Reference no: EM132363490
Assignment - Exploring the Effects of Data Scaling, Classification, Aggregation and Misuse of Environmental Data
AIM
The goal of this activity is for you to think about how data which is used for environmental resource management can be manipulated - either intentionally or unintentionally to produce misleading results.
Task 1.
On pages 5 - 7 you will find six identical images. You are required to classify the image(s) into landuse or feature type maps of varying detail and resolution. Divide each image into low, medium and high spatial resolutions and use the following class schemas to classify your image(s). The low and medium images have been gridded for you.
2-class schema = grassland; other.
5-class schema = grassland; woody vegetation / trees, roads; built structures / housing; other.
Color or shade-in each cell according to the dominant class type that occurs within the cell for each image. Do this for each spatial resolution and number of classes combination. Make sure you provide appropriate documentation for your maps such as legends and class labels. The resulting maps should consist of:
Page 5. Low spatial resolution (e.g. 3x3 grid) for 2-class and 5-class classifications
Page 6. Medium spatial resolution (e.g. 6x6 grid) for 2-class and 5-class classifications
Page 7. High spatial resolution (e.g. 12x12 grid or higher) for 2-class and 5-class classifications
Questions:
1. Calculate the percentage of area covered by each class in each map (the number of cells in the class / the total number of cells * 100).
2. Describe how the percentage of area covered by each class varies between the maps of different resolutions. Discuss the effect that increasing the spatial resolution and number of classes has in respect to the accuracy of each of the final classified images. (Hint: look at how features (or objects) in your image, and transitions between features in your image can be lost depending on the mapping scale used.)
3. Which map (specify Figure number) offers the map reader the greatest amount of information. Briefly explain your answer.
TASK 2
Figures 7 - 9 (attached) show grids which describe the number of hollows, the number of old-growth trees in a forested area and also the average number of hollows per tree. The area of forest has been field surveyed and divided into 24 quadrats (6 columns x 4 rows).
|
100
|
10
|
5
|
10
|
5
|
10
|
|
20
|
10
|
20
|
10
|
10
|
5
|
|
10
|
20
|
10
|
400
|
10
|
20
|
|
20
|
40
|
20
|
40
|
20
|
40
|
Figure 7. Number of tree hollows.
| 200 |
20 |
10 |
20 |
10 |
20 |
| 20 |
10 |
20 |
10 |
20 |
10 |
| 10 |
20 |
10 |
400 |
10 |
20 |
| 10 |
20 |
10 |
20 |
10 |
20 |
Figure 8. Number of old-growth trees.
| 0.5 |
0.5 |
0.5 |
0.5 |
0.5 |
0.5 |
| 1 |
1 |
1 |
1 |
0.5 |
0.5 |
| 1 |
1 |
1 |
1 |
1 |
1 |
| 2 |
2 |
2 |
2 |
2 |
2 |
Figure 9. Number of hollows per old-growth tree.
Consider the following scenario(s). The government has granted permission for the private forest industry to log a small area of the forest shown in Figures 7-9. The conditions of the agreement are:
a) Total area available for logging cannot exceed 2x2 cells.
b) Areas with a ratio equal to, or greater than, 1.0 tree hollow per old growth tree must be excluded from logging.
In addition, the government has specified that decision-making will be performed at a lower spatial resolution than what the data was collected at, and as such any reporting and mapping may be done at an aggregated scale of 2x2 cells (that is, use the average values for any 2x2 cells).
You have been employed by a logging company to identify the area that meets the above government conditions but also maximizes the number of old-growth trees able to be harvested.
Questions.
1. Identify and label in Figure 8, the one 2x2 area that best meets government and employer requirements. In order to identify the ‘best' area you need to consider (a) the total area available cannot exceed 2x2 cells (b) the ratio limit of tree hollows is not exceeded and (c) the area maximizes the number of old growth trees to harvest.
(HINT: Your decision as to the best 2x2 area should be based on the ‘mean' or ‘average' value for the 2x2 cells - not the individual cells. In other words, calculate the ‘average' value for your chosen 2x2 area for each grid and use these aggregated values to make your final decision.)
2. In the area proposed for logging, what is the aggregated amount of:
a. Number of tree hollows?
b. Number of old growth trees?
c. Average number of hollows per old growth?
3. Consider the scenario whereby the acceptance of aggregation was revoked, which means you may no longer use aggregated averages and the original data must be used. Identify and label in Figure 8, the one 2x2 area that best meets government and employer requirements given aggregation is no longer permissible. Is it the same or different? Explain?
4. In the new area proposed for logging, what would be the aggregated amount of:
a. Number of tree hollows?
b. Number of old growth trees?
c. Average number of hollows per old growth?
5. How do the results in Question 2 differ from the results in Question 4? Discuss how the use of numerical averages (or other simple statistical measures) and data aggregations may be used to provide misleading information. What kinds of data or information do you think could be included ore provided in reports and/or maps to present a more "honest" answer when data summary techniques are used?
Attachment:- Exploring the Effects of Data Scaling.rar