How do levels of urban stress, such as noise levels and pollution, vary when moving further away from the CDB of Bandung, Indonesia?
School: Mutiara Nusantara International School
Candidate Name: Constantijn Louis Pennekamp
Word count: 2,410Table of contents:
Fieldwork question and Geographic context……………………………2
Justification and procedures……………………………………………………5
Investigation and Presenting of Data………………………………………6
Data and Figures…………………………………………………………………….8
Analysis of Data and figures………………………………………………… .10
Fieldwork Question: How do levels of urban stress, such as noise levels and pollution, vary when moving further away from the CDB of Bandung, Indonesia?
Bandung, the capital city of West java province (Figure 1), is known to be one of the top three largest cities in Indonesia and consists of a population of over 7 million people. In addition, just like all large
cities, the presence of a central business district (CDB) is inevitable. Bandung’s limited amount of land and the lack of infrastructure and innovative planning have caused the city to sprawl instead of progressing upwards. Since Indonesian declared Independence against the Dutch in 1945, Bandung has experienced rapid development and urbanization, transforming Bandung from a very small town into a densely populated metropolitan area (Tamin, Sjafruddin, and alim 738). Figure one shows the borders of Bandung and determines its CBD. Coincidentally, Bandung’s Railway goes straight through the CBD, therefore, experiments were taken based on where A.S. Permana beliefs the CBD is located and held at railway stations.
This experiment will focus on how urban stress levels vary when moving outwards to the periphery of the CBD of Bandung Indonesia. Results will be compared to the typical factors influencing CBD decline, which is stated in Geography Course Sompanion. “The central business District (CBD) is the commercial and economic core of a city. It is the heart of the city, the area most accessible to public transport, and the location with the highest land values.” (Nagle and Cooke 286-315).
I hypothesized that urban stress levels would generally be higher when closer to the CBD compared to being further away from the CBD. And so we used the following Indicators to test the hypothesis. When the CBD was ascertained to a specific location, and all the stations were tracked down to its location, planning of how to gather primary data essential for constructing and concluding a hypothesis took place. Our initial goal was to find out different indicators for urban stress levels, which are: Pollution, noise, speed of cars, Infrastructure, and head count.
Indicator 1: Pollution
For measuring the amount of pollution, a Vaseline sample was used (Figure 2). In theory, the more pollution in the air, the more dust and other polluted particles that will adhere to the Vaseline (the darker the shade). It was assured that the same surface area and amount of Vaseline was used, also the amount of time that the Vaseline samples were exposed to open air which was 2 weeks. Pictures of each Vaseline sample will be taken after retrieving them and further Photoshoped to blend the colors in order to get a mean “shade” Indicator 2:
Speed of cars
A simple technique was also used to measure the speed of cars on the main roads closest to each train station: Using a measuring tape, a stopwatch and two other people to assist. This process is shown in Figure 3. Indicator 3: Noise levels
To measure noise, the original plan was to use a decibel meter; however, that piece of equipment did not fit into our price range. Instead, we used a high quality sound recorder to record the noise caused by civilization, with the intention to measure the decibels on a sound editing software later on. Indicator 4:
Amount of cars/Headcount
I used the cannon 7D to take pictures from the highest elevation possible in order to get a “bird’s eye view” on the area we were testing. I took 10 pictures with a 10 second interval between each one. Later during the processing of the data, I will count the amount of cars (only) visible on the picture, and calculate an estimate of cars passing per hour. ( 350) The reason what we resorted to simple techniques to obtain necessary data is simply because high tech and accurate equipment is too expensive to buy, especially in Indonesia, where recourses are very limited. This would however lead to more variables than one might like, but this could easily be handled by controlling other key elements, making the data as reliable as possible. Materials used
-HQ sound Recorder
1. Prepared 9 Vaseline Samples by stapling a white sheet of paper on cardboard and applying Vaseline on a pre measured circular area 2. Went to each of the 9 sites and: sticking the Vaseline samples vertically in a Dry
and safe place as near to the read as possible, measure the speed of 20 cars by counting mow much time it took for a car to travel 50 meters and calculate the mean speed of cars within the area, take 10 pictures from the highest elevation possible with an interval of 10 seconds of the nearest road from the station, and recording the sound produced by the area for 5 minutes long and later on measuring the mean decibels in a computer software. Investigation
The data analyzed in this IA was collected from the 15th until the 29th of February. All data was collected from 9 different train stations in Bandung, Indonesia. The rail track that goes through the Bandung area as a whole is estimated to be 21.54Km long. Random sampling was used in order to determine where the different locations of experiments took place; however, the locations were strategically picked (Figure4). There were a number of qualitative and quantitative results obtained at the end of this experiment. The qualitative data included: The Vaseline samples (Figure 6), and the urban land use map as secondary data (Figure 5). The quantitative data included: The Mean speed of cars, Amount of cars within the area, and the mean decibel count within 5 minutes. Data Presentation and Processing Nagle, Garrett, and Briony Cooke. Geography Course Sompanion. Oxford: Oxford University Press, USA, 2011. 286-315. Print. Bandung, Indonesia. July 27, 2011. Copyright Google Ink. June 2, 2012. kh.google.com
Initially it was predicted based on the land use map (Figure 5) that the further away from the CBD, the higher the urban stress will become (as land use increases within proximity to the CBD), however, this was not always the case. From the positive and negative correlation graphs (Figure 10, 11, 12 and 13) I can conclude that, in general urban stress levels increase the closer one gets to the CBD. The statements which were obtained from these graphs are: Formula for Spearman’s Rank Coefficient:
as the Distance from the CBD gets greater, The speed of cars generally increases (Figure 10) with a spearmans rank Corelation of 0.433. as the Distance from the CBD gets greater, The number of cars generally decreases (Figure 11) with a spearmans rank Corelation of -0.575. As the number of cars decreases, the speed at which they travel at increases (Figure 12) with a spearmans rank of -0.350. As the speed at which cars travel at increases, noise levels increase as well (figure 13) with a spearmans rank of 0.616. The exceptions to this experiment are the pollution levels from each site.
This was clear on the day the Vaseline samples were retrieved as the samples from site 8, which is shown in Figure 6, was visually the filthiest sample retrieved. This was a shock to me because site number 8 is located approximately 11 kilometers away from the CBD (one of the sites furthest away from the CBD). The sample from site 5 initially seemed fairly clean, and after being processed in Photoshop, it was ranked 3rd in proximity to being the least polluted. The statement “as distance from the CBD increases, pollution levels increase as well.” could not be justified because Vaseline sample from site 2 and 1, are ranked 1st and 2nd in proximity to being the least polluted (figure 9). The reason behind this inconclusive data might lie in the infrastructure of Bandung itself. As I have noticed during the experimentation period, the further away we moved from the CBD, the more frequent signs of poverty, such as illegal housing and bad infrastructure, become.
In addition, there was a fair amount of trucks passing through most of the sites. The roads in site 8 were quite wide and therefore some amount of industrial activity might be taking place there, causing it to be ranked 9th as the most polluted site. In contrast, site number 1 (Figure 14) is ranked 2nd in proximity to being the least polluted. Being approximately 8 km away from the CBD, the site was surprisingly clean and had well built buildings. A well structured shopping mall was also at site. Greenery must have played a huge role in the cleanliness of this area (Figure 14).
Another Interesting site is site 7 (Figure 15), also known as Cikudapateuh. This area is virtually very dirty and messy. The reason I say this is because the build quality of the buildings is very poor and old. However, this is not the reason why this site is interesting. The two ways how waste is handled in the area juxtaposes each other in a very humors way, as shown in figure 15. Cikudapateh is the only 1 of the 9 sites visited where we located a recycling site. The recycling site recycled mostly plastic cups and scrap metal. This could have played quite of a role in keeping the roads clean and reducing the amount of pollution. On the other hand, the areas local Traditional market is a huge source of organic waste. The mountain of trash visible in figure 15 has proved the previous statement. This mountain has also probably been lying around for about 3 days. I did not compare and annotate all the sites according to the amount of waste produced because dumpsters in the other sites were hard to locate, thus insufficient data for this was obtained. Be that as it may, I did use the method of disposal as a way to reason the sites ranked pollution.
The Spearman’s Correlation Coefficient is a formula used to determine whether a linear graph has a negative or positive correlation. This is different than the visual representations as it shows a numerical figure, -1 meaning the perfect negative correlation, 0 meaning no linear correlation, and +1 meaning a perfect positive correlation, in order to determine how strong the correlation is. For example, the relation of the distance from Bandung’s CDB and the Number of cars passing per minute (Figure 11) has the highest negative correlation with the Spearman’s rank Correlation of -0.575 while the Relation between the noise levels and the speed cars travel at (Figure 13) has the highest Positive correlation with a Spearman’s rank correlation of 0.616. With this information within my grasp, I can justify that figure 13 is more statistically significant compared to figure 12 and that figure 11 is more statistically significant compared to figure 10.
As stated in the paragraph above I have decided to only use figure 11 and 13 to analyze as they were the two most statistically significant. The Positive correlation between noise levels and the speed at which cars travel at suggest that high noise levels are generally caused by the speed of cars. In other words, cars which travel faster produce more noise due to the energy usage by the engine.
Figure 11 shows that there are more cars in proximity the CBD, which also influences the noise levels of the area. When there are more cars in the area, traveling at high speeds, the more noise produced, therefore increasing the urban stress levels. However, cars are not the only variables that contribute in producing noise. People and motorcycles also contribute to it; sadly, I did not use the headcount as there were no pictures taken which clearly shows the amount of overcrowding. This is the reason why the amount of cars passing per minute was counted instead of headcount (excluding motor vehicles).
The overall data obtained is sufficient to support my initial hypotheses, which is: Urban stress levels would generally be higher when closer to the CBD compared to being further away from the CBD. Data shows that cars travel faster further away from the CBD, Noise levels also increases in vicinity to the CBD. A strong negative Correlation between noise levels and speed of cars can support all of the above statements. However, there are some exceptions to this experiment.
The Vaseline samples was the only piece of information retrieved which did not support the hypothesis but the results from them is supported by this quote “…the travel length, fuel consumption, emission and transport cost at urbanized-area-road still the highest among others…” (Lubis, Isnaein, and Nurjaya 1709) This may be due to the fact that the CBD is more taken care of compared to the vicinity of the CBD, in addition, newer and more energy efficient cars are more frequent within the CBD causing less air pollution. On the other hand, the majority of data obtained further supports my hypothesis.
It seems that the data gathered is disappointingly insignificant, as correlations between the data do not have a high coefficient. However, some resources provide information I can use to reason the reason behind the results I obtained. Because the whole railway track was approximately 21.5 km long, the experiment took around 2 days to complete. Moreover, this might have altered the results as each site was not tested on the same day at the same time.
There was a minimum 30 minute interval between each train stop, and the testing period it took around 20-30 minutes. The accuracy of the results could have been more accurate if better technology, such as ones that measures co2 levels at ppm, and a speedometer, was used and measured at the same time of day. Having had a more controlled time frame would have given more accurate results. An interesting study for the future could be the amount of gasoline usage between different types of cars as distance from the CBD increases and compare that to the pollution levels within that area.