University of Oregon cycling counts alongside daily class enrollment averages. The months of April and May are being analyzed exclusively due to the nature of the data source. Cycling count data are being fed live to the University of Oregon server via a magnetic inductive loop + microcontroller which stores the numeric count data. The sensor is installed at the West entrance of UO campus at East 13th Avenue, arguably the most popular port of entry to the University campus.

The top graphic illustrates UO cycling counts alongside precipitation rates (area). The bottom graphic is class population according to the University’s Registrar office for reference. The sudden decline in ridership from the last data point (June 2) appears to be a collective speculation among cyclists that rainfall was going to rival precipitation rates of the day before (June 1). Thus, it is possible that nearly half of typical Tuesday riders chose another method of transportation, particularly one that decreased the individual’s exposure to the environment.

Daily UO cycling counts alongside average daily temperature for the latter half of April and the month of May up to date. Cycling counts experienced a dramatic increase from mid to late April while assuming a higher average over the course of May, bike counts appear to have less volatility than daily temperatures.

 

The Sensor

Located beneath the pavement at the University’s Western entrance are several  magnetic inductive loops. As a bike wheel passes above the loop, the magnetic field of the loop is disturbed. The disturbance then triggers a single ‘count’ recorded on the microcontroller (large box situated to the South of the entrance). These data, in the form of 1-4 bytes of numeric characters, are then sent to the University of Oregon server via a wireless connection. Data transfer occurs approximately once per minute.

The Visualization

I chose Days as my temporal resolution for both environmental and population data. At the resolution of days, fluctuations within a week can be clearly identified. Repetitive patterns then emerge across the past two months which form an average. Months would have muddled the patterns seen per week which would have disabled the viewer from noticing nuanced responses, or lack thereof (in bicycle counts), to on campus events and inclement weather.

Overall, cyclists seem most responsive to discrete environmental events that are not displayed in the data explicitly.