Lab 2 Q’s
Q1: Waypoints must be recorded in the correct datum so they appear in the correct places when uploaded to qgis and mapped with the original campus data we were given. As the lab continues, having waypoints in the same datum as the rest of the layers will also be important for communicating between attributes, using ‘difference,’ or ‘joins.’
Q2: The campus data layers are identified in QGIS as “User Defined Coordinate Systems” under a “Generated CRS.” Generated CRS description: +proj=lcc +lat_1=42.33333333333334 +lat_2=44 +lat_0=41.66666666666666 +lon_0=-120.5 +x_0=1500000 +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=ft +no_defs There are many other Coordinate Reference Systems (CRS) to choose from to put the campus data layers into. By opening this data in ArcMap, the CRS is more comprehensibly presented as an NAD 1983 Harn State Plane Oregon South, in a Lambert Conformal Conic projection. The GPX coordinate system I added is in EPSG:4326, WGS 84). I made sure my GPS unit was set to NAD83, so this was slightly surprising to me.
Q3: I decided only to buffer the callboxes by 10 feet because they are sometimes difficult to see amongst foliage if you are any further away, and if a person was attacked I don’t think their response time would allow them to travel much more than 10 feet, depending on the distance of the assailant of course. To identify areas of sidewalk outside both call box and streetlight buffers I would merge the callbox and light buffers, then take the difference with the walklines. After mapping this, I realized (logically!) that my NoLights layer is identical to my NoLights_NoCallbox layer because I chose a small radius for my callboxes and there is a lamppost next to every callbox. So, the callbox and it’s reachable radius fall within the light radius of the lampposts.
Q4: The perimeter of buildings and main walkways are adequately lit, except for around the Cemetery and to the Millrace Studios. The lamp posts average only 31 feet apart in these areas. The Near Neighbor (NN) index of 0.3701, however, indicates lampposts are clustered. Visually this is obvious when looking at walkways by the Millrace Studios along the highway, behind the Music Building, and between University Street and the Cemetery. I would recommend the University install more lights along the interface of University Street and the Cemetery, around the soccer field behind Hayward Field, behind the Cemetery to the music building, and along the walkways to the Millrace Studios. Callboxes also appear slight clustered, primarily around the Lorry I. Lokey Education and HEDCO Buildings, and in the Millrace Studios. The average observed distance between callboxes is 246 feet. The expected distance is only 263 feet (close to the observed). Yet, some areas such as the walkways between the main Millrace Studios and the Innovation Center, the area between Lillis to Hendricks Hall appear to have relatively less callbox access, and the entire cemetery. For better callbox coverage of campus, the University should consider installing callboxes in the cemetery, between 1600 Millrace Drive and the Innovation Center, and in the quad by Chapman. If the University installed lights and callboxes in the above-mentioned areas, particularly around the cemetery and distant walkways at Millrace, they could greatly strengthen their safety measures on campus.
Q5: One of the issues with the available crime location data is that it does not always match the locations provided in the buildings layer. Some of this was easy to fix by changing the Clery report names to match the buildings layer (e.g. Lokey Ed Bldg to Lorry I. Lokey Education Building). However, some crime locations were too ambiguous to match with the buildings layer. For example, one instance of harassment was ‘Telephonic’ and its location was recorded as simply ‘On Campus.’ Since it did not match any names (and I couldn’t find anything similar to modify it too in the buildings layer) in the building layer attribute table, it did not survive the join, and was not mapped. I have included the two cases of harassment that did not appear on the map for this reason at the bottom of the map. I didn’t want to arbitrarily place them on the map, but I still think it’s important to include they happened. The available crime locations also do not specify exactly where a crime happened. The general area of the building it occurred closest to is helpful, but knowing which side or any other details would be much more informative for placing additional lamplights or callboxes.
I used orange triangles to indicate theft, and red triangles to symbolize harassment. These colors both stand for alarm and are easily visible on my map.
The biggest judgement calls I had to make while entering data was what to include under the ‘theft’ and ‘harassment.’ I chose to stick with those incidents that specifically included the words ‘theft’ and/or ‘harassment.’ However, I felt a number of the other crimes such as ‘Burglary,’ ‘Assault,’ and ‘Fondling,’ would also fall into these categories. Compromises I had to make were putting my triangles on top of the building footprint since I do not know the exact location it occurred around the building, and only mapping specific cases of ‘theft’ and ‘harassment,’ leaving out those mentioned above that seemed similar. Including other crimes would begin to introduce personal bias of what I felt qualified as theft or harassment.
Q6: Knowing areas with highest and lowest foot traffic would provide another layer of completeness in this report to aid in identifying areas where the University can improve on its safety. In areas of lower foot traffic there would be fewer witnesses of a crime and fewer people to prevent a crime from happening, or assist the person being harassed or robbed. In these locations more safety callboxes could be installed or more lights added if it is not sufficiently lit. By combining foot traffic information with distances from light and callbox buffers, a new layer of risk areas could be created that shows areas of seclusion in darkness away from callboxes. Knowing the times the crimes occurred would also be extremely helpful. If crimes are occurring during the day when light is not an issue, proximity to callboxes may be the main problem. If they are happening at night in low lit areas more lights should be added to these places. (If they crimes are happening at night in well-lit areas near callboxes – this could tell UOPD areas they should increase surveillance.) While using buffers is extremely useful to identify areas of concern, it also has short comings. First of all, it is circumstantial whether being inside a light or callbox buffer zone will prevent a crime from occurring. Many students wear earbuds, even at night, and could be attacked from behind right below a lamppost or a few feet from a callbox if they do not hear their assailant coming. These buffers may be better at predicting areas of theft than harassment. One instance of harassment recorded by the Cleary Report was ‘Telephonic harassment’ that could occur anywhere. Another limitation of buffers is the perfect circles it forms around each post or callbox. Many of these lampposts are against a wall or building and actually form a semicircle of light rather than a circle. I tried to represent this by putting the buffer features below the building layer to cut off the light radius where it obvious would not be shining through a solid building. Trees and shadows of trees or .other objects can also interfere with the light radius buffers. These shadowed areas could be areas of concern, but will not be represented on this map because it is excluded by the use of buffers.