The Vancouver Institute Of Leprechaun Research

Data Sources Spatial Analysis Results and Problems Background

Methodology

    This section focuses on the steps taken in completing the analysis. All steps were completed using IDRISI GIS software. The analysis was divided into two sections. The first part of the analysis is to determine the most suitable habitat for leprechauns. The second part of the analysis focuses on finding pathways that are highly traveled by leprechauns.

Analysis 1 - Suitable habitat selection
    The first step in the study was to decide on which characteristics of the environment would most influence a leprechaun’s choice of habitat. First of all, elevation would have to be limited to lower levels because of the colder temperatures and thinner air at higher elevations. As leprechauns are used to the elevation of Ireland, I selected a maximum survival elevation of 1041 m, which is the maximum elevation in Ireland. The second characteristic would have to be areas where there are no humans living or traveling. This includes areas where no people are living as well as areas as far away as possible from roads. Next, I looked at land use. This characteristic combines with the idea of leprechauns staying away from people as well as the idea that they prefer certain habitats such as forests.

    The next step was to convert some of the data so that it could be used in the analysis. Some of this was done in the MCE, which is discussed later in this section, but some of it was done using the reclass option in IDRISI. The land use data is the data that was converted using reclass to give each land use a suitability factor from 0 to 255 (0 being least suitable). The following table lists the reclassification.

1211

CODE	DESCRIPTION	SUITABILITY
1	agriculture	150
2	harvesting and research	200
3	extractive industry	0
4	residential - single family	50
5	residential - rural	75
6	residential - townhouse and lowrise apartments	0
7	residential - highrise apartments	0
8	commercial	0
9	industrial	0
10	institutional	0
11	transport, communications, utilities	0
12	parks and protected natural areas	225
13	lakes	0
14	open spaces and undeveloped	175
15	protected watershed	225

    The third step was to create a multiple criteria evaluation. This is a way to avoid the rigidity of Boolean operations and to create a layer based on several criteria, but each criterion has a relative importance rather than an absolute, acceptable or unacceptable value. Reclassifying each factor on the same scale so that they can be comparable does this.

    There was one absolute criterion or constraint, which was the human habitation criterion. Leprechauns can obviously not live where humans are living. I then incorporated three factors in the MCE. The first was the land use factor, which was already reclassified by hand. The second factor was elevation, which decreases in suitability up to 1041m and then levels off, as any elevation above this is just as unsuitable as another. The final factor is the distance from roads factor (vector layer converted to raster in order to do MCE), in which suitability increases away from a road up to 300m, after which there is no increased suitability.

    The next step is to decide on there relative importance to selecting a habitat. Using the weighting system shown below accomplishes this.

	gvrd_dem	land use	road distance
gvrd_dem	1
land use	5	1
road distance	3	1/3	1

Weights
gvrd_dem	0.1047
land use	0.6320
road distance	0.2583

consistency ratio	0.03

The higher the weight value the more important the factor. For our analysis the land use factor is the most important. This MCE result is shown in the Spatial Analysis section. We also created a Boolean image of the highest suitability areas (255) and then digitized those areas to show suitable regions.

Analysis 2 - Determining highly used leprechaun pathways.
This section of the analysis looks at leprechaun pathways and involves several steps. The first step was to create a friction surface. This is a surface where every pixel has a value that corresponds to a level of traveling difficulty. For our analysis there are many factors that affect traveling ability so we used a backward MCE to create our friction surface. We use opposite values because instead of a high number for the most suitable we want a low number for the most suitable areas. We also added two factors. The human habitation factor was moved from a constraint to a factor because it is not impossible for a leprechaun to travel where people live, but it is risky. We also included a water factor in which it is difficult, but not impossible for a leprechaun to travel over water (by sneaking on a boat or over a bridge).

	gvrd_dem	land use	water	road distance	no people
gvrd_dem	1
land use	3	1
water	9	7	1
road distance	5	3	1/5	1
no people	7	5	1/3	3	1

Weights
gvrd_dem	0.0333
land use	0.0634
water	0.5128
road distance	0.129
no people	0.2615

consistency ratio	0.05

In this case water is the least suitable traveling factor. In other words it has the highest friction.

module

    This chart shows the steps taken in finding the most likely pathways. This was repeated substituting different origins and destinations for each one of the six pathways.

    The next step was to digitize the shoe repair stores and shoe supply stores as well as origin points located in the center of each of the three regions. Then using our friction surface and origin we create a costdistance layer. Finally, using this costdistance surface and our destination we come up with a least cost pathway, or the safest way for a leprechaun to go from his home to a shoe store to get materials. This is repeated for each origin and destination.