دانلود رایگان ترجمه مقاله رویکردهای موجک و گاوسی برای تخمین تغییرات آب های زیرزمینی (نشریه وایلی 2015)

شیوه های موجی و گاوسی برای تخمین تغییرات آب زیرزمینی با استفاده از داده های GRACE

Wavelet and Gaussian Approaches for Estimation of Groundwater Variations Using GRACE Data

Abstract

In this study, a scheme is presented to estimate groundwater storage variations in Iran. The variations are estimated using 11 years of Gravity Recovery and Climate Experiments (GRACE) observations from period of 2003 to April 2014 in combination with the outputs of Global Land Data Assimilation Systems (GLDAS) model including soil moisture, snow water equivalent, and total canopy water storage. To do so, the sums of GLDAS outputs are subtracted from terrestrial water storage variations determined by GRACE observations. Because of stripping errors in the GRACE data, two methodologies based on wavelet analysis and Gaussian filtering are applied to refine the GRACE data. It is shown that the wavelet approach could better localize the desired signal and increase the signal-to-noise ratio and thus results in more accurate estimation of groundwater storage variations. To validate the results of our procedure in estimation of ground water storage variations, they are compared with the measurements of pisometric wells data near the Urmia Lake which shows favorable agreements with our results.

1 Introduction

From all of the available water in the world, about 97% is distributed in the oceans and seas where the value of salinity is very high and not suitable for specific civilian use. From the 3% of the remaining water, about 2% is frozen in the polar ice sheets, unattainable to most, and only 1% is terrestrial water storage (TWS). TWS is the most important component of the global water cycle comprising the water stored in soil, snow over land, and the so-called groundwater storage (Chen et al. 2014). TWS variations reflect the accumulation of precipitation, evaporation, canopy, and runoff in a region. The estimation of TWS variations is a powerful tool for investigation of forecast flood, natural phenomena such as drought, and the other uses of water supply. Among the aforementioned ingredients in the TWS, groundwater storage is an important parameter in water resource management, land-surface processes, and hydrological cycle.

TWS variations could be estimated using spacebased data, such as remote sensing images or satellite gravimetric measurements, as well as ground-based data such as pisometric wells observations or climatological experiments. The remote sensing images have a good spatial coverage, but their sensing regions reached down to a depth of many centimeters. The data of pisometric wells could provide useful information of groundwater down to a depth of about 100 m and more, but they suffer from drawback of point-wise measurement which does not provide a good spatial resolution. Satellite gravimetric measurements of time variable gravity field are a new data type which is capable of modeling and detecting global mass transfer within the Earth. This subject, in its present form, began with the launch of the GRACE mission (Gravity Recovery and Climate Experiments). The GRACE mission provides a useful apparatus to study the time variation of the gravity field of the Earth. GRACE is able to monitor changes in a TWS from the land surface to the base of the deepest aquifer (surface water, soil moisture, groundwater, and snow) (Tapley et al. 2004b).

Numerous studies have shown that GRACE can offer useful constraints on TWS, including ocean mass change (e.g., Chambers et al. 2004; Lombard et al. 2007), mass balance of the ice sheets (e.g., Lutchke et al. 2006; Ramillien et al. 2006; Velicogna and Wahr 2006), polar ice sheet melting (e.g., Velicogna and Wahr 2006), and groundwater variations (Moiwo et al. 2012; Jin and Feng 2013; Lenk 2013).

GRACE consists of twin satellites which provide a unique opportunity to estimate the groundwater storage and its temporal variations. Because GRACE observations are affected by all sources of TWS, extraction of the groundwater component from GRACE observations requires the estimation of other impacts from auxiliary data sets. The useful data sets that can be used for this purpose are hydrological models such as GLDAS (the Global Land Data Assimilation Systems) or WGHM (WaterGAP Global Hydrology Model) (Rodell et al. 2004). In this study, the GLDAS outputs that are soil moisture, snow water equivalent, and total canopy water storage are used.

Recent outcomes of the GRACE mark a major step forward in assessing groundwater storage variations in our study location. Forootan et al. (2014) derived an estimation of TWS in Iran using combination of GRACE, altimetry, and hydrological data and analyzed a mass decrease with an average linear rate of 15 mm/year and linear trend of groundwater storage for the drought period of 2005 to March 2011. Joodaki et al. (2014) estimated the human contribution to groundwater depletion in the Middle East using GRACE and land surface models and found that the largest groundwater depletion is occurring in Iran, with a mass loss rate of 25 gigatonne (GT)/year and showed that over half of the groundwater loss in Iran may be attributed to human withdrawals. Moreover, Voss et al. (2013) estimated approximately 91 km3 groundwater depletion in Middle East particularly in Iran using GRACE observations from 2003 to 2009.

In this study, we apply GRACE TWS changes in conjunction with GLDAS outputs to resolve groundwater storage changes in Iran during the period of January 2003 to April 2014. Two different low pass filtering schemes were applied to refine the GRACE spherical harmonic (SH) coefficients, that is, the traditional Gaussian filtering and an innovated wavelet analysis. To show the performance and accuracy of the results, the estimated groundwater changes are compared with groundwater level obtained from in situ measurements of pisometric wells data over the Urmia Lake.