Geodatabases, Attributes and Domains

Introduction

• This week we recorded points on a GPS and recorded aspects that had to do with microclimate. At these points we recorded things such as temperature, wind direction, dew point, wind speed, relative humidity and of course the date. These details are considered attribute data, which help "describe the geographic characteristic of features" and are sorted into a table (Esri). Attributes tie notes and facts to a point, so it is not just a point floating in space. We also discovered the importance of domains and how they set rules and boundaries to what values are acceptable to enter into the attribute table. This entire system of the geodatabase is the newer object oriented model that has completely transformed the geographic information system. It is completely organized with the attributes and domains tied to each point. Ensuring data integrity and accuracy. 

Study Area

• The area we studied is located in Eau Claire, Wisconsin on the UW-Eau Claire college campus, outside the front door of the Phillips building. We took a few points in the campus mall area with the arcpad collector. On March 23rd at about 5:00 pm the day was relatively warm out, around 40 degrees fahreinheit and raining slightly; it was an overcast sky with a slow breeze ranging from 1-2 miles per hour. 

Methods

• Before setting out and gathering our points we set up a geodatabase and made sure it was well organized. This is easier said than done, there are many specific details that you must make sure are correct before you can continue. Within a geodatabase there is a feature dataset and within that there is a feature class. Domains further organize a feature class by defining the valid values to each attribute. Values are predefined as to ensure that correct data has been entered. Some issues I came across was that "type" field must be the same when you are entering domains for the attributes. For example if you say the type of a certain field is "text" in the domain than that corresponding field in the attributes must be "text" in order for the domain to work. Without proper set up in the geodatabase this will not be transferred over to the Trimble Juno which will then cause issues in the field when collecting data. You may be able to gather points, but attribute qualities and information may be missing which then tells you nothing about what you collected. Re-Collecting data would have to be done. So if you don't want to create more work for yourself, make sure the geodatabase, domains and attributes are set up correctly. 
• Once we had the geodatabase set up we went out to collect a couple of points on the Trimble Juno. We had to load the geodatabase with a basemap onto the arc collector. We then set out to collect a few points in the campus mall to see how the Juno was working. Dr. Hupy explained to us how to collect a point on the map, measure the values on the kestrel device and then record those values in the arc collector attributes. 

Figure 1: Kestrel device that measures many aspects of the weather

Figure 2: Trimble Juno device we used to collect point and record attribute data.

Results/Discussion

• I only recorded two GPS points which limited the results I had. We also did not define the domain for wind speed well enough because we did not state that there could be the option of no wind which also meant no wind direction. For these values I just had to enter something and then go back manually and correct those values. From these mistakes I have learned a lot and how to correct these for next time. If we could combine our own data points with other groups we would have a better understanding of the microclimate on campus. To do this we would upload the data to ArcMap and then use the merge tool. This would bring in all the data from all different groups (as long as they put it in a shared folder) and place them into one feature class with all the same attributes. To merge all the data there would have to be data collecting standards. We would have to know what each value represented (dew point, wind speed, average wind speed, highest wind speed, relative humidity, temperature etc) because there are many values the kesrel device can measure. We would then have to make a standard of where we would hold the device, 2 feet off the ground 4 feet off the ground, it wouldn't make sense to hold it at shoulder height because obviously we're all different heights. 

Conclusion

• Collecting a point from a GPS doesn't tell you anything about that specific point. Was it hard to get to? What day was it? How was the weather? Is there anything that should be noted about that point? Collecting a point on a GPS and then mapping it with domains and attributes allows you to see data collected with that point and you can make sure that it is valid from the domains you entered beforehand. Obviously database setup has to be correct and accurate in order for all of this to happen. Data normalization is important in these steps as well, this is "dividing one numeric attribute value by another to minimize differences in values based on the size of areas or the number of features in each area" (Esri). For example this is very important with population density (diving the population by the area). Data setup and data normalization must occur in order for the data that you collected to have data integrity (meaning the data is trust worthy and was collected with standards.