Spatial interpolation is a process whereby with the use of known data we are able to predict values for areas where we do not have data for reasons such as shortage of funding, lack of access or missing data. Using spatial Interpolation techniques I designed a total of six maps for the County of Los Angeles. Using data from the counties Water Resources website, http://ladpw.org/wrd/Precip/index.cfm, I was able to create an excel sheet that contained the known rainfall totals for both the normal rainfall totals and the season rainfall totals based on stations around Los Angeles County. With this information I then created maps using two different Spatial Interpolation techniques; Inverse Distance Weighting (IDW) and Spline, which are more exact techniques. With maps created for both normal and season rainfall totals in each of these spatial interpolation techniques I was able to use the raster calculator to determine the difference between the normal rainfall and the season rainfall to create two new maps.
As you can see on the IDW Normal map the heaviest areas of rainfall are in the eastern central part of the state with very precise definition. In the Spline Normal map below, you see the higher levels of rainfall in the same area, but with larger areas, less defined. In the Season maps again you see more defined areas of rainfall in the IDW map and more general in the Spline map. We observe the comparison maps created using the raster calculator we see more variation in the IDW map than in the Spline, however they is a general pattern that they share with above average rainfall in the eastern portion of the county and less in the west.
When creating these maps I had to consider which spatial interpolation technique to use. I opted for the IDW and the Spline methods as they are considered more exact. When using the IDW method I opted to use the default power of 2 and the variable radius type. I considered my data after evaluating the locations of the stations and since some areas of the county had sparse data, I opted to use number of points for my evaluation and used the default of 12. When using the Spline method, I tried the default regularized type however the data seemed extreme, so I tried the tension type which gave me more reasonable results. Based upon the final results using these two spatial interpolation techniques, I would tend to use the Inverse Distance Weighting as I feel like it gave more specific results.
No comments:
Post a Comment