|
|
|
| Precise comparison of spatial interpolation for precipitation using KRIGING and TPS (Thin plate smoothing spline) methods in Loess Plateau |
| Wan Long,Ma Qin,Zhang Jianjun,Fu Yanling,Zhang Xiaoping |
| College of Resources and Environment, Northwest Agriculture & Forest University; Insititute of Soil and Water Conservation,Chinese Academy of Sciences and Ministry of Water Resources; State Key Laboratory of Soil Erosion and Dry land Farming in Loess Plateau: 712100, Yangling, Shaanxi, China |
|
|
|
|
Abstract Loess Plateau covers large area in China, where precipitation shows complicated spatiotemporal variability. The interpolation method is an integral component for analyzing the spatial and temporal characteristics of precipitation in Loess Plateau, as well as they are crucial conditions for obtaining reliable results of hydrological and erosion modeling. In this paper, the precise of KRIGING and TPS (Thin plate smoothing spline) methods,were compared and analyzed according to the observed data including average annual rainfall, yearly and monthly rainfall for 20 years from 1981 to 2000 from 50 meteorological stations in the He-Long Section and the nearby regions. The performance of the two different algorithms, KRIGING and TPS, were verified and compared by cross-validation test based on the data of 27 stations. The results indicated that: 1) For the interpolation of the average annual precipitation, yearly and monthly precipitation, both the KRIGING and TPS methods can reflect the spatial distribution of precipitation in the He-Long Section. The absolute value difference of average A (the index of agreement) for whole period for both of the methods are less than ±0.01. There is no significant difference for the precipitation interpolation precision between the KRIGING and TPS methods. 2) The differences of the interpolation precision for both of the methods at different time scales show nearly the same trend. The average A for the average annual precipitation is better than that of April about 14%, better than that of the annual precipitation and the July about 19%, 35% respectively.
|
|
Received: 15 December 2010
|
|
|
|
|
|
|
