Q&A: Spatial Analysis in ArcGIS Online Map Viewer
On May 9, 2023, the ArcGIS Online team hosted a webinar titled, Spatial Analysis in ArcGIS Online Map Viewer. In this post, you will find the recording, resources, and answers to the questions asked.
We recommend that you watch the recording first before diving into the Q&A.
Analysis Tools and Capabilities
Q: What is required to use Map Viewer analysis?
To perform feature analysis in Map Viewer, a foundational user type and credits are required. To perform raster analysis, in addition to the foundational user type and credits, you also need to have the ArcGIS Image for ArcGIS Online user type extension.
Q: What is required to view the analysis results?
When you perform analysis in Map Viewer, the results are saved as hosted feature layers in your ArcGIS Online organization. By default, the analysis results are not shared. You may choose to share the results with either selected groups or individuals within your organization or share them with the public which means everyone with the link can view the content.
Q: Can you provide some additional resources on Map Viewer analysis?
You can find detailed help topics on Map Viewer analysis from ArcGIS Online help. Here are also some recent blog articles, and ArcGIS learn lessons that you can follow at your own pace.
• Perform Feature Analysis Online Documentation
• Perform Raster Analysis Online Documentation
• Introducing Analysis in Map Viewer
• Chart Viewer: A New Instant App
• Analysis Case Studies (samples + demos that you can try out)
Q: Are there different tools in Map Viewer Classic and Map Viewer?
Yes. The initial release of analysis in Map Viewer includes 22 feature analysis tools and 24 raster analysis tools. Some of the tool names have been updated in Map Viewer to better reflect their capabilities. For example, the Create Drive-Time Areas tool in Map Viewer Classic has been renamed to Generate Travel Areas in Map Viewer. There are also new raster analysis tools. For now, seven tools are available only in Map Viewer Classic. The outstanding analysis tools and functionality will migrate to Map Viewer in upcoming ArcGIS Online updates.
In addition to migrating the existing tools from Map Viewer Classic, analysis in Map Viewer adds additional features. For example, the ability to search tools by name or keyword, the ability to set the coordinate system of the analysis results, and additional options for setting the processing extent of analysis. Additionally, analysis in Map Viewer now has Analysis history, which stores information about the tools that have been run, their results, the parameters used, and any errors or warnings. From History, you can reopen and re-run any of the tools with the same or modified parameters. Analysis history is saved with the web map and can be shared with others, allowing you to document the workflows used to create the data in your web map project.
Q: Is Analysis History available in both Map Viewer and Map Viewer Classic?
Analysis history is a new feature that is only available in Map Viewer.
Q: Can workflows saved in Analysis History be deleted individually?
Yes! You can remove any entries in Analysis History that you do not want to keep in your web map. Note that this only removes the tool run from analysis history. It does not delete the resulting layers.
Q: Can I create a web app to include just one or a subset of Map Viewer tools?
Yes. You can develop your own web applications using Esri’s REST API, ArcGIS API for Python, and ArcGIS Notebooks.
Q: How can I control who can use spatial analysis tools in my organization and limit the number of credits used?
Administrators can create custom roles for members, including privileges to access tools they need to use and to restrict access to tools they do not need. For example, a member may be assigned with a privilege to use Spatial Analysis tools but not to use the Enrich Layer tool. In addition, administrators can use credit budgeting to set a limit of credits that a member can use. As a good practice, we encourage you to always check the required credits prior to running a tool by clicking the Estimate credits button.
Q: Can I schedule a tool to run automatically in ArcGIS Online?
You can access Map Viewer analysis tools and other spatial data science libraries via ArcGIS Notebooks. An ArcGIS Notebook can be scheduled to run automatically, with no user interaction, at a set time or on a recurring basis.
Q: Can I run custom notebook tools in Map Viewer?
Not at this point but it is on our roadmap.
Q: Can I save Map Viewer analysis workflows in a model?
Not at this point but it is on our roadmap.
Q: Can I create a hosted feature layer view with the Join Features tool in Map Viewer?
The joined view layer option is currently planned to be added to the create view layer in the next ArcGIS Online update (June 2023).
Q: Will the sketch layer be an option for an input layer for analysis tools?
Yes! Map Viewer sketch layers will be a supported input type for analysis tools in the upcoming ArcGIS Online update.
Q: When will Map Viewer analysis be available in ArcGIS Enterprise?
It is planned for ArcGIS Enterprise 11.2.
Charts in Map Viewer
Q: What is required to use charts in Map Viewer?
To create and share charts in Map Viewer, a foundational user type is required. To view charts in Map Viewer, only a viewer user type is needed.
Q: How do I configure charts in Map Viewer?
You can configure chart titles, axes, and legend under the General tab 
Q: Can I share Map Viewer charts with others?
Yes! Charts authored in Map Viewer can be saved and shared with either the feature layer or the web map, and discovered through the chart list. Layers published from ArcGIS Pro with any of the five supported chart types will be available in Map Viewer with the published layer.
Raster Analysis and Deep Learning
Q: What extension do you need to use Map Viewer raster analysis including the deep learning tools?
To run raster analysis tools in Map Viewer, you need to have ArcGIS Image for ArcGIS Online user type extension.
Q: What types of pretrained deep learning models are available in ArcGIS Living Atlas?
In addition to the car detection model that was shown in the demo, the list of ready-to-use pretrained models in ArcGIS Living Atlas of the World is growing rapidly. Those models can be applied to a wide range of industry applications, and used to perform tasks, such as feature extraction, classification, redaction, detection, and tracking, to derive meaningful insights from large amounts of imagery.
Q: How do you create the hosted imagery layers to use in Map Viewer raster analysis tools?
Hosted imagery layers are image services that are hosted in ArcGIS Online. Both tiled and dynamic imagery layers are supported for analysis. To publish a hosted tiled or dynamic imagery layer in ArcGIS Online, use the Create imagery layers window in ArcGIS Image for ArcGIS Online that guides you through the steps.
Q: Can I analyze drone images using the deep learning tools in Map Viewer?
Yes! Once the images are published as hosted imagery layers in ArcGIS Online, you can use them with the deep learning tools in Map Viewer.
Q: Do you need an enterprise GIS system to run deep learning tools? Can they be used in ArcGIS online?
With ArcGIS Image for ArcGIS Online, you have access to several raster analysis tools in Map Viewer to perform deep learning inferencing, which is the process of using the information learned during the deep learning training for image classification and object detection.
ArcGIS has multiple deployments for different users. You can run deep learning using tools in ArcGIS Pro, you can scale out your processing using ArcGIS Enterprise, or you can run deep learning in a SAAS based environment through ArcGIS Online where no infrastructure setup is required.
Q: Can I use the pretrained deep learning models in ArcGIS Pro?
Yes! The pretrained models are designed to support and enhance your workflows regardless of your ArcGIS deployment. You can run these models in ArcGIS Online, ArcGIS Pro (Image Analyst extension for all forms of imagery, and 3D analysis for point cloud classification), or ArcGIS Enterprise.
Q: Is there a minimum quality requirement for the raster images used in the pretrained deep learning models?
Deep learning models are typically trained at a particular image resolution. For example, the Building Footprint Extraction – USA deep learning package is designed to work with high-resolution images (10–40 cm). Other pretrained models in ArcGIS Living Atlas may have different recommended resolutions – check the package’s item details page for more information.
Q: Can I use custom deep learning models in Map Viewer?
Yes, you can train a deep learning model and create the ArcGIS Pro deep learning model package (.dlpk) using ArcGIS Pro Image Analyst extension, ArcGIS Notebooks/ArcGIS API for Python, and ArcGIS Imager Server Deep Learning Studio. The model package can then be uploaded to ArcGIS Online as a DLPK item and used as the input to deep learning raster analysis tools.
Q: Does it cost credits to use the pretrained models in Map Viewer?
Yes, credits are consumed with ArcGIS Image for ArcGIS Online raster analysis including using deep learning tools. You can use the credit estimator to calculate the expected number of credits the process will use prior to running the tool.