Habitat Corridor Modeling Process
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Habitat Corridor Modeling Process as PDF for free.
More details
- Words: 1,755
- Pages: 12
Grid Modeling and Map Algebra: Calculating a Habitat Corridor
Geomatics 2022 Cartography: Digital Mapping February 15, 2009 Joe Fraser
Description of the Exercise Calculating a habitat corridor requires a number of variables. First, two areas of suitable habitat must be determined. Secondly, a path between these two areas must be mapped, connecting the two in a corridor. The suitable habitat grid is made up of four variables in this exercise: proximity to dirt roads, water, and forest edge, and the fourth variable is percentage of slope. Each of these is ranked from 1 to 10, with lower numbers assigned to the cells most favourable to
the habitat of deer. Corridor impedance is a grid made from variables detrimental to the habitat of deer. Paved roads serve vehicle traffic, and deer need to be a certain distance from a source of water the protection of the forest. The overall goal of the exercise is to find the best path according to the variables of suitability and impedance to connect two existing habitats.
Part 1: Building a Deer Suitability Grid Proximity to Resource Road Grid Step 1: Selecting Unpaved Roads Because deer are drawn to the edges of forest clearings, low-traffic roads provide two areas where they are likely to be found. To include unpaved roads in the deer suitability grid, select them from the “q_rd” file. Shown below at left is the resultant unpaved roads grid, at right the unpaved roads grid is compared to the paved roads proximity grid created later on, to make sure that a paved road and an unpaved road do not overlap. In this case two roads were removed because of this comparison.
Figure 2: “unpaved_rd1”
Figure 1: “select” command
Figure 3: “unpaved_rd1” vs. “proximity to paved roads
Step 2: Buffer the Unpaved Roads A temporary grid is needed to show the graduated distance from the paved roads. This grid will be reclassed to be used as the resource road proximity suitability grid.
Figure 4: “eucdistance” command
Page 2
Cartography: Digital Mapping GEOM 2022
Figure 5: “road_suit_tmp”
Joe Fraser
Step 3: Reclass Using a Remap Table Using the remap table, the proximity to the center of the dirt road is given a ranking, with 1 as the most suitable, 10 as the least. In this remap table, the actual road is given a value of 100 to discount that area. The best areas assigned in the remap table are from 1m to 75m, with up to 150m as preferred. Further than that the area is less valuable, and from 250m to the end of the eucdistance at 1205 are set to 10. Below is the grid.
Figure 6: “reclass” command
Figure 7: “unpavedroad_suit.txt” remap table
Figure 8: “road_suit”
Step 4: Proximity to Riparian Areas Suitability The process of determining suitable locations according to proximity to watercourses begins with a eucdistance on a grid with the streams and rivers.
Figure 9 “eucdistance” command
Page3
Cartography: Digital Mapping GEOM 2022
Figure 10 “water_temp”
Joe Fraser
Step 5: Using the Remap Table to Reclass Similar to the unpaved road remap table, water_suit.txt is used to classify the distance away from the stream. A value of 10 is given to anything beyond 250m, assuming that deer need to stay relatively close to a source of water. The stream itself is given 100 as well, and the area within 100m is the most suitable. The grid water_suit will be used as a variable in the deer suitability grid.
Figure 12 “reclass” command
Figure 11 “water_suit.txt” remap table
Figure 13 “water_suit”
Step 6: Field Edge Proximity Suitability Here is an example of a nested command. Using the select function, a eucdistance of 10 is created on only the value of 10 from the previously created field grid. The value of 10 represents the edges of the field.
Figure 14: nested commands “select” and “eucdistance
Page 4
Cartography: Digital Mapping GEOM 2022
Figure 15 “water_suit”
Joe Fraser
Step 7: Buffering the Edges of the Fields Eucdistance provides a buffer from the edge of the fields. Again, this grid will be reclassed to create a ranked suitability grid.
Figure 16 “eucdistance” command
Figure 17 “f_edge_buf”
Step 8: Creating a Field_suit Grid Field_suit.txt is used to reclassify the values in the field_buf grid. Areas further than 250 metres are classified as being undesirable. This way, land close to orchards is weighted higher, assuming that the food Figure 19 “reclass” command using a remap table deer like to eat best grows within a short range of the clearing edge. Below is field_suit, the grid that will be the third variable in the habitat suitability grid. Figure 18 “field_suit.txt” remap table
Figure 10: “streams_ord”
Figure 20: “field_suit”
Page 5
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 9: Slope Suitability Using the dem1 grid, Spatial Analyst can create a slope grid with percentages.
Figure 23: “dem1”
Figure 22: Spatial Analyst Dialogue
Figure 21: Spatial Analyst Menu
Step 10: Reclassing the Slope Percentage Grid Using the assumption that deer are fairly sure-footed animals, the slope_suit grid was created. Grades of 0 to 10% were given a value of 1, and up to 50% slope was characterized as acceptable. Values of 50 to 75% aren’t prohibitive if other values in the deer suitability grid are favourable, but grades greater than 75% were ruled out. Slope is an important factor to consider when terrain is varied, as a chasm or rock outcrop can interrupt a corridor and render calculations incorrect if forgotten.
Figure 24: “Slope_Percent”
Figure 26: “reclass” command
Figure 25: “slope_suit.txt”
Figure 27: “slope_suit”
Page 6
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 11: Creating the Deer Suitability Grid The deer suitability grid is made up of the four above grids: road_suit, water_suit, field_suit, and slope_suit. At right, the map algebra formula shows the weighting of these variables, with water and field proximity being 3 and 2 times more important, respectively, than road and slope suitability. The resultant grid is shown below.
Figure 28: adding weighted grids
Figure 29: “deer_suit”
Part 2: Creating Hexagon Patches Step 1: Create a Point Grid The first step in creating the hexagon patches is to make the point grid with the queens_clip boundary extent. The distance between the triangulated points is 537.29, to create a final output of 25 hectares. Figure 30a: GeoWizards menu
Figure 30c: GeoWizards menu
Figure 30b: GeoWizards menu
Figure 30d: GeoWizards menu
Figure 31: “point_537”
Page 7
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 2: Build Thiessen Polygons Based on the points grid created in step 1, Thiessen polygons can be made. These are the hexagonal patches that will be filled with an average value using the zonal mean command. Figure 32b: Build Polygons Menu Figure 32a: Build Thiessen Polygons Menu
Figure 32c: Build Polygons Menu
Figure 33: “hex_537”
Step 3: Cleaning the Polygons The polygons need to be cleaned using the ET GeoWizards add-on before they are converted to a grid.
Figure 34a: Clean Polygons Menu
Figure 34c: Clean Polygons Menu
Figure 34b: Clean Polygons Menu
Figure 34d: Clean Polygons Menu
Figure 35: “hex_537c”
Step 4: Converting Polygons Finally, the polygons can be converted to grid using the ET_ID field. This ensures each hexagon has a unique identifier. The reason a Figure 36: Convert Features to Raster standardized hexagon patch is used is because of its shape unlike a square, the hexagon is equidistant each adjacent hexagon, making it ideal for costpath analysis or direction grids. Using a square shape favours the squares above, below and to the side.
Page 8
Cartography: Digital Mapping GEOM 2022
Figure 37: “hex_537c2”
Joe Fraser
Part 3: Prepare a Zonal Mean Grid Create a Zonal Mean Grid Using the deer suitability grid, a zonal mean command fills in the hexagon patches with the average value. What is created is in essence an average of the areas deer are likely to prosper, ranked from low (most suitable, 2.73) to high (unsuitable, 47). Figure 38: “zonalmean” command
Figure 39: “Patchmean1”
Part 4: Selecting the Largest Habitats Step 1: Recoding the Patchmean1 Grid Patchmean1 is now recoded to show the groups of habitats within 5.5 metres of the best habitats. The area is small and precise Figure 40: recoding Patchmean1 because all of the ranking values were kept relatively close together.
Figure 41: “hab_patch6”
Step 2: Assign a Unique Number The regiongroup command can assign each habitat cluster a unique number. The grid is shown at left.
Figure 42: “regiongroup” command
Figure 43: “hab_clust1”
Page 9
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 3: Selecting the Largest Habitat Clusters The two largest habitat clusters are cluster 3 and cluster 17. Using the select command, two new grids are created containing only the values of 3 and 17, respectively. Below, these two clusters are show above the rest of the clusters.
Figure 44a: “select” command
Figure 44b: “select” command
Figure 45: Clusters 3 and 17
Part 5: Creating a Corridor Impedance Grid Corridor Impedance The easiest way to figure out the barriers to a clear corridor. In this equation, the mean is Figure 46: “mean” command found from the grids road_cor_suit, water_suit, and deer_suit. Road_cor_suit was made from a reclass of a buffer of all the paved roads, shown on page 2. This ensures that paved roads are given a low suitability ranking. This grid will be used to figure out the costdistance paths between the two habitat areas. Figure 46: “cor_imp” grid
Page 10
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Part 6: Cost Distance Grids Cost3 and Cost17 Grids Using the cor_imp grid, a costdistance command provides a way to show the path of least resistance for grids cost3 and cost17. The results are shown below
Figure 47a: “costdistance” command
Figure 48: “cost17” grid
Figure 47b: “costdistance” command
Figure 49: “cost3” grid
Cost3 and Cost17 Grids Once the costdistance grids have been created, the corridor command can create the best path between the two clusters. Figure 50: “corridor” command Below is the final corridor grid cor3-17 shown with the best 5% of the contoured results highlighted.
Figure 51: “cor_3-17” grid
Figure 52: “cor_3-17” with contours
Page 11
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Model Builder
Figure 53: Model Builder Diagram
Sources and Software Software: - ArcGIS 9.3 with Spatial Analyst - CorelDRAW X4 - CorelCAPTURE X4 Source Data: - Terrain Data DEM from NSCC AVC LAN - NSGC Contours and Hydrography - Department of Natural Resources: Forcov.shp and Biosys.shp - LANDSAT DATA 7TM for deriving Orchard.shp
Page 12
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Geomatics 2022 Cartography: Digital Mapping February 15, 2009 Joe Fraser
Description of the Exercise Calculating a habitat corridor requires a number of variables. First, two areas of suitable habitat must be determined. Secondly, a path between these two areas must be mapped, connecting the two in a corridor. The suitable habitat grid is made up of four variables in this exercise: proximity to dirt roads, water, and forest edge, and the fourth variable is percentage of slope. Each of these is ranked from 1 to 10, with lower numbers assigned to the cells most favourable to
the habitat of deer. Corridor impedance is a grid made from variables detrimental to the habitat of deer. Paved roads serve vehicle traffic, and deer need to be a certain distance from a source of water the protection of the forest. The overall goal of the exercise is to find the best path according to the variables of suitability and impedance to connect two existing habitats.
Part 1: Building a Deer Suitability Grid Proximity to Resource Road Grid Step 1: Selecting Unpaved Roads Because deer are drawn to the edges of forest clearings, low-traffic roads provide two areas where they are likely to be found. To include unpaved roads in the deer suitability grid, select them from the “q_rd” file. Shown below at left is the resultant unpaved roads grid, at right the unpaved roads grid is compared to the paved roads proximity grid created later on, to make sure that a paved road and an unpaved road do not overlap. In this case two roads were removed because of this comparison.
Figure 2: “unpaved_rd1”
Figure 1: “select” command
Figure 3: “unpaved_rd1” vs. “proximity to paved roads
Step 2: Buffer the Unpaved Roads A temporary grid is needed to show the graduated distance from the paved roads. This grid will be reclassed to be used as the resource road proximity suitability grid.
Figure 4: “eucdistance” command
Page 2
Cartography: Digital Mapping GEOM 2022
Figure 5: “road_suit_tmp”
Joe Fraser
Step 3: Reclass Using a Remap Table Using the remap table, the proximity to the center of the dirt road is given a ranking, with 1 as the most suitable, 10 as the least. In this remap table, the actual road is given a value of 100 to discount that area. The best areas assigned in the remap table are from 1m to 75m, with up to 150m as preferred. Further than that the area is less valuable, and from 250m to the end of the eucdistance at 1205 are set to 10. Below is the grid.
Figure 6: “reclass” command
Figure 7: “unpavedroad_suit.txt” remap table
Figure 8: “road_suit”
Step 4: Proximity to Riparian Areas Suitability The process of determining suitable locations according to proximity to watercourses begins with a eucdistance on a grid with the streams and rivers.
Figure 9 “eucdistance” command
Page3
Cartography: Digital Mapping GEOM 2022
Figure 10 “water_temp”
Joe Fraser
Step 5: Using the Remap Table to Reclass Similar to the unpaved road remap table, water_suit.txt is used to classify the distance away from the stream. A value of 10 is given to anything beyond 250m, assuming that deer need to stay relatively close to a source of water. The stream itself is given 100 as well, and the area within 100m is the most suitable. The grid water_suit will be used as a variable in the deer suitability grid.
Figure 12 “reclass” command
Figure 11 “water_suit.txt” remap table
Figure 13 “water_suit”
Step 6: Field Edge Proximity Suitability Here is an example of a nested command. Using the select function, a eucdistance of 10 is created on only the value of 10 from the previously created field grid. The value of 10 represents the edges of the field.
Figure 14: nested commands “select” and “eucdistance
Page 4
Cartography: Digital Mapping GEOM 2022
Figure 15 “water_suit”
Joe Fraser
Step 7: Buffering the Edges of the Fields Eucdistance provides a buffer from the edge of the fields. Again, this grid will be reclassed to create a ranked suitability grid.
Figure 16 “eucdistance” command
Figure 17 “f_edge_buf”
Step 8: Creating a Field_suit Grid Field_suit.txt is used to reclassify the values in the field_buf grid. Areas further than 250 metres are classified as being undesirable. This way, land close to orchards is weighted higher, assuming that the food Figure 19 “reclass” command using a remap table deer like to eat best grows within a short range of the clearing edge. Below is field_suit, the grid that will be the third variable in the habitat suitability grid. Figure 18 “field_suit.txt” remap table
Figure 10: “streams_ord”
Figure 20: “field_suit”
Page 5
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 9: Slope Suitability Using the dem1 grid, Spatial Analyst can create a slope grid with percentages.
Figure 23: “dem1”
Figure 22: Spatial Analyst Dialogue
Figure 21: Spatial Analyst Menu
Step 10: Reclassing the Slope Percentage Grid Using the assumption that deer are fairly sure-footed animals, the slope_suit grid was created. Grades of 0 to 10% were given a value of 1, and up to 50% slope was characterized as acceptable. Values of 50 to 75% aren’t prohibitive if other values in the deer suitability grid are favourable, but grades greater than 75% were ruled out. Slope is an important factor to consider when terrain is varied, as a chasm or rock outcrop can interrupt a corridor and render calculations incorrect if forgotten.
Figure 24: “Slope_Percent”
Figure 26: “reclass” command
Figure 25: “slope_suit.txt”
Figure 27: “slope_suit”
Page 6
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 11: Creating the Deer Suitability Grid The deer suitability grid is made up of the four above grids: road_suit, water_suit, field_suit, and slope_suit. At right, the map algebra formula shows the weighting of these variables, with water and field proximity being 3 and 2 times more important, respectively, than road and slope suitability. The resultant grid is shown below.
Figure 28: adding weighted grids
Figure 29: “deer_suit”
Part 2: Creating Hexagon Patches Step 1: Create a Point Grid The first step in creating the hexagon patches is to make the point grid with the queens_clip boundary extent. The distance between the triangulated points is 537.29, to create a final output of 25 hectares. Figure 30a: GeoWizards menu
Figure 30c: GeoWizards menu
Figure 30b: GeoWizards menu
Figure 30d: GeoWizards menu
Figure 31: “point_537”
Page 7
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 2: Build Thiessen Polygons Based on the points grid created in step 1, Thiessen polygons can be made. These are the hexagonal patches that will be filled with an average value using the zonal mean command. Figure 32b: Build Polygons Menu Figure 32a: Build Thiessen Polygons Menu
Figure 32c: Build Polygons Menu
Figure 33: “hex_537”
Step 3: Cleaning the Polygons The polygons need to be cleaned using the ET GeoWizards add-on before they are converted to a grid.
Figure 34a: Clean Polygons Menu
Figure 34c: Clean Polygons Menu
Figure 34b: Clean Polygons Menu
Figure 34d: Clean Polygons Menu
Figure 35: “hex_537c”
Step 4: Converting Polygons Finally, the polygons can be converted to grid using the ET_ID field. This ensures each hexagon has a unique identifier. The reason a Figure 36: Convert Features to Raster standardized hexagon patch is used is because of its shape unlike a square, the hexagon is equidistant each adjacent hexagon, making it ideal for costpath analysis or direction grids. Using a square shape favours the squares above, below and to the side.
Page 8
Cartography: Digital Mapping GEOM 2022
Figure 37: “hex_537c2”
Joe Fraser
Part 3: Prepare a Zonal Mean Grid Create a Zonal Mean Grid Using the deer suitability grid, a zonal mean command fills in the hexagon patches with the average value. What is created is in essence an average of the areas deer are likely to prosper, ranked from low (most suitable, 2.73) to high (unsuitable, 47). Figure 38: “zonalmean” command
Figure 39: “Patchmean1”
Part 4: Selecting the Largest Habitats Step 1: Recoding the Patchmean1 Grid Patchmean1 is now recoded to show the groups of habitats within 5.5 metres of the best habitats. The area is small and precise Figure 40: recoding Patchmean1 because all of the ranking values were kept relatively close together.
Figure 41: “hab_patch6”
Step 2: Assign a Unique Number The regiongroup command can assign each habitat cluster a unique number. The grid is shown at left.
Figure 42: “regiongroup” command
Figure 43: “hab_clust1”
Page 9
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Step 3: Selecting the Largest Habitat Clusters The two largest habitat clusters are cluster 3 and cluster 17. Using the select command, two new grids are created containing only the values of 3 and 17, respectively. Below, these two clusters are show above the rest of the clusters.
Figure 44a: “select” command
Figure 44b: “select” command
Figure 45: Clusters 3 and 17
Part 5: Creating a Corridor Impedance Grid Corridor Impedance The easiest way to figure out the barriers to a clear corridor. In this equation, the mean is Figure 46: “mean” command found from the grids road_cor_suit, water_suit, and deer_suit. Road_cor_suit was made from a reclass of a buffer of all the paved roads, shown on page 2. This ensures that paved roads are given a low suitability ranking. This grid will be used to figure out the costdistance paths between the two habitat areas. Figure 46: “cor_imp” grid
Page 10
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Part 6: Cost Distance Grids Cost3 and Cost17 Grids Using the cor_imp grid, a costdistance command provides a way to show the path of least resistance for grids cost3 and cost17. The results are shown below
Figure 47a: “costdistance” command
Figure 48: “cost17” grid
Figure 47b: “costdistance” command
Figure 49: “cost3” grid
Cost3 and Cost17 Grids Once the costdistance grids have been created, the corridor command can create the best path between the two clusters. Figure 50: “corridor” command Below is the final corridor grid cor3-17 shown with the best 5% of the contoured results highlighted.
Figure 51: “cor_3-17” grid
Figure 52: “cor_3-17” with contours
Page 11
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Model Builder
Figure 53: Model Builder Diagram
Sources and Software Software: - ArcGIS 9.3 with Spatial Analyst - CorelDRAW X4 - CorelCAPTURE X4 Source Data: - Terrain Data DEM from NSCC AVC LAN - NSGC Contours and Hydrography - Department of Natural Resources: Forcov.shp and Biosys.shp - LANDSAT DATA 7TM for deriving Orchard.shp
Page 12
Cartography: Digital Mapping GEOM 2022
Joe Fraser
Related Documents
Habitat Corridor Modeling Process
April 2020 19
Chap 9 - Process Modeling
June 2020 11
Business Process Modeling
October 2019 28
Process Modeling Dfd 2
June 2020 8
Process Analysis Modeling
December 2019 21
Bdsm-ch3 Modeling Process
June 2020 3More Documents from "Pard Teekasap"
Mochelle, Nova Scotia
December 2019 14
Habitat Corridor Modeling Process
April 2020 19
Workers Employed In Natural And Applied Science, Nova Scotia (2006)
December 2019 18
Annapolis River, Nova Scotia Watershed
December 2019 10
Nova Scotia Population By Gender, Age And County (2006)
December 2019 14