A2-DaisyWang
From CS294-10 Visualization Fa07
The question that I am interested in is: How light changes at different positions of a 70m Redwood tree?
The dataset is from swivel. It captures a month in the life of a redwood tree by measuring temperature, humidity, and light simultaneously at many levels of the 70m tree. You can see variation in these environmental parameters throughout the height of the tree, and then see how this profile changes over time. The dataset is captured with a research wireless sensor network system developed at UC Berkeley, called TinyDB, back in 2004. Each column represents a sensor, and the column header describes where the sensor was placed within the tree. Each wireless sensor network node had a temperature, light, and humidity sensor attached to it. As time went on, sensor nodes failed for various reasons, and not all the possible data was collected.
The dataset contains 79 columns. The first column contains the timestamp of at which the sensor readings are captured. The rest of the columns records light, humidity and temperature readings which are collected at 26 different positions (in terms of height and distance from the trunk) of a redwood tree.
The question that I initially posed cannot be answered by directly splot the existing data in the dataset. We have to parse the header (schema) to find out the height and the width of the sensor point, and also compute the average light over all time except the zero and missing values. The step that I took are the following:
1. I first strip the height and the width from the header row.
For example, for "Light (lux) at 22.9m up 1m out SW dir",
height = 22.9m and width = 1m.
2. I compute the average and standard deviation of the light readings
(non-zero, non-empty) for each column (>2).
3. Thus, I result in a table with 4 columns
(Height, Width, Light Avg, Light StDev),
and 26 rows with different Height and Width of a redwood tree.
This visualization captures three dimensions: - Height: the height of the sensor in a redwood tree - Width: the distance of the sensor horizontally from the main trunk - Light: the brightness of the light sensor readings
The Height is encoded in the Y axis, Width is encoded in the x axis, thus, each square in the graph is the (x,y) position, where each sensor is placed. The lightness of the square encodes the brightness of the light readings.
As we can see from the graph, the higher in the tree, more light is sensed by the sensors. The Width seems to have little effect on the light readings. We can also see how the sensors are positioned in the tree: between 45m to 55m, the sensors are placed on the far branches of the tree.
