Inputting the General LUT Configuration parameters is the first step to define the configuration of a LUT in ALG, independently of the selected atmospheric RTM.
This step appears when the user press the New LUT configuration button () or, alternatively, through the menu bar by
clicking on File→New→LUT config..
The user starts by introducing the LUT file path/name in the corresponding Save LUT file(s) text box or by pressing the … button (1). All LUT-related files and folders will have the introduced name but different extensions (e.g., .h5, .hdr, .bat, .xml). By default, ALG sets the LUT path to the LUTs_folder folder but this can be edited in the software preferences.
Then, the number of parallel CPU cores used to run the RTM is specified in the # parallel runs pop-up menu (2). This number is automatically limited by the maximum CPU cores of the PC where ALG is installed. By selecting a value of the N parallel runs, ALG will automatically distribute all the RTM simulations in N different sub-folders with the path/name introduced in the Save LUT file(s) text box and a suffix _i (with i from 1 to N).
Next, the user level (Basic, Medium or Advanced) is selected in the radio-button group at the bottom-left of the General LUT Configuration panel (3). This parameter allows users to have access to more advanced LUT configuration parameters and options such as the key input variables and the LUT sampling strategy.
The general LUT configuration follows with the selection of the RTM used to generate the LUT data and its operation mode (4). The available atmospheric RTMs are described in the Third-party Software section of this Help guide. With respect to the operation mode, the following options are currently implemented:
Notice that not all these operation modes are available for all atmospheric RTMs. Particularly, the transmittance mode is only available for MODTRAN5 and MODTRAN6.
Once selected the RTM and its operation mode, the user must choose the sampling strategy on its pop-up menu (5). This parameter defines the distribution of the LUT nodes in the input variable space. In ALG, the current available options are:
Depending on the chosen sampling strategy, different options will then be available to evaluate the stop condition, i.e., to determine the number of nodes included in the LUT (6):
When the user selects the cross-validation error option to evaluate the stop condition, two additional parameters will be visible.
This first of them allows users to introduce the relative error threshold (in %) (7). The second, allows selecting the surface reflectance spectral (8). In combination with with the atmospheric transfer functions in a LUT, this parameter allows generating the TOA radiance used to evaluate the interpolation error through cross-validation. Two options are available: Vegetation (conifers) (Baldridge et al., 2009) and Sea water.
Once introduced all these parameters, the user should click on the Next button at the bottom-left corner to proceed introducing the key input atmospheric RTM variables (or the alternative version when the Read distribution option is selected in the sampling strategy). This will create a LUT configuration .xml file with the selected file path/name (e.g., C:\ALG\LUTs_folder\LUTfile.xml), storing these parameters for their later use in the execution of the atmospheric RTMs and generation of input data.
McKay, M.D., Beckman, R.J., & Conover, W.J. (1979), "A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code." Technometrics. Vol. 21, No. 2, pp. 239‒ 245.
Bratley, P., & Bennet, L.F. (1988), "Algorithm 659: Implementing Sobol's quasirandom sequence generator." ACM Transactions on Mathematical Software (TOMS). Vol. 14, No. 1, pp. 88‒ 100.
Halton, J.H. (1964), "Algorithm 247: Radical-inverse quasi-random point sequence." Communications of the ACM. Vol. 7, No. 12, pp. 701‒ 702.
Baldridge, A., Hook, S., Grove, C., & Rivera, G. (2009), "The ASTER spectral library version 2.0," Remote Sensing of Environment. Vol. 113, No. 4, pp. 711‒ 715.
Vicent, J., Alonso, L., Martino, L., Sabater, N., Verrelst, J., Camps-Valls, G. & Moreno, J. (2018), "Gradient-based Automatic Look-Up Table Generator for Radiative Transfer Models," IEEE Transactions on Geosciences and Remote Sensing. Vol. 57, No. 2, pp. 1040‒ 1048.