Usage Guide
Scientific Computing Platform for Geophysical Geodesy · Usage Guide · 2026

How to Use PAGravf4.5 and ETideLoad4.5 to Achieve Proficiency, Capability Building, Problem Solving and Integrated Application

Drawn page by page from the official website www.zcyphygeodesy.com/en/ and its subpages and links; all names and technical terms follow the official wording, and all data and engineering results are quoted from publicly accessible pages of the official site.

2 core systemsPAGravf4.5 steady-state + ETideLoad4.5 time-variable
10 subsystemsfive in each system
about 7 working daysto complete all example exercises
5 typical applicationsengineering cases published by the official site
G

Guide to This Report: How to Use It and How to Read the Website

First build an overall picture of “the four main threads plus the official-site map”, then choose a reading path according to your role; terminology is compared in Appendix 8.4.
One-Minute Overview The Scientific Computing Platform for Geophysical Geodesy comprises two core modules: the Precise Approximation of Earth Gravity Field and Geoid (PAGravf4.5) and the Earth Tide, Load Effect and Deformation Monitoring Computation (ETideLoad4.5). Both were independently developed over a decade by Prof. ZHANG Chuanyin of the Chinese Academy of Surveying and Mapping, representing the culmination of his twenty years of pioneering research in this field. The platform and its core source codes were open-sourced and made freely accessible in September 2021, in order to improve the science-education environment in which such computational resources have been critically scarce worldwide.

0.1Official-Site Map: Sections and Subpages

Every key point of this report can be traced back to a specific page of the official website. The table below lists the top-level entries and the content they carry (all pages were checked for accessibility and return HTTP 200).

SectionPage addressContent covered
Homewww.zcyphygeodesy.com/Platform overview, introduction to the two systems, code-level hybrid programming and program scale, the example mechanism, five typical applications, and the entry to Academic and Technical Exchange
PAGravf4.5/Geoid.htmlScientific objectives, the complete algorithmic framework for terrain effects, technical characteristics of gravity field integral algorithms, spectral-domain Spherical Radial Basis Function (SRBF) approximation, typical computation programs and classroom demonstrations
ETideLoad4.5/Tide.htmlScientific objectives, the five major subsystems, basic principles for representing, approximating and monitoring the load deformation field, the principle of CORS-network time-series InSAR collaborative monitoring of vertical deformation, and the seamless continuous quantitative monitoring scheme of regional ground stability variations
Downloads and Resources/h-col-104.htmlLists of program packages and User References, algorithm and process documents, computation processes and exercises, external data resource links, and installation notes
Ideas & Philosophy/h-col-105.htmlConcepts of geodesy and physical geodesy, the metrological essence and normative constraints of the discipline, geodetic datum constants, the “four-in-one” cornerstone, Analytical Compatibility, error analysis and accuracy assessment, height datum and reference system theory
Fortran Codes/h-col-110.htmlModule-by-module Intel Fortran source codes and module introductions (basic functions, geophysical models, algorithm formulas, DOS test program codes)
Academic and Technical Exchange/h-nr.html (with ?m396pageno=2…17 pages)Article index, categorized as “News / Software Teaching / Work Processes / Source Codes”, linking to subpages
Codes & Contact/h-msgBoard.htmlOnline communication and messages
English site/en/English description of the platform and the software
Subpage naming convention: /h-nd-*.html are individual article pages; /h-nr--0_31_*.html and /h-nr--0_396_*.html are category list pages.

0.2Four Main Threads and Reading Paths

Main threadProblem addressedCorresponding chaptersCorresponding official-site pages
① ProficiencyStarting from scratch, can you independently run an official process from beginning to end?Chapter 1, Chapter 2Home, Geoid.html, Tide.html, h-col-104.html
② Capability BuildingCan you compute correctly and judge accurately — algorithmic judgement and error awareness?Chapter 3, Chapter 3 (Supplement)h-col-105.html, h-nd-162.html, h-nd-163.html
③ Problem SolvingGiven a concrete task, can you immediately locate the subsystem and the process?Chapter 4Geoid.html, Tide.html, h-nd-109/110/111
④ Integrated ApplicationCan you coordinate the steady-state and the time-variable, and unify multi-source heterogeneous data?Chapter 5Tide.html, h-col-104.html, h-nd-87.html
O

Process Overview: From Knowing the Platform to Integrated Application

Skim the global path first, then go deeper chapter by chapter; all terms and technical names follow the wording of the official website.

O.1Overview of the Four Main Threads

Main threadGoalKey actionsMain outputsCorresponding chapters
① ProficiencyRun an official process independently from beginning to endInstall the platform → complete all example exercises (about 7 working days per package) → build a knowledge map with the teaching coursewareThe ability to use the software systems independentlyChapter 1, Chapter 2
② Capability BuildingCompute correctly and judge accuratelySelf-test algorithms with an ultra-high-degree geopotential coefficient model → choose parameters by the quantitative selection criteria for terrain effects → perform error analysis and accuracy assessmentAlgorithmic judgement, quality-control awareness, Spatial Representativeness AccuracyChapter 3, Chapter 3 (Supplement)
③ Problem SolvingFind the entry point immediately when a task arrivesDecide steady-state or time-variable first → then the system → then the subsystem → apply the official process and quality controlSpecialized results such as gravity field, geoid and deformation monitoring productsChapter 4
④ Integrated ApplicationUnify multi-source heterogeneous data and the two systems in a coordinated wayUnify models and numerical standards → unify the spatiotemporal monitoring datum and reference epoch → deep fusion by the five principles → cross-checkGround stability, time-variable gravity field, and numerical prediction of geological hazard riskChapter 5

O.2End-to-End Process (Ten Steps)

1Download the resources: obtain the complete main program packages of both systems, the User References, the algorithm and process documents, and the example process data.
2Complete the installation: place the split-volume files in the same folder before installing (Windows 10 platform); if security software isolates a file, restore trust in its quarantine.
3Reproduce the examples: following the operation process and the program interface screenshots in each example folder, reproduce the outputs one by one to confirm that the environment and the operations are correct.
4Study the courseware: use Chapters 3–8 of the Physical Geodesy teaching courseware to build a knowledge map of terrain effects, reference systems, height datums, modeling and time-variable monitoring.
5Run through the official processes: practise them in turn — terrain effects and gravity exploration modeling (four steps), the expedited process for all-element gravity field modeling in full external space using SRBFs (six steps), the complete computation process of high-resolution surface load deformation field time series (four steps), and the collaborative monitoring process of groundwater and surface environmental load deformation field (five steps).
6Prepare the external data: obtain global geopotential models, land and seafloor topography, hydrological and meteorological data, satellite gravity, sea level and ocean tide models, and other data listed on the official resources page.
7Self-test with an ultra-high-degree model: take degrees 1–540 of EGM2008 as the reference gravity field and degrees 541–1800 as the residual disturbing gravity field, and test the performance of approximation programs and parameter settings.
8Fix parameters and schemes: select the terrain effect type by the quantitative selection criteria, and satisfy the two restrictive requirements — identical terrain mass model and integration radius, and fully analytically compatible algorithms.
9Organize modeling and monitoring: use PAGravf4.5 to establish the steady-state skeleton (reference gravity field, regional gravity field and geoid model, height datum), then use ETideLoad4.5 to compute tidal and polar shift effects, non-tidal effects and the load deformation field.
10Fusion and cross-check: fuse epoch by epoch according to the five principles for deep fusion of multi-source heterogeneous data, unify the spatiotemporal monitoring datum and the reference epoch, and cross-check with analytical compatibility relations and independent external data.

O.3Quick Reference of Common Entry Points

Task typeSystemPreferred subsystem
Processing diverse external terrain effectsPAGravf4.5② Computation of Diverse Terrain Effects on Various Gravity Field Elements
All-element gravity field and geoid modelingPAGravf4.5③ High-Precision Gravity Field Approximation and Full-Element Modeling
Regional height datum optimization, unification and GNSS height conversionPAGravf4.5④ Optimization, Unification, and Application of Regional Height Datums
Correction of solid Earth tides, ocean tides, atmospheric tides and polar shift effectsETideLoad4.5① Computation of Various Tidal and Polar Shift Effects on All Geodetic Elements
Load deformation field, time-variable gravity field and terrestrial water storage monitoringETideLoad4.5③ Load Deformation Field Approximation and Monitoring from Heterogeneous Variations
Ground stability tracking and numerical prediction of geological hazard riskETideLoad4.5④ CORS/InSAR Collaborative Monitoring and Ground Stability Estimation
The complete “problem–subsystem” quick reference is given in Section 4.8.
↑ Back to top
1

Knowing the Platform: Composition, Subsystems, Technical Foundation and Capability Boundaries

Corresponding official-site pages: Home, Geoid.html, Tide.html, h-col-105.html. Terminology follows the official wording throughout.
Key Points of This Chapter ① One platform = two systems; ② each system is organically composed of five major subsystems, ten in total; ③ when a task arrives, answer three questions first — steady-state or time-variable? Which subsystem should be called? Where does the data come from?

1.1One Platform, Two Systems

SystemOfficial full name and positioningMain problems addressedOfficial source
PAGravf4.5 (steady-state branch)Precise Approximation of Earth Gravity Field and Geoid (PAGravf4.5) — a scientific computing package for geodesy grounded in steady-state gravity field theory, comprehensively covering the fundamental principles, main methods and all formulas of physical geodesy and Earth gravity field theoryAnalytical computation of diverse external terrain effects · all-element analytical modeling of the gravity field in full external space · gravity exploration modeling from multi-source heterogeneous data · optimization and unification of regional height datums · 1 cm refinement of the steady-state geoid · determination of external accuracy metrics and control of computational performance/Geoid.html
ETideLoad4.5 (time-variable branch)Earth Tide, Load Effect and Deformation Monitoring Computation (ETideLoad4.5) — a large Windows program package for scientific computing of geophysical geodetic monitoring, covering the fundamental principles, main formulas and key methods of geodesy for a deforming EarthAnalytical computation of tidal and polar shift effects on all elements in full space · processing and analysis of non-tidal geodetic time series · load deformation field approximation and monitoring from heterogeneous variations · CORS/InSAR collaborative monitoring and ground stability estimation · collaborative monitoring of terrestrial water storage and the time-variable gravity field/Tide.html
Official wording: the platform was "independently developed over a decade by Prof. ZHANG Chuanyin of the Chinese Academy of Surveying and Mapping, representing the culmination of his twenty years of pioneering research".

1.2Technical Foundation and Program Scale

DimensionOfficial wording
Development approachDeveloped in the Visual Studio 2017 x64 integrated environment using code-level hybrid programming: QT C++ (Visual C++) for the user interface, Intel Fortran (Fortran90, 132-column fixed format) for the core function modules, and mathGL C++ for geodetic data file visualization
Scale of PAGravf4.5Composed of nearly 50 win64 executable programs and more than 500 function modules
Scale of ETideLoad4.5Composed of more than 50 win64 executable programs and nearly 600 function modules
Example mechanismBoth packages are provided with complete computation examples; each program example folder contains the operation process, input-output data and a series of program interface screenshots
Proficiency standardAfter completing all the example exercises (about 7 working days per package), users are basically able to use the software system independently
Intended usersIntegrates classroom instruction, self-study exercises, applied computing and scientific research; suitable for senior undergraduates, graduate students, scientific researchers and engineering technicians in geodesy and geosciences, geology and geophysics, geomatics and geographic information, aerospace and satellite dynamics, and seismology and geodynamics
Installation requirementThe _win64cn.exe and _win64cn.2 files must be downloaded into the same folder before installation (Windows 10 platform)
Open accessThe platform and its core source codes were open-sourced and made freely accessible in September 2021

1.3Capability Map: Five Major Subsystems in Each System

No.PAGravf4.5 (steady-state)ETideLoad4.5 (time-variable)
①Analysis and Preprocessing of Earth Gravity Field DataComputation of Various Tidal and Polar Shift Effects on All Geodetic Elements (elsewhere on Tide.html worded as “analytical computation of tidal and polar shift effects on all geodetic elements”)
②Computation of Diverse Terrain Effects on Various Gravity Field ElementsProcessing and Analysis of Non-Tidal Geodetic Variation Time Series
③High-Precision Gravity Field Approximation and Full-Element ModelingLoad Deformation Field Approximation and Monitoring from Heterogeneous Variations
④Optimization, Unification, and Application of Regional Height DatumsCORS/InSAR Collaborative Monitoring and Ground Stability Estimation
⑤Corresponding subsystems: PAGravf4.5 — Toolkit for Geodetic Data File Editing, Calculation, and Visualization; ETideLoad4.5 — Editing, Calculation, and Visualization Tools for Geodetic Data FilesSame correspondence on the time-variable side (editing, calculation and visualization of geodetic data files)
The subsystem division and names come from Geoid.html and Tide.html. Note: the official website writes “CORS/InSAR”, without a hyphen.

1.4Functional Commitments: Two Sets of Lists

SystemOfficial functional commitments
PAGravf4.5① Constructs analytically harmonic terrain effect fields outside the geoid and develops an algorithmic framework for diverse terrain effects that suits various field elements at different altitudes in space, air, land and sea; ② develops a scientifically complete algorithmic framework of spatial-domain forward and inverse integral operations and spectral-domain Spherical Radial Basis Function (SRBF) approximation, realizing all-element analytical modeling of the regional gravity field in full external space from multi-source, heterogeneous, varying-altitude, cross-distributed land-sea observations, and realizing closed-loop analytical operations among field elements; ③ proposes quantitative selection criteria for terrain effects, clarifies the analytical relations between the gravity field and height datums, creates analytical modeling methods for gravity exploration from multi-source heterogeneous data, and optimizes and unifies height datums; ④ provides gross error detection for gravity field observations in complex situations, determination of external accuracy metrics, all-element analytical modeling and computational performance control.
ETideLoad4.5① Employs consistent geophysical models and rigorous uniform numerical standards to construct analytically compatible geodetic and geodynamic algorithms, and uniformly computes various tidal and non-tidal effects on all types of geodetic elements; ② precisely approximates global and regional load deformation fields from surface environmental observations; ③ strictly follows the principles of geophysical geodesy to constrain and assimilate the deep fusion of multi-source heterogeneous monitoring data; ④ quantitatively tracks variations in the surface dynamic environment and ground stability, scientifically supporting the deep fusion of multi-source heterogeneous geodetic data and collaborative monitoring across heterogeneous geodetic techniques.
Sources: Geoid.html and Tide.html / h-col-104.html respectively.

1.5Theoretical Foundation: Four Groups of Concepts

Theoretical pointCore content on the official siteSignificance for use
The metrological essence and normative constraints of geodesyGeodetic elements must satisfy objective uniqueness and precise measurability; geodetic datum constants must satisfy unique invariance, precise measurability and conceptual freedom from specific epochs and measurement errorsDecides whether a quantity can serve as a “datum” or a “solution”
Geodetic datum constantsThere are exactly nine: the origin of the Earth-fixed reference system (mean center of mass, 3 constants), the reference pole (mean figure pole, z-axis orientation, 2 constants) and the zero meridian (reference meridian, x-axis orientation, the Non-Rotating Origin of the GCRS, 1 constant), together with the global geopotential W₀, the geocentric gravitational constant GM and the mean Earth angular velocity ωThe basis for unifying the numerical standards of the systems
The “four-in-one” cornerstoneForms the “four-in-one” cornerstone of geodetic theory, breaking the long-standing pattern in which geometric and physical geodesy developed relatively independently and achieving the intrinsic unification of the entire disciplineExplains why the two systems can work in coordination
Analytical CompatibilityThe analytical relations between different or same-type observations and geodynamic influence quantities; it is the minimum requirement for multi-source heterogeneous data fusion and collaborative monitoring across heterogeneous techniquesTests whether integrated results are correct

1.6Platform Positioning and Capability Boundaries

Platform Positioning To improve the science-education environment in which such computational resources have been critically scarce worldwide; to build a cognitive framework and knowledge architecture for multi-source heterogeneous data fusion and collaborative multi-technique geodesy; to demonstrate the scientific elegance and potential of geodesy; and to promote the strategic transformation of the discipline from a “descriptive measurement tool” into a “dynamic Earth system analyzer”, and of geodetic education from “technology-oriented” knowledge transfer to “problem-driven” systems thinking and conceptual innovation.
Boundary dimensionDescriptionNote
Operating platformThe official installation notes specify the Windows 10 platform and the win64 build; cross-platform use requires a self-built environment
Distribution and sizeSplit-volume files must be placed in the same folder before installation; both main program packages are of the order of gigabytes
External data dependencyThe platform does not embed global topography, hydrological, tidal or satellite gravity data; these must be obtained separately from the lists on the official resources page
Parameter automationMultiple optional algorithms, parameters and technical routes are provided; choosing among them requires the user's own testing and analysis — results are not produced at the push of a button
Result verificationThe built-in case studies “are intended primarily to illustrate computation processes and do not optimize or analyse the algorithms themselves or their main parameters”; greater potential must be explored by users for their specific situations
Precision definitionWhat the platform provides is the Spatial Representativeness Error / Spatial Representativeness Accuracy definition, which is not equal to the actual observation accuracy at measurement points
Chapter Summary Once a task arrives you can answer immediately — steady-state or time-variable? Which subsystem should be called? Where does the data come from? — and the first step is secure.
↑ Back to top
2

Proficiency: Resources, Installation, Examples and Running Through the Processes

Corresponding official-site pages: Home (example mechanism and proficiency standard), h-col-104.html (Downloads and Resources), h-col-110.html (Fortran Codes), h-nd-166…173 (teaching courseware).
Pathway to Proficiency (Six Steps) Obtain all resources → install the platform → run the examples → build a knowledge map with the courseware → practise with the official processes → prepare the external data. Official standard: after completing all the example exercises (about 7 working days per package), you are basically able to use the software system independently.

2.1Step 1: Collect All Resources by Category

CategoryOfficial list
Main program packagesPAGravf4.5: PAGravf4.5_win64cn.exe (620.0 MB) + PAGravf4.5_win64cn.2 (462.42 MB); ETideLoad4.5: ETideLoad4.5_win64cn.exe (845.0 MB) + ETideLoad4.5_win64cn.2 (792.92 MB). The packages were updated on 1 August 2026
User ReferencesPAGravf4.5 User Reference (PDF, 40.6 MB), ETideLoad4.5 User Reference (PDF, 68.48 MB)
Algorithm and process documentsTen topical PDFs for PAGravf and ten for ETideLoad; additionally the JPLEPH.440 planetary ephemeris file (JPL Development Ephemeris DE440) required by ETideLoad
Example dataEach process is provided with an example process data package (e.g. the all-element gravity field modeling in full external space using SRBFs, load deformation field approximation, and collaborative monitoring examples)
Page: /h-col-104.html.

2.2Step 2: Three Key Points of Installation and Deployment

Key pointOfficial requirement and practiceNote
Split volumes in the same folderBe sure to download the .exe and .2 files into the same folder before installing (Windows 10 platform)
Security software interceptionAfter downloading, if security software isolates the .exe file, find it in the quarantine area and restore trust
Installation pathA dedicated directory is recommended; paths containing Chinese characters or spaces may cause read/write anomalies in individual modules

2.3Step 3: Learn by “Examples”, Not by “Functions”

Key Design Both software packages are provided with complete computation examples; each program example folder contains the operation process, input-output data and a series of program interface screenshots; after completing all the example exercises, users are basically able to use the software system independently.
1Read the examples through: first study the operation process and the program interface screenshots to build an intuitive impression of “what goes in, what to click, what comes out”.
2Reproduce the examples: run the example results with the input data supplied with the examples and compare each output file to confirm that the environment and the operations are correct.
3Break the examples open: replace some parameters or input data, observe how the outputs change, and understand the role and sensitivity of each parameter.
4String the examples together: link several examples within the same subsystem into one complete process and understand the data flow between modules.
DayTask
Days 1–2Familiarize yourself with the program interfaces and data file formats
Days 3–6Run the examples through, subsystem by subsystem
Day 7Link the processes together and carry out one end-to-end recomputation

2.4Step 4: Use the User Reference as a “Dictionary” and the Courseware as a “Map”

MaterialOfficial sourceKey point
User Referenceh-col-104.htmlLook-up of functions, parameters and processes; large and information-dense, consult it on demand
Teaching courseware (PPT-01…08)h-nd-166…173Chapter 3 Terrain effects, tidal and load effects; Chapter 4 Earth-fixed reference system and relativistic unification in Earth space; Chapter 5 Gravity field modeling from heterogeneous observations; Chapter 6 Height datums and their gravity field theory; Chapter 7 All-element modeling of the local gravity field and time-variable monitoring (based on PAGravf4.5 and ETideLoad4.5); Chapter 8 Vertical datums unified in the Earth-fixed reference system
Textbooks and monographsh-nd-164 (Physical Geodesy, 2026); h-nd-156 (Geodesy for a Deforming Earth, 2025)The theoretical foundations of the two systems respectively
Official wording: the courseware serves multi-level needs — “classroom instruction, self-study exercises, applied computing and scientific research”.

2.5Step 5: Practise with the Official Processes and Algorithm Documents

Official process / exercise StepsTraining value
Expedited process for computing land-sea Bouguer and isostatic anomalies from a gravity field modelFour stepsQuickly and conveniently practise the computation of land-sea unified Bouguer and isostatic gravity anomalies (disturbances)
Terrain effects and gravity exploration modeling exerciseFour stepsThe complete chain of gravity exploration modeling from analytical fusion of heterogeneous observations
Expedited process for all-element gravity field modeling in full external space using SRBFsSix stepsOmits terrain effect processing; quickly conveys the essentials of gravity field observation analysis, computational quality control and gravity field modeling
Complete computation process of high-resolution surface load deformation field time seriesFour stepsTaking regional terrestrial soil moisture as an example, realizes high-precision, high-resolution all-element grid time series computation of the load deformation field in near-Earth space over an arbitrary region
Collaborative monitoring process of groundwater and surface environmental load deformation fieldFive stepsJointly uses the CORS network, the solid Earth tide station network and the groundwater monitoring network to realize collaborative monitoring across techniques
Example: the four-step method for terrain effects and gravity exploration modeling
1Determine the computation scope, the computation surface (a terrain equi-elevation surface is recommended) and the gravity exploration model type, and acquire (collect) gravity field exploration and geodetic data within and around the computation scope.
2Call the relevant programs of “High-Precision Gravity Field Approximation and Full-Element Modeling” to precisely determine the high-resolution grid model of the target gravity field element on the computation surface.
3Call the relevant programs of “Computation of Diverse Terrain Effects on Various Gravity Field Elements” to precisely compute the high-resolution grid model of the terrain effect consistent with the gravity exploration model type.
4Subtract the terrain effect grid model from the target gravity field element grid model to obtain the gravity exploration grid model that fully fuses all available gravity field data information.
PAGravf4.5 algorithm and process documents (10)ETideLoad4.5 algorithm and process documents (10 + ephemeris)
Updating the concepts and philosophy of height systems and height datumsUnified computation of solid Earth tidal effects on all elements on the ground and external to the solid Earth
Analytical interpretation of the concept of terrain effects and its restrictive requirementsGlobal load tide harmonic analysis and spherical harmonic synthesis of load tide effects
Algorithmic framework for terrain effects on all-element gravity field quantities external to the EarthComputation of rotation polar shift effects and tidal effects on Earth rotation parameters
Integral algorithm formulas for the disturbing gravity field in the spatial domainComputation of Earth's center of mass variation and figure pole shift effects
Spherical radial basis function theory and algorithms for gravity field approximationGlobal load spherical harmonic analysis and spherical harmonic synthesis of the load deformation field
Expedited process for computing land-sea Bouguer and isostatic anomalies from a gravity field modelLoad Green's function and spatial-domain integral algorithms for load effects
Process for computing Complete Bouguer gravity disturbances on a mean terrain elevation surfaceSRBF approximation of surface loads and SRBF synthesis of load effects
Exercise process for all-element gravity field modeling by spatial-domain integral methodsRadial basis function approximation of the load deformation field from multi-source heterogeneous monitoring quantities
Expedited process for all-element gravity field modeling in full external space using SRBFsComplete computation process of high-resolution surface load deformation field time series
Principles and requirements for regional gravity field and geoid modelingCollaborative monitoring process of groundwater and surface environmental load deformation field
Document names are taken verbatim from /h-col-104.html.

2.6Step 6: Prepare the External Data Environment

CategorySourceAddress
Gravity field modelsICGEM time-variable model series / long-term modelsicgem.gfz-potsdam.de/series · icgem.gfz-potsdam.de/tom_longtime
Regional geoidISG (regional models and software)isgeoid.polimi.it/Geoid/reg_list.html · isgeoid.polimi.it/Software/software.html
Marine gravity field, etc.DTU scientific data and model downloadsspace.dtu.dk/english/Research/Scientific_data_and_models/downloaddata
Satellite gravityGRACE / GRACE-FO / AOD1B / PODAAC / CSR ephemerideszcyphygeodesy.com/grace/ · zcyphygeodesy.com/grace-fo/ · gfz-potsdam.de/en/aod1b/ · podaac-tools.jpl.nasa.gov · filedrop.csr.utexas.edu/pub/slr/
Earth rotation and reference framesIERS Conventions (2010), rapid EOP prediction, GGFC, SOFA, ITRF2014iers.org/IERS/EN/DataProducts/Conventions/conventions.html · hpiers.obspm.fr/eop-pc/index.php · loading.u-strasbg.fr/GGFC/ · iausofa.org · products/reference-systems/terrestrial/itrf/itrf2014/
Topography and bathymetryETOPO1 (NOAA), GEBCO (BODC), Sandwell merged topography, SRTMngdc.noaa.gov/mgg/global/ · bodc.ac.uk/data/hosted_data_systems/gebco_gridded_bathymetry_data/ · topex.ucsd.edu/WWW_html/mar_topo.html · srtm.csi.cgiar.org/srtmdata/
Hydrology and meteorologyNASA GES DISC (GLDAS), NOAA PSL (CPC soil moisture, reanalysis)disc.gsfc.nasa.gov/datasets · psl.noaa.gov/data/gridded/data.cpcsoil.html · psl.noaa.gov/data/gridded/index.html
Sea level and oceanOpenADB, Aviso sea level, CMEMSzcyphygeodesy.com/pub/OpenADB · aviso.altimetry.fr/…/gridded-sea-level-anomalies-mean-and-climatology.html · resources.marine.copernicus.eu/
Ocean tide modelsTPXO8 / DTU global ocean tide modelstpxo.net/global/tpxo8-atlas · space.dtu.dk/english/research/scientific_data_and_models/global_ocean_tide_model
Shared software and servicesSHTOOLS spherical harmonic analysis library, Chalmers ocean tide loading online serviceshtools.oca.eu/shtools/public/ · holt.oso.chalmers.se/loading/
All links are taken from the official “Downloads and Resources” section. Note: the availability of external links changes over time; before downloading, please refer to the latest status of the official site and of each source site.

2.7Proficiency Self-Check List

√Self-check item
□Can recite verbatim the names and the division of labour of the five major subsystems of each system
□Can complete the installation independently and correctly handle split-volume files and security software interception
□Can independently reproduce results in the example folders by following “operation process + program interface screenshots”
□Can describe the step structure of at least three official processes
□Can quickly locate the needed resources and parameters in h-col-104 and the User References
□Can list the types of external data required by a task and the links for obtaining them
Chapter Summary Proficiency is not about “clicking fast”, but about being able to independently run an official process from beginning to end and explain what each step is doing. About 7 working days of example exercises are needed, after which you move on to “Capability Building”.
↑ Back to top
3

Capability Building: From Using It, to Computing It, Judging It and Innovating

Corresponding official-site pages: h-col-105.html; h-nd-162.html (error analysis and accuracy assessment in PAGravf gravity field approximation); h-nd-163.html (principles, methods and restrictive requirements for terrain effects on PAGravf gravity field quantities); Geoid.html.
Key Points of This Chapter A four-level capability model plus five training methods: use an ultra-high-degree geopotential coefficient model to “health-check” algorithms · use the quantitative selection criteria for terrain effects to ground your parameter choices · use Analytical Compatibility and the Spatial Representativeness Error to bound credibility · use spectral-domain Spherical Radial Basis Function (SRBF) approximation for performance assessment · use the boundary value problem classification logic to choose the right solution method.

3.1Capability Layering: A Four-Level Progressive Model

LevelCapability markerCorresponding training leverage on the platform
L1 Using itCan reproduce results following the examplesExample folders (operation process + input-output data + program interface screenshots)
L2 Computing itCan choose the right subsystem, set the parameters properly and produce self-consistent resultsAlgorithm and process documents, User References, quantitative selection criteria for terrain effects
L3 Judging itCan assess accuracy, identify gross errors and control uncertaintySelf-testing with ultra-high-degree models, closed-loop analytical operations, error simulation, error analysis and accuracy assessment
L4 InnovatingCan embed the platform in original research or complex engineering and produce new methods or resultsFortran code modules, monograph theory, cross-system combinations
The bottleneck for most users lies between L2 and L3: the program runs, but they cannot explain “why these parameters were chosen, how accurate the result is, and where the errors come from”.

3.2Training Method 1: Use an Ultra-High-Degree Geopotential Coefficient Model to “Health-Check” Algorithms

Official Scheme The performance of most gravity field approximation algorithms and parameter settings can be tested and verified by users with an ultra-high-degree geopotential coefficient model; many PAGravf4.5 program examples take degrees 1–540 of EGM2008 as the reference gravity field and degrees 541–1800 as the residual disturbing gravity field for self-testing.
1Take residual disturbing field elements of certain types as the observations.
2Call the PAGravf4.5 program or function under test to obtain computed values of residual field elements of another type.
3Take the field element values given by the model as the reference values.
4Compare the computed values with the reference values to assess the technical performance of the program and its parameter settings.
Advanced techniqueOfficial practice and value
Closed-loop analytical operationsFrom one type of field element, all-element field elements in the whole external space outside the geoid can be computed, and the same type of field element at the same point can be computed cyclically; by comparing the differences and similarities between the observed field elements and the cyclically computed field elements, the algorithmic characteristics and performance of the programs and functions invoked can be analysed
Error simulationBecause the theory of gravity field approximation in external space is founded on a linear space, all terrain effect and gravity field approximation algorithms implemented in the platform are linear; simulated spatial noise can therefore be used as observations or as land-sea terrain input, so that the outputs directly express the error distribution properties of the target field elements (or terrain effects)
High analyticity of gravity field informationGravity field approximation is essentially a one-dimensional linear transformation between gravity field quantities; gravity field information is highly analytical, so weak signals can be extracted from strongly noisy observation environments — for example, satellite gravity exploration can effectively extract gravity field signals from noise with a signal-to-noise ratio below 10⁻³

3.3Training Method 2: Put the “Criteria” and “Restrictions” for Terrain Effects to Work

First Locate the Position: the Treatment of Terrain Effects on Gravity Field Quantities in Physical Geodesy Has “Exactly Two Purposes” First, the estimation or gridding of discrete disturbing gravity field quantities, so as to facilitate numerical spatial-domain gravity field integral computation or gravity exploration modeling; second, the separation of the ultra-short-wavelength components of gravity field quantities during gravity field approximation, so as to improve the approximation performance for the ultra-short-wavelength gravity field. Methods and technical requirements should revolve closely around these two purposes.
Three key elements of terrain effectsDescription
① Terrain/crustal mass adjustment methodThis distinguishes the types of terrain effects: Local Terrain Effect, Terrain Bouguer Effect, Seawater Bouguer Effect, Isostatic Effect, Helmert Condensation Effect, Residual Terrain Effect, etc.
② Type of gravity field quantityThat is, the object of the terrain effect (height anomalies, gravity disturbances, vertical deflections, gravity gradients, etc., outside the geoid)
③ Positional relationship between the gravity field quantity and the terrain massesE.g. ground gravity disturbances, gravity disturbances external to the geoid, gravity disturbances on the geoid
No.CriterionTestStatus
①Smoothness criterionAfter removing the terrain effects, the standard deviation of the discrete field quantities should decreaseA technical requirement, to be observed
②Ultra-short-wavelength criterionThe standard deviation of the disturbing field quantities decreases after removal, and the mean value of the terrain effect quantity over tens of kilometres is small enough to be neglectedA technical requirement, to be observed
③Outlier criterionThe ratio of range to standard deviation, D/ε, reflects the outlying character of the ultra-short-wavelength signal; a large D/ε indicates that the ultra-short-wavelength share is small but its magnitude is large, so using that terrain effect to process the field quantity is advantageousA technical reference, to be applied flexibly
④Short-wavelength richness criterionWhen D/ε is of comparable magnitude, the larger the ratio of the standard deviation of the gravity terrain effect to that of the height anomaly terrain effect, the richer the short-wavelength componentsA technical reference, to be applied flexibly
Official limitation: when terrestrial gravity data are scarce and the spatial representativeness of the statistical results for terrain effect quantities is seriously insufficient, the above criteria no longer apply.
The Three Steps of Remove-Restore and the Two Restrictive Requirements Three steps: ① compute and remove the terrain effect of the source field quantity to obtain the residual value of the source field quantity; ② from the residual value of the source field quantity, compute the residual value of the target field element by the approximation method; ③ compute and restore the terrain effect of the target field element to obtain the approximate solution of the target field element.
Two restrictive requirements: ① the terrain mass model (including its adjustment method) and the integration radius used for computing the terrain effect of the source field quantity and that of the target field element must be exactly the same; ② the terrain effect algorithms for the source field quantity and the target field element must be fully analytically compatible (their analytical functional relation must be the same as the analytical relation between the field elements).
If either requirement is not satisfied, the gravity field approximation result will contain uncontrollable uncertainty, degrading the approximation performance and the quality control capability.
Role and performance-control principleKey points of the official wording
Cannot serve independently as an observation quantityA terrain effect of any kind on any type of field element cannot independently serve as an observation quantity for gravity field approximation — this is a restrictive requirement and must be strictly observed
Constructing densified observation quantitiesThrough the remove-restore method, using the current observations as constraints, the approximation method can improve the resolution or the observation type and construct densified observation quantities containing terrain ultra-short-wavelength information
Keeping the approximation observation model unchangedKeep the approximation (parameterized) observation model of the current observation quantities unchanged, supplement it with densified observation quantity models and reduce their sensitivity to the parameters, so as to achieve ultra-high-resolution approximation by least squares
Cumulative residual approximationThe cumulative residual gravity field approximation technique can be used to improve the approximation performance for the ultra-short-wavelength gravity field
Official special note: at a time when Earth gravity field exploration technology is developing rapidly and observation resources are becoming ever richer, “it is unscientific to overstate the role and position of terrain effects in gravity field approximation without thorough and meticulous testing and analysis and without concrete evidence”.

3.4Training Method 3: Build Awareness of Both “Analytical Compatibility” and “Spatial Representativeness Error”

Analytical compatibility (two classes)Content and examples
Class 1: theoretically binding requirementCompatibility between different types of observations/parameters and same-type geodynamic influence quantities. Example: the load effect on the height anomaly should equal the Hotine integral of the load effect on gravity disturbances; the solid Earth tidal effect on the normal height at a station should equal the difference between the solid Earth tidal effects on its ellipsoidal height and on the geoid
Class 2: determined by dynamicsCompatibility between same-type observations/parameters and different types of geodynamic influence quantities, determined by the geodynamic equations of a deforming Earth (including the constitutive equations)
Spatial Representativeness Error: Two Key Definitions ① The error of an observed field quantity is the error of the spatial average of that quantity corresponding to a given spatial resolution — the Spatial Representativeness Error — not the actual observation error at the measuring point; ② the accuracy of a target field element is its Spatial Representativeness Accuracy corresponding to the effective resolution, and depends on the approximation accuracy, the spatial resolution and the local gravity field structure at the grid cell (especially the short- and ultra-short-wavelength structure).
Quantitative example: when the gravity measurement accuracy at the measuring points is already better than 0.1 mGal while the Spatial Representativeness Error of 1′×1′ gravity disturbances is 1 mGal, the Spatial Representativeness Error is mainly determined by the density of measuring points, the local gravity field structure and the performance of the gridding algorithm — and further improving single-point measurement accuracy is then useless. In the era of terrestrial geodesy it was often assumed that the short-wavelength structure was directly related to topographic relief, so the Spatial Representativeness Error was also called the “terrain representativeness error”.
Error analysis and accuracy assessment schemeSteps
Spatial-domain integral method · Scheme I① Estimate the Spatial Representativeness Error of the integrand observation field quantity grid on the boundary surface (including errors introduced in obtaining the grid from discrete observations); ② derive the error propagation from the integral formula, neglect the reference model error, and estimate the Spatial Representativeness Accuracy of the target field element
Spatial-domain integral method · Scheme II① Construct an integral algorithm taking the target field element as the integrand quantity and the discrete measured field elements as the integrand values; ② compute the integral values of the observed field quantities at the measuring points, statistically analyse their differences from the measured values and obtain the Spatial Representativeness Error of the observed field quantities; ③ then estimate the accuracy of the target field element by Scheme I (taking the full-band measured field quantities as the reference truth; the reference model error generally need not be considered)
Spectral-domain least squares approximation① Using the spectral-domain coefficients obtained by approximation (geopotential coefficients or SRBF coefficients) and their series expansion, compute the estimated values of each observed field quantity at the measuring points, statistically analyse their differences from the measured values and obtain the Spatial Representativeness Error of each observed field quantity; ② derive the error propagation formula from the least squares approximation model; ③ compute the accuracy of all spectral-domain coefficients and the Spatial Representativeness Accuracy of the target field element
Official conclusion: gravity field approximation methods can effectively determine the Spatial Representativeness Error of observed gravity field quantities of any type from space, air, land and sea, but cannot determine the direct measurement accuracy specifications of measuring instruments; the decisive factor for the level of gravity field modeling can only be the Spatial Representativeness Error of the observed field quantities.
Additional Catalogue Entry: Accuracy Assessment of Geoid Results The official site also presents an assessment method based on GNSS/leveling height anomalies and the spectral error characteristics of the gravity field. Key points: the practical accuracy of ground height anomalies should be expressed by a height anomaly difference error curve that varies nonlinearly with distance; four elements — two error metrics plus two error curves — are recommended to express quasi-geoid accuracy completely; when the spacing of two GNSS/leveling points approaches or is smaller than the average spacing within the network, the GNSS/leveling height anomaly plays the main role in the practical ground height anomaly; at larger spatial scales, accuracy depends mainly on the control provided by gravimetric ground height anomalies.

3.5Training Method 4: Understand the Three Key SRBF Measures and Be Able to Assess Performance

Key measureProblem addressed and practice
① Edge effect suppression as an alternative to normal equation regularizationWhen an SRBF center (node) lies at the edge of the computation region, its SRBF coefficient is set to zero as an observation equation, improving the stability and reliability of coefficient estimation; once introduced, the normal equations no longer need regularization, thus preventing the analytical relations between field elements from being affected by observation errors
② Cumulative SRBF approximation for optimal field recoveryA target field element is the convolution of the observed field element with an SRBF filter; a single SRBF can hardly match the spectral centers and bandwidths of both the observed and the target field elements at the same time, which causes spectral leakage. This scheme combines multiple SRBFs with distinct spectral centers and bandwidths to fully resolve the spectral signal of the target field element and achieve optimal recovery of the target field element in the spatial domain; each stage of residual approximation essentially treats the previous result as the reference field and refines the residual target field element by the remove-restore principle
③ Normal equation normalization for fusing heterogeneous observation systemsBy normalizing the normal equations of different observation groups, the fusion of their covariance structures is controlled, decoupling the influence of the “observation system model” from that of “observation quality (errors, gross errors)”, so that the fusion is unaffected by observation errors, observation type or differences in point distribution. This is particularly advantageous for fusing data with extremely different spatial distributions (e.g. very few astronomical vertical deflections or GNSS/leveling points)
The Multi-Source Heterogeneous SRBF All-Element Gravity Field Approximation Program: Four Technical Features ① The observed field elements, the target field elements and their mutual relations have rigorous analytical relations, so the algorithm performance is unaffected by observation errors; ② it analytically fuses in one step heterogeneous, varying-altitude, cross-distributed, land-sea coexisting observations of many kinds, without reduction, continuation or gridding; ③ it simultaneously realizes all-element analytical modeling of the geoid and of the gravity field in full external space, and can effectively fuse very few astronomical vertical deflections or GNSS/leveling data; ④ it provides powerful gross error detection, external accuracy metrics, determination of height datum differences and computational performance control. The official site also calls it an extremely adaptable universal tool for all-element gravity field modeling and quality assessment.
How to Use SRBF Performance Assessment By inputting different single types of observed field elements, selecting different types of radial basis functions (spherical point-mass kernel, Poisson kernel, m-order radial multilevel subkernel, Poisson wavelet kernel, etc.) and their parameters, and computing different types of target field elements, the spatial-domain, spectral-domain and analytical properties of the radial basis functions and of the approximation algorithms can be tested comprehensively. An advanced use is to examine the spectral centers and bandwidths of the observed field elements, the radial basis functions and the target field elements and, on the principle of “fully resolving the spectral structure of the target field element”, to optimize the design of the best regional approximation scheme in the light of actual observation conditions. The program can also be used for extreme cases such as verifying the analytical fusion of very few high-precision astronomical vertical deflections or GNSS/leveling data, gross error detection and external accuracy assessment of multi-source, multi-generation aliased data, and boundary current detection and sea surface topography separation in shallow-water satellite altimetry.

3.6Training Method 5: Choose the Right Solution Method with the “Boundary Value Problem (BVP) Classification Logic”

ClassDefinition (classified by whether the boundary surface is an equipotential surface)Typical solutions
Stokes-type problemThe boundary surface is an equipotential surface and the boundary value is any linear combination of partial derivatives of the disturbing potential in the local horizontal systemRegularized Stokes integral, Hotine integral, Vening-Meinesz integral, radial gradient integral, etc.
Molodensky-type problemThe boundary surface is not an equipotential surface and the boundary value is any linear combination of partial derivatives of the disturbing potentialFour approaches — analytical continuation, correction, direct integral solution, and spectral-domain least squares approximation
Official recommendation: gravity field boundary value problems should be addressed mainly with Stokes-type solution methods; the Molodensky boundary value theory is used for the reduction of gravity field data to an equipotential surface and for the quantitative analysis of approximation errors in gravity field information introduced when processing data on non-equipotential surfaces. Also: residual field elements are insensitive to position, so an ellipsoidal height accuracy of no less than 10 m for representing the boundary surface position is sufficient.

3.7Capability Matrix and Self-Check

Capability areaCore capabilityCorresponding subsystems and resources
Gravity field modeling capabilityTerrain effect processing, spatial-domain boundary value integrals and spectral-domain SRBF approximation, all-element modeling, Spatial Representativeness Accuracy assessmentSubsystems ② and ③ of PAGravf4.5 + the corresponding algorithm and process documents
Datum and height capabilityGeoid refinement principles, GNSS/leveling fusion, stepwise control, uniqueness of geodetic datum constantsSubsystem ④ of PAGravf4.5 + the Chapter 6 courseware
Deformation monitoring capabilityTidal/polar shift corrections, non-tidal time series processing, load deformation field approximation, ground stability computationSubsystems ①②③④ of ETideLoad4.5 + the corresponding algorithm and process documents
√Self-check item
□Can perform an independent “computed values vs model values” test on an approximation program using an ultra-high-degree geopotential coefficient model
□Can state the four quantitative selection criteria for terrain effects and identify which two are technical requirements that must be observed
□Can restate the two restrictive requirements and explain the consequences of not satisfying them
□Can distinguish the “Spatial Representativeness Error” from the “actual observation error at measuring points” and give an example of accuracy saturation
□Can describe the steps of the three error analysis and accuracy assessment schemes
□Can explain what problem each of the three key SRBF measures solves
Chapter Summary The essence of capability building is turning “black-box calls” into “white-box understanding”: use ultra-high-degree models to health-check algorithms, use the quantitative selection criteria to ground parameter choices, use Analytical Compatibility and the Spatial Representativeness Error to bound credibility, and use the boundary value problem classification logic to choose the right solution method.
↑ Back to top
3+

Theoretical Threads: Three Underlying Logics Running Through the Platform

Corresponding official-site pages: h-col-105.html, h-nd-160.html, h-nd-161.html, and courseware PPT-03/PPT-04 (h-nd-168/h-nd-169).
Summed Up in Three Sentences ① Where does the reference system come from — it is determined directly by the Earth's gravity field; ② what is the datum constant of the height datum — the global geopotential W₀, uniquely invariant; ③ how are the moving parts classified and corrected — tidal effects can be modeled, while non-tidal effects require measurement and epoch reduction.

3.9Thread One: the Theory of Orientation and Positioning of the Terrestrial Reference Frame

ItemDilemma of the conventional International Terrestrial Reference System (ITRS)Original path of this platform
Orientation conditionThe temporal evolution of the ITRS orientation is constrained by the “No-Net-Rotation (NNR)” condition — an untestable artificial conventionCompletely discards geophysical conventions, the Earth's rotational motion and external agreements
Reference poleThe IERS reference pole (axis) is conventionally anchored near the mean rotation pole of 1900–1905, merely to preserve historical continuityThe second-degree, first-order terms of the geopotential uniquely determine the position of the figure pole (the direction of the polar axis of inertia)
OriginNo direct, rigorous scientific linkage has been established with the Earth's intrinsic physical properties (mass distribution, moments of inertia, Earth's gravity field), giving rise to fundamental problems of “conventionality” and “non-uniqueness”The first-degree terms of the geopotential uniquely determine the position of the Earth's center of mass in the Earth-fixed reference system
Result: the origin and the z-axis of the reference system can be determined directly and uniquely from the observable gravity field, yielding an Earth-fixed reference system that does not depend on any external convention, is independent of the Earth's rotational motion, and is defined entirely by geodetic theory itself.
Publication-Level Result Derives a theoretical method for the positioning of the Earth's center of mass and the figure pole through the synergy of spatial geometric and physical geodesy: it can precisely monitor the time series of Earth center of mass variations and figure pole shifts, and can position and orient the current Earth-fixed reference system to the mean center of mass and the mean figure pole of a specified period. It also provides monitoring models and algorithms for the “joint determination of the Earth's center of mass, the figure pole and various Earth rotation dynamic parameters” through the synergy of satellite observations, VLBI kinematic measurements, and radial displacements and gravity variations at ground stations.

3.10Thread Two: the Theoretical Foundation of the Height Datum Gravity Field Unified in the Earth-Fixed Reference System

Traditional deficiencyNature of the problem
Classical Helmert orthometric heightIts definition relies on the assumption of an average crustal density between the ground point and the geoid; because crustal density is unknown and spatially heterogeneous, this directly renders the Helmert orthometric height “non-unique” at the centimeter level
The so-called “true geoid”A modified equipotential surface obtained by adjusting topographic masses into the interior of the geoid; it depends severely on the specific mass adjustment method, and different methods yield discrepancies of up to the decimetre level, so it lacks both uniqueness and measurability
Quasi-geoidObtained indirectly from height anomalies at the Earth's surface; because the actual heights of measuring points above the ground have random uncertainties and the digital elevation models used to represent the terrain are inherently diverse, the quasi-geoid cannot be uniquely defined
Six Main Conclusions ① Orthometric (and normal) heights are defined by gravity field theory in the Earth-fixed reference system and possess uniqueness and precise measurability, but their physical meaning (the geopotential number) is approximate, while the geometric elevation has a typically local character;
② a theoretical method for the positioning of the Earth's center of mass and the figure pole through the synergy of spatial geometric and physical geodesy is derived; it does not rely on geophysical assumptions or dynamic protocols, and realizes reference-system positioning and orientation entirely through geodetic theory;
③ a method of determining the global geopotential W₀ that is analytically compatible with the gravimetric geoid is recommended, resolving the problem of consistency between the Gaussian geoid and the gravimetric geoid; it is also demonstrated that the normal gravity field has only 3 degrees of freedom and is likewise a unique and precisely measurable element of physical geodesy;
④ the concept of the height starting datum surface is updated: after neglecting the geometric deformation of the geoid, the height starting surface is the (global or regional) geoid for the orthometric height, the normal height and the geopotential number systems alike; and it is demonstrated that at decimetre-level accuracy requirements the quasi-geoid should not be used as the starting surface of normal heights;
⑤ the theoretical foundation of the height datum gravity field unified in the Earth-fixed reference system is perfected, the datum conditions and implementation principles for GNSS Replacing Leveling are derived, and the Analytical Orthometric Height is recommended as more suitable for height datum purposes than other types of orthometric height;
⑥ surfaces of constant orthometric height are parallel to the geoid, and the orthometric height is the height of the straight-line distance from a ground point to the geoid along the vertical direction — the customary notion that “the orthometric height is the length of an irregular curve along the plumb line” has no geodetic basis and is incorrect.

3.11Thread Three: Classification and Dynamic Framework of Tidal and Non-Tidal Deformation of the Solid Earth

ConceptKey points of the official wordingSource
Tidal deformation and tidal effectsPeriodic deformations of the solid Earth and associated changes in the Earth's gravity field, excited by the celestial tide-generating potential, ocean tides and atmospheric tides, are collectively termed the tidal deformation of the solid Earth; the corresponding periodic changes in geodetic elements are termed the tidal effects on those elements, comprising solid Earth tidal effects and tidal load effectsh-col-105
Can be modeledGeodetic tidal effects can be modeled and accurately removed or restored at any time and place; a tidal effect equals the negative value of its tidal correctionh-col-105
Non-tidal load deformationNon-tidal variations in surface-system environmental loads (soil and vegetation moisture, river-lake-reservoir water, glaciers-ice sheets-snow mountains, groundwater, atmosphere, sea level, etc.) cause geopotential variations and simultaneously excite solid Earth deformation, manifesting as ground vertical deformation, ground gravity and ground tilt variationsh-col-105 · Tide.html
Plastic isostatic vertical deformationGroundwater use, underground resource extraction, engineering construction, glacier melting, etc. cause shallow rock and soil layers to lose their original equilibrium and to tend slowly toward a new equilibrium state under self-gravity or internal stress; it is plastic deformation, with the dynamic action located in the subsurface rock and soilTide.html
Stationary / dynamic frameA frame corrected only for tidal effects and neglecting non-tidal effects remains a stationary (time-invariant) frame (e.g. a precision leveling network or gravimetric control network corrected for tidal effects); a frame that accounts for non-tidal effects can only be a dynamic frame, whose reference value corresponds to a unique reference epoch: reference value at the current epoch = reference value at the reference epoch + (non-tidal effect) epoch reductionh-col-105

3.12Common Tone of the Three Threads: Critical Inheritance and Strategic Simplification

Simplification propositionBasis and purpose
The “spherical approximation” error can be completely avoidedBy rigorously distinguishing the Earth-fixed coordinate system from the local coordinate system, it is demonstrated that the spherical approximation error inherent in traditional gravity field approximation theory can be completely avoided, which both clarifies a misconception and avoids introducing the mathematical complexity associated with ellipsoidal geodesy
The Stokes boundary value theory is sufficient for external approximationBy introducing the Analytical Geoid, it is demonstrated that the Stokes boundary value theory is sufficient to solve the external gravity field approximation problem, thereby comprehensively de-emphasizing the Molodensky boundary value theory and its complex conceptual framework (the telluroid, the quasi-geoid, etc.)
Unified treatment of terrain effects, tidal effects and load effectsKey point of Chapter 3 of the courseware: the terrain effect, tidal effect and load effect on Earth gravity field quantities refer specifically to the contributions to the gravity field from terrain masses, celestial tide-generating forces and ocean-atmosphere tidal loading, and internal mass adjustments of the Earth (non-tidal variations of atmosphere, ocean and terrestrial water, etc.) respectively; they therefore coexist in all gravity field quantities in Earth space. Using the same standards and analytically compatible algorithms to treat uniformly the terrain effects and the tidal and load deformation effects on all geodetic and gravity field quantities in space, air, land and sea is a binding principle and restrictive requirement of physical geodesy (h-nd-168)
All of these aim to “reduce unnecessary complexity in theoretical methods” and to steer the discipline back to a focus on the measurability and the mathematical-physical essence of the gravity field itself.
Section Summary The three threads answer respectively “where does the reference system come from”, “what is the datum constant of the height datum”, and “how are the moving parts classified and corrected”; together they support the steady-state and time-variable branches of the platform, and they are also the theoretical premise for understanding the later “mathematical isomorphism” and the principles for deep fusion of multi-source heterogeneous data.
↑ Back to top
4

Problem Solving: A Solution Library for Seven Typical Problems

Corresponding official-site pages: Geoid.html, Tide.html, h-nd-109/110/111, h-nd-87, h-nd-104/105.
General Routine Decide steady-state or time-variable first → then the system → then the subsystem → finally apply the process and the quality control; when unsure, consult the quick-reference table in 4.8 first and then self-check with the pitfall list in 4.9.

4.1Terrain Effect Processing and Multi-Source Heterogeneous Gravity Exploration Modeling

ItemContent
Problem characteristicsLand, marine, airborne and satellite gravity / gravity gradient / vertical deflection / GNSS-leveling data at different altitudes, cross-distributed and of differing accuracy must be fused into a unified refined gravity field or gravity exploration model
Subsystems involvedSubsystem ② of PAGravf4.5 (Computation of Diverse Terrain Effects on Various Gravity Field Elements) + subsystem ③ (High-Precision Gravity Field Approximation and Full-Element Modeling)
Recommended process — the official four-step method① Determine the computation scope, the computation surface (a terrain equi-elevation surface is recommended) and the gravity exploration model type, and obtain the data → ② use the approximation programs to precisely determine the high-resolution grid model of the target gravity field element on the computation surface → ③ precisely compute the high-resolution grid model of the terrain effect consistent with the model type → ④ subtract the terrain effect grid model from the target gravity field element grid model
OutputsLand-sea unified Complete Bouguer gravity anomalies / Complete Bouguer gravity disturbances / Complete Bouguer vertical deflections / Complete Bouguer gravity gradients, as well as Classical Bouguer and Isostatic anomalies (disturbances)
Key control pointsThe terrain effect type is selected by testing against the quantitative selection criteria; the terrain mass model (including its adjustment method) and the integration radius used for removal and restoration must be exactly the same; the terrain effects of the source field quantities and of the target field elements must be fully analytically compatible
Technical highlightTaking the analytical functional relations between field elements as a strong constraint, the computation process contains no gravity reduction, continuation or gridding operations, which avoids the problems of signal attenuation, non-analytical distortion and difficult control of terrain effects inherent in conventional methods
Official demonstration pages/h-nd-111.html (PAGravf4.5 terrain effect exercise demonstration) · /h-nd-100.html (PAGravf4.5 terrain effect analytical algorithm case)

4.2Regional Gravity Field and 1 cm Geoid Refinement

ItemContent
Problem characteristicsConstructing a high-precision gravity field and geoid model for a province, city or engineering area, or improving the accuracy of an existing model
Subsystems involvedSubsystems ② and ③ of PAGravf4.5 (terrain effects + approximation modeling), together with subsystem ④ for datum applications
Recommended process① Select a suitable global geopotential coefficient model as the reference gravity field and as a constraint on the regional outer boundary → ② integrate observation gravity field resources in and around the region → ③ carry out fused modeling by spatial-domain boundary value integrals or spectral-domain SRBF approximation → ④ cross-check with multi-type field elements such as height anomalies / gravity disturbances / vertical deflections / gravity gradients
Key control point — stepwise controlThe geoidal height at any point can only be determined precisely with observation data covering the globe, and regional observations cannot fully contain the medium- and long-wavelength information; moreover, the longer the wavelength of a field element determined by approximation, the higher its accuracy generally is. Regional refinement should therefore use a suitable global model as the reference field and as the regional outer boundary to control medium- and long-wavelength errors; for local areas such as a province or a city, a regional gravity field model should serve as the reference field to control the accumulation of medium- and short-wavelength errors and to improve short- and ultra-short-wavelength accuracy
GNSS/leveling fusionLeveling is essentially the measurement of the geopotential difference between two points, and its relative error is in principle independent of the length of the leveling line (over long distances its relative accuracy is actually lower than that of a gravimetric level surface); the fusion algorithm should integrate the high-precision medium- and long-wavelength components of the gravimetric height anomaly with the high-precision short- and ultra-short-wavelength components of the GNSS/leveling control network, while suppressing short-wavelength gravity errors and controlling the accumulation of long-distance GNSS/leveling errors
Empirical referenceIn the 2011 coastal zone project, a remove-restore technique combining Local Terrain Effects with a reference gravity field precisely determined the 2.5′×2.5′ land-sea unified ground gravity disturbances, vertical deflections, gravimetric quasi-geoid and sea surface topography of China, bringing the gravimetric geoid of China's coastal zone to centimeter level for the first time. Checked by the National Quality Inspection and Testing Center for Surveying and Mapping Products with more than 280 GNSS/leveling points, the root-mean-square error of the land-sea gravimetric quasi-geoid was 3.9 cm; an additional external check against pre-2010 GNSS/leveling results in 12 coastal provinces and municipalities showed that, apart from subsidence of several centimetres in Tianjin and Shanghai, the maximum difference among the mean residuals of the provinces and municipalities did not exceed 3 cm, and the national 1985 height datum propagated to Hainan Island showed no significant difference at the 1 cm level

4.3Regional Height Datum Optimization, Unification and GNSS Height Conversion

ItemContent
Problem characteristicsUnifying a regional height datum, converting GNSS ellipsoidal heights into normal or orthometric heights, or handling height connections across regions and across the sea
Subsystems involvedSubsystem ④ of PAGravf4.5 (Optimization, Unification, and Application of Regional Height Datums)
Recommended process① Establish that the global geopotential W₀ is the unique, invariant datum constant of the height starting surface → ② determine the Analytical Geoid model from the external gravity field → ③ following the principle of “incorporating it into the Earth-fixed reference system and reducing it to the global height datum”, let GNSS/leveling height anomalies participate in modeling as ordinary observed field quantities → ④ use the GNSS/leveling control network for short-wavelength control and external checking
Key control points — three error-prone spots① The platform de-emphasizes the concept of the quasi-geoid and does not treat it as the starting surface of normal heights, and height anomalies are strictly in one-to-one correspondence with their spatial positions; ② actual measuring points often do not lie on the particular digital elevation model surface used to generate the height anomaly model, so centimetre-level applications require an additional height-dependent height anomaly gradient (or gravity disturbance) correction; ③ for the orthometric height, the normal height and the geopotential number systems alike, the height starting surface can only be the geoid
Further reading on the official site/h-nd-161.html (theoretical foundation of the height datum gravity field unified in the Earth-fixed reference system and updating of classical concepts) · Chapter 6 of the courseware, Height datums and their gravity field theory (/h-nd-171.html)

4.4Precise Removal of Tidal and Polar Shift Effects

ItemContent
Problem characteristicsGNSS, leveling and gravity observations contain solid Earth tidal effects, ocean tidal load effects, atmosphere tidal load effects and polar shift effects, which must be precisely corrected according to uniform standards, or epoch-reduced in datum maintenance
Subsystems involvedSubsystem ① of ETideLoad4.5 (Computation of Various Tidal and Polar Shift Effects on All Geodetic Elements) + subsystem ② (Processing and Analysis of Non-Tidal Geodetic Variation Time Series)
Recommended process① Uniformly compute the solid Earth tidal and tidal load effects on each element according to consistent numerical standards → ② compute rotation polar shift effects and tidal effects on Earth rotation parameters, and Earth's center of mass variation and figure pole shift effects → ③ apply corrections to the time series observations → ④ when necessary, perform non-tidal effect epoch reduction on the dynamic reference frame
Key control points① Tidal effects can be modeled and accurately removed or restored at any time and place, and equal the negative value of their tidal corrections; ② non-tidal effects are difficult to model, are generally measured by geodetic techniques, and in rapid or real-time applications the measured time series can be extrapolated by short-term prediction; ③ a frame corrected only for tidal effects and neglecting non-tidal effects is still a stationary frame — the reference value of a dynamic frame must correspond to a unique reference epoch
Official demonstration pages/h-nd-101.html (ETideLoad4.5 tidal and polar shift computation demonstration) · the algorithm documents “Unified computation of solid Earth tidal effects on all elements on the ground and external to the solid Earth” and “Computation of rotation polar shift effects and tidal effects on Earth rotation parameters”

4.5Load Deformation Field and Terrestrial Water / Time-Variable Gravity Field Monitoring

ItemContent
Problem characteristicsInverting ground vertical deformation, ground gravity and the time-variable gravity field from surface hydrological, meteorological and other environmental load data or from various monitoring time series, and obtaining high-resolution grid time series
Subsystems involvedSubsystems ② and ③ of ETideLoad4.5 (Processing and Analysis of Non-Tidal Geodetic Variation Time Series + Load Deformation Field Approximation and Monitoring from Heterogeneous Variations)
Recommended process A — four-step methodTaking regional terrestrial soil moisture as an example, a remove-restore scheme combining a load spherical harmonic coefficient model with regional load SRBF approximation: global surface environmental load spherical harmonic analysis → spherical harmonic synthesis of the load deformation field → spectral-domain SRBF analysis of the regional residual surface environmental load → SRBF synthesis of the residual load deformation field
Recommended process B — five-step methodJointly using the CORS network, the solid Earth tide station network and the groundwater monitoring network, realizing collaborative monitoring across techniques of the all-element grid time series of the groundwater and surface environmental load deformation field by the known surface load remove-restore method
Monitoring quantities that can be fused① (GNSS/leveling points) height anomaly variation, mm; ② (mobile gravimetry co-located with GNSS or solid Earth tide CORS stations) gravity disturbance variation, μGal; ③ (mobile gravimetry or solid Earth tide stations) ground gravity variation, μGal; ④ (CORS stations or mobile GNSS points) ellipsoidal height variation, mm; ⑤ (mobile leveling points) normal height variation, mm; ⑥ (hydrological monitoring stations) equivalent water height variation, cm
Key control pointsAll monitoring quantities entering the computation should first use the same surface load model and analytically compatible non-tidal effect (epoch reduction) algorithms, a unified monitoring datum and reference epoch, and the long-wavelength part of the load deformation field should be removed to satisfy the conditions for local SRBF approximation; global and regional soil moisture time series must each have their annual mean subtracted to unify the monitoring datum; the parameters of groundwater monitoring stations must first be calibrated with the regional groundwater EWH weekly-variation grid time series computed from CORS/solid Earth tide data
Official demonstration pages/h-nd-108.html (high-resolution load deformation field approximation demonstration) · /h-nd-109.html (SRBF all-element modeling process) · /h-nd-110.html (deformation field collaborative monitoring process) · /h-nd-104.html (load deformation field approximation and collaborative monitoring)

4.6CORS/InSAR Fusion, Ground Stability and Numerical Prediction of Geological Hazard Risk

ItemContent
Problem characteristicsContinuous tracking of ground stability variations at regional scale, assessment of the capability to capture geological hazard precursors, and numerical prediction of risk
Subsystems involvedSubsystem ④ of ETideLoad4.5 (CORS/InSAR Collaborative Monitoring and Ground Stability Estimation) + subsystems ② and ③
Recommended process① With the CORS network as control, constrain and assimilate hydrological and meteorological monitoring according to deformation dynamics and gravity field laws → ② compute the regional grid time series of ground vertical deformation, ground gravity, ground tilt and ground stability variations → ③ compute the regional overall numerical model of ground stability and assess susceptibility → ④ take the “moment of accelerated reduction in ground stability” as the moment when a hazard precursor appears, and carry out numerical prediction of risk
Principle of CORS-network time-series InSAR collaborative monitoring of vertical deformation ① Through gross error detection, spatial filtering and time-series analysis, the time-series InSAR monitoring quantities are separated into two parts — the vertical variations of the ground surface and of the soil within a few metres of the surface (expansion and contraction caused by temperature and precipitation), and the vertical deformation of the rock and soil layers at depths of several metres; only the latter is compatible with other geodetic monitoring; ② the ellipsoidal height variation time series of the CORS network constrain the multi-source time-series InSAR monitoring quantities and control the precise separation of the vertical deformation signal; ③ the vertical variations of the ground surface and shallow soil separated from the time-series InSAR monitoring quantities can be used for soil, wetland and water environment monitoring, geological environment survey monitoring, disaster monitoring and post-disaster assessment, and hydro-meteorological research
Quantitative criteria for ground stability degradationFrom the grid time series of ground vertical deformation, gravity and tilt variations fused from multi-source heterogeneous data, deterministic objective regularities are constructed: increase in ground ellipsoidal height / decrease in gravity; large horizontal gradients of ellipsoidal height or gravity variations; large inner product of tilt variation and terrain slope, etc.; by further referring to hazard events that have occurred, a ground stability variation monitoring capability adapted to the local environmental geology is established
Empirical referenceFor Lishui–Wenzhou (38 CORS stations, 39 meteorological stations) the rate of advance capture of geological hazard precursors reached 92.5%. For Baoshan–Dali (72 CORS stations, 44 groundwater stations and 41 hydrological stations over an area of about 60,000 km²), precursors appeared more than 2 days in advance at 186 of 197 hazard sites, an agreement rate of 94.4%; for the 90 events in 2020, the prediction rates at 1 week / two weeks / 1 month were 92.2% / 83.3% / 75.6% respectively. A comparison of the overall ground stability model with potential hazard sites as of December 2016 showed that baseline stability was poor at 83% of the sites, and 37 new potentially susceptible zones were interpreted
Official pages/Tide.html (seamless continuous quantitative monitoring scheme of regional ground stability variations) · /h-nd-87.html (CORS network ground stability variation monitoring and prediction)
Basis for Judging the Nature of Deformation: Spatiotemporal Dynamic Characteristics of the Three Types of Ground Vertical Deformation
TypeCauseSpatiotemporal quantitative characteristics
Elastic load-induced vertical deformationExcited by mass variations of surface environmental loads and characterized by load Love numbersThe time-varying characteristics resemble those of load variations (nonlinear, quasi-periodic) and are synchronous in time
Plastic isostatic vertical deformationInduced by human activity or the natural environment; the dynamic action lies in the subsurface rock and soil and slowly approaches a new equilibriumThe spatial influence angle is about 45° (influence radius ≈ burial depth); the duration is approximately proportional to the burial depth; on a scale of several years the deformation quantity and the acceleration rate have opposite signs
Local tectonic vertical deformationDriven by horizontal plate motion and acting only near compressional fault zonesDecays rapidly to zero with distance from the fault zone, with an influence radius comparable to the tectonic depth; on a centennial scale the rate remains essentially constant

4.7Data Editing, Gridding and Accuracy Assessment (General Support)

ItemContent
Problem characteristicsObservation data have inconsistent formats, uneven distribution and gross errors, requiring preprocessing, gridding, visualization and accuracy assessment
Subsystems involvedSubsystem ⑤ shared by both systems (editing, calculation and visualization of geodetic data), together with the quality control functions of each system
Common capabilitiesCoordinate and time system transformation, interpolation of irregularly sampled data, weighted basis-function gridding of multi-source heterogeneous unevenly distributed data, low-pass filtering of digital grids, estimation of first- and second-order horizontal gradients of grid values, quasi-stable adjustment, and quantitative computation of ground stability variations from vertical deformation grids; both systems provide gross error detection and determination of external accuracy metrics
Key control points① Before gridding, clarify the target resolution and the Spatial Representativeness Error definition; ② gross error detection is best carried out within the spectral-domain least squares framework by setting the corresponding observation coefficient weights to 0; ③ accuracy assessment should preferably use the algorithm chain consistent with the final product

4.8Problem–Subsystem Quick-Reference Table

Your problemSystemPreferred subsystem
Observation data are in a chaotic format and require preprocessing and griddingPAGravf4.5① Analysis and Preprocessing of Earth Gravity Field Data
Unsure which terrain effect to choose or whether its treatment is advantageousPAGravf4.5② Computation of Diverse Terrain Effects on Various Gravity Field Elements
Modeling with aliased multi-source heterogeneous data, gross error detection, accuracy assessmentPAGravf4.5③ High-Precision Gravity Field Approximation and Full-Element Modeling
Unifying height datums, converting GNSS ellipsoidal heights to normal heightsPAGravf4.5④ Optimization, Unification, and Application of Regional Height Datums
Solid Earth tide / ocean tide / atmospheric tide / polar shift correctionsETideLoad4.5① Computation of Various Tidal and Polar Shift Effects on All Geodetic Elements
Analysis of the non-tidal components of GNSS and other time seriesETideLoad4.5② Processing and Analysis of Non-Tidal Geodetic Variation Time Series
Inversion of the load deformation field / time-variable gravity field / terrestrial water storageETideLoad4.5③ Load Deformation Field Approximation and Monitoring from Heterogeneous Variations
Ground stability tracking, numerical prediction of geological hazard riskETideLoad4.5④ CORS/InSAR Collaborative Monitoring and Ground Stability Estimation
Batch editing, plotting and visualization of result filesBoth⑤ Editing, Calculation and Visualization of Geodetic Data

4.9Common Pitfalls and Troubleshooting List

Misconception / symptomCause and correction
“Treating terrain effects as a universal tool and removing them whenever data appear”The treatment of terrain effects has exactly two purposes; moreover, a terrain effect of any kind on any type of field element cannot independently serve as an observation quantity for gravity field approximation
“Using different terrain models or integration radii for removal and restoration”This violates the two restrictive requirements and the result contains uncontrollable uncertainty; the terrain mass model (including its adjustment method) and the integration radius must be exactly the same, and the algorithms must be fully analytically compatible
“Improving single-point observation accuracy will necessarily improve grid accuracy”This is constrained by the Spatial Representativeness Error and is ineffective once the density of measuring points and the local gravity field structure become the bottleneck
“Taking the quasi-geoid as the starting surface of normal heights”This contradicts the definition of normal height; the platform does not regard the quasi-geoid as the starting surface of normal heights, and height anomalies are strictly in one-to-one correspondence with spatial positions
“Once the tidal corrections are done, the frame is dynamic”A frame with only tidal corrections is still a stationary frame; a dynamic frame additionally requires non-tidal effects and epoch reduction
“Directly treating time-series InSAR deformation values as ground subsidence”The variations of the ground surface and of the soil within a few metres of the surface must first be separated from the vertical deformation of the rock and soil layers at depths of several metres; only the latter is compatible with other geodetic monitoring
“Example results are the standard answers”The case studies are intended primarily to illustrate computation processes and do not optimize the algorithms or parameters; their statistical results reflect only basic performance
Chapter Summary Decide steady-state or time-variable first, then the system, then the subsystem, and finally apply the process and the quality control; add a pitfall list for self-checking, and the vast majority of tasks will quickly find their entry point and avoid the pitfalls.
↑ Back to top
5

Integrated Application: Unified Models, Analytical Compatibility and Multi-Source Fusion

Corresponding official-site pages: Tide.html (basic principles and coordination schemes), h-col-104.html (formats and resources), Home (the five typical applications), h-nd-87.html.
Key Points of This Chapter Integrated application = unified models and numerical standards (premise) + Analytical Compatibility (bottom line) + deep fusion of multi-source heterogeneous data (means) + coordination of the two systems (path) → Earth system analysis (goal).

5.1Why Integration Is Essential

Driving levelReason
Physical necessityThe Earth is deforming at every moment, and the steady state is only a special case of the time-variable; the height and gravity of the same measuring point require both a steady-state datum and time-variable corrections
Technical necessityCORS, GNSS/leveling, gravity, time-series InSAR and hydro-meteorological observations each have their strengths; no single observation type can satisfy both spatiotemporal resolution and the diversity of elements
Methodological necessity“First establish the datum and the medium- and long-wavelength skeleton with the steady state, then characterise the short wavelengths and dynamics with the time-variable” — neither is dispensable

5.2Methodological Bottom Line: Consistent Models, Identical Standards, Compatible Algorithms, a Unified Epoch

Technical Premise of Integration Adopt consistent geophysical models and identical numerical standards, construct compatible geodetic and geodynamic algorithms, and unify the spatiotemporal monitoring datum and the reference epoch — thereby creating the mathematical-physical foundation for collaborative monitoring across multiple geodetic techniques in space, air, land and sea. The official site also points out that adopting consistent geophysical models, identical numerical standards and compatible analytical algorithms, and treating uniformly the ubiquitous non-tidal geodetic effects, is an important basis for achieving geodesy at the 1 cm (20 μGal) precision level, and a necessary condition for the deep fusion of multi-source heterogeneous data, collaborative monitoring across heterogeneous techniques, and the construction and maintenance of high-precision geometric and physical geodetic datums.

In other words, integration is not about “stitching together” the results of different software; it is about organizing the computation from the very beginning with the same coordinates, the same geodetic datum constants and the same analytical relations. The two systems of the platform come from the same team and the same theoretical framework and naturally share the geophysical models and numerical standards; and “Analytical Compatibility” is the lowest criterion for testing whether an integration is correct.

5.3Deep Fusion of Multi-Source Heterogeneous Data: the Official Five Principles

No.Principle Implementation
1Construct scientifically consistent geophysical models, rigorously uniform numerical standards and analytically compatible geodetic and geodynamic algorithms, and unify the spatiotemporal monitoring datum and the reference epochUnify the coordinate frame / reference epoch, the geodetic datum constants and the correction standards (tidal effects, load effects, polar shift effects); “align” before computing
2On the principles of geophysical geodesy, deeply fuse or constrain-assimilate multi-source heterogeneous data and reconstruct the geodetic and geodynamic spatiotemporal correlations among the various monitoring quantitiesUse physical relations (the load–deformation–gravity analytical relations) as constraints, not purely statistical interpolation
3For time series of same-type multi-source heterogeneous monitoring quantities, achieve deep epoch-by-epoch fusion through fundamental geodetic constraints or joint measurement adjustment (with additional monitoring datum parameters as required)Different observations of the same physical quantity (e.g. CORS and GNSS/leveling) are fused epoch by epoch by adjustment
4For time series of multiple monitoring quantities of different types, achieve deep analytical fusion through constraints from physical geodesy, solid Earth geophysics or environmental geodynamics (with additional dynamic parameters as required)Different physical quantities (deformation, gravity, hydrology) are fused under dynamic equation constraints
5Through fusion and coordination, simultaneously improve the spatiotemporal resolution of monitoring and enrich the variety of monitored elements, revealing the geophysical structure and dynamic characteristics of the monitored objectsThe acceptance goal is “more complete elements, higher resolution, clearer mechanisms”

5.4Theoretical Pivot of Coordination: the “Mathematical Isomorphism” of the Load Deformation Field and the Earth's Gravity Field

Earth gravity field (steady state)Load deformation field / time-variable gravity field (time-variable)
The analytical relations between Earth gravity field parametersThe analytical relations between the load effects on Earth gravity field parameters are entirely consistent with the analytical relations between the gravity field parameters themselves
The global gravity field can be represented by a geopotential coefficient model (GCM); given that model, various gravity field parameters at any point external to the Earth can be computed; the geoid can be computed from gravity data by the Stokes integralThe global load deformation field and the time-variable gravity field can be represented by a Global Surface Load Spherical Harmonic Coefficient Model; given that model, the load effects on various geodetic parameters external to the solid Earth can be computed; surface environmental load effects can be computed from their load equivalent water height by the Green's function integral
The Earth gravity field approximation theory is linear and can be approximated by the remove-restore scheme and the cumulative iteration methodThe elastic (linear viscoelastic) load deformation theory is also linear and can likewise be refined by the remove-restore scheme and cumulative approximation
The methods are summarized as spatial-domain gravity field boundary value integral methods and spectral-domain spherical basis function (surface spherical function, radial basis function, spline function, etc.) approximation methodsFor load deformation field (time-variable gravity field) approximation or monitoring there are likewise two methods: the load Green's function integral constraint and spectral-domain spherical basis function approximation
An easy-to-remember corollary: the sum of the deformation effects of various environmental loads (atmospheric pressure, terrestrial water, sea level variations, etc.) equals the deformation effect of the sum of those loads.

5.5Interface Foundation of Integration: Five Classes of Time-Series Data File Formats and Monitoring Quantity Conventions

FormatContent stored and example
① Ground monitoring quantity time seriesStores the time series of several attribute monitoring quantities at a station, baseline or route, with identical sampling epochs for all attributes (e.g. CORS coordinate solutions, solid Earth tide station observations and analysis results, GNSS baseline solution time series)
② Monitoring network station record time seriesStores the time series of one type of monitoring quantity at multiple ground stations of a monitoring network (e.g. CORS network coordinate solutions (one-dimensional), leveling network point heights, solid Earth tide station networks, tide-gauge networks, time-series InSAR monitoring quantities)
③ Monitoring network observation record time seriesStores time series of one-dimensional baseline solutions of the CORS (GNSS) network, leveling line height differences of the leveling control network, and gravity segment differences of the gravimetric control network
④ Digital model (vector) grid time seriesConsists of grid files of one type of monitoring quantity (vector) with epoch information; the seventh attribute in the grid header file is the sampling epoch of the grid (e.g. terrestrial water EWH, sea level variations, various ground load deformation field grids, time-variable gravity field grid time series)
⑤ Spherical harmonic coefficient model time seriesConsists of spherical harmonic coefficient (Stokes coefficient) model files with time information; the file header contains at least three attributes — GM (×10¹⁴ m³/s²), the ellipsoid semi-major axis a (m) and the sampling epoch — where GM and a are called the scale parameters
ConventionContent
Definition of monitoring quantitiesA geodetic monitoring quantity = the geodetic observation or parameter at the current epoch minus its mean over a period (or its value at a certain reference epoch)
Unit conventions (examples)Height anomaly / geoid variation, mm; ground / gravity disturbance variation, μGal; ground tilt / vertical deflection variation, mas; ground horizontal displacement, mm; ellipsoidal height variation, mm; orthometric (normal) height variation, mm; gravity disturbance gradient variation, 10 μE; terrestrial water EWH variation, cm; sea level variation, cm; ocean tidal height, cm; atmospheric pressure variation, hPa
Vector-direction conventions (examples)Ground tilt / vertical deflection variation, SW: the first component points south and the second west, forming a right-handed system with the gravity direction (natural coordinate system); station horizontal displacement, EN: the first component points east and the second north, forming a right-handed system with radial displacement (east-north-up coordinate system); horizontal gravity gradient variation, NW: the first component points north and the second west; tidal constituent harmonic constants: the first component is the in-phase amplitude (cosine) and the second the quadrature amplitude (sine)
Recommendation: before a project starts, assign each type of monitoring data to its correct format among the five classes and unify the epoch and units; only when this step is solid can all the subsequent fusion algorithms be correct.

5.6Paradigms of Typical Integrated Application: the Official Five Typical Applications

Typical application Data combinationCollaborative outputs and results
Seamless continuous monitoring of regional terrestrial water storage, ground deformation field and gravity field variations — Three Gorges area, 2015–201626 CORS stations, 8 gravity stations and 1 ground tilt station in and around the Three Gorges area, together with daily means from river hydrological and meteorological stations and observations from various satellite altimetry, Ziyuan-3 satellite remote sensing and GRACE satellite gravityMonthly grid time series of 2′×2′ ground vertical deformation, ground gravity, ground tilt and groundwater variations; monitoring accuracy 5 mm for ground vertical deformation and 10 μGal for ground gravity variations; annual variation amplitudes in the Three Gorges area of 36 mm for ground vertical deformation, 28 mm for the geoid and 117 μGal for ground gravity; after the impoundment of the Yangtze River the groundwater variation in the reservoir area increased by nearly 0.5 m³ per square metre, with the affected range extending more than 150 km from the Yangtze centreline
Assessment of ground stability variations and of geological hazard monitoring and early-warning capability — Lishui–Wenzhou, 2017–2018Continuous GNSS observations from 2015–2017 at 38 CORS stations in and around Lishui–Wenzhou, Zhejiang Province, together with daily mean atmospheric pressure data from 39 meteorological stationsMonthly grid time series of 1′×1′ ground vertical deformation, ground gravity and ground stability; quantitative tracking of ground stability variations and assessment of the geological hazard precursor capture capability of the CORS network; advance capture rate of geological hazard precursors up to 92.5%
Refinement of the coastal-zone gravity field and geoid — 2011Terrestrial / airborne / shipborne / satellite gravity + various satellite altimetry2.5′×2.5′ land-sea unified ground gravity disturbances, vertical deflections, gravimetric quasi-geoid and sea surface topography of China; the coastal-zone gravimetric geoid reached centimeter level for the first time, with a root-mean-square error of 3.9 cm checked against more than 280 GNSS/leveling points
Ground subsidence and height datum monitoring and maintenance — Shandong 2016 / Hanzhong, Shaanxi 2017 / Tianjin 2018 / Taizhou, Zhejiang 2019Ground subsidence monitoring with the national GNSS reference station network, and ground subsidence and regional height datum stability monitoring with the CORS network; Taizhou used the fusion of time-series InSAR and the CORS networkA demonstration of ground subsidence and regional height datum stability monitoring, regional time-variable gravity field and groundwater storage monitoring, and vertical datum monitoring and maintenance. Conclusions: CORS stations can continuously monitor ground vertical deformation at millimetre-level accuracy and can monitor small variations in the regional gravity field at higher spatiotemporal resolution; after unifying the spatiotemporal monitoring datum of time-series InSAR and CORS, atmospheric delays can be corrected and the vertical deformation effect of tidal loading compensated
Regional coordinated monitoring of geological hazards and numerical risk prediction — Baoshan–Dali, 202172 CORS stations, 44 groundwater monitoring stations and 41 hydrological stations, together with soil moisture and meteorological data (1 January 2018 to 31 December 2020)A grid time series model of regional ground stability variations and numerical risk prediction at 1 km / 1 week; precursors appeared more than 2 days in advance at 186 of 197 hazard sites (agreement rate 94.4%); for the 90 events in 2020 the prediction rates at 1 week / two weeks / 1 month were 92.2% / 83.3% / 75.6%; 37 new potentially susceptible zones were identified
Titles are taken verbatim from the “Typical Applications” section of the official-site home page.

5.7General Workflow for Organizing an Integrated Task

①Fix the datum: use PAGravf4.5 to settle the steady-state skeleton first — the reference gravity field, the regional gravity field and geoid model, and the height datum and its unified framework; confirm the definition of the geodetic datum constants.
②Build the models: use ETideLoad4.5 to compute the various tidal and polar shift effects and non-tidal effects and obtain the dynamic corrections and the load deformation field time series; unify the data interface and the reference epoch according to the five file formats.
③Fuse the data: fuse same-type monitoring quantities epoch by epoch according to the five principles, analytically fuse different types of monitoring quantities, unify everything to the same spatiotemporal monitoring datum, and verify online with the analytical compatibility relations.
④Produce the products: output grid time series and thematic products such as the load deformation field, the time-variable gravity field and ground stability, recording at the same time the quality control metrics and the Spatial Representativeness Accuracy.
⑤Verify afterwards: cross-check with the analytical compatibility relations and with independent external data (GNSS/leveling, independent gravity, independent leveling), assess the accuracy and trace back the sources of error.

5.8From “Technical Data Service” to “Earth System Analysis”

Purpose The official site expresses the platform's goal as promoting the strategic transformation of the geodetic discipline from a “descriptive measurement tool” into a “dynamic Earth system analyzer”, and of geodetic education from “technology-oriented” knowledge transfer to “problem-driven” systems thinking and conceptual innovation. At the level of use: do not stop at “I computed a grid”, but answer “what geophysical process does this grid reveal” — turning scattered measurement data into explanations and predictions of the surface dynamic environment, ground stability and material transport in the Earth system.
↑ Back to top
6

Continued Advancement: Source Codes, Teaching, Research and Outputs

Corresponding official-site pages: h-col-110.html (Fortran Codes), h-nd-166…173 (teaching courseware), h-nr.html (Academic and Technical Exchange), h-nd-156/164 (monographs).

6.1Reading the Source Code: the “Deep Water” of Capability Building

Description of the official “Fortran Codes” section: module-by-module Intel Fortran source codes, in which comments, integration and testing are refined unit by unit; they are uploaded successively and freely downloadable. The modules are relatively independent and easy to call. A module introduction mainly includes the module name, basic functions, geophysical model, DOS test program code, module source code, calling-module source code and algorithm formulas.

Source-code module Official page
Fortran codes for all-element field modelling using SRBFs from heterogeneous observations/h-nd-154.html
Fortran codes for local gravity field approach using SRBFs and its performance test/h-nd-152.html
Fortran codes for Molodensky boundary value correction to Stokes on arbitrary shape surface/h-nd-51.html
Fortran codes for forecast of tidal effects on Earth rotation parameters (ERP)/h-nd-150.html
Source codes for load deformation effects by SRBF approximation / codes for land water and load deformation time series by Green's integral / codes for land water and load deformation field time series by SRBF/h-nd-135.html · /h-nd-144.html · /h-nd-146.html
Source codes for building load tide spherical harmonic coefficient models / global load spherical harmonic analysis and spherical harmonic coefficient model time series/h-nd-140.html
Fortran codes for Earth center of mass variation effects / codes for forecasting ocean tidal load effects on the Earth's center of mass / codes for computing Earth ellipsoid constants/h-nd-122.html · /h-nd-127.html · /h-nd-118.html
Fortran codes for rigorous integration of Local Terrain Effects / Fortran codes for land-sea Residual Terrain Effect integration / source codes for generalized Vening-Meinesz numerical integration/h-nd-124.html · /h-nd-131.html · /h-nd-137.html
The module list grows as modules are uploaded; please refer to the current list in the h-col-110 section. Suggested way to advance: read “module source code + calling-module source code + algorithm formulas” side by side, then perform an independent test of that module using the “model value comparison” method with an ultra-high-degree geopotential coefficient model.

6.2Using the Platform for Teaching

PracticeKey point
Problem first, formulas secondIntroduce the knowledge points with real problems (regional geoid refinement, judging ground subsidence from a station time series) and then verify them with the platform
Verification instead of lecturingHave students self-test algorithms with ultra-high-degree models and check processes with closed-loop analytical operations, so that the “why” becomes clear
Linked to textbooks and monographsPhysical Geodesy (2026, h-nd-164) and Geodesy for a Deforming Earth (2025, h-nd-156) are the theoretical foundations of the two systems respectively, and can be combined into a four-piece set of “theory–courseware–software–examples”
The official site provides nine courseware PPTs in total; it also states explicitly that “after completing all the example exercises, users are able to use the software independently”.

6.3Using the Platform for Research: Reusable Points of Leverage

DirectionPlatform capabilities that can be usedOfficial-site leads
Regional ultra-high-resolution gravity field modelingCumulative SRBF approximation + terrain effect remove-restore + deep fusion of multi-source heterogeneous datah-nd-154, h-nd-152
Land-sea unified gravity field and geoid refinementStokes-type solution + remove-restore combining Local Terrain Effects with a reference gravity fieldGeoid.html, the typical applications on the home page
Dynamic height datum maintenance methodsGeodetic datum constant theory + non-tidal effect epoch reduction + GNSS/leveling fusionh-nd-161, h-nd-163
Regional water storage and load deformation inversionLoad Green's function integral + load SRBF approximation + time series analysish-nd-144, h-nd-146
Ground stability and numerical prediction of geological hazard riskCORS/InSAR fusion + quantitative criteria for ground stability degradation + time series modelsh-nd-87, Tide.html
Terrestrial reference frame and Earth center of mass / figure pole monitoringFirst- and second-degree geopotential terms and computation of Earth center of mass variation and figure pole shift effectsh-nd-160, h-nd-122

6.4Outputs and Citation: Making Work Accumulate

Recommended record itemDescription
Platform version and programs usedRecord the platform version (PAGravf4.5 / ETideLoad4.5) and the subsystems, programs and function modules invoked
Data sources and versionsSources and versions of external data (Earth gravity field models, land and seafloor topography, hydrology, satellite gravity, ocean tide models, etc.)
Key parameters and schemesTerrain effect scheme and type, integration radius, radial basis function type and parameters
Accuracy assessment resultsIn particular the Spatial Representativeness Error / Spatial Representativeness Accuracy and the checking method
Basis for citation: the two systems were independently developed by the team and have been open-sourced and made freely accessible since September 2021; the software copyright registration number of ETideLoad V4.5 is 2024SR2012207 (copyright holder: Chinese Academy of Surveying and Mapping); the technical scheme of the platform also appears in published papers and academic reports (e.g. the h-nd-160 paper in Acta Geodaetica et Cartographica Sinica and the h-nd-91 international forum report).

6.5Continued Advancement Self-Check List

√Self-check item
□Can understand at least one published source-code module and state its geophysical model and algorithm formulas
□Can perform an independent performance test of an approximation program using an ultra-high-degree geopotential coefficient model
□Can design and run through a complete task chain that links the steady-state and the time-variable
□Can record the version, data sources, parameters and accuracy assessment in results in a standardised way
□Can search the official “Academic and Technical Exchange” section and keep track of related articles and courseware updates
↑ Back to top
7

Closing and Role-Based Action Checklists

The four main threads build on one another, but they need not be strictly sequential: you can read algorithms while running examples, and fill in gaps while solving problems.
One Main Line Build proficiency through the examples → establish judgement through self-testing and criteria → connect to concrete problems through the solution library → realize integrated value through coordination of the two systems and deep fusion of multi-source heterogeneous data.

7.1Role-Based Action Checklists

RoleKey actionsExpected outputs
Undergraduates / graduate studentsInstall the platform → run all the examples (about 7 working days) → study Chapters 5–8 of the courseware → carry out a self-test with an ultra-high-degree modelThe ability to use the software systems independently, and to complete gravity field / deformation computations in course projects and theses
Course teachersDesign problem-driven classes around the courseware → assign the example exercises → use platform results for classroom verificationA four-piece teaching scheme of “theory–courseware–software–examples”
Surveying / geophysical engineersLocate the subsystem with the quick-reference table → organize the work with the official four-, five- and six-step processes → implement quality control and accuracy assessmentReproducible and deliverable gravity exploration / geoid / deformation monitoring results
ResearchersRead the open-source modules → reuse the approximation and fusion algorithms → innovate methods for new regions or new elements → cite properlyPublishable algorithmic improvements or regional application results
Teams / project managementUnify models, numerical standards and the reference epoch → organize multi-technique coordination by the five principles for deep fusion of multi-source heterogeneous data → establish quality and accuracy assessment stepsEngineering-oriented monitoring and modeling schemes integrating space, air, land and sea

7.2Five Most Easily Overlooked Recommendations

No.Recommendation
1Read the Ideas & Philosophy first, then touch the software — the official site explicitly recommends “frequently reading, analysing and reflecting upon the content of this section while using the platform”
2Do not skip terrain effects and Analytical Compatibility — this is where loss of accuracy most often occurs, and it is the core value that distinguishes the platform from “black-box software”
3Treat verification as part of the process — the built-in case studies reflect only basic performance; the real accuracy must be tested with your own data
4Unify the “Spatial Representativeness Error” definition before discussing accuracy — when discussing 1 cm or 20 μGal you must state the spatial averaging accuracy at which resolution
5Make the data interface the first task — aligning data to the five classes of time-series data file formats and to the monitoring quantity unit and direction conventions is the prerequisite for integrated application to run through
Remember the Platform in One Sentence The steady state fixes the skeleton (PAGravf4.5) + the time-variable describes the dynamics (ETideLoad4.5) = an integrated geodetic computation and monitoring system unified in the Earth's gravity field.
To become proficient, rely on the examples; to master it, rely on self-testing and criteria; to use it well, rely on the solution library; to go beyond it, rely on coordination of the two systems and deep fusion of multi-source heterogeneous data.
↑ Back to top
8

Appendix: Page Index and Terminology Conventions

Every key point of this report can be traced back to an official-site page; terms and technical names follow the official wording throughout.

8.1Official-Site Page-Level Index

CategoryPageKey contentLevel
Top-level sections/ HomePlatform overview, introduction to the two systems, code-level hybrid programming and program scale, example mechanism, five typical applications, exchange entries
/Geoid.htmlPAGravf4.5: scientific objectives, the complete algorithmic framework for terrain effects, technical characteristics of gravity field integral algorithms, spectral-domain Spherical Radial Basis Function (SRBF) approximation, typical computation programs and classroom demonstrations
/Tide.htmlETideLoad4.5: scientific objectives, the five major subsystems, basic principles for representing, approximating and monitoring the load deformation field, the principle of CORS-network time-series InSAR collaborative monitoring of vertical deformation, and the seamless continuous quantitative monitoring scheme of regional ground stability variations
/h-col-104.htmlDownloads and Resources: lists of program packages and User References, algorithm and process documents, computation processes and exercises, external data resource links, installation notes
/h-col-105.htmlIdeas & Philosophy: concepts of geodesy and physical geodesy, metrological essence and normative constraints, geodetic datum constants, the “four-in-one” cornerstone, Analytical Compatibility, error analysis and accuracy assessment, height datum and reference system theory
/h-col-110.htmlFortran Codes: module-by-module Intel Fortran source codes and module introductions
/h-nr.html (with ?m396pageno=2…17 pages)Academic and Technical Exchange: article index (categorized as News / Software Teaching / Work Processes / Source Codes)
/h-msgBoard.htmlOnline communication and messages
/en/English site
Theoretical and methodological articles/h-nd-88.htmlIntroduction to Prof. ZHANG Chuanyin
/h-nd-160.htmlMonitoring method for the joint determination of the Earth's center of mass, figure pole and various Earth rotation dynamic parameters (published in Acta Geodaetica et Cartographica Sinica)
/h-nd-161.htmlTheoretical foundation of the height datum gravity field unified in the Earth-fixed reference system and updating of classical concepts
/h-nd-162.htmlError analysis and accuracy assessment in PAGravf gravity field approximation
/h-nd-163.htmlPrinciples, methods and restrictive requirements for terrain effects on PAGravf gravity field quantities
/h-nd-84.htmlGeoid accuracy assessment method based on GNSS/leveling and the gravity field error characteristics
/h-nd-87.htmlCORS network ground stability variation monitoring and prediction (Baoshan–Dali pilot)
/h-nd-91.htmlIntroduction to the Earth Tide, Load Effect and Deformation Monitoring Computation system (report at the 2021 International Forum on Earth System Spatial Monitoring, with courseware attachment)
/h-nd-83.htmlGeodetic software, a key to university education
/h-nd-104.html · /h-nd-105.htmlLoad deformation field approximation and collaborative monitoring · all-element analytical gravity field modeling
/h-nd-108.html · /h-nd-109.html · /h-nd-110.html · /h-nd-111.htmlHigh-resolution load deformation field approximation demonstration · SRBF all-element modeling process · deformation field collaborative monitoring process · PAGravf4.5 terrain effect exercise demonstration
/h-nd-100.html · /h-nd-101.html · /h-nd-156.html · /h-nd-164.htmlPAGravf4.5 terrain effect analytical algorithm case · ETideLoad4.5 tidal and polar shift computation demonstration · Geodesy for a Deforming Earth (2025) · Physical Geodesy (2026)
Teaching courseware/h-nd-166.html … /h-nd-173.htmlPhysical Geodesy teaching courseware PPT-01…08 (Chapter 3 terrain effects, tidal and load effects; Chapter 4 Earth-fixed reference system and relativistic unification in Earth space; Chapter 5 gravity field modeling from heterogeneous observations; Chapter 6 height datums and their gravity field theory; Chapter 7 all-element modeling of the local gravity field and time-variable monitoring; Chapter 8 vertical datums unified in the Earth-fixed reference system)
/h-nr--0_31_8.html · /h-nr--0_31_4.html · /h-nr--0_31_5.htmlSoftware Teaching section · Work Processes section · Source Codes section list pages
Source-code modules/h-nd-154.html · /h-nd-152.html · /h-nd-51.htmlAll-element field modelling using SRBFs from heterogeneous observations · local gravity field approach using SRBFs and its performance test · Molodensky boundary value correction to Stokes on arbitrary shape surface
/h-nd-118 / 122 / 124 / 127 / 131 / 135 / 137 / 140 / 144 / 146 / 150Earth ellipsoid constant computation · Earth center of mass variation effects · rigorous integration of Local Terrain Effects · forecasting of ocean tidal load effects on the Earth's center of mass · land-sea Residual Terrain Effect integration · load deformation effects by SRBF approximation · generalized Vening-Meinesz numerical integration · building of load tide spherical harmonic coefficient models · land water and load deformation time series by Green's integral · land water and load deformation field time series by SRBF · forecast of tidal effects on Earth rotation parameters
The page list is compiled from official-site crawling and online verification; the source-code module section and the exchange sections are still being updated, so please refer to the current list on the official site.

8.2External Resource Link List

Note Section 2.6 already lists, under the categories “gravity field models / regional geoid / marine gravity field / satellite gravity / Earth rotation and reference frames / topography and bathymetry / hydrology and meteorology / sea level and ocean / ocean tide models / shared software and services”, the full text of every external resource link given on the official download page (pointing to IERS, ICGEM, ISG, DTU, GFZ, NASA, NOAA, CMEMS, TPXO, SHTOOLS, etc.).

8.3Quick Reference of Subsystem and Process Names

ItemOfficial name
Five major subsystems of PAGravf4.5① Analysis and Preprocessing of Earth Gravity Field Data ② Computation of Diverse Terrain Effects on Various Gravity Field Elements ③ High-Precision Gravity Field Approximation and Full-Element Modeling ④ Optimization, Unification, and Application of Regional Height Datums ⑤ Toolkit for Geodetic Data File Editing, Calculation, and Visualization
Five major subsystems of ETideLoad4.5① Computation of Various Tidal and Polar Shift Effects on All Geodetic Elements ② Processing and Analysis of Non-Tidal Geodetic Variation Time Series ③ Load Deformation Field Approximation and Monitoring from Heterogeneous Variations ④ CORS/InSAR Collaborative Monitoring and Ground Stability Estimation ⑤ Editing, Calculation and Visualization of Geodetic Data
Main official computation processesTerrain effects and gravity exploration modeling exercise (four steps) · expedited process for computing land-sea Bouguer and isostatic anomalies from a gravity field model (four steps) · expedited process for all-element gravity field modeling in full external space using SRBFs (six steps) · complete computation process of high-resolution surface load deformation field time series (four steps) · collaborative monitoring process of groundwater and surface environmental load deformation field (five steps)
Monitoring network station typesCORS network · solid Earth tide station network · groundwater monitoring network · hydrological monitoring network · tide-gauge network
Monitoring quantity types (six)height anomaly variation · gravity disturbance variation · ground gravity variation · ellipsoidal height variation · normal height variation · equivalent water height variation

8.4Terminology and Wording Conventions

Official standard wordingCommon deviationNote
CORS/InSAR“CORS-InSAR”The official site does not use a hyphen
Helmert Condensation“terrain Helmert condensation”The official site does not insert a space
global geopotential W₀ / geoidal potential W₀“geoidal potential constant W₀”The official site uses both of the former expressions
Spatial Representativeness Error“terrain representativeness error”The latter was the customary name in the era of terrestrial geodesy; the official site makes the current name explicit
Analytical Compatibility / fully analytically compatible“analytical compatibility” used loosely“Compatibility” is the fixed term on the official site
epoch reduction“epoch normalization”The official site uniformly uses “reduction”
gravity field quantity (terrain-effect context) / gravity field element (analytical-relation context)Mixing “field quantity” and “field element”Both usages coexist on the official site and should be distinguished by context
1 cm steady-state geoid refinement; geodesy at the 1 cm (20 μGal) precision levelVague “centimetre-level accuracy”The official site gives the precise definition and the parenthesised form
the platform and its core source codes were open-sourced and made freely accessible in September 2021“released in 2021” / “free and open source”The official site uses “open-sourced and made freely accessible”
Prof. ZHANG Chuanyin of the Chinese Academy of Surveying and Mapping“Professor ZHANG Chuanyin” etc.The official designation is research geodesist
Terrain effects and gravity exploration modeling exercise“terrain effect and gravity exploration computation process exercise”Follow the official file name
time-series InSAR monitoring quantitiesthe general term “InSAR deformation values”The official site emphasises the separation of InSAR monitoring quantities
This report has been worded uniformly according to the table above; if it differs from the User References supplied with the packages or from the latest pages of the official site, the latter shall prevail.