This week's lab was all about interpolation methods for surface data. Interpolation of spatial data is essentially assigning values across surface data at unmeasured/unvalued locations based on the values at measured/values sampling locations. During this lab, I worked with three primary forms of surface data interpolation. Thiessen Polygons, IDW or Inverse Distance Weight and Spline Interpolation. For the Spline interpolation method, there are two types of Spline Interpolation: Regular and Tension.
For this week's lab I worked in two parts. For part A, I compared Spline and IDW interpolation to create a Digital Elevation Model. While this part of the lab and assessing the differences between the two data methods was interesting, I'm going to share with you Part B. For Part B, I was provided 41 sample station water quality measurement points sampled in Tampa Bay, Florida. These data points essentially focus on the water quality and specifically the Biochemical Oxygen Demand (mg/L) at each sample.
Thiessen Polygons
The Thiessen Polygon interpolation method is fairly straight forward. Thiessen polygons contruct polygon boundaries where the value throughout the polygon is equal to the value of the sample point. Overall this method is fairly simple and widely used but for the nature of water quality data, the drastic shifts in polygon values and their clunky look do not reflect the data.
IDW (Inverse Distance Weight)
This method is much better for the nature of the data I was interpolating. Essentially, the values associated with the points directly affect the interpolation while the value decreases the further it gets away from the points. Points that are clustered together tend to push the overall data distribution higher in the clustered areas of concentration. For this data, this method still felt too clunky and did not reflect water quality.
Spline (Regular and Tension)
Spline Interpolation, the smoothest method employed in this lab essentially tries to smoothly go through the data sample points while reducing the curvature of the surface data. Regular spline interpolation is much more dynamic with value ranges (I had negative values even though my data contained none) with lower lows and higher highs. Tension spline interpolation attempts to reduce the factor of data values outside the initial range. For the nature of the data, I believe that Tension spline interpolation (Below) is the best method to visualize the surface data. Water is a smooth continuous medium and water quality can change constantly. Interpolation of this quality of data needs to be loose, but not exceed the data itself making tension spline interpolation the best method for this week.
~Map On!
No comments:
Post a Comment