Suitable Habitats for Tiger Reserves

Thailand can provide a healthy landscape for tiger reserves, however they do require a very specific set of criteria.

Study Area

We were assigned to investigate Thailand for protected zones with suitable conditions for tiger reserves. We are specifically interested in locations 1km from the road network, with proper vegetation (Dipterocarp, Evergreen, Mixed Deciduous, Non-Forest, Pine Forest) and proper mammals (Sambar, Eld_deer, Gaur). With respect to the boundaries containing the correct ecological and environmental landscapes we also require a reasonable variable for the roundness of the areas themselves. 


Analysis

For Criteria 1 we began with placing a buffer region of 1km around the road network, this is to give us enough area away from the roads. Then we use the erase feature to remove this area from the areas of consideration. 

For Criteria 2 we are looking to find forest areas with Dipterocarp, Evergreen, Mixed Deciduous, Non-Forest,Pine Forest so we select these attributes from the forest layer and use the intersect feature tool to merge this criteria area with the layer in criteria 1.

For Criteria 3 we will have to use the Protect_ID to join the protect layer to the mammals layer. Then we will create a new feature layer based on the new joined table. Next we will have to use the Feature Class to Feature Class tool in order to create a feature class where Sambar and Eld_deer is one or more and Guar is less than or equal to one. 

For Criteria 4 we are looking to find the remaining polygons with a roundness value of 1800 or more, this is to ensure the polygon shape is sustainable for a tiger habitat. So we will first need to create a new column for roundness and next we will fill out the column with the ‘calculate field tool’. Next we will select the polygons with a roundness of at least 1800 and then subset the features. 

For Criteria 5 we test for spatial autocorrelation between the criteria 4 zones by running a global test for spatial autocorrelation (Global Moran’s I). A test for spatial autocorrelation can help us to see if there is clustering within the roundness feature of the habitat zones. This feature is chosen due to it being the last criteria for a well defined and suitable habitat for the reserves. We choose the inverse distance for the ‘Conceptualization of Spatial Relationships’ because we are interested in a mean distance between the zones as opposed to the squared distance which would lead us toward the median of the distances. 

Upon further analysis we may want to check other factors such as inverse distance squared and or locate a local point or zones to perform local tests for spatial autocorrelation. 

Criteria 1

To ensure a safe distance from the known road networks, we load the data from our data set. Next we place a 1km buffer zone around the road vectors and use the erase feature to remove these areas from our prospective results

Criteria 2

To exclude all vegetation types other than those that met the criteria, we selected only the desired vegetation attributes from the data. Then we intersected the Criteria 1 layer with the Criteria 2 layer.

Criteria 3

We need to identify Criteria 2 regions with proper and natural prey are essential to a well balanced habitat. Thus we needed to include proper mammals and ratios to ensure the areas would be sufficient sources of living conditions.

Criteria 4

We needed to ensure the areas found in Criteria 3 we found are a proper degree of Roundness based on the shape area and shape length.

Criteria 5

After we get our zones in Criteria 4 of suitable environments we will be looking to further explore the habitats welcoming to tigers and their place in the Thailand ecosystem. Thus we run a test of spatial autocorrelation to see if the zones themselves are a result of some underlying phenomena.

Results

After we applied to Criteria 1-4 to the study area, the end result is left only 86 zones for reservation. We then applied a measure of spatial autocorrelation to find that there is not an indication of positive or negative autocorrelation. Based on the lack spatial autocorrelation we can conclude there is not a relationship about roundness and ecology to investigate.

Following studies will look to replicate the use of criteria to discern tiger reserve habitats in alternate location and further investigate the underlying phenomena that may cause a clustering of suitable habitats.

The total study area in comparison to the total areas that provide the necessary criteria for the tiger reserves.

Resources

All of the data used in map creation and spatial analysis.

boundary.dbf

UCSB

boundary.sbn

UCSB

boundary.sbx

UCSB

boundary.shp

UCSB

boundary.shp.xml

UCSB

boundary.shx

UCSB

forest.dbf

UCSB

forest.sbn

UCSB

forest.sbx

UCSB

forest.shp

UCSB

forest.shp.xml

UCSB

forest.shx

UCSB

mammals.dbf

UCSB

mammals.Protect_ID.atx

UCSB

protect.dbf

UCSB

protect.sbn

UCSB

protect.sbx

UCSB

protect.shp

UCSB

protect.shx

UCSB

roads.dbf

UCSB

roads.sbn

UCSB

roads.sbx

UCSB

roads.shp

UCSB

roads.shx

UCSB

The total study area in comparison to the total areas that provide the necessary criteria for the tiger reserves.