By: Jessie Smith
Geovis Project Assignment @RyersonGeo, SA8905, Fall 2018
My Geovisualization Project focused on the use of LiDAR to create a 3D Basemap. LiDAR, which stands for Light Detection and Ranging, is a form of active remote sensing. Pulses of light are sent from a laser towards the ground. The time it takes for the light pulse to be returned is measured, which determines the distance between where the light touched a surface and the laser in which it was sent from. By measuring all light returns, millions of x,y,z points are created and allow for 3D representation of the ground whether it be just surface topography or elements such as vegetation or buildings etc. The LiDAR points can then be used in a dataset to create DEMs or TINs, and imagery is draped over them to create a 3D representation. The DEMs could also be used in ArcPro to create 3D buildings and vegetation, as seen in this project.
ArcGIS solutions are a series of resources made available by Esri. They are resources marketed for industry and government use. I used the Local Government Solutions which has a series of focused maps and applications to help local governments maximize their GIS efficiency to improve their workflows and enhance services to the public. I looked specifically at the Local Government 3D Basemaps solution. This solution included a ArcGIS Pro package with various files, and an add-in to deploy the solution. Once the add-in is deployed a series of tasks are made available that include built in tools and information on how to use them. There is also a sample data set included that can be used to run all tasks as a way to explore the process with appropriate working data.
The tasks that are provided have three different levels: basic, schematic and realistic. Each task only requires 2 data sources, a las(LiDAR) dataset and building footprints. Based on the task chosen, a different degree of detail in the base map will be produced. For my project I used a mix of realistic and schematic tasks. Each task begins with the same steps: classifying the LiDAR by returns, creating a DTM and DSM, and assigning building heights and elevation to the building footprints attribute table. From there the tasks diverge. The schematic task then extracted roof forms to determine the shape of the roofs, such as a gabled type, where in the Basic task the roofs remain flat and uniform. Then the DEMS were used in conjunction with the building footprints and the rooftop types to 3D enable buildings. The realistic scheme created vegetation points data with z values using the DEMs. Next, a map preset was added to assign a 3D realistic tree shape that corresponds with the tree heights.
Basic Scene Example
The newly created 3D basemap, which can be seen and used on ArcGIS Pro, can also be used on AGOL with the newly available Web Scene. The 3D data cannot be added to ArcGIS online directly like 2D data would be. Instead, a package for each scene was created, then was published directly to ArcGIS online. The next step is to open this package on AGOL and create a hosted layer. This was done for both the 3D trees and buildings, and then these hosted layers were added to a Web Scene. In the scene viewer, colours and basemaps can be edited, or additional contextual layers could be added. As an additional step, the scene was then used to create a web mapping application using Story Map template. The Story Map can then be viewed on ArcGIS Online and the data can be rotated and explored.
You can find my story map here:
This type of project would be very doable for many organizations, especially local government. All that is needed is LiDAR data and building footprints. This type of 3D map is often outsourced to planners or consulting companies when a 3D model is needed. Now government GIS employees could create a 3D model themselves. The tasks can either be followed exactly with your own data, or the general work flow could be recreated. The tasks are mostly clear as to the required steps and processes being followed, but there could be more reasoning provided when setting values or parameters specific to the data being used inside the tool. This will make it easier to create a better model with less trial and error.