Separation of large scale water storage patterns over Iran using GRACE, altimetry and hydrological data
Separation of large scale water storage patterns over Iran using GRACE, altimetry and hydrological data
复制标题
DOI：
10.1016/j.rse.2013.09.025
复制
发表时间：
2014-01-01
影响因子：
13.5
通讯作者：
Famiglietti, J.
中科院分区：
文献类型：
--
作者：
Forootan, E.;Rietbroek, R.;Famiglietti, J.
提取大规模蓄水(WS)模式对于了解水循环和改善伊朗的水资源管理至关重要，因为伊朗面临着水资源有限的挑战。重力恢复和气候实验(GRACE)任务提供了监测总蓄水量(TWS)变化的独特可能性。然而，从GRACE-TWS产品中准确估计地面和地表WS的变化需要适当的信号分离程序。为了实现这种分离，本研究提出了一种统计方法，该方法使用水文模型和高程测量数据中的地面和地表WS变化的先验空间模式。然后使用最小二乘平差(LSA)程序将图案调整为GRACE-TWS产品，从而最大限度地利用可用数据。在2002年10月至2011年3月期间，GRACE-TWS的月度变化是在包括伊朗在内的广泛区域内得出的。通过将以下辅助数据分解成统计上独立的分量，得到了先验模式：(I)全球陆地数据同化系统(GLDAS)的陆地WS变化输出；(Ii)经空间校正的里海地表WS变化；(Iii)波斯湾和阿曼海湾的WS变化；(Iv)咸海的WS变化；以及(V)选定区域的小湖泊的WS变化。最后，将(I)到(V)的模式调整为GRACE-TWS图，以估计它们的贡献并分离GRACE-TWS信号。分离后，我们的结果表明里海WS的年变化幅度为152毫米，波斯湾和阿曼湾均为101毫米，咸海为71毫米。自2005年1月以来，伊朗大部分地区，特别是中部和西北部的陆地WS呈现出质量下降的趋势，平均线性速率接近15毫米/年。2005年至2011年3月干旱期间，伊朗的六个主要盆地：哈扎尔盆地、波斯湾盆地和阿曼海湾盆地、乌尔米亚盆地、马尔卡齐盆地、哈蒙盆地和斯拉赫斯盆地的地下水储量估计线性趋势分别为-6.7、-6.1、-11.2、-9.1、-3.1和-4.2毫米/年。分离后的估算结果与256个现场测压观测结果吻合较好。(C)2013 Elsevier Inc.保留所有权利。
Extracting large scale water storage (WS) patterns is essential for understanding the hydrological cycle and improving the water resource management of Iran, a country that is facing challenges of limited water resources. The Gravity Recovery and Climate Experiment (GRACE) mission offers a unique possibility of monitoring total water storage (TWS) changes. An accurate estimation of terrestrial and surface WS changes from GRACE-TWS products, however, requires a proper signal separation procedure. To perform this separation, this study proposes a statistical approach that uses a priori spatial patterns of terrestrial and surface WS changes from a hydrological model and altimetry data. The patterns are then adjusted to GRACE-TWS products using a least squares adjustment (LSA) procedure, thereby making the best use of the available data. For the period of October 2002 to March 2011, monthly GRACE-TWS changes were derived over a broad region encompassing Iran. A priori patterns were derived by decomposing the following auxiliary data into statistically independent components: (i) terrestrial WS change outputs of the Global Land Data Assimilation System (GLDAS); (ii) steric-corrected surface WS changes of the Caspian Sea; (iii) that of the Persian and Oman Gulfs; (iv) WS changes of the Aral Sea; and (v) that of small lakes of the selected region. Finally, the patterns of (i) to (v) were adjusted to GRACE-TWS maps so that their contributions were estimated and GRACE-TWS signals separated. After separation, our results indicated that the annual amplitude of WS changes over the Caspian Sea was 152 mm, 101 mm over both the Persian and Oman Gulfs, and 71 mm for the Aral Sea. Since January 2005, terrestrial WS in most parts of Iran, specifically over the center and northwestern parts, exhibited a mass decrease with an average linear rate of similar to 15 mm/yr. The estimated linear trends of groundwater storage for the drought period of 2005 to March 2011, corresponding to the six main basins of Iran: Khazar, Persian and Oman Gulfs, Urmia, Markazi, Hamoon, and Srakhs were -6.7, -6.1, -11.2, -9.1, -3.1, and -4.2 mm/yr, respectively. The estimated results after separation agree fairly well with 256 in-situ piezometric observations. (C) 2013 Elsevier Inc. All rights reserved.