GRACE月时变模型缺失值的填补方法与精度分析

Filling method and accuracy analysis of missing values of monthly GRACE temporal gravity model

长时间序列的GRACE时变重力场对研究全球地表质量变化具有重要的意义.部分月份的GRACE卫星观测数据质量不佳导致了相应月份的时变重力场模型缺失,为了保持时变重力场模型的连续性,可采用一定的插值方法填补.本文以ITSG-Grace2016时变重力场模型序列为研究对象,详细分析了三次样条插值、三次埃尔米特插值和三次多项式插值等3种方法用于填补GRACE时变模型序列的精度,实验结果表明:(1)利用3种插值方法获取空缺1个月或连续空缺2个月的时变重力场模型时,插值时变模型与实测时变模型比较,阶误差均较小,且三次埃尔米特插值的精度稍好;(2)利用插值时变模型分析区域质量变化时,在空缺1个月数据的情况下,插值时变模型与实测时变模型符合度较高,但在连续空缺2个月数据的情况下,插值时变模型与实测时变模型的计算结果差异较大,说明利用阶误差评定模型精度具有一定局限性;(3)对区域质量变化的趋势项进行分析时,区域质量变化的复杂程度决定了模型内插的精度,当时间序列的长度在3年或3年以上时,插值时变模型的精度对区域质量变化分析的影响较小.在分析区域质量变化时,三次样条插值方法的插值结果与实测结果更为接近,建议采用该方法填补GRACE月时变重力场模型的缺失值.

The long-term sequence of GRACE temporal gravity field is of great significance for studying global surface mass changes.The poor quality of GRACE satellite observation data leads to the absence of temporal gravity field models in some months.In order to maintain the continuity of the models,we can fill it by some interpolation methods.In this paper,we selected the ITSGGrace2016 temporal gravity field models as the research object,and analyzed the interpolation accuracy of filling the GRACE temporal model sequence by cubic spline interpolation,cubic Hermite interpolation and cubic polynomial interpolation.The conclusions were as follows:(1)Using three interpolation methods to obtain the temporal gravity field model when missing one-month or two-month data,the degree error of the interpolation model obtained by each method is small,and the accuracy of the cubic Hermite interpolation is slightly better;(2)When using the interpolated models to analyze the regional surface mass changes,the interpolated models and the measured models have higher consistency for missing onemonth data.However,the calculation results of the interpolated model and the measured model are quite different for missing twomonth data.It indicated that using the degree error to evaluate the accuracy of the model is limited.(3)The accuracy of model interpolation is affected by the complexity of regional quality changes when analyzing the trend.However,when the length of the time series is more than 3 years,the accuracy of the interpolation model has little effect on the analysis of regional quality changes.When analyzing the regional mass change,the interpolated result of the cubic spline interpolation is closer to the measured result,so we propose to use cubic spline interpolation to fill the missing value of the GRACE monthly temporal gravity field model.