Isotropic and Anisotropic Variogram Models for Interpolating Monthly Mean Wind speed Data of Six Selected Wind Stations in Nigeria

Abstract

Abstract Mean Wind speeds often exhibit directionality in which they are increasing or decreasing across a surface; however, microclimatological effects sometimes produce high or low wind speed over a surface that can create confusion during kriging surface construction. The aim of this study was to investigate the appropriateness of anisotropic variogram models within ordinary kriging for interpolation of monthly mean wind speed data of six selected wind stations which include: Sokoto, Maiduguri, Ilorin, Ikeja, Port Harcourt and Enugu in Nigeria. Four types of isotropic and anisotropic variogram models were fitted: Linear, Spherical, Exponential, and Gaussian. Each model was described using the following parameters: the nugget variance, the sill, and the range. Three statistics to aid the interpretation of model output: the residual sum of square (RSS), R2 and proportion C/(C0+C) were provided to give the best fitted model for each wind station. The study found that the six wind stations could be best fitted by linear, Gaussian and exponential anisotropic models. Sokoto wind speed showed the strongest spatial distribution (>7.8 m/s), Maiduguri and Enugu, Ikeja and Port Harcourt showed similar wind speed patterns (3.1-4.0) m/s and (2.1-3.0) m/s respectively whereas Ilorin showed a pattern of low wind speeds (<2.0m/s) . These results may assist in identifying wind stations that are suitable for exploitation of wind energy for electricity generation as well as in mitigating losses to structures due to excessive wind events.