Tag: GEOB270

For the final group project of the course, I formed a group with the people sitting immediately next to me in the lab and we went with the first idea on which we agreed. We heard that Uber may be coming to Vancouver later this year and wanted to work with that. Uber is a sort of taxi service that uses a phone app to conveniently catch Uber taxi rides in your city. My group mate came up with the idea to try and map out where Uber hub zones might be placed throughout Vancouver. The idea behind the concept of a hub zone is that it is a dedicated spot where Uber customers can wait to be picked up, or Uber drivers can wait to pick up customers. These hubs can make Uber a much more convenient service to use, especially in parts of town where it  may be difficult to give an exact address of where you wish to be picked up, like downtown for instance.

Here is a link to the report my team created: GEOB270-Final-Project-Report-WED5

Here is the map we created:

To determine the best locations for Uber hubzones in Vancouver, we used data on daytime attractions, night time liquor primary establishments, transit stations, and population. We wanted the hubs to be in zones where daytime and night time activities overlapped, while remaining at least 100m away from transit stations, and tried to place the hubs in census tracts containing at least a medium level of population density.

Collecting the data for this project was tedious, but not too difficult. There was some data that came only in an excel file, so we had to run it through google spreadsheets before we could input it into ArcGIS and that did take some time. Once we had inputted all of our layers onto the map, we realized we did not exactly know what to do next. Determining the “best” locations for hubs was not something we could easily do with the click of one button on ArcMap and it did involve some subjective judgements on our part. We used our discretion in this final stage of making the map.

I definitely found working in a group difficult. I do not enjoy group projects; I already knew this about myself. Since our lab work is normally done individually, I was used to working on GIS alone and found adjusting to working with three other people was not enjoyable. Four people sharing one computer was not exactly conducive to much work getting done. I also noticed there were many moments where the “two heads are better than one” phrase did not work at all. We would all find ourselves stuck, wondering, unable to properly think and put our heads together.

Something I learnt is that it is extremely important to really think hard about your project before you start working on it. Our project ended up not working out so well because we did not properly determine whether or not it was something we could feasibly do; it turns out it sort of wasn’t. None of us have enough experience with GIS to perform things like network analysis, which would really have come in handy for a project to do with cars and roads. I think we should have picked a simpler topic that we could have gone more in depth on, and figured out all the necessary steps to complete the task before embarking on it.

What could have made our project better is if we had more data, and less assumptions/simplifications of reality. We only had two sets of data for reasons why people might take Uber; daytime attractions and bars/clubs. Of course in reality there are probably countless other places where customers might take Uber: shopping malls, their friend’s house, the airport… So our simplification of reality meant that the map might not end up conveying the actual “best” places to install Uber hubs. As well, we did not take into consideration if there is actually enough physical space in area to build a hub and have cars waiting to pick people up.

All in all, it was quite testing to have to figure out all on our own an idea, and then all the steps to make it become a reality. Through the completion of this lab I gained confidence in the use of ArcGIS and successfully used the software to complete a map for a social geography issue.

Quantitative Data Classification:

In this lab, I used multiple classification methods when creating maps to learn that different methods of data classification can influence the interpretation of data on maps. The maps below that I created of the cost of living in Vancouver show four different ways of displaying the same data.

The way you show the data matters. You might interpret the cost of living in Vancouver to be different than what it actually is depending on which map you are shown. The natural breaks and manual breaks maps make Vancouver look severely unaffordable, but the equal interval map makes it seem as though Vancouver is actually quite affordable. So if I was a real estate agent I would most likely use a map that looks like the equal interval map in order to convince people to buy property in Vancouver by presenting it as cheaper than it really is.

Housing Affordability:

Here are maps that I have made in order to compare housing affordability in Vancouver to London:

Housing affordability is different than just house cost. Affordability is the house cost normalized by residents’ income. This makes it more fair to compare housing cost between cities whose residents have different levels of income. And since Vancouver is a big city and London is a smaller city, people will have generally higher income in Vancouver. So we would expect also that housing prices would be higher where there is higher income, but looking at the map, it is clear that even though people in Vancouver have higher income levels, the houses are still priced way too high to be considered affordable.

The way that affordability was determined was by using the ratings provided on the Demographia International Housing Affordability Survey 2017. You can see the 5 categories in the legend on my map. I think that, for what they are, these can most likely be trusted, and Demographia seems like a reliable source. However, the question I have is whether or not using price normalized by income is the best way to determine affordability. In Vancouver, it is well known that housing prices are driven up by foreign buyers with massive wealth, and their income is not reflected in calculations of local income. And income is not the only factor playing into what house a person can afford: think about savings, or inheritances. So we have to keep in mind that while Demographia’s affordability categories can give us some idea of the affordability of a region, there are other factors to keep in mind.

Affordability can be a good indicator of a city’s livability but it is definitely not the only indicator. Other indicators you may want to look for would be low crime rates, good health statistics, low levels of unemployment and homelessness, good quality of education, and even very specific things like if you want to live in a place with lots of bike lanes. Affordability is a big factor that plays into whether or not you can live in a region, but even if you can afford to live there, you may not want to, because of crime rates or poor employment opportunities.

In this lab, we used data on elevation as well as distance to create a map to determine what parts of Vancouver would be affected by a tsunami.

The education and healthcare facilities that would be affected by a tsunami in Vancouver are as follows: Emily Carr Institute of Art & Design, Henry Hudson Elementary, False Creek Elementary, St Anthony of Padua, Ecole Rose des Vents, False Creek Residence, Villa Cathay Care Home, Broadway Pentecostal Lodge, and Yaletown House Society. To figure this out, I used select by location, and selected the education and healthcare facilities that fell within the danger zone. I then created new layers to show only those facilities in danger.

As shown in this map, in dark purple, the proposed new location of the St Paul’s hospital falls directly in the tsunami danger zone. This region is referred to as the “tidal flats” for a reason. It is possibly not such a good idea to build a hospital in this spot, especially considering that Vancouver is due for an earthquake at any time. In fact it would be quite ironic to see a hospital fall right to the ground shortly after being built.

To fix misaligned and improperly referenced spatial data, the first important step is to look at the properties of your layers and figure out the coordinate system information: what is there, and what is missing. If the coordinate system for a layer is unknown, then the data is improperly referenced and will be misaligned when placed on the map. To fix this problem, all you need to do is go into the properties of the improperly referenced layer, find the coordinate system tab, go into geographic coordinate systems, and select a coordinate system. The coordinate system you select should be the one that was originally assigned to the data before it was lost. The project-on-the-fly feature of arcmap will make the layers line up nicely even if they have different coordinate systems. If you want to match up the coordinate systems of the layers so that they are all the same, you can use the project tool to permanently alter a layer’s coordinate system.

An advantage to using remotely sensed Landsat data for geographic analysis is that new information is released every 16 days, so it is always possible to have up to date data. Another advantage is being able to compare images all the way back to 1972. The best part? All the data is free to access and use.