adjust

Identifying potential land-use Conflict

To identify potential land-use conflict, it is proposed that three steps should be taken. , Firstly, remove lands whose use will not change; secondly, normalize and collapse preference results, and thirdly, combine the normalized and collapsed preference results to identify areas of conflict. This is a standard procedure we must adopt in identifying potential land-use conflict; however, there are three ways—natural breaks method, equal interval method, standard deviation method—to classify the target into preference results.

What are the deviations among these methods? The experiment is conducted to explore the question.

Three methods are used here to compare the deviations.

Figure 1 the Flowchart of Identifying the Conflicts

Figure 2 the characteristic of different conflict

Figure 1 shows taht all the models we have created to identify the conflicts have gone through three steps—normalization, reclassification, and combination. Normalization is the initial step for the model, in which all the AHP results are divided by 9 to get 0-1 normalized scale. Subsequently, three different methods are used to capture three different results, which is the purpose of this paper. Finally, by using raster calculator, we could combine the three different range rasters into one raster through varied digit positions to represent three distinct types, and by the characteristic of the combined raster, the potential conflicts are identified (Figure 2). The first and third steps are completely the same for all the models we have created, so more details are discussed below regarding the second step.

Natural breaks method is used by identifying naturally occurring groupings within the cell values of the input raster. Equal interval method is the easiest to understand, and the range of values in each raster is divided into three equal intervals. Standard deviation method groups cells according to how far their individual values vary from the mean value of all cells within the dataset.

Results

Figure 3 three different results using three classification methods

From the above result, we can see that the entire distribution is similar in terms of the five conflict classes, and the extent of each conflict is different. Furthermore, for the major conflict, the results from natural breaks method and equal interval method are (I don’t understand) the same as distribution; however, the result of standard deviation method is totally different.