Systematic Approch to a GIS

Systematic Approach To
A GIS Concluding With
An Accuracy Analysis
2016
DATA COLLECTION OF NATIONAL GEODETIC SURVEY MONUMENTS
USING BAD ELF’S BLUETOOTH GPS TO DETERMINE IF THE DEVICE
PRODUCES CONSISTENT RESULTS AT ONE METER CEP
ELIZABETH RUNKLE
BAD ELF, LLC | https://bad-elf.com/

i | P a g e
Contents
Introduction ----------------------------------------------------------------------------------------------------------------------- 1
Required Apparatus ----------------------------------------------------------------------------------------------------------- 1
Procedure ------------------------------------------------------------------------------------------------------------------------ 1
Constructing the Geodatabase------------------------------------------------------------------------------------------------ 2
Geodatabase-------------------------------------------------------------------------------------------------------------------- 2
Domains-------------------------------------------------------------------------------------------------------------------------- 2
Feature Datasets--------------------------------------------------------------------------------------------------------------- 3
Feature Classes----------------------------------------------------------------------------------------------------------------- 3
National Geodetic Survey Monuments-------------------------------------------------------------------------------- 4
New NGS Feature Class---------------------------------------------------------------------------------------------------- 4
Bad Elf Data Collection Feature Class ---------------------------------------------------------------------------------- 5
ArcToolbox----------------------------------------------------------------------------------------------- 6
Constructing a Map for Publication------------------------------------------------------------------------------------------ 7
Joining Feature Classes------------------------------------------------------------------------------------------------------- 7
Adding Fields from a Table -------------------------------------------------------------------------------------------------- 8
Feature Class Properties ----------------------------------------------------------------------------------------------------- 9
Layer Properties---------------------------------------------------------------------------------------------------------------10
Feature Templates -----------------------------------------------------------------------------------------------------------10
Publishing to ArcGIS Online ------------------------------------------------------------------------------------------------11
Constructing a Map for ESRI Collector -------------------------------------------------------------------------------------11
ESRI Collector---------------------------------------------------------------------------------------------------------------------12
Preparation for Analysis -------------------------------------------------------------------------------------------------------13
Accuracy Analysis Calculation ------------------------------------------------------------------------------------------------14
Cartographic Design ------------------------------------------------------------------------------------------------------------15
Results of Analysis---------------------------------------------------------------------------------------------------------------16
Appendix 1 - Bad Elf Accuracy Data-----------------------------------------------------------------------------------------17
Appendix 2 - Postprocessed Accuracy Data-------------------------------------------------------------------------------18

1 | P a g e
Introduction
This white paper is intended for professionals looking to start a GIS for data
collection using ESRI Collector in conjunction with Bad Elf’s Bluetooth GPS
receiver. This paper will provide a systematic guide to:
 Setting up a Geodatabase
 Starting a GIS
 Publishing to ArcGIS Online
 Using ESRI Collector
 Accuracy Analysis
 Cartographic Representation
 Final Results
Required Apparatus
An ESRI ArcGIS Online account is required, licensing to ArcGIS Desktop and the ESRI
Collector App. Bad Elf’s handheld Bluetooth GPS receiver [https://bad-elf.com/] in
correlation with the Bad Elf App were used for the study.
Procedure
There are three ESRI products necessary for preparation in using Bad Elf’s handheld
Bluetooth GPS receiver for data collection in concurrence with ESRI Collector. First,
using ArcCatalog to create a geodatabase, domains, feature datasets and feature classes.
Second, the feature classes must be prepared for publishing to ArcGIS Online using
ArcMap. Third, once published online the project then must be shared for viewing in
ESRI Collector. Fourth, the ESRI Collector app can be downloaded on a smartphone or
tablet for data collection in the field.
The goal of this project is to determine if Bad Elf’s handheld Bluetooth GPS receiver can
produce consistent results at 1 meter CEP. To follow through with this task, using known
and documented point locations are necessary. The location points used are National
Geodetic Survey monuments researched at http://www.ngs.noaa.gov/.
There are two tasks involved for field collection. First, to survey at least 15 to 20
monuments using ESRI Collector. To supply at least 20 surveys, 40 monuments were
chosen to allow for destroyed or unlocated marks. Second, to provide a minimum of 15
minute raw logs of those monuments using the Bad Elf app.
Finally, two simple delta positional errors have been calculated using horizontal accuracy
between NGS monuments and Bad Elf’s Bluetooth GPS surveys, along with the
computation of the raw logs to those monuments. A gradational point model using size to
represent distance of the delta differences have been published as a final product in

2
ArcGIS Online. The final cartographic map representation can be viewed at
http://bit.ly/ERMCCGIS.
Constructing the Geodatabase
Constructing a geodatabase can be accomplished in ArcMap, though the use of ArcCatalog
through ArcMap has limited functionality. Therefore, the following procedures will be
completed directly in ArcCatalog.
Geodatabase
Setting up a geodatabase provides a central location for GIS data storage. Within a
geodatabase, domains are used to make dropdown menus for adding consistent naming
conventions in a data field. Also datasets are useful for organization and keeping
consistent geographic coordinates and projections between feature classes and shapefiles.
Steps to create a new geodatabase are as followed:
First, connect to the project folder.
Right click on [Folder Connections]
and select [Connect to Folder]. In the
window that follows, browse to the
intended project folder and click
[OK].
Second, right click on the project
folder and select [New Personal
Geodatabase] and give the
geodatabase a name. There is also
the option to connect to an already
existing database.
Domains
Domains are important for assigning attributes and keeping a consistent labeling scheme
within feature class fields. When adding or editing data, if a domain is applied to the
feature class, a dropdown menu will be provided for uniformity. Domains are created
within the properties of the geodatabase. Once the [Database Properties] window is open
click on the [Domains] tab.

3
Add fields that have been predetermined to
contain a dropdown list. In this project, there
are six domains that will be applied to
different fields in the feature classes. Once
[Domain Name] and [Description] are
added, the [Field Type] must be designated.
The [Field Type] is the data type of the
domain and it is essential that it matches the
feature class field type when created. Next
[Coded Values] must be added. Since the
[Field Type] is [Text], alphanumeric codes
can be assigned. The [Code] and
[Description] are required to be entered
simultaneously before moving on to the next
code.
Feature Datasets
Feature Datasets within a geodatabase are very useful for grouping feature classes that
share the same spatial reference. However, while editing, the dataset is locked to multiple
users so only one person can implement edits at a time. Three separate datasets were
constructed in this project; the Bad Elf
data collection, the National Geodetic
Survey data, and an accuracy analysis
dataset.
To add a feature dataset to the
geodatabase right click on the
geodatabase and select [New]. In the
dropdown menu select [Feature
Dataset]. When the [New Feature
Dataset] window appears, enter a name
for the dataset. In the following window
select the coordinate system that will be
specific to the project. In this case the
projected coordinate system
NAD_1983_HARN_UTM_Zone_12N
was chosen. No vertical coordinate
system was used in the following window. Finally select XY, Z and M tolerances or
leave the default values provided.
Feature Classes
In the initial stages of this project, feature classes are added to two of the three feature
datasets. The accuracy analysis dataset will be used in conjunction with the field data to

4
determine whether the handheld Bad Elf Bluetooth GPS receiver produces consistent
results at 1 meter CEP.
National Geodetic Survey Monuments
The NGS monuments dataset can be used to import the National Geodetic Survey
shapefile acquired from http://www.ngs.noaa.gov/ into its own feature class. If
there are any differences in the spatial reference for the shapefile, the data will
automatically be converted to the spatial reference designated for the feature
dataset. Alternatively, ArcToolbox can be used to change the datum manually. To
import the NGS monument data, right click on the feature dataset and select
[Import] then [Feature Class].
Within the [Feature Class to
Feature Class] window click on
the folder icon located next to
[Input Features]. Click on the
[Connect to Folder] icon and
browse to the NGS shapfile for
importation. Verify the [Output
Location] is correct and assign a
name to the feature class under
[Output Feature Class]. A review
of the NGS Monument metadata
shows much of the data is not
necessary for this project.
New NGS Feature Class
Since the monuments are located within eight different cities of the eastern
suburbs of Phoenix Arizona, a simple way to locate the monuments and add a few
attributes within ESRI Collector will be helpful. To do this, a new feature class is
required. This feature class will also contain fields to identify whether the
monument is located, accessible or destroyed and if the condition of each
monument that is located happens to be good or poor. The most important field to
be added to all the feature classes is the permanent identifier field or PID. The
PID will be the primary key and the only way to join the data for analysis. These
fields will be designated domains created in the geodatabase.
To create the new feature class right click on the feature dataset and navigate to
[New] then [Feature Class]. In the [New Feature Class] window name the feature
class and select [Point Features] under [Type of features stored in this feature
class] since the NGS monuments are point data.

5
In the following window within the
[Fields] tab enter the [Field Name] and
[Data Type] for each field within the
feature class. The [Field Properties] box is
where null values are allowed, a domain is
added if necessary, and the length of the
text within the field is designated.
Next the NGS metadata must be loaded to
this new feature class. The coordinates are
assigned to the point data when PID is
loaded. Right click on the new feature
class and navigate to [Load] then [Load
Data]. Within the [Simple Data Loader]
window click the folder icon for [Input
data] and browse to the initial feature class
loaded in the dataset. Add this to the [List of source data to load].
When [Target Field] and [Matching Source Field] are displayed, the common
field in both feature classes is PID (Permanent Identifiers). This will be the
connection between both feature classes so the NGS metadata will be loaded to
the new feature class. The fields
[Locate] and [Condition] should be
[Null] for each of the monument
PID’s. These fields will be edited and
updated through ESRI Collector.
Bad Elf Data Collection Feature Class
The second feature dataset is for the Bad Elf data feature classes. There will be
two feature classes in this dataset but the second feature class won’t be loaded till
after the field collection. The second feature class will contain the RINEX output
derived from the 15 minute raw logs with Bad Elf’s handheld Bluetooth GPS
receiver at each monument where accessible to do so.
The first feature class will contain fields loaded from a utility that must be
inserted into ArcToolbox along with some additional descriptive fields.

6
Initially a new feature class must be added
to the dataset. This feature class will also
contain point features and will designate
fields with domains assigned for PID,
City, Weather, and Interference. Once the
feature class is created the utility that will
be inserted to ArcToolbox will load the
fields required to import data directly from
the GPS receiver. For more information on
loading feature classes, refer to New NGS
Feature Class above.
ArcToolbox
The utility to be loaded into ArcToolbox is designed for all GPS receivers
and is compatible in conjunction with ESRI Collector. To open
ArcToolbox click the icon with the red toolbox on the menu bar.
The toolbox can be loaded by
right clicking in the [ArcToolbox]
window and selecting [Add
Toolbox]. Navigate to the location
of the toolbox called
CollectorUtils.tbx and select to
add the utility.
To complete the Bad Elf data
collection feature class open
ArcToolbox. Navigate to [Add
GNSS Metadata Fields] and
browse to the Bad Elf feature class
within the geodatabase Bad Elf
feature dataset.

7
Once complete verify no
errors occurred and the fields
were added to the feature class as
shown in the image to the right.
Constructing a Map for Publication
To publish feature classes to ArcGIS Online, the layers
must be modified and templates must be created. Also,
the property extents of the main layer must be altered
after adjusting the scale range of the feature classes.
This is accomplished in ArcMap.
After adding the Bad Elf data collection feature class
and the NGS feature class to ArcMap, it is helpful to add a bookmark of the project location. If
the map goes blank from accidental panning away from the project area, a bookmark can easily
zoom back to the geographic location. To add bookmarks simply go to the [Bookmarks] shortcut
menu at the top of ArcMap.
Joining Feature Classes
Since the NGS feature class has no data other than [PID] and [NULL] values in the other
two fields, joining this feature class to the original National Geodetic Survey feature class
will add all the data to the attribute table. Then create a new feature class with all the data
joined and finally remove any fields not needed for the project.
In the NGS collection feature class shortcut menu
under [Joins and Relate] select [Join]. In the [Join
Data] window select [Join attributes from a table].
In step one under [Choose the field in this layer
that the join will be based on], choose [PID] since
this field is the common identifier between both
feature classes. Select the National Geodetic
Survey feature class to join with in step two and
in step three select the common identifier of the
feature class to be joined. Again, in this case the
field of commonality is [PID]. In the [Join
Options] box check [Keep all records] then
validate before completing the join. Index the join
field if requested.
Once the feature classes are joined, in the short
cut menu of the NGS collection feature class
under [Data] select [Export Data]. Keep all
default properties except the [Output feature

8
class]. Select the browse icon and in the popup window [Saving Data], change the [Save
as type] option to [File and Personal Geodatabase feature classes]. Navigate to the project
geodatabase and save in the appropriate feature data set. Add the layer to the map when
prompted. Remove the National Geodetic Survey and NGS collection feature classes
from the map which are no longer required.
Open the attribute table in the new feature class’ shortcut menu and delete any
unnecessary fields by selecting [Delete Field] in the field name shortcut menu.
Adding Fields from a Table
The accuracy analysis will be based on horizontal accuracy. The horizontal accuracy is
not an attribute in the National Geodetic Survey feature class but can be found in the
NGS datasheets from http://www.ngs.noaa.gov/. By researching through each Permanent
Identifier’s datasheet an FGDC 95% confidence horizontal accuracy is obtained. Tabulate
each horizontal accuracy with the PID in excel and
save as a csv file. Add the csv file to ArcMap.
Open the attribute table in the feature class short cut
menu. Within the table options short cut menu select
[Add Field]. In the [Add Field] window, enter the
name for the field, the data type which is [Double]
and allow null values.
This field will be used to calculate the horizontal
accuracy from the csv file.
Join the NGS feature class to the csv table using PID to link the table to the feature class.
Open the attribute table and in the dropdown menu of the horizontal accuracy field, open
the [Field Calculator]. In the [Fields] box of the [Field Calculator] window, scroll down
to find the horizontal accuracy field from the table and double click to add the field to the

9
calculation box. Once applied the column will be populated with the data from the table.
Remove all joins when complete.
Feature Class Properties
The properties can be found in the
menu of each feature class. Once
the [Layer Properties] window is
open select the [General] tab. The
[Out beyond] scale range must be
added to prevent errors when
publishing. To enter the [Out
beyond] scale range verify the
option [Don’t show layer when
zoomed] is checked.
Next, select the [Symbology] tab in
the [Layer Properties] window.
Within the [Show] box, under the [Categories] drop down menu highlight [Unique
values]. In the neighboring box choose a [Value Field]. For this project the value field
chosen for the NGS feature class is [Locate] and the Bad Elf feature class is
[Interference]. The templates for
both feature classes will be
assigned from these fields. Once
the [Value Field] is assigned click
on the [Add All Values] box to add
all the values assigned to the field
from the domain.
Furthermore, within symbology is
where the cartographic design of
the symbols are allocated. To do
this, double click on the symbol to
edit.

10
Layer Properties
To prevent an error during publishing, the layer extent needs
to be modified. Open the main [Layers] property window and
select the [Data Frame] tab. Within the [Extent Used By Full
Extent Command] box, check [Other] and open [Specify
Extent]. In the [Full Extent] window check [Current Visible
Extent] to set the Extent.
When selecting the [Current Visible Extent] option, it is
important to verify the whole project area is visible in the
[Data View] window. Otherwise a portion of the project will
be discarded when published online.
Feature Templates
Feature templates are critical for generating
features and are necessary for publication
online. All feature classes must have an
associated template. To add a template to a
feature class, within the feature class menu
under [Edit Features], open [Organize
Feature Templates]. In the [Organize
Feature Templates] window, select [New
Template] then click [Select Visible
Layers]. Verify the templates are assigned
properly before closing.

11
Publishing to ArcGIS Online
To publish the map to ArcGIS Online, signing into an online account is essential. In the
main menu bar under [File], select [Sign In]. Once signed into an online account the
project must be shared as a service. This is also found in the [File] dropdown menu,
under [Share As] and [Service]. In the [Share as Service] window, check [Publish a
service] and provide a [Service name]. Once the [Service Editor] window appears,
display the [Feature Access] options and check the [Create], [Delete], [Query], [Sync]
and [Update] boxes. Next
display [Item
Description]. The
[Summary] and [Tags]
are required to be
completed for
publication. Finally
display the [Sharing]
options to share with the
public or an organization.
Once completed, it is best
to [Analyze] and
[Preview] the map in the
[Service Editor] window
to verify there are no
errors prior to
publication.
Constructing a Map for ESRI Collector
Publishing a service to ArcGIS Online yields a [Feature Layer] and [Service Definition] within
[My Content] of an account. Viewing the item details of the feature layer displays an overview
of the published layers and is where the details of the layers can be edited. This is also where the
data can be viewed to verify the layers loaded properly, settings can be adjusted, and layer
property attributes can be changed.
Open the project in [Map Viewer] after reviewing the
feature layer. It may be necessary to zoom in to see the
locations of the National Geodetic Survey points on the
map. Under each layer name there are icons to display the
legend, attribute table, change the style of the legend,
filter, perform analysis, and an options dropdown menu as
shown in the image to the right.
Once the map and features are satisfactory, save the map. Enter the map title, tags, and summary.
After the map is saved it is important to share the map either publicly or with a group account.
As soon as the map is shared it is ready to be viewed in ESRI Collector.

12
ESRI Collector
ESRI Collector was used with an iPhone and Bad Elf’s handheld
Bluetooth GPS. Open the app and sign into an account to select the
project. The two layers published from
ArcMap are now ready to receive data. The
National Geodetic Survey data collection
layer is used to easily locate and drive to
each monument and enter data by using the edit feature. Select a
monument point and a popup window displaying the NGS data
collection layer will appear. On the right
of the window there is an icon of a box
with an arrow pointing up. Select this icon
to open the menu and select [Edit].
Next enter [Location] and [Condition]
by tapping on the fields and selecting
the appropriate options from the
dropdown menus. These dropdown
menus are the same attributes that
were created when adding the
domains from the geodatabase to the
feature classes prior to publication.
Select [Update] when complete.
When the layer is updated, the new feature point will be displayed on
the map and can be viewed by those who have access.
Since ESRI Collector is connected
to the Bad Elf Bluetooth GPS
through the iPhone, data can easily
be added to the Bad Elf collection
layer. After placing the GPS on the
monument, allow the satellite
reading to stabilize for a minute to
acquire a better positional
accuracy. Satellite coverage can be
viewed in the Bad Elf app.

13
The [+] symbol in the menu bar of ESRI Collector opens the attribute
window to start the process of recording the location. A list from all
template attributes become available. Select
the appropriate option for the [Interference]
field of the Bad Elf data collection layer. Next
enter the [PID] from the menu. Refer to the
NGS layer to select the accurate PID label.
Likewise, select the appropriate attributes for
the [City] and [Weather] fields. The remainder
of the fields are
automatically
populated from ESRI
Collector in relation
to the Bad Elf GPS connection. Hence the
necessity for importing the GNSS metadata
fields to the feature class from ArcToolbox
prior to publication.
After striving to survey 40 National Geodetic
Survey monuments; 23 were located, one was
visibly pulled out of the ground and destroyed
near a construction site and the remainder were
either inaccessible or no longer viable.
Preparation for Analysis
Recall that the goal of this GIS is to determine if Bad Elf’s handheld Bluetooth GPS receiver can
produce consistent results at 1 meter CEP. Now that all the data is collected and noted for
possible interference, a simple analysis can be completed in ArcMap. For the analysis, two
feature classes will be added into the accuracy analysis dataset within the geodatabase. The
second feature class will be based on the postprocessed data with the accuracy analysis identical
to the Bad Elf data collection feature class analysis. The analysis is simply to calculate the delta
difference between the NGS and Bad Elf horizontal accuracy and then visually display the
results.
The data from each layer must be exported
as shapefiles so manipulation of the
attribute tables in ArcMap can be
permitted. Export each layer from the
[View item details] page of the feature layer
found within [My Content] of the ESRI
Online account. Select the [Export Data] dropdown menu and
[Export to Shapefile]. In the [View item details] of the
shapefile select [Download]. Save the zipped file in the project
folder and extract the zipped data in preparation for ArcMap.

14
Open ArcMap and add both shapefiles with all the data collected to a new project. In the
dropdown menu of each shapefile select [Data] then [Export Data]. Using the browse icon, save
as a [File and Personal Geodatabase feature classes] and navigate to the appropriate dataset to
save. Add the exported data to the map as a layer for each feature class and remove the
shapefiles.
Next, join the attribute table of the Bad Elf data to the NGS feature class. Under the dropdown
menu of the NGS layer select [Joins and Relates] and [Join]. Again, the [PID] field is the
common identifier of both feature classes to join properly. Index the join field if requested.
Export the joined attribute data as a new feature class in the accuracy analysis dataset the same
way the shapefiles were exported into the geodatabase. Be sure to save as a feature class and add
the layer to the map. Remove the join and the Bad Elf feature class when finished.
Accuracy Analysis Calculation
Open the attribute table to the new accuracy analysis feature class and remove all unnecessary
fields. Other than permanent fields, at a minimum, the fields required are [PID], [HorizAccy_m]
and [ESRIGNSS_H]. Once complete add three new fields to better identify each horizontal
accuracy and for the calculation of the absolute difference.
To add the fields, under [Table Options] select [Add Field]. The first field will be labeled to
identify the NGS horizontal accuracy. Likewise, the second field is will be used for identifying
the Bad Elf horizontal accuracy. The final field is for the absolute difference between the
horizontal accuracies of NGS and Bad Elf surveys. Make sure each new field contains a
[Double] data type.
Next select the new NGS horizontal accuracy field and open the [Field Calculator]. In the
[Fields] box double click on the [HorizAccy_m] field to add to the calculation box. Run the
program when ready. The same process is used for the Bad Elf horizontal accuracy populated
from the [ESRIGNSS_H] field. Once both fields are calculated, delete the originals.

15
To calculate the absolute difference in the
remaining field, open the [Field Calculator] and
within the [Functions] box, double click [Abs()]
to add to the calculation box. Next, within the
[Fields] box input the [NGS_HorizAccy_m]
within the parenthesis of the absolute value
function. Select the [-] icon then add the
[BadElf_HorizAccy_m] field. Run the
procedure when finished.
Cartographic Design
After all fields are calculated, a cartographic
representation of the delta difference field is
pertinent. This is accomplished within the
[Symbology] window of the feature class
properties. An easy way to visualize the delta
difference of the horizontal accuracy is to use a
gradational point model using size to represent
distance of the delta differences.
In the [Symbology] window within the [Show] box under [Quantities] select [Graduated
symbols]. Select the calculated absolute difference horizontal accuracy field from the dropdown
menu for [Value] within the
[Fields] box. Under [Classification],
enter the number of [Classes] for
the quantity of symbols desired. For
this cartographic representation 3
ranges are used to display the data.
The intervals for each symbol can
be changed by selecting the values
under [Range] and [Label]. Instead
of using the [Range] default values,
to be more intelligently pleasing
and help answer the question of
consistent results less than one
meter CEP, intervals were chosen at
0.50, 0.75, and greater than 1.00
meter. Finally Enter the [Symbol
Size from] range and double click
on each symbol to change style and color.
Once the cartographic design is complete, publish the results online. To publish the results, all
feature classes must be designated a template, scale range and an extent to the layers property.
Then share as a service for publishing. Review publishing the data collection feature class for
more details.

16
Lastly, after publishing, add the new accuracy feature layer to the online map for completion.
Results of Analysis
Overall, Bad Elf's handheld Bluetooth GPS receiver was consistent with results less than 1 meter
CEP. Although the analysis resulted in two locations having a horizontal accuracy greater than 1
meter CEP, this can be explained due to very high interference. The delta difference of these two
monuments, AJ3656 and AJ3671, ranged between 1.13 and 1.26 meters respectively.
Furthermore, AJ3656 was located within about two feet of a 6 foot block wall and AJ3671 was
directly under a large tree with several other large trees within a few meter proximity. Do to the
results of the study it would be beneficial to add a field containing an interference scale that
would portray the severity of potential satellite hindrance.

17
Appendix 1 - Bad Elf Accuracy Data

18
Appendix 2 - Postprocessed Accuracy Data

Systematic Approch to a GIS

More Related Content

What's hot (20)

Viewers also liked (8)

Similar to Systematic Approch to a GIS (20)

Systematic Approch to a GIS