Hello, and welcome to the first week discussing homeland security type topics in GIS. We are done with the natural disasters and GIS responses for now. Its time to look at how we can help out our law enforcement agencies. Today we will look at GIS as it applies to mapping crime data in both proximity to police stations as a whole, and percentage of crimes occurring closest to individual police stations in the Washington DC area, for crime data as of 2011. The crimes data was provided by UWF, as available through the DC Metro Police Department (Https://data.dc.gov). The overall objectives for this week being exercised by this lab are as follows:
- Analyze data stored in a Microsoft Excel Database
- Create data using the Display XY tool
- Create an address locator using street data
- Geocode tabular address data to point features
- Prepare data for processing in a geodatabase
- Use Field Calculator to calculate attribute table values
- Perform multiple ring buffers and create spatial joins in the attribute tables
- Create Multiple data frame maps to show various crime distributions
- Use Kernel Density to display crime clusters
- Compile and present results for real world problem solving
These objectives were used to develop the two maps below:
There is a multitude of things being explored in this map simultaneously, most of which are highlighted in the central map. The data for the police stations and crimes was given as excell spreadsheet data which required conversion to ArcMAP points. This were done by adding the X and Y coordinates and importing the information in the excel table. Then I moved forward with the use of multi ring buffers located around the DC police stations to show the areas examined for crimes occurring within 0.5, 1, and 2 miles of police stations. Percentages of all crimes occurring within these distances are displayed in the small chart next to the appropriate legend items. Also, looking at the individual police stations it should become apparent that the police station marker is of varying sizes. This is a result of taking all of the crimes shown and determining which police station they are closest too. With that we then portray it as a percentage of the total crimes. The three largest police stations at 13, 12, and 11 percent show as the largest symbol, while stations only close to 1-2% of total crimes are the smallest. Both this and the buffer layer wouldn't be possible without the use of spatial joining tools. That is tools that allowed me to join the attribute from one dataset like the police stations with another, that of the crimes, to compare which stations were closest to what crimes. With this analysis comes a proposal. Where should one or two new police stations ideally be places. The two green markers represent my pitch for new police stations. The addition of these two stations would decrease the two closest and highest percentage stations as well as not raise the new stations overall percentage above 9%. This is explained in the narrative on the map. Additionally you can see overall population totals by tract area and also see a graph of total crimes by offense.
Next we will look at a distribution of some of these particular crimes.
Again with the multi product theme, three different density analyses (based on the crime points generated for the previous map) are being compared in this page. These are kernel density analysis which involve taking specific point locations and grouping them together by proximity to one another, and then gives a variable input raster. This the varying cells are given values based on proximity to points using a search radius for points. This method of analysis gives you a look at where the crimes were occurring spatially, and can also be used as a predictive tool for leveraging resources. In general, I tried to make things as simple as possible with this, under a less is more. You're left with just map overlays for the majority of what you need to process, with simple one size fits all legends. Population density is represented underneath all of the crime density layers, but for more clarification another inset of just this feature was provided. These are just a couple of the ways we can look at how GIS can affect and help the realm of law enforcement. We will continue to explore this topic in the coming weeks. Thank you.
No comments:
Post a Comment