 |
| Fig 1. The Geographic Inquiry Process. |
The Objective of this weeks lab was to investigate a question using the Geographic Inquiry Process. This would be a question that could be answered by geospatial data or phenomenon as well as analyzing that data to flush out a geospatial relationship. After determining the relevant data and information that would be need to answer this question, a geodatabase would be needed in order to capture it. Our task was to create, and deploy a geodatabase with domains to ArcCollector and go out into the field to capture data that could be analyzed in the geographic inquiry process.
The question arose from the distribution of House Sparrow nests on the outside of Phillips Hall which have been built on the ledges in various places, under the roof of the building. Some of the nests have been abandoned, some have been removed (presumable due to disrepair), while some are still inhabited. The nests are spread around both the outer walls of the building as well as inside the courtyard and on the inlets where the doors of the hall are located. This pattern of nest placement and status lead to the question, do House Sparrows choose to nest on South facing slopes, being that south facing slopes receive the most solar radiation and heat during the day?
To begin with, a geodatabse and domains would be needed. As stated in previous blog posts, proper design of geodatabases and implementation of domains is extremely important when capturing data. The use of domains ensures that the person capturing the data has a standard form of data input which will make analysis and mapping of that data easier, as well as more accurate and precise.
While the importance of domains is reiterated a few times, their importance may not be overstated. Standard data format ensures that the data can be constrained to cretin variables, not entered in a incorrect form (numbers rather than words, etc), be missing key components of information, and most importantly, reduces the risk of data being misinterpreted when the data is analyzed.
It may seem like a small detail, but if data is not normalized or entered in a standard fashion then the analysis of that data can will be plagued with issues. Particularly if the the people who collect the data are not the same people who will be analyzing the data. In the same way a game of telephone scrambles basic sentences, lack of domains can cause the same errors when analyzing data....In short, implementing domains when collecting field data to constrain data input and normalize data results is very important!
Study Area:
The Study Area was the exterior of Phillips Hall on the UW- Eau Claire campus. The structure of the building lends itself to be rather unique on the UWEC campus, as the interior of the building contains a rather large courtyard. This courtyard effectively increases the exterior surface area, and creates many more places for house sparrows to construct their nests.
 |
| Figure 1. Phillips Hall the House Sparrow Nests are tucked under the upper edge of the building, between the face of the building and the roof. |
The nests of the House Sparrows are on the top most area of the walls of the building, which is roughly 15 meters high. The nests are formed out of mud and fixed to the surface of the exterior face of the building and the underside of a roof ledge, which protrudes a few feet over the exterior walls.
Methods:
For the Geodatabase, the domains that would be most useful for this project, and the ones that were ultimately used were tilted; Face: which gave the direction of the face of the building (North, East, South, West, North East..ect), Height: which was constrained between 0 and 50 meters (this was overkill for the height of Phillips Hall, Number: the number of nests, and the Status: whether the nests were active, abandoned or if the status of the nest was unknown.
 |
| Fig 2. Picture of the implemented domains. |
After deploying the map to ArcGIS online via this handy guide we were off to collect data!
The method for collecting the data in the field was simple, observe the locations of the House Sparrow nests, and capture the required domain fields which would allow the analysis of the geosspatial question. Walking around the building and through the courtyard, a few errors were noticed in the domains, a text field that had been previously added to the geodatabase had a domain applied to the field, rendering it useless. However it was determined (not for the first time) that adding a domain to a text field is not to be done, it always causes an error when it comes time to enter notes. Secondly, a field had been added that would count the number of individuals in the nests, with out binoculars or some other equipment it would not be possible to ascertain this number with out guessing, so that field was deleted.
Results/Discussion:
After collecting the data and uploading it to ArcGIS online the map was published so that the analysis of the distribution of House Sparrow nests could be interactive. As one can see on the map below, the data for the nests does seem to indicate that the House Sparrows are building nests on the South Faces of Phillips Hall, but include nests that are facing West, and East facing as well. The distribution of the nests is limited to two nest of unknown habitation status on the South Side, while the majority of the nests are located in the courtyard as well as on the West side of the building. Their are also a higher number of abandoned nests in the courtyard. The spatial distribution would suggest that the House Sparrows prefer the inner courtyard to nests that are on the exterior of the building, regardless of weather the nests face south or not.
While some nests in the courtyard do face south, they seem to be limited in sunlight exposure by the shadows of the building. The next highest abundance of nests appears on the West side of the building, and would face the setting sun, but would be delayed in receiving direct sunlight for a large portion of the day.
It may be that the House Sparrows are not distributing their nests based on solar energy absorption, but are seeking a more protective environment such as the interior courtyard, which is buffered from wind and other weather events. More information would need to be captured to support this theory, as well as increasing the sample size to more conclusively determine an outcome.
The method for collecting the data in the field was simple, observe the locations of the House Sparrow nests, and capture the required domain fields which would allow the analysis of the geosspatial question. Walking around the building and through the courtyard, a few errors were noticed in the domains, a text field that had been previously added to the geodatabase had a domain applied to the field, rendering it useless. However it was determined (not for the first time) that adding a domain to a text field is not to be done, it always causes an error when it comes time to enter notes. Secondly, a field had been added that would count the number of individuals in the nests, with out binoculars or some other equipment it would not be possible to ascertain this number with out guessing, so that field was deleted.
Results/Discussion:
After collecting the data and uploading it to ArcGIS online the map was published so that the analysis of the distribution of House Sparrow nests could be interactive. As one can see on the map below, the data for the nests does seem to indicate that the House Sparrows are building nests on the South Faces of Phillips Hall, but include nests that are facing West, and East facing as well. The distribution of the nests is limited to two nest of unknown habitation status on the South Side, while the majority of the nests are located in the courtyard as well as on the West side of the building. Their are also a higher number of abandoned nests in the courtyard. The spatial distribution would suggest that the House Sparrows prefer the inner courtyard to nests that are on the exterior of the building, regardless of weather the nests face south or not.
While some nests in the courtyard do face south, they seem to be limited in sunlight exposure by the shadows of the building. The next highest abundance of nests appears on the West side of the building, and would face the setting sun, but would be delayed in receiving direct sunlight for a large portion of the day.
It may be that the House Sparrows are not distributing their nests based on solar energy absorption, but are seeking a more protective environment such as the interior courtyard, which is buffered from wind and other weather events. More information would need to be captured to support this theory, as well as increasing the sample size to more conclusively determine an outcome.
While data was captured that helped answer the geospatial question of House Sparrow nest distribution, there dose seem to be a few gaps in the data that would help us to draw a specific conclusion. If the opportunity arose to redo this project, adding additional information about the nests location
Specifically something that was realized after the start of collecting the data was that the pillars that make up the outside of Phillips Hall play a crucial roll in nesting placement. The pillars actually create different faces and increase the surface area that House Sparrows can nest on. More importantly, it creates faces that face different aspects and changes the exposure of the House Finch Nests to the sun. Adding another domain to the geodatabase which would account for this would make the capture and thus the analysis of this data more complete and paint a more accurate picture of nest placement across the faces of Phillip's Hall. Additionally, knowing more about House Finch nests and their construction could prove to supplement the information of active and abandoned nests in relation to the geospatial phenomenon of solar radiation and slope aspect when comparing nest locations on Phillips Hall. Another aspect of this maybe that the form of heat the House Finchs are most wanting is in the form of heat radiating from the building itself, not from the sun directly. Again, more information would be needed to answer these questions.
A very important take away from this project was that it is a good idea to test deploy and capture field data in small quantities to know how well your domain design is and if all of the field that need to be captured to begin to analyze the data are present before attempting massive data capture in the field. But again this dilemma highlights proper geodatabase design and the ability to understand domains as an important aspect of project management that can be easily overlooked.
Sources:
http://doc.arcgis.com/en/collector/
No comments:
Post a Comment