This layer starts to address the issue that all parks are not of equal value. We decided to give parks a higher value based on two initial factors:
- Parks that are close to a lot of people are more important than those far away from people (all else being equal).
- If there are a lot of parks in an area, each park is of less importance than if it was the only park in the area (all else being equal). Being one of a few parks is still important, just slightly less so.
Granted, there are other factors affecting the valuation of parks, and maybe those can come in future years.
Detailed Methods: A parks and reserved layer (i.e. publicly accessible open spaces) was created by combining LTSBC conservation easements with public access, CPAD lands with gap statuses of 1-3, and county parks and reserves all larger than 1300 square meters (in order to remove coastal rocks included in the dataset). For each park the number of people that could get to the park easily was estimated. This was done by first creating a human population surface for the county.
Then, for each park, the euclidean distance to the park was calculated for every place in the county. The furthest distance to a park, of any park, was then assigned a value of 0, and all other distances scaled up to 1, reverse logarithmically. (In other words, after about 20 miles of distance, the relative value of the park was already low.) The direction to the park from every place was also determined, and the standard deviation of all the directions for a park was determined. The highest resulting standard deviation of all the park/direction surfaces was then assigned a value of 1, and the min a value of (max-min)/min. This is x. The min of all these was determined, and that value was divided by x for each park. This is a form of a centrality index, and compensates for the fact that parks at the edges of the region could only accumulate population visitation in one or a few directions. This index was multiplied by the distance surface for each park, and by the population surface. Then all the values were summed, for the region, associated with that point. This was done for every park, to get the relative value estimate for every park and reserve.