TY - JOUR
T1 - Efficient kriging for real-time spatio-temporal interpolation
JF - Proceedings of the 20 th Conference on Probability and Statistics in the Atmospheric Sciences
Y1 - 2010
A1 - Srinivasan,B.V.
A1 - Duraiswami, Ramani
A1 - Murtugudde,R.
AB - Atmospheric data is often recorded at scattered stationlocations. While the data is generally available over a long period of time it cannot be used directly for extract- ing coherent patterns and mechanistic correlations. The only recourse is to spatially and temporally interpolate the data both to organize the station recording to a reg- ular grid and to query the data for predictions at a par- ticular location or time of interest. Spatio-temporal in- terpolation approaches require the evaluation of weights at each point of interest. A widely used interpolation ap- proach is kriging. However, kriging has a computational cost that scales as the cube of the number of data points N, resulting in cubic time complexity for each point of interest, which leads to a time complexity of O(N4) for interpolation at O(N) points. In this work, we formulate the kriging problem, to first reduce the computational cost to O(N3). We use an iterative solver (Saad, 2003), and further accelerate the solver using fast summation algo- rithms like GPUML (Srinivasan and Duraiswami, 2009) or FIGTREE (Morariu et al., 2008). We illustrate the speedup on synthetic data and compare the performance with other standard kriging approaches to demonstrate substantial improvement in the performance of our ap- proach. We then apply the developed approach on ocean color data from the Chesapeake Bay and present some quantitative analysis of the kriged results.
ER -