Introduction
When starting a large GIS project, it is smart to create a
file geodatabase. File geodatabases are
a useful way of storing multiple files and pieces of information in one
place. They are defined by ESRI as
being, “a collection of various types of GIS datasets held in a file system
folder.” They are advantageous in that
they can hold a very large amount of data, from 1 to 256 TB. File geodatabases allow you to work with very
large file sizes without having to compromise on speed, a very important
factor, as having large files open at once can slow a system down
dramatically.
What to consider when
creating a geodatabase for deployment in the field
When setting up a geodatabase, it is important to stop and
consider the purpose of the geodatabase.
Obviously, you are setting this up so that all of your information can
be held and later accessed in one convenient place, but the considerations must
go further than that. You must consider
the data collection component and the needs of the field agent who will be
doing the collection. There are several
questions to consider.
What types of data will be collected? Will they be points,
lines, areas, or a combination? For example, if the field agent is collecting
GPS coordinates of trees, he or she will be collecting point data.
How much information will need to be entered about each
piece of data? For the above example regarding trees, what does the field agent
need to record about the tree? Tree type and relative size or age might be
example of information he or she needs to enter in addition to simply capturing
the GPS coordinates.
Next, to ensure the accuracy of the data collected in the field by
the agent, domains can be set up in the geodatabase. Domains are rules that govern the data being
entered. For example, if your field
agent is collecting air temperature data at various locations, you can set a
domain for the temperature field such that a temperature value cannot be
recorded outside of a set range. That
way, if the agent is trying to record a temperature of 100 degrees and types
1,000 by mistake, he will get an error message and be able to enter the correct
value instead. Domains can be very
valuable at preventing faulty numbers from being entered and ensuring accurate
data collection.
The number and type of feature classes to create is also an
important consideration. While a GIS
user may be used to incorporating multiple feature classes into a single
project, and would have no issue with simply creating a number them, the field
agent collecting the data must again be remembered. It is easy to display and edit multiple
feature classes on a computer with a large screen and the pinpoint control of a
mouse, but the field agent does not have this luxury. His or her data collecting device likely has
a small screen and buttons. This makes
switching feature classes a tedious endeavor, especially if it must be done on
a regular basis. To maximize the time a
field agent has for collecting data it is important to keep the number of
feature classes to a minimum. This will
also make the data collecting process easier on the field agent.
How to set up a file
geodatabase
To demonstrate, here is an explanation of the creation of a
file geodatabase used in this class. The
geodatabase will store information collected in the field by a group of
students including myself. Our project
goal is to create a microclimate map of a section of the UWEC campus. First, we need to set up our file geodatabase
for the project. We open up Arc Catalog
and pick the location where the geodatabase will be created. This location should have enough free space
as is predicted to be needed for the entire project. In Arc Catalog, find the folder where you
will be working and right-click on it.
Mouse over “New” and click on “File Geodatabase.” Your file geodatabase is created.
There are two steps now to get the data collection ready to
go. We need to create our feature class
and we need to define our domains. It
will be easiest if we start with the domains.
We know that we want to record information on the following items:
Temperature, Wind Speed, Wind Direction, Relative Humidity, Dew Point, Snow
Depth, Time, Group Name, and Notes. By
setting up our domains, we can define each of these further and set the limits
on their ranges. We will be defining,
for each domain, its name, description, and field type, along with the range of
acceptable values. For instance, for the
Temperature domain, “Temperature” will be typed in for Name and “Air
Temperature” will be typed in for the Description. The description can be anything that helps explain
the name further. For Field Type, there
are several options: Short, Long, Float, Double, Text, and Date. The first four of these describe numeric
values. Short and Long refer to whole
numbers while Float and Double refer to numbers with decimals. Short can contain a number value from -32,768
to 32,767. Long can contain a number
value from -2,147,483,648 to 2,147,483,647.
Float can contain a number value with a decimal after it, and has a
range far larger even than Long. Double
can also have a number value with a decimal, with a range way larger than Float
has. The range of Double is so large
that it will probably not ever be necessary to use. Essentially, if you are going to be entering
whole numbers, you will probably use Short, unless its range is not large
enough for the data you need to enter.
If you need the exactness of a decimal point, you should use Float. If the data you are entering is not a number,
but is a calendar date, such as the date that you are entering the data, choose
Date as the Field Type. If what you are
recording is text, choose Text.
For example, for my domain “Temperature,” I know that the
temperature variations I will be recording will be very small due to the small
size of my study area, and therefore I need the precision of a decimal. Therefore, for my Field Type, I choose
“Float.” On the other hand, for my
domain “Notes,” I know that I will need to be able to enter text. Therefore, I choose “Text” as my Field
Type. After you have chosen the field
type, you can set the range of values.
For “Temperature,” I know that with it being a Wisconsin winter day that
we will be doing the data collection, I may have temperatures in the
negative. Therefore, I set my minimum
value to -20. Realistically, the
temperature should not exceed 50 degrees, so I can set my maximum to 50. That’s it.
I now do this for all of my domains.
Wind Speed has a description of “Wind Speed in mph” and will be a Short
field with values from 0 to 60. Wind
Direction will be a Text field, and have no set range. Relative Humidity and Dew Point will each by
Short with a range of 0 to 100. Snow
Depth, which is described as “Snow Depth in inches,” will be a Float field with
range from 0 to 36. Notes is a Text
field. Time will be a Short field, and
will contain the time of day in military notation, with 0 as the minimum and
2400 as the maximum value. Group Number
will be a Short field, and at present will have no set range. Later, when I know what my group number is, I
can change the range to be that exact number; if my group number is 2, I can
make the range be from 2 to 2. When all
of these domains have been entered, you can click OK to close the window. These domains can be reopened and edited
later if you need to.
Lastly, I have to create my feature class that will be
loaded into the data collecting device my field agents and I will use to record
our data. The feature class does not
have to contain all of the domains you created, but it certainly can. In our case, it will. However, if our project had been larger, and
we had created more domains than we would be using for this particular data
collecting excursion, we could pick which domains we would need and leave the
rest alone for now. This is part of the
glory of creating domains: they are created for the geodatabase as a whole, not
for the individual feature class, and can therefore be utilized in any feature
class created in the geodatabase. To
create our feature class, we right-click on our geodatabase and choose “New”
then “Feature Class.” We give it a name,
“Microclimate.” Since the features we
will be creating will be points on campus with our various domain-linked
attributes associated with them, we choose “Point” as the type of feature. Click Next.
Choose the coordinate system.
Click Next. Click Next a second
time. Now we will add our fields. This process will be much like what we did
for defining our domains. There are two
fields already present; these are required for the feature class. Below them, click in a box and enter your
field’s name. We will make our first
field “Temperature.” We made Temperature
Float field, so choose Float as the type.
This allows you to select “Temperature” from the drop-down menu where
you can choose the field’s domain, as Temperature was one of the domains you
created that has the Float type. Now the
correct domain is linked to this attribute in the feature class. Repeat this step until you have all of the
attributes you need and the correct domains associated with each attribute. When you are finished, click Finish and your
feature class will be created. It is now
ready to be imported into your team’s GPS and used in the field.