Comparison of accuracy assessment methods for quasi-geoid model
Accurately assess the accuracy of the quasi-geoid model is an important reference for its appli-cation.As to the problem of the indicator differences calculated by existed accuracy assessment methods,the applicability of each method is analyzed in this paper.The model accuracy is derived from the differ-ence between the measured and the model elevation anomaly,and it is taken as the true value to analyze the reliability of the assessments by the empty point method,the cycling empty point method,and the er-ror characteristic-based method under different terrain types,as well as their applicability under different land types.Three testing areas with typical topographic characteristics in China are selected for analy-zing.The results show that in areas with flat terrain and relatively consistent surface types,the empty and the cycling empty point method can be used to evaluate the model accuracy,while the former has higher requirements for point selection.When the terrain is complex and the surface morphology is variable,the accuracy determined by the cycling empty point method is closer to real conditions.The error characteris-tic-based method is less affected by terrain,and the calculated accuracy by it is relatively higher.
quasi-geoidempty point methodcycling empty point methoderror characteristic-based methodaccuracy assessment