IDRISI Result
*Assume the project is using CA_MARKOV & MARKOV, cellular automata.
Although i cannot complete the project by using IDRISI because of some technical problem encounter, by the experience of using cellular automata, i can still compare the result with ArcView spatial decision analysis. In the project, original data was came from ArcView shapefiles. I have to import the data into IDRISI. Image show in figure 1.
Figure 1 IDRISI image on Coquitlam
All polygon has unique ID. I suppose to reclass them and assign a new unique value then we can reclassify the land use. As i mention in Methology section. Reclass function cannot perform. I have tried to fix the problem, but still, do not work.
An original idea of the project is using MCE to generate suitability map and then we may use the MARKOV & CA_MARKOV to have a prediction on future growth. By setting constraint and factors in MCE, the highest suitability place for residential grow will be up on the northern Coquitlam, where has a huge amount of open land. Although i may not have a result image to support my assumption, the most likely place to growth is up on northern Coquitlam. If two years data are collected, we can also use MARKOV & CA_MARKOV to predict.
Cellular Automata can predict a future growth on development. Unlike ArcView, CA can have a long term prediction based on probability matrix. Probability matrix is generated from MARKOV, it requires two different years data as an input. Also, CA use filter window to determine the cell future value. Different filter and size affect the expansion on land use. Default filter is 5 x 5 and with diamond shape.
| 0 | 0 | 1 | 0 | 0 |
| 0 | 1 | 1 | 1 | 0 |
| 1 | 1 | 1 | 1 | 1 |
| 0 | 1 | 1 | 1 | 0 |
| 0 | 0 | 1 | 0 | 0 |
Default filter 5 x 5. These 13 cells are used to determine and examine the future values. An origin cell future value is based on these 13 cells and see how they may affect the origin cell. And obviously, different style such as Moore, and different size, such as 3 x 3, will totally affect the result of expansion.
| 1 | 1 | 1 | 1 | 1 |
| 1 | 1 | 1 | 1 | 1 |
| 1 | 1 | 1 | 1 | 1 |
| 1 | 1 | 1 | 1 | 1 |
| 1 | 1 | 1 | 1 | 1 |
Moore 5 x 5
Cellular Automata used a simple process and perform a complex analysis. It is such a good spatial decision tool because it gives user a better visualization on future image. However, it requires a lot of data conversion because it does not have a direct file import to IDRISI.