​Habitat Suitability Model

To view all maps, please refer to 

 

Comparing the present MCE model to the 2080 model shows that Suitable Habitat for Mountain Caribou in the Hart Ranges study area decreases significantly over the next 70 years. The amount of habitat that is ranked in the top 20% of suitable areas decreases by 1,542 sq.km. by 2080, which corresponds to 12.4% of the total study area of 12,465.6 sq.km. This means that suitable habitat for Mountain Caribou in southeastern BC will decrease by 26.5% by 2080!

 

Weighted Results and Sensitivity Analysis

The sensitivity analysis shows how much the weighted MCE and the equally-weighted MCE overlap and how sensitive our results are to the arbitrary weights we have assigned to the parameters. The Weighted Analysis maps and Sensitivity Analysis maps for 2013 and 2080 are shown below. As shown in the Weighted Analysis, the most suitable habitat (dark blue) occurs less frequently in 2080, and the less suitable habitat (yellow) is more spread out over the study area. The maps demonstrate how Suitable Habitat areas change over the study period due to the movement of BEC zones as a result of climate change. As well, the most suitable habitat areas become more spread out over the study area by 2080, making it difficult to locate an effective park. In both the sensitivity maps, the white area shows where overlap exists between the weighted MCE and the equally-weighted MCE. There is not a significant amount of white area compared to areas of non-overlap, showing that the results are slightly sensitive to the weights assigned.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Parks Analysis 

The present park boundaries are well placed within the study area to contain the most suitable habitat, but by 2080, the park no longer protects the most suitable areas. Because of the changing BEC zones, the suitable habitat for Mountain Caribou will also change, therefore affecting the effectiveness of the present park placement. Of the 12,465 sq.km. included in the study area, 1,356 sq.km. are designated park or conservation area. The largest of these, Kakwa Provincial Park, contains 63% (860 sq.km.) of the total study area. Therefore this park was chosen as the model for determining optimal park placement. Since the cell size is 1 sq.km. by 1 sq.km., we identified the top ~860 cells on the weighted raster through reclassification. By separating these top-ranking cells, we effectively created the most suitable park, but depicted through cells spread over the study area. In 2013, many of the pixels fall within the Kakwa borders - indeed it appears that one of the most concentrated areas of these pixels is within the park. The 2080 map, however, tells a different story. The park borders now contain more unsuitable than suitable habitat. This should be kept in mind when determining protected areas for caribou, because the natural features of the land will be changing within the next century, and suitable areas today may be unsuitable in the future.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Possible error

There may be several sources of uncertainty for this project, as the sensitivity analysis reveals. Possible sources of uncertainty could be in converting all the layers into raster format. This resamples the data, which changes the information contained within the layer; this might have altered our results. While there may be bias in the data collection, we cannot test this, and must therefore rely upon the government to have provided data that is as accurate as possible. There also may have been bias in the factors that we used, but we tried to include the most important factors for the habitats of Mountain Caribou based on background research. We attempted to reduce the bias in the MCE analysis by performing the analysis multiple times with different weights, including equally weighting the factors. This allowed us see whether the factors were accurately weighted for this particular scenario.