Module 1.2 - Data Quality - Standards

     This week's module involved determining the positional quality of two Albuquerque road networks against the city's aerial imagery. This module follows the procedures outlined by the National Standard for Spatial Data Accuracy (NSSDA). 

 

   The blue road network was sourced from the City of Albuquerque Planning Department, and the red road network was sourced from StreetMap USA. The red points are randomly chosen StreetMap USA intersections. Corresponding intersections in the city's planning dept. dataset and aerial imagery reference were mapped as well.

   In order to determine the horizontal accuracy, I first divided the study area into four quadrants using the Subdivide Polygon tool. This was necessary in order to obtain an ideal distribution of 20 test points as outlined in the 1999 Positional Accuracy Handbook by the Minnesota Planning, Land Management Information Center. After digitizing these points, I used the AddXY Tool to add the X and Y values to the three attribute tables. These tables were then exported using the Table to Excel conversion tool and entered into the Horizontal Positional Accuracy Statistic Worksheet.

     Using the National Standard for Spatial Data Accuracy, the City of Albuquerque Planning Dept. dataset tested 14.2 feet horizontal accuracy at 95% confidence level. Using the National Standard for Spatial Data Accuracy, the StreetMap USA dataset tested 163.8 feet horizontal accuracy at 95% confidence level. 

Module 1.1 - Fundamentals

     The first lab of this course focused on determining horizontal and vertical accuracy and precision of GPS-collected waypoints. Accuracy was defined as the subjective closeness of a database representation of an object to the true value, and precision as the consistency of the measurement method (Bolstad, 2022).

   Horizontal accuracy and precision were determined to be 0.8 m and 4.3 m respectively. Vertical accuracy and precision were determined to be 6.04 m and 0.14 m respectively. In this module accuracy was calculated by using the distance between the average waypoint and the reference waypoint. The precision was determined by analyzing the distribution of 68% of the waypoints around the average waypoint using buffers horizontally, and the attribute table vertically.

Module 6 Part 2: Least-Cost Analysis

 

    This post focuses on the Least-Cost Analysis part of the module. We were given 3 criteria to work with for this study area: Distance to road, elevation and landcover. To obtain the Road Suitability, I used the Euclidean Distance Tool and the Reclassify Tool. To obtain the elevation and landcover suitability rasters, I used the Reclassify Tool. Then I used the Weighted Overlay Tool on the three reclassified rasters. Afterwards, I created a cost surface by inverting the combined raster layer using the Reclassify Tool. Then, I used the Cost Distance Tool twice, once for each protected area as a source. Finally, I used the Corridor Tool on the two coronado source cost distance rasters to obtain the corridor raster.

Module 6 Part 1: Suitability Analysis

     The last module of this course (Suitability and Least-Cost Analysis) will be split into two different blog posts, with this post placing emphasis on Suitability Analysis. This map layout compares two suitability scenarios, with each weighing 5 criteria differently. The 5 suitability criteria were Land Cover, Soils, Slope, River Distance, and Road Distance.

    The first 3 criteria feature layers were obtained using the Reclassify Tool, and the distance criteria were obtained using the Euclidean Distance Tool. The final rasters were obtained using the Weighted Overlay Tool. The Alternate scenario is greater in the 2, 4 and 5 suitability category areas, whereas the Equal Weight is greater in the 3 category. This is most likely due to the differing weights, as the Alternate Scenario places greater emphasis on the Slope criteria, and less emphasis on the Distance criteria.

Module 5 - Damage Assessment

     This week's lab had us perform damage assessment using aerial imagery of a Hurricane Sandy impacted study area. I had to use ArcGIS Pro's Mosaic Dataset feature in order to start comparing the pre- and post-storm imagery. The Swipe feature in the Appearance Tab was very useful in performing the assessment.

    In order to determine the extent of the structural damage, I first had to create a coastline polyline feature layer using the pre-storm imagery as a guide. Then, I used the Select By Location tool to select houses within the required distances from the coastline, then I added a 1 or 0 to their respective distance fields I previously added to the Structure Damage table. In order to obtain the counts, I used the Select By Attributes tool. Structures within 100 m would be 1 / 1 / 1 / x, 200 m would be 0 / 1 / 1 / x and 300 m would be 0 / 0 / 1 / x for the Expression selections of "100m / 200m / 300m / Structure Damage" within the Attribute Table, respectively.

     The structure damage counts found suggest that buildings closer to the coast have a higher probability of major damage or complete destruction (42% destroyed @ 100 m vs 13% destroyed @ 200 m vs 7% minor damage @ 300 m). This damage appears to have been primarily caused by storm surge.

Module 4 - Hazards: Coastal Flooding

     This week's module focused on the necessary procedures for coastal flooding assessment. Using DEMs, we assessed destruction, storm surge/flood risk and flood zones.


    I created this map starting from USGS and LiDAR DEM rasters by using various geoprocessing tools, including the Reclassify tool and the Region Group tool. This map omits low-lying areas that were not connected to the larger USGS/LiDAR DEMs. Exclusion of these low-lying disconnected areas could lead to an underestimation of storm surge risk. These areas could still flood if they are connected through drainage systems. Hydrological modeling could help identify flow paths of watersheds to help with increasing accuracy in the assessment of flood risk.

Module 3 - Visibility Analysis

     This week's module involved completing select courses from the "Get Started with Visibility Analysis" Esri Learning Plan. I learned the difference between the different elevation types (on ground, relative to ground, and absolute height) from the "Introduction to 3D Visualization" course. The course also offered examples of extrusion types for 3D symbology, such as polygon features being extruded vertically to represent buildings.

    The "Performing Line of Sight Analysis" course listed the inputs required for performing a line of sight analysis, and described the workflow for doing so. The image below is one of the example images used in the course, where the green lines represent areas that are visible, and the red lines represent areas not visible. The course also described how to use the Construct Sight Lines and Line of Sight tools.

 

    The "Performing Viewshed Analysis in ArcGIS Pro" course describes how to create an output raster that represents the visible area from an observer point, similar to the previous course. The main difference between the two analyses is that the Line of Sight analysis determines obstructed and unobstructed view along a line, and the Viewshed analysis symbolizes an area as visible or not visible. This course instructed me on how to use the Viewshed tool.

    The final course, "Sharing 3D Content Using Scene Layer Packages" mainly taught me how to create and use Scene Layer Packages.