Solid Earth Response To Environmental Variation

Mass above Earth's surface are redistributed due to temporal variations of atmospheric, continental hydrologic and oceanic loading. Mass redistribution causes variation of Earth's surface displacement, gravity, tilt and strain, as well as geocenter, geoid and Earth's rotation. Vertical displacement of Earth's surface due to environmental loading can reach several centimeters, and horizontal displacements have amplitudes of several millimeters with accompanying gravity variation of several tens u gal. These deformations may add noise to geodetic observations, so the effect of environmental variation can not be ignored for modem geodesy. In many instances, the deformation is large enough to be detected with space based geodetic techniques as well as terrestrial gravity observations.This dissertation focuses on physical mechanism and computing method of solid Earth response to environmental variation. Software for computing environmental variation effect is developed. Some cases are investigated using weather and altimetry data, and related analyses are done. The main products and conclusions of this dissertation are:(1)Theory and computing method of environmental effect on Earth's surface are addressed. The computing methods can be divided into three kinds: linear model approach, spherical harmonic function approach and Green's functions approach. Principles of each kind of methods are put forward, advantages and disadvantages of each kind of methods are evaluated. Detection methods of displacement and gravity variation are summarized.(2)Computing method for crust's displacement and strain variation caused by temperature variation are presented, some related software are developed, and some instances have been investigated. From the results, we can see that the maximum value of vertical displacement variation does not exceed 0.5mm and that of strain does not exceed 0.4nstrain.(3)Atmospheric gravitational load Love numbers (AGLLN) are defined and a set of asymptotic equations of AGLLN are put forward. Green's functions of atmospheric gravitational loading, describing the Earth deformation which includes horizontal and vertical displacements, horizontal and vertical accelerations and strain tensor, are derived out by evaluating the properly weighted sums of AGLLN. Atmospheric pressure loads Love numbers (APLLN) are defined and the formulas of Green's functions for pressure are deduced. Some related programs are developed and the numerical values of Green's function are computed using the preliminary reference Earth model (PREM).(4)Based on ideal atmospheric model, relationship between atmospheric density fluctuation in the air and pressure fluctuation at the Earth's surface is established. Formula and numerical values of gravitation Green's function only depending on pressure are presented. Together with Green's function for atmospheric pressure, we get the practical form of atmospheric Green's function. There is no observable difference between Green's function of atmospheric loading and the corresponding one of ocean loading. The effects of terrain, temperature variation and atmospheric model are discussed.(5)For computing the quantities describing Earth's deformation caused by atmospheric loading, it needs to be evaluated convolutions of the Green's function with the fluctuant pressure. In practical computation, we always adopt the pressure data at certain grid at certain time interval provided by various organizations, so we can transfer integral operation into operation of sum. The inner zone should be specially treated because the Green's function generally has larger absolute value with smaller angular distance. High-resolution land-sea data base is one of the important factors for the accurate loading estimation, especially for adjacent-sea sites. Details of process and computing flow chart are addressed, and related C++ programs for computing the deformation due to atmospheric loading are developed. Deformation time series of 40 GPS sites due to atmospheric loading, from Jan 1, 1952 to Jul. 6, 2003, are computed using atmospheric pressure data from National Center for Environmental Prediction (NCEP). Inverted barometer (IB) hypothesis and non-inverted barometer (NEB) hypothesis, as the oceanic response to pressure forcing, are investigated. From the results, we can see that the amplitude of vertical displacement due to atmospheric loading can reach ±20mm, the amplitude of horizontal displacement is about ± 5mm, and the amplitude of gravity variation is about ± 50 u gal. The power spectrum, time-frequency distribution, annual period and semi-annual period of the deformation series are analyzed. We have computed differences of deformations for different Green's function numerical results and different land-sea database as well as different atmospheric response models. The admittance factors are computed by least square estimation, and most of results are between -0.1 and -0.8 mm/mbar.(6)Continental water models are presented, and numerical recipes for computing the deformation due to Continental water variation are addressed. The programs for computing the deformation due to snow loading and soil moisture changes have be developed. Displacement and gravitation variation series due to continental water loading are computed using the about 50 years data from NCEP. It is shown that the amplitude of vertical displacement due to snow loading can reach ±8mm, the amplitude of horizontal displacement is about ±2mm, and the amplitudeof gravity variation is about ±10fj,gal. It also shown that the amplitude of vertical displacement due to soil moist variation can reach rtlOmm, the amplitude of horizontal displacement is about ±3mm, and the amplitude of gravity variation is about ±20figal .The power spectrum, time-frequency distribution ,annual period and semi-annual period of the deformation are analyzed. We have computed differences of deformations for different Green's function numerical results and different land-sea database as well as different data sources.(7)The modern global tide models can be categorized into three groups: hydrodynamical model, empirical model and assimilation model. Main ocean tide models are evaluated, and numerical recipes for computing Earth deformation due to ocean tide loading are addressed. The C++ programs for computing the effect of ocean tide loading are developed, and displacements of 40 sites response to M2, S2, N2, K2, K^,O\, Pi, Qi,MF, MM, and ssa are computed. M2, S2, N3, Ki,Oi and Pi have important effect on Earth's surface , even for island and adjacent-sea sites.(8)Deformation series of 40 sites, due to sea level anomalies of nearly 10 year's Topex/Poseidon and ERS1/2 altimetry data from Centre National d'Etudes Spatiales(CNES), are computed. From the results we can see that the maximum value of vertical displacement due to sea level anomalies can reach ±5mm, and that of horizontal displacement is about ±2mm. From power spectrum analysis, it is shown that annual term is the main period in displacement series. As the trend of sea level rising, The trend is also existed in the displacement series of each site. The trend for horizontal displacement is less than ±0.15mm/yr, and that of vertical displacement is about ±0.35mm/yr.(9)Load moment of atmospheric loading, continental hydrologic loading and sea level anomalies are computed. Potential and vertical displacement, due to environmental variation of every 2 months in 2000, are computed using the theory of Blewitt(2001). It is shown that the potential variation is about l~3gal and the vertical displacement is about lmm.(lO)Geoid variation, which is caused by environmental variation, can be computed by spherical harmonic function approach . Geoid variations, due to environmental variation of Feb. 1, Mar. 29, May 31, Aug. 2, Oct. 4, and Nov. 29 in 2000, are computed. From the results, we can see that maximum value of Geoid variation can reach 25mm.(1 l)Power spectrum and periods of pole motion(PM) and of length of day(LOD) variation, from geodesy and from effect of environmental variation (inverted barometer,wind and continental water), are computed. From the results, we know that the main periods are annual termand semi-annual term, and the amplitudes of annual term and semi-annual term are reduced after removing environmental effect.(12)Vertical annual and semi-annual terms of 40 IGS GPS sites are analyzed. The amplitudes of annual and semi-annual terms are deduced after pole tide, ocean tide loading, atmosphere, notidal ocean, snow, and soil wetness corrections. So environmental loading corrections should be added to time series of GPS coordinate.(13)GPS baseline time series of "GNS-JIAN" , "GNS-CTK" and "WUHN-GNS" are computed. The corrections of tide and environmental effect for vertical component of those baselines are investigated. From the results,we know that the corrections of tide and environmental effect should be considered when the length of baseline is much longer and accuracy is required higher.(14)The software SEREV 0.1 (Solid Earth Response to Environmental Variation, Version 0.1) is developed. The Earth's deformation due to environmental effect and tide can be computed using SEREV 0.1. It can compute power spectrum, time-frequency distribution and period. SEREV 0.1 also provide utility for coordinate transferred and data formats changed.