In dschlaep/ecotoner: ecotoner - ecotones / ecological boundaries across environmental gradients

The algorithm searches different sized windows which are centered around a search point, whether it is possible to establish a transect across an ecotone. That means that a transect can be uniquely named after the search point (location), the neighborhood window size, and the random number generator seed.

• Generation of search points

• Loop over each search point

  • Call function 'crop_to_neighborhood': crop raster grids to maximum neighborhood extent
  • Loop over each neighborhood window size
    1. Set random seed (either arbitrarily or with a seed unique to search point and neighborhood window size)
    2. Call function 'establish_ecotone_transect'
       1. Call function 'locate_candidate_transect_lines'
          1. Establish candidate transect lines of a specified transect type
             • If type 1 or type 3, then call 'locate_max_value' to locate the point with the maximum value of the environmental grid within the neighborhood window.
               • If type 1, then the one candidate line is the direct line between the maximum point and search point which is extend by factor 2 past the maximum point and from which the section between the minimum and maximum values is extracted by calling 'elongate_linear_transect'.
               • If type 3, then the one candidate curve is the flowpath generated with a call to 'calc_flow_path', starting from the maximum point until it reaches the minimum point within the neighborhood window.
             • If type 2, then call 'locate_steepest_slope' to locate the point which has the steepest mean slope when connected with the search point in a line. The one candidate line is the direct line between this point and the search point which is extend by factor 2 past both points and from which the section between the minimum and maximum values is extracted by calling 'elongate_linear_transect'.
             • If type 4, then call 'locate_candidate_THAs' to sample candidate lines of sufficient length along homogeneous areas of aspect
          2. Loop over each candidate line until the first successful candidate
             1. Call function 'orient_transect_line'
             2. Call function 'locate_candidate_transect_band'
                • Extract band transect along elevation transect with a width of 200 cells
                • Calculate outline of maximal band transect along elevation transect and how to transform from original coordinates to band transect coordinates
             3. Call function 'determine_ecotone_linear_extent': Apply definition of zone of ecological boundary to BSE and TF
             4. Call function 'trim1_band_to_ecotone': Crop band transect to limits1_ZoneEcolBoundary
             5. Determine majority BSE and majority TF types within limits1_ZoneEcolBoundary
             6. Call function 'determine_ecotone_linear_extent': Apply definition of zone of ecological boundary to majority of BSE and majority of TF within limits1_ZoneEcolBoundary
          3. If a candiate successful
             1. Call function 'trim_line_to_ecotone': Crop the linear transect to the ecological boundary zone
             2. Call function 'extract_tband_grids': Crop the band transect to the ecological boundary zone
             3. Call function 'trim2_band_to_ecotone': Crop band transect to zone of ecol boundary and set start of 'my' coordinate system
             4. Environmental gradient along transect
                • Extract environmental conditions along transect band
                • Quality/Homogeneity of environmental gradient along transect
                  • Human footprint
                  • Quality of aspect: local (RMSE), patches (clumps), and global (mean) aspect
                  • Quality of elevation
             5. Vegetation data along transect: Need to recalculate veg1, veg2, veg1density, and veg2density based on patches
                • Quality: vegetation patch sizes
    3. Call function 'identify_migration_patches':
       1. AllMigration: all vegetation in transect
       2. OnlyGoodMigration
          • all vegetation minus patches whose flowpath connects to beyond y-border indicating non-x-axis migration origin
          • identify patches that 'drain' out either at y=1 or at y=200; assumption: most likely migration route = flowpath
       3. Call function 'calc.MigrationRoutes_EstimateFlowpaths':
       4. Call function 'calc.Identify_GoodvsBadMigration':
       5. Call function 'calc.RemoveBadMigration_fromVeg':
  • end of loop for each neighborhood
• end of loop over each search point

dschlaep/ecotoner documentation built on May 15, 2019, 2:57 p.m.