REFTYM = 1.05; OFFTYM = 1.05; TXON = [0.0, 0.001, 1.0492, 1.05]; TXAMP = [0.0, 1.0, 1.0, 0.0]; TOPN = [0.006000, 0.007625, 0.009750, 0.012500, 0.015880, ... 0.020250, 0.025880, 0.033000, 0.042130, 0.053750, ... 0.068500, 0.087380, 0.111400, 0.151700, 0.181100, ... 0.231000, 0.294600, 0.375900, 0.479500, 0.611600]; TCLS = [0.007625, 0.009750, 0.012500, 0.015880, 0.020250, ... 0.025880, 0.033000, 0.042130, 0.053750, 0.068500, ... 0.087380, 0.111400, 0.151700, 0.181100, 0.231000, ... 0.294600, 0.375900, 0.479500, 0.611600, 0.780100]; SXE = [5, -5, -5, 5]; SXN = [5, 5, -5, -5]; RXE = 15; RXN = 0; RXZ = 0;

NLYR = 3; NPLT = 0; NLITH = 3; LYTH = [9.9773, -1, 1, 1, 0, 0, 1; ... 2035.84, -1, 1, 1, 0, 0, 1; ... 887.959, -1, 1, 1, 0, 0, 1]; LITHL = [1, 2, 3]; THK = [94.2462, 34.1019]; UNITS = [4]; ```

Then, to run the model in Matlab:

[X, Y, Z] = LEROI_TEM(NCHNL, REFTYM, OFFTYM, TXON, TXAMP, TOPN, TCLS, SXE, SXN, ... RXE, RXN, RXZ, NLYR, NPLT, NLITH, LYTH, LITHL, THK, UNITS)

This will return three output double vectors, X, Y and Z.

If the UNITS value is omitted, a default value of 4 will be used:

[X, Y, Z] = LEROI_TEM(NCHNL, REFTYM, OFFTYM, TXON, TXAMP, TOPN, TCLS, SXE, SXN, ... RXE, RXN, RXZ, NLYR, NPLT, NLITH, LYTH, LITHL, THK)

If a single output argument is provided, only the Z values will be returned:

Z = LEROI_TEM(NCHNL, REFTYM, OFFTYM, TXON, TXAMP, TOPN, TCLS, SXE, SXN, ... RXE, RXN, RXZ, NLYR, NPLT, NLITH, LYTH, LITHL, THK, UNITS)

The Fortran code will validate input values before running, e.g.:

``` ...

NLYR = 3.3;
NPLT = 0; NLITH = 3; LYTH = [9.9773, -1, 1, 1, 0, 0, 1; 2035.84, -1, 1, 1, 0, 0, 1; 887.959, -1, 1, 1, 0, 0, 1]; LITHL = [1, 2, 3]; THK = [94.2462, 34.1019];

LEROITEM(NCHNL, REFTYM, OFFTYM, TXON, TXAMP, TOPN, TCLS, SXE, SXN, ... RXE, RXN, RXZ, NLYR, NPLT, NLITH, LYTH, LITHL, THK) Error using LEROITEM Argument 13 (NLYR) should be an integer. ```

Units

Valid values for the UNITS are:

• 1: volts (V)
• 2: millivolts (mV)
• 3: microvolts (mu-V)
• 4: nanovolts (nV)
• 11: nanoteslas / sec (nT/s)
• 12: picoteslas / sec (pT/s)
• 21: nanoteslas (nT)
• 22: picoteslas (pT)
• 31: ratio
• 32: percent
• 33: parts per thousand (PPT)
• 34: parts per million (PPM)
• 35: parts per billion (PPB)
• 41: volts per metre (V/m)
• 42: millivolts per metre (mV/m)
• 43: microvolts per metre (mV/m)
• 44: nanovolts per metre (nV/m)

Limitations

The values which are hard coded in the Fortran source, and passed to the READ_SYSTEM_AND_SURVEY_DATA subroutine, affect the way that model runs, and what input is required.

If these values were to be changed, additional changes would likely also be required elsewhere in the code. For example, changing the SURVEY_TYPE value will likely change the type of input data required.