Activity #6: Microclimate Geodatabase Construction

Introduction

Later in the semester, we will be collecting data from around UWEC in order to make a microclimate map of campus. To do this, ten groups of two will gather GPS data from different locations on campus. The activity we are engaged in today, however, involves getting ready for and planning organizational methods for that data collection. 

When collecting data in the field, it is always wise to decide in advance what type of data we will need for the project. Skipping this step can lead to more time needed out in the field because of poor planning, mistakes on data entry, and loss of data.

In the case of using ArcGIS, this planning stage involves creating a geodatabase and setting rules on what types of data are allowed to be used, as well as the ranges that that information can fall in. These rules are referred to as domains, and are very important in maintaining data integrity (especially when there is a great deal of data being collected). These domains are then applied to feature class fields and exported to the GPS devices that will be used in data collection. For example, if technicians were going out into the field to collect wind speeds at certain points, the feature class that the data is applied to could contain a domain that restricts the information entered to numbers between 0 and 60 (if using mph). Without the domain, a technician might accidentally enter 90 instead of 9. This would be a huge error in data, but caused by a simple human mistake. Domains ensure that errors are kept to a minimum, makes data collection in the field quicker, and forces GIS personnel to think about data entry before going into the field.

 For our purposes, our class brainstormed what kinds of data would be useful for analyzing microclimates. Air temperature, wind speed and direction, relative humidity, and dew point were all early ideas. We also included Group (our groups are numbered 1-10) and Notes (general notes to leave for points taken, such as 'standing by a heating vent' to explain data that doesn't seem to fit) to fields we would be using. During this discussion, several ideas cropped up that would have been skipped had we just went out into the field to collect general data. Snow Depth and Time of Day were added to our list. Time is important, as some groups might collect data in early morning, when the weather is coldest, and others might collect in the afternoon, when it is warmest. Snow depth may reveal interesting patterns that have gone unnoticed.

Tutorial

Figure 1 - Creating a new File Geodatabase

In ArcCatalog (or in catalog within ArcMap), right click on the folder you want your geodatabase to be, select New, then click File Geodatabase (Figure 1). 

Figure 2 - Domain tab with no domains entered yet

The new geodatabase will appear in that folder simple titled New File Geodatabase.gdb. To rename it, right click the geodatabase and choose rename.

Now we will set up the domains for the geodatabase. Right click the geodatabase again and select Properties. The properties dialog box contains two tabs: General and Domains. Clicking on the Domains tab will bring up a window that looks like figure 2.

Figure 3 - Domain name and description

Now we can enter new domains. Under Domain Name, enter a name that avoids using spaces. Something short will be fine, as you can type out an explanation under the Description field. For example, a domain for relative humidity might have RH under the domain name and Relative humidity % for the description (Figure 3).

Figure 4 - Field Type and Do

 Next you'll have to decide the Field Type and Domain Type. This depends on what the domain will be used for. For the the relative humidity example, we would set the field type to Short Integer. This is used when the data being input must be a non-decimal number. Long Integer is used for integers that are very large (or very negative), and won't be used during this activity. Since relative humidity is represented as a percentage, we set the domain type a Range between 0 and 100 (Figure 4). When collecting data in the field, a number outside of that range will not be able to be entered, increasing data integrity.

Figure 5 - Text field type and Coded Values domain type
for coding cardinal direction

Another field type is Text, and is used for the WindDir_C domain. This is the domain for recording wind direction using cardinal directions instead of Azimuth. Obviously, integers will not help here. If we set the field type to Text and the domain type to Coded Values, however, we can make a code for each direction (Figure 5). This way, someone in the field won't have to type 'Northeast' if the wind is from the northeast, they simply have to type 'NE.' Coded vales will also be used for the Group domain. Instead of Group 1 or Group 2, we will used coded values text so that we only have to enter 1 or 2. A text domain does not have to be coded values, as in the case with the Notes section. This will use a text field, but no coded values will be entered.

Figure 6 - Temp domain using Float field type and
Range domain type.

The Temp domain will use the Float field type. This field type should be used for decimal data. Since the Temp domain will be used for both air temperature dew point data, we need to set the range accordingly. I set the range between -20 and 100 (F). I can only assume that data will not reach either of those limits, but it's good to be prepared.













Figure 7 shows all of the domains I created along with the field type, domain type, and range (if applicable) associated with them. Figure 8 is a table with the coded values for WindDir_C (wind direction using cardinal directions) and figure 9 is the coded values for Group. Figure 10 shows the dialog window with all of the domains created.

Figure 7 - Table showing my domains with field types, domain types, and range.

Figure 9 - Coded values for Group

Figure 8 - Coded values for
WindDir_C

Figure 10 - End product of domain creation

Figure 11 - New feature class window

Now it is time to create a feature class that will utilize these domains. In the catalog, right click your geodatabase and select New ---> Feature class. This opens the dialog box allowing you to name your feature class and select what kind of features you want (line, point, polygon, etc.). For this exercise, I named my feature class microclimate and went with point features (Figure 11).

Figure 12 - Choosing coordinate system for new
feature class

Next you will have to decide the coordinate system that will be applied to the new feature class. I chose NAD_1983_2011_UTM_Zone_15N since I will be dealing with data in Eau Claire (Figure 12). The next two screens (XY Tolerance and Configuration Keyword) I always leave at default settings.

Figure 13 - New feature class field list

The screen shown in figure 13 is where we enter the fields that will be in our new feature class. For me, it ended up very similar to my list of domains. The only real difference was that I used the Temp domain for both dew point and air temperature.

In this table, enter the different fields into the Field Name column. This includes temperature, dew point, snow depth, etc. The Data Type column should match the Field type column from when we were setting up the domain for that field (float, short integer, text, etc.). When you have the data type set, you can then choose the domain at the bottom using the pull down menu. All domains within the geodatabase that where set with that same data type will populate the menu (Figure 14). Figure 15 shows a completed field list.

Figure 14 - Setting data type and domain for each field

Figure 15 - Finished field list for new feature class

Once we click finish, the feature class is added to the geodatabase with no data. We can open up the attribute table for the shapefile by right clicking it and choosing Open Attribute Table. This shows all the fields we created (Figure 16). We are now ready to export the feature class to a GPS device for data collection in the field.

Figure 16 - Attribute table for new feature class

Conclusion 

I have learned how to create geodatabases and set up domains in other classes. What was helpful about this lab, though, was getting in the habit of planning ahead. Thinking thoroughly about the project is crucial before jumping into the data collection. It is always a good idea to know exactly what data you will be collecting so that you can set up domains to protect against error. You will still be able to add new things when in the field, but cutting down on the amount of information you add on the spur of the moment will greatly reduce the chances of error.