联合GRACE、GRACE-FO重力卫星数据监测华北地区水储量变化

Monitoring the variation of terrestrial water storage in North China using GRACE and GRACE-FO gravity satellite data

本文在三维加速度点质量模型法的基础上,融合不等式约束处理条带噪声,模拟计算结果表明该方法相比零阶Tikhonov约束,信噪比提高4.23%.利用不等式约束的三维加速度点质量模型法,采用GRACE与GRACE-FO重力卫星数据对2002年4月至2021年4月华北地区水储量进行了估算.研究结果表明,2002年4月至2021年4月,附有不等式约束的三维加速度点质量模型法计算的华北地区水储量以-1.36 cm/a的速率下降,Mascon与球谐系数法所获得的亏损速率分别为-1.52 cm/a、-0.80 cm/a,表明华北地区水储量处于明显的亏损趋势,并由亏损速率空间分布可知,水储量在河北省与山西省交界处亏损最为明显,亏损中心区域超过-3.0 cm/a.将研究时间段分为两部分分别进行拟合后发现,2011年至2021年华北地区水储量亏损趋势大于2002年至2010年.研究时间段内,2003年水储量存在明显的上升趋势,结合TRMM降雨数据对该时间段降雨时空分布进行研究,结果表明重力卫星数据计算结果相比降雨数据较为滞后,2003年春季时期的水储量的上升主要由于2002年下半年降雨的影响.

As an important resource for the development of human society, Terrestrial Water Storage(TWS) play an important role in industry, agriculture and daily life. However, with the population growth, industrial and agricultural development and urban construction, TWS are exploited and utilized in large quantities, resulting in some coastal areas such as seawater back irrigation and land salinization. Due to the rapid economic development in North China, a large number of TWS were exploited. In 1960, a settlement center began to form in Tianjin. Therefore, it is necessary to monitor TWS in North China. Based on the three-dimensional acceleration Point-Mass Modeling Approach(3 D-PMA), a 3 D-PMA with inequality constraints is proposed in this paper. The simulation signal calculation results show that the Signal-to-Noise Ratio(SNR) of this method is increased by 4.23% compared with the 3 D-PMA with zero order Tikhonov constraints. We calculated TWS in North China using GRACE and GRACE-FO CSR RL06 data by 3 D-PMA with inequality constraint and Spherical Harmonic coefficient method(SH). In addition, CSR RL06 Mascon data were used to extract TWS signals in the same spatio-temporal range to compare the calculated results. The results show that:(1) the SNR ratio of three-dimensional acceleration point-mass modeling approach with inequality constraint is better than that of three-dimensional acceleration point-mass modeling approach with zero-order Tikhonov constraint. Among all the constraint forms, the 400 km linear-type weighting has the smallest standard deviation.(2) The results of the 3 D-PMA with inequality constraints show that the TWS loss rate reaches^(-1).36 cm/a from 2002 to 2021, and the results of Mascon and SH are similar. According to the segmented fitting of TWS changes based on time series characteristics, TWS loss rate in North China from 2011 to 2021 is greater than that from 2002 to 2010, indicating that the TWS loss in North China accelerates in the later period. In terms of spatial distribution of TWS, the deficit is mainly located at the border of Shanxi Province and Hebei Province, and the deficit is most serious in Taihang Mountain range and its piedmont area.(3) TWS in north China in 2003, a significant upward trend, combined with eliminate seasonal signals of TRMM precipitation data found that gravity satellite data to calculate the TWS compared with rainfall data with a delay, when after the occurrence of rainfall, gravity satellite data need time to respond. Therefore, the increase of TWS in 2003 is related to the positive abnormal rainfall in autumn and winter 2002 and summer 2003.