Final Project: What determines the Library Location in Los Angeles County?

Introduction

Public libraries provide access to any residents for the use of books and other educational materials. It also provides a place for residents for internet access and reading. They are as important as schools because they are invaluable institution the education and literacy levels of a country. Since library access is so important, we will examine the accessibility of library to the schools location in the Los Angeles County. If the public libraries are well-located near schools, children especially in the age of 5-17 are more willing to access to these locations. This will promote literacy and education development of the children. In this final project, we will also examine the factors determining the location of the library with race, age, population density and household income median level analysis. For example, are the public libraries mainly located in high household income median level? A GIS-based analysis is able to determine the relationship between schools and the library, and democratic factors and locations of library.

Method

First, to show the numbers and locations of public libraries within each city in the Los Angeles County, more than 200 public libraries were geocoded and the newly-created shapefile by the ArcGIS address locator of library location was joined to the city boundary shapefile.

Second, to determine the accessibility of library to schools, there are two sets of buffers were created. The first one is a two-mile buffer representing the walking distance from the library. The second is a ten-mile buffer representing the driving distance from the library. Some library locations are so near the coast or the edge of the county that the buffers extended over the boundary; both sets of buffers were clipped to the boundary of the LA county city boundary layer.

Third, using the selection tools, we can find out how many schools are within the walking and driving distance of the library buffer. In the select by location tool, the target layer will be the school and the source layer will be the geocoded library location result. Using target layer features are within 10 miles of the source layer, several school locations will be selected. The percentage of number of schools that are within walking and driving distance can be determined. The number of schools that are left out or not within the buffer can be viewed on the map as well.

Forth, to determine several democratic factors that may affect the distribution of public libraries in the Los Angeles County, we are using cluster analysis on the race and age census tracts data in the Los Angeles County. Cluster Analysis is a spatial statistics technique that using spatial autocorrelation to determine the clustering of a data. The age group 5-17 population was picked because it represents the population of children studying in schools before college level. The Black, White and Asian population was picked because they are the most common types of race in Los Angeles County. The overall clustering will be overlay with the geocoded library locations to determine what factors determine the location of the library.

Besides, two other maps were created to have a better understanding the relationship between the library location and democratic factors,

One map of population density by census tracts was made by joining an excel spreadsheet containing all the census tracts data of the population density to the census tracts shapefile.

One map of household income median by dollars was made by joining an excel spreadsheet containing the income data for each city to the city boundary shapefile.

Results

The results will be divided into two parts, the school/library accessibility and democratic factor/library distribution.

1.     School and Library Accessibility

To determine how well the Los Angeles County library reaches the schools location, a buffer analysis was performed on the library locations.  Both “walking distance” and “driving distance” buffer were created. A “walking distance” buffer is a distance of 2 miles away from the library. This may represent a 20-30 minute walking distance or a 15 minute bike distance. A “driving distance” buffer is a distance of 10 miles away from the library. This may represent approximately 20-30 minute of drive. Figure 1 shows the result of the library accessibility to the surrounding school locations. Using the select by locations technique in ArcGIS, school locations within the buffer can be examined. There are 2387 out of 2400 schools selected within the 10 miles of driving distance buffer and there are 1773 out of 2400 schools selected within the 2 miles of walking distance buffer. After a simple calculation, the percentage of the library accessibility over school in 10 mile driving distance will be 99.46% and 2 mile walking distance will be 73.875%.

2.     Democratic factors and library distribution

There are several maps created regarding to the democratic factors to determine the distribution of the library.

First, the cluster analysis shows the sense of clustering of the population.  The HH areas represent tracts with high numbers of individuals surrounded by areas with high numbers of individuals; these are the hot spots. The LL areas represent tracts with low numbers of individuals surrounded by areas with low numbers; these are the cold spots. The remaining two categories are the outlier. White generally clustered in the west, Asian generally clustered in the east and the Black generally clustered in the South. Most of the library locations are concentrated in the Black clustered area. As shown in the map, most library locations are located in the Black clustered area with some library in both Asian and White clustered area.

On the other hand, the age group 5-17 population is mostly clustered in the south-east. But most of the library locations are on the LL area which is the cold spots of age group 5-17 population.  

The remaining two maps are population density with census tracts and household income median with city boundary. The population density map shows that there will be library located in high/very high population density area such as the downtown area, while there will be no library located in very low population density area such as the central valley area. The household income median map shows that with high household income median, there will be less library located, such as the west or Santa Monica area. The map shows a poor area may have more library located, for example, center of city Los Angeles, the downtown area.

Conclusion/Discussion

In terms of the library accessibility over school, the result suggested that it has a high accessibility to the library over school in the Los Angeles County. However, there are around 8 schools that have low accessibility because they are not within both walking and driving distance buffer. These schools are mostly located in the northern part of the Los Angeles County. I would also suggest that the percentage of walking distance access should be increased to 80% because most children do not drive. If we want to provide a better education development to them, more libraries should be built in the 2 miles walking distance buffer zones.

On the distribution of library, it is hard to tell what the exact factor to determine the library location. But from this project, there are several things were found. Library locations are mostly located in Black clustered area and poor (low household median income). The reason may due to the poor are more needed to get access to the library because they can’t afford to buy books to read. Library would be a best location for the poor because they can enjoy free access to the internet and it is a free entertainment for them.

However, there is lack of library for the age group of 5-17 population. In my opinion, they are the most needed for access to the library. Library offers book and educational materials. Apart from the formal reading, it is very important for them to read outside the school. It is needed to build more libraries in the age group 5-17 clustered area to meet the population needs. They are the future of the country.

There are several limitations for this study. City boundary in the Los Angeles County is complicated. I found several city boundary layer but most of them do not have all the city names/boundaries in the layer. Because of this, most of the data is using the census tracts because there will be a more accurate and precise view on the result. Besides, the study only includes city and county library in the Los Angeles County. The total number of library may be more than that because there is different library system in every place.

Sources:

California counties and census tracts shapefiles were downloaded from UCLA GIS data resource website.

The data and census tracts data were downloaded from US Census Bureau

Household median income data were downloaed from www.laalmanac.com/LA/la09.htm

Los Angeles County/ City of Los Angeles Library address from

www.colapublib.org/ and www.lapl.org  

Lab 6: Interpolation

In this final lab, we are using the interpolation technique to estimate the rainfall amount between stations that recorded the amount of precipitation. Spatial Interpolation is a useful tool to estimate unknown values with known values. In doing so, I picked around 55 precipitation station where they have both season normal and season total data. These geographical location values are called sample points or control points. We are using these 55 already known values to predict/estimate the rainfall in the other unknown value area. There is no need to collect data from every single point in the country to get the result.

There were two methods in analyzing the data in this lab, IDW and Spline. IDW uses the 40 surrounding points that I picked to determine the new value for in-between points. This method works well when the point density is high, just like this case, the calculations for in-between points will be more accurate when the points are closed together. In other words, spline estimates values using a mathematical function that minimizes overall surface curvature.

There are slightly different result in IDW and Spline. I think the IDW shows the a better result than Spline. From the IDW maps, in terms of the normal and total rainfall, the north-east of the LA County has the least amount of rain fall. This is pretty close to the result in reality and what I entered from the precipitation data. The most "heaviest" rain fall located in the mid-east of the LA County where most of the dams are located. The dam would be one of the factor that provide climate factors that causing more amount of rainfall. The IDW shows a more "normal" result. For Spline, although there is heaviest rainfall located in the dam area, there is rainfall located in the west. This is not so true that compare to common sense and what I entered from the precipitation data. Normal and Total rainfall map both have influences on the difference of rainfall map. The difference in result may due to the differences between the interpolation method.  IDW calculated the value of  in-between points and spline estimates values using a mathematical function that minimizes overall surface curvature. Precipitation data would be best to use IDW to represent. 

Quiz 2

1. China, India, USA, Indonesia, Russia, Brazil, Pakistan, Japan, Bangladesh, Nigeria. Select by attribute from cntry02 layer, type in query “POP_CNTRY” and get unique value. Get from the highest one and  to the 10^th.
2. 15 rivers in Amazon. Open the attribute table in rivers layer, select by attribute, enter query system="amazon".
3. 61 cities 
4. 516500000
5. The most populous landlocked country is Vatican City and the least populous landlocked country is Ethiopia. Open the attribute table in cntry02 layer, select by attribute, enter query landlocked=’Y’ and then Popcntry<=5000 on least populous and >=5000000 on most populous to see the result.

6.    6.  Poland, Czech Republic,Slovakia,Austria,Slovenia,Hungary,Romania,Croatia,Bosnia & Herzegovina, Yugoslovia

7. Libya, Niger, Sudan, Nigeria, Central African Republic, Cameroon. Selected Attribute using query CNTRY_NAME=’Chad’ found out which country share the common border on the map. Select by location and using the selected features touch the boundary of the source layer to find the answer.
8. Russia, United States, Thailand,  Turkey, Vietnam. 
9. There is a build in function called calculate the geometry in the attribute table. First, select attribute to locate where the country of Sudan is, then select attribute by location to locate the river inside the country of Sudan. I wonder if I can use the calculate the geometry in the attribute table to actually calculate the length of the river.
10. First, I join the table in lakes and country layer. I opened the attribute table, trying to use the build in function of field calculator, but it didn't work for me on several times. I know I need to use the field calculator to choose several of the countries. 
11. Again, I joined the table in lakes and country layer. I opened the attribute table, trying to use the build in function of field calculator and entering the query of equation to calculate the total area of lakes because there is a field with area. I know I need to put an equation in the field calculator to calculate the total area of lakes. 
12. 
    

Lab 4: Station Fire Hazard Analysis

Write-up on process and challenges

First of all, we need to look for data for this lab. I downloaded the digital elevation model from the USGS seamless viewer website. I picked the area where the station fire boundaries and the surrounding areas were and downloaded the DEM. I downloaded the station fire perimeters from the Los Angeles County GIS Data Portal. I was a bit confused at first on what data I should use as they have surface fuel and surface fuel rank. I finally used the surface fuel data from the California Department of Forestry and Fire Protection to finish my map.

The first map I created was the Slope Hazard map. First, I changed the projection of the DEM data to the NAD Projection UTM Zone 11. Then I got an accurate slope map. I followed the tutorial and then reclassify the slope value into NFPA hazard points. This is the first reclassification in this lab. The result image was then overlay with the hill shaded image to get a better visual appearance.

Second, we need to create and reclassify the surface fuel model data. When I opened the surface fuel data, there are about 13 classes of vegetation and they were all assigned with the numbers. I had to go back to the FRAP website to check on the definition of each and add on some descriptions beside those numbers. Then I reclassified the 13 vegetation classes into Non-Fuel, Light, Medium, Heavy and Slash, five different classes. And then I assigned these five classes into the NFPA hazard points as demonstrated in the tutorial.

The last map is the product and purpose of this lab. We need to use the raster calculator to “combine” both reclassified slope and surface fuel data into one map. After having both slope and surface fuel layer activated, we need to enter the formula into the raster calculator. This is the most important part of the lab because it will reclassify the data into a slope/surface fuel map. After summing both classified data, slope/fuel hazard can be assessed. The slope/surface map can be a useful tool in reality such as the safest locations to build residences can be determined, past fire extents can be assessed and fire suppression opportunities can also be explored.

I had a big challenge in doing this lab because I changed all the data projections to NAD 1983 UTM Zone 11. When I reclassified the surface fuel to NFPA hazard points, the image became blurry and could not be used for further investigation or analysis.  I had to re-download the data and do it again. One of the things I learned about in this lab is that I would only change the projection to do the slope map because of the projection would affect the calculations of the slope. The other challenges will be reclassification. Since the reclassification result really base on what I want to reclassify or what I want to group, it’s important to know different types of vegetations and their ability to catch fire. This is so important because the hazard map is completely based on a personal reclassification. Overall, I like using the raster calculator to combine the data and create a new map. This could be really useful to produce a map with two types of related data in GIS. 

Week 3 Lab: Suitability Analysis with Raster Data

Final Project: Introduction

What does it determine in the library location across Los Angeles County and the accessibility of library?

The project will look at several things. First, library locations geo-coding, located all the libraries across Los Angeles county. Second, schools  locations geo-coding, the schools is to determine whether library will be located near schools, whether students in the school can enjoy a walking distance to the library. Third, the median/average of household income across Los Angeles County. This would be an interesting factor that will there be more/less library accessible to rich/poor. Those are the three things that the project will focus on.

Quiz #1

I am against the decision that requires medical marijuana dispensaries in the City of Los Angeles to be at least 1,000 feet from places where children congregate, such as schools, parks and libraries. I support the statement by the given map. The star represents most types of the schools including elementary school, middle school, Junior and Senior High school but excluding colleges and universities. The green polygon represents the park location or area in the city of Los Angeles. The triangle represents the location of the library and the circle represents the location of dispensaries location that is allowed to operate according to the L.A. City Attorney.  Under the Los Angeles medical marijuana ordinance, only those dispensaries that registered with the city by Nov. 13, 2007, will be allowed to operate. I particularly focus on the location of both library and schools because there will be most numbers of children in the two locations. As shown on the map, the schools and libraries have the 1000 feet buffer. Of the 39 dispensaries that are mapped, there are around 12 of the dispensaries are completely located in the 1000 feet buffer.   What does this tell us? That means almost half of the dispensaries are located within 1000 feet buffer. These dispensaries have to find another place for relocation. These may cause a lot of economic problems for the dispensaries because they may have to consider about the rent and relocation fee and it would be unfair for some of the dispensaries.  It is because they are not “purposely” open or locate their dispensaries that are closed to school, libraries or even children. This causes more problems because dispensaries are also important in the community.I am strongly against that because it would be the choice of the children to get into these dispensaries but not because of moving further away to these locations that can fix the problems.

There is another map showing the dispensaries that are within 700 feet buffer of schools and libraries. I think for those super close to the schools and libraries. I consider 700 feet as a super close distance because it would be a normal walking distance for the children in my point of view. Those dispensaries that are within 700 feet should be relocated because children may get to these dispensaries really easily.  It is very important to limit the access of children to dispensaries because they can be really easily to get the drugs and medicines. It is needed to ask those dispensaries which are close to these children "concentrated" places to relocate to some places. The big issue on that will be dispensaries are not only used by children but also adults. It may cause a lot of problems as well especially in the community level because adults have to go further way to look for dispensaries. I have a suggestion that it should limit the distance between each dispensaries so it won't be too many or too concentrated number of stores in one location but then local people are still happy with that because they can still reach the dispensaries.