Lab 2: Geospatial Web Services

Introduction:

Integral to understanding Web GIS is the role of a web service in distributing data.  This lab was designed to familiarize the students with geospatial web servers by guiding us through the creation and hosting of services with ArcGIS Online and ArcGIS server.  First, ArcGIS online will be used to host feature services and create web-based maps in the cloud.  Then, the ArcGIS server hosted by the UW-EC ClaireDepartment of Geography and Anthropology will be used to publish a tiled map service through the University.  By the completion of these exercizes, I will demonstrate proficiency (or at least competency) in distributing GIS across web platforms.

Methods:
       Part 1: “Publishing Feature Hosted Services through ArcGIS Online”
            Having publisher rights to their website provided by the university for this course, ArcGIS Online is used to publish a large package of feature classes, make a map of them, then publish feature services using a comma delimited value file (.csv) and finally again by using ArcMap.


   Sections 1 and 2: Publishing a feature hosted service using a Shapefile, and creating a web-based Map
            As the data being published in this section of the lab is larger than 5 GB, the first step in this process is to compress the data.  ArcGIS Online will not accept files larger than this and it is best practice to compress large files before transfer in any case.  After compressing four features (representing Wisconsin counties, cities, interstates and highways) in a .zip file, the data was uploaded to ArcGIS Online.  Uploading data to AGO is a simple process: one merely signs into their account (with publishers permissions) and under “My Content” click “Add Item”.  After selection of the compressed files, tags were added to the content for search optimization (fig 1) along with a description and summary of what was uploaded.

figure 1: adding tags to the transportation features.

            After the data was successfully uploaded, it was used to create a web-based map.  The data was added to a new map document on ArcGIS Online, which already had a topographic basemap layer present.  Symbology and Pop-Ups were toyed with here, in order to make a user-friendly and cartographically pleasing layout.  Finally, the product was shared with the class.

  Section 3: Publishing a feature hosted service using a CSV file
            The procedure for this section of the exercise is similar to the previous one, in fact it is identical after the data is uploaded.  The major difference here is that instead of uploading shapefiles, data from a Microsoft Excel file is being uploaded (Wisconsin Fire Occurrence 2004).  In order to accomplish this, first the workbook (.xls) is converted into a comma delinieated file format (.csv); in Excel this is an easy task accomplished by “Saving As” and then where naming a file selecting “.csv” in the dropdown menu where “.xlsx” is selected by default.
            Having completed the task of converting the desired table to .csv, the data is uploaded to the cloud in the same fashion as the earlier data (sign in, “My Content” à “Add Item”).  A description, a summary, and tags were added to the metadata for this content as before.  Once published, the pop-up attributes were edited (fig 2) because they showed superfluous data to the user: feature ID, longitude and latitude were removed from the pop-up data leaving potential end users with only the date of the fire as information accessible by click.

figure 2: the pop-up editor

  Section 4: Publishing a feature hosted service using an ArcMap document
            This task is slightly different from the others in that this data will be distributed to ArcGIS Online directly from ArcMap, fully employing ESRI integration to easily serve local data worldwide.  This process will host all vector data present in a map document.
            Step one is to open the features to be uploaded in ArcMap: our features are Wisconsin Rivers_and_Streams as well as WI Lakes.  The map then was saved locally.  Next ArcMap can be used to sign in to ArcGIS Online: “File” à “Sign In”.  Then, to place the data on the web, navigate “File” à “Share As” à “Service”.  Following the dialogue, my data was hosted as a service from my (re: UWEC Geography and Anthropology) hosted services.  Deletion permissions were disallowed in the “Capabilities” tab, and a description, summary, credit, and access constraints were added to the service in “Item Description”.  Of course, tags were also added here. The map was next shared with the class with the “Sharing” tab.  Lastly, the analyze tab was run to check the data for any potential errors in online publishing.  It may be a good idea to preview the feature service here as well before publishing.

Not all symbology accessible in ArcMap is available online, as many browsers do not support ESRI’s more advanced cartography, therefore it is important to assess the data after uploading it by viewing it using AGO’s mapping services.  Most other errors should be found by running the analyze tool before publishing.  The Wisconsin Water service was at last published after modifing the data for cartography.

            Part 2: “Publishing a Tiled Map Service using ArcGIS Server”
            In this part of the lab, data will be published using the Department’s own ArcGIS Server rather than ArcGIS Online.  Each student has their own personal database on the server, and will be uploading a tiled landuse/landcover raster of the Lower Chippewa Valley region.  Then we shall attempt to access and manipulate the data using arcgis.com

  Section 1: Publishing a tiled map service using ArcGIS Server
            First, the local workspace must be connected to the database which houses the data, which in this case for students was located on a University SQL server.  After practicing with our local distributed GIS by connecting and authenticating ourselves, the .tif landuse/landcover file was loaded into ArcMap.  This file is a classified Landsat image of the area provided by the professor, and some of the classes needed trimming from the symbology of the map in order to clean the legend.  Then the data was shared: “File” à “Share As” à “Service”.  Following the dialogue, this service was shared using ArcGIS server as a GIS service.
            In order to improve performance in serving large datasets online, cached tiling is often used as a way of accelerating mapping services by distributing processor workload from a server into a client’s browser cache.  Uploading data with ArcGIS Server provides for this capability in its “Caching” tab of the service editor.  For our purposes, Tile Format was switched to “mixed” from “PNG” and the default tiling scheme and levels were left.  Finally, the data is analyzed for errors much as before, and then published.
 
  Section 2: Consuming the map service in a web application
            Finally, the served landuse/landcover data was accessed on the University’s server and after navigating to it there it can be viewed directly in AGO, overlain by default on a topographic basemap.

Results:
            These are the final mapped results taken from the lab.


figure 3: Part 1 Section 2

 

This map (fig. 3) is the final representation of the data from parts 1 and 2 of the lab.  It displays Wisconsin Interstates, major Highways, cities and counties over a topographic base.  The counties were made transparent to convey more information to the viewer.



figure 4: Part 1 Section 3

 

The map above (fig 4) displays locations of fires in Wisconsin during 2004.  A pop-up is displayed, notice that no superfluous locational or database object data is displayed; only information that is useful to the audience is included in the pop-up.



figure 6: Part 1 Section 4

 

This map (fig 5) displays wisconsin waterways.  The viewer is centered on the majestic Green Bay.

The final map (fig 6) is landuse-landcover data, viewed here in ArcGID Online but hosted by the University of Wisconsin - Eau Claire Department of Geography and Anthropology.


Sources:

Part 1, Section 1

Environmental Science Research Institute, ArcMap USA geodatabase.

Part 1, Section 3

Wisconsin Department of Natural Resources.

Part 1, Section 4

Lakes- Created by Dr. C. Wilson, 2013.
Rivers and streams- United States Geological Surveys (USGS), 2008.

Part 2, Section 1

Earth Resources Observation and Science Center (EROS). 2011.
Processed by Dr. C. Wilson. 2012.

Figures 1 and 2 were pulled directly from the lab assignment posted by Prof. Wilson.