Lab 5: Best Path Analysis to Optimize Likelihood of Encountering Tigers or Tiger Signs
in the Huay Kha Khaeng Wildlife Sanctuary, Thailand

In order to optimize the likelihood that a researcher would encounter a Tiger and/or Tiger signs, best path analysis can be extremely useful.

Objectives

---

Given the data, application and output from such application, best path analysis can assist in determining best path for likelihood that one would encounter a tiger and/or tiger signs while traversing through Huay Kha Khaeng Wildlife Sanctuary, in Thailand.

Objective 1: Run 3 different analysis using slightly different parameters in order to test best conditions for finding tigers and/or tiger signs

Conditions and Assumptions

---

Several assumptions were made in order to perform the analysis:
· There are, in fact, tigers inside the protected habitat
· Male and female tigers have the same habitat preferences
· Tiger migration habits inside these protected habitats is not determined by elevation and landcover
· Level of available prey to the tiger will remain constant throughout the wildlife sanctuary
· No migratory restrictions will be placed on tigers inside the protected habitats

Methodology

---

Spatial Data needed:

Non-spatial Data needed:

Non-spatial Operations:

Spatial Operations:

Implementation

---

1. Enter parameters into the following models:

	variable	distance (m)	weight	effect on the# of tigers
Run 1	villages	5000	1	decrease
	roads	5000	1	decrease
	boundary	5000	1	decrease
	rivers	5000	1	increase
Run 2	villages	6000	8	decrease
	roads	5000	2	decrease
	boundary	10000	3	decrease
	rivers	4000	1	increase
Run 3	villages	10000	1	decrease
	roads	5000	7	decrease
	boundary	3000	10	decrease
	rivers	7000	3	increase

2. Evaluate outuput supplied by Sue Grady of Hunter College, Dept. of Geography.

 

Evaluation

---

Three runs were outputted for interpretation. Run 1 (see Figure 1) represents a baseline analysis. Weighting the distance from villages, roads, boundary, and rivers equally (weight of 1) with equal distance (5,000 m) results a path very close to the rivers inside the park.

Figure 1: Output of Run 1

 

Comparing Run 2 to that of Run 1 (see Figure 2), we can see that setting the weight of the villages more than the other variables seems to cause a very slight variation of the path away from the villages but does not seem significant by eye. In the center of the park, the path does seem to alter slightly from Run 1 to Run 2, at the point of intersection of several rivers (see Figure 1 and box inside Figure 2). This is caused by decreasing the distance for optimization of rivers from 5,000 m to 4,000 meters, thereby forcing the path to hug closer to the rivers. Since most of the rivers are deep inside the park, and only 2 villages to contend with, increasing the distance from 5,000 m to 10,000 m for boundaries will not affect the run. This is due to the closeness of the distance to rivers required. Roads were not changes, therefore did not affect the model of Run 2.

Figure 2: Output of Run 2

 

Observing the results from Run 3 and comparing them to Run 1 and 2, we notice much similarity between the three runs, with Runs 2 and 3 being very close (including intersection of rivers). Setting the strongest weight to the boundaries does not seem to affect the model (10 in Run 3; 3 in Run 2; and, 1 in Run 1). Once again, this is probably due to the location of rivers, roads and villages inside the park. Although the distance from rivers was increased from 4,000 m in Run 2, to 7,000 m in Run 3, does not seem to affect the model whatsoever.

Figure 3: Output of Run 3

 

 

Limitations and Summary

---

Although the parameters of the model were altered at every run, no significant changes of path were found. Distance calculations may result in determining the final path, as shortest path would be preferred. Additional data layers are needed to do a more extensive model. Landcover, prey and predatory animals, and elevation data are such a layers. Different elevation and landcover types can influence tiger habitat (see previous analysis). In addition to influencing the numbers of tigers expected to be found in different elevation and landcover types, consideration of these two variables may influence the final path taken, as land with least slope and low vegetation will help traverse through the park.