The combination of Python and IDL software is used for the operational monitoring of thermal anomalies over the selected volcano. Python script takes care of automatic data and results transport over FTP connections. The hotspots are detected and characterized using IDL software.
The system has been tested on a computer with 2.4 GHz processor and 2 Gb of RAM using:
•    operating system MS Windows Professional (ver. 2002), SP2
•    Python 2.41
•    RSI IDL 7.0.0
•    IDL Windows console application wincl_7.1255.0.0.zip by Abraham Campbell

3.1 Installation on Windows

Considering that you have previously already installed a compatible version of Python and IDL you still have to set up some details. First of all, the IDL Windows console application has to be installed. There is no direct link to download the file, go to www.ittvis.com/UserCommunity/CodeLibrary.aspx and search for the author “Abraham Campbell” or "wincl_7.0.zip". Open the zip file and copy “idl.exe” into the “…IDL70\bin\bin.x86” directory. For more references see the “wind.doc” file within the zip.
Secondly, copy the “run_modis.pro” IDL script to your “home” folder (start command line and the folder where you currently are is the one).
Copy all other IDL files to a selected folder on a local disk. This path has to be set in just mentioned “run_me.pro”.
Set the global variables in the “global_data.pro”. At least next variables have to be considered: s_project, s_volcano, d_lon, d_lat.
Set the folder where the data will be downloaded in the “run_non_stop.py”. You also need to set up the volcano longitude and latitude. Optionally, set up the interval used to check if new data are online.
To finally run the script, simply double click the “run_non_stop.py”. The Python scripts runs in the command line modus. If it successfully downloads data then it calls IDL Windows console application that runs the IDL procedures. The results (XML and GEOTIFF files) are then copied to the ZMAW FTP where Exupéry server accesses them. If the script hangs (because of the network connection), just close it and restart.

3.2 Installation on Linux

To be included soon.

3.3 Description of processing parameters

Within the near real time monitoring for Exupéry VFRS we used only MODIS (sensor aboard Terra and Aqua satellite) data. These are available on NASA FTP. Usually you get four images per day (depends on the latitude). We used level 1b data that are radiometrically calibrated. The important parameters for data processing are listed below. These parameters can be set up in the different parts of the code. In the case of MODIS malfunction or using some other satellite data Python script must be rewritten; some parts of IDL code have to be then adapted too.

GLOBAL_DATA.PRO

In this IDL file you can set up the most important data that do not depend on the satellite characteristics but contain some technical information (name of the project, etc.) or some physical constants (lava density, etc.).

Project name (decision by the Exupéry team in the project beginning)
    s_project = 'ExuperyTest'
    Volcano code (http://www.volcano.si.edu/)
    s_volcano = '1802-09='

Predefined emissivity in the appropriate channels (I used 0.95 for USEReST)
    d_e_SWIR= 0.97
    d_e_MIR = 0.97
    d_e_TIR = 0.97

Lava parameters (needed to estimate the effusion rate): difference between the eruption temperature and the temperature where the lava motion ceases [K], lava density [kgm^-3], heat capacity [Jkg^-1K^-1], crystallization percentage, latent crystallization heat [Jkg^-1]
    d_tstop = 200.
    d_rolava = 2500.
    d_cplava = 1225.
    d_cryst_perc = 0.03
    d_lat_crst_heat = 3.5*10.^5

First guess for hotspot temperature [K]
    d_gues_t = 700

Volcano centre lontitude and latitude in decimal degrees
    d_lon =-25.47
    d_lat = 37.77

Resolution of the output images in decimal degrees; together with the parameter below, you can set up the image size (using the proposed parameters, 301 * 301 pixels large)
d_resolution = 0.005

Distance between the image centre and image border in decimal degrees (be aware that in Polar Regions the image is not isotropic – e.g. at 60 degrees the scaling factor in N–S direction is twice smaller than in E–W direction).
d_volc_radius = 0.75     

READ_MODISL1B.PRO

In this IDL file you can set up some sensor specific data. The most important are listed below. There are more small parameters that are integrated into the code. Such parameters are e.g. thresholds (together with the rules) used for cloud and sun glint mask.

Wavelengths [μm]
    d_l_ch2 = 0.8585
    d_l_ch6 = 1.640
    d_l_ch21 = 3.959
    d_l_ch22 = 3.959
    d_l_ch29 = 8.550
    d_l_ch31 = 11.030
    d_l_ch32 = 12.020

Nominal nadir resolution [km]
    d_resolution_km = 1.

NTI threshold for day and night time (if NTI is greater than the threshold, then the code detects an anomaly – CRITICAL PARAMTER)
    d_NTI_treshold_day = -0.6
    d_NTI_treshold_night = -0.80

RUN_NON_STOP.PY

In this Python file you can set up some general data required to download the data from FTP. In addition to those you can of course adjust the FTP connections in the rest of the script.

Volcano centre lontitude and latitude in decimal degrees
    d_lon =-25.47
    d_lat = 37.77

Path on the local disk to the satellite data
    path_satellite_data = 'F:\\Exupery\\data_results\\'

Time delay in seconds before running the python script again – the Python first searches if there are some new data for the last nm_days (the parameter below), it might also run IDL, etc. and when it is over, it takes a time of for ftp_time.
    ftp_time = 3600

Number of previous days to check if new data for these days exist on NASA FTP
    nm_days = 5