Hourly surface wind monitor consistency checking over an extended observation period

摘要: A consistency checking methodology is presented to aid in identifying biased values extended historical records of hourly surface wind measurements obtained from a single station. The method intended for screening observation periods which do not fail physical checks (i.e., standard or complex quality assurance methods), yet nonetheless exhibit statistical properties differ the bulk record. Several specific types inconsistencies common monitoring datasets are considered: annual biases, unexpected values, and discontinuities. The purely empirical self-consistency temporal distribution by explicitly modeling diurnal variability. Each year data modeled using principal component analysis (PCA) (or orthogonal functions, EOF), then hierarchical clustering with nearest neighbor linkage used visualize any biases existing measurements. distributions speed direction additionally estimated visualized determine time inconsistent typical cycle given monitor. The robust applied set 44 monitors operating San Joaquin Valley (SJV) Central California over 9-year period. Monitors SLAMS, CIMIS, RAWS networks considered. Similar detected all three networks; however, network-specific found as well. Copyright © 2008 John Wiley & Sons, Ltd.