Lab #7: Spatial Interpolation

            The normal seasonal rainfall values represent average values of rainfall, calculated using data from many years of measurements. The total seasonal rainfall represents the current precipitation patterns. According to LADPW, this total rainfall is a sum of precipitation from storm events occurring between October 2010 and March 2011. When comparing normal and total rainfall in Los Angeles County using an inverse distance weighted (IDW) interpolation, it is apparent that this year’s total seasonal rainfall is unexceptional and mostly follows the same patterns as the normal rainfall.

            The regions with the highest rainfall are the mountainous areas on the east and west side of the county. The desert areas in the north region of the county receive significantly less precipitation than the rest of the county. Finally, the south and northwest regions of LA County receive a more moderate amount of rainfall.

            Spline interpolation predicts the rainfall values of areas surrounding the gage stations by using an algorithm that minimizes overall surface curvature. This function prioritizes the position of the surface so that it exactly passes through the data points, instead of making estimations based on spatial patterns. Therefore, this interpolation is not ideal for predicting rainfall. For example, the normal and total rainfall maps analyzed with the spline interpolation have areas that were estimated to have negative rainfall values (represented by the grey areas).

            Inverse distance weighted interpolation makes predictions about precipitation rates by assuming that things that are close to one another are closer in value than those that are farther apart. Therefore gage stations that are closer to the prediction location are weighted greater than those farther away. This strategy is very effective in estimating rainfall due to the way in which storm systems move across terrain. Furthermore, the interpolation only generates appropriate values, unlike the spline interpolation, which can predict negative precipitation values.

Citations

"Precipitation." Department of Public Works. Los Angeles County. <http://ladpw.org/wrd/Precip/index.cfm>.

Childs, Colin. "Interpolating Surfaces in ArcGIS Spatial Analyst." ArcUser: ESRI Education Services. (2004): 32-35.