Sept. 20, 2020

Improving the methods of land hydrology monitoring in North America

Researcher: Dimitrios Piretzidis | Supervisor: Prof. Michael G. Sideris

Short description: The need for accurate monitoring of terrestrial water is of great importance, especially in overpopulated areas that already suffer from freshwater shortage. Contemporary methods of land hydrology monitoring on global scale depend on satellite observations that suffer from several uncertainties. My work focuses on the development of improved processing techniques for the mitigation of errors contained in the measurements of the Gravity Recovery and Climate Experiment (GRACE) twin-satellite system. GRACE measures the temporal variations of the Earth’s gravity field that occur for the most part due to the redistribution of water masses. Its measurements, along with an established and well-tested theoretical framework, provides a unique method for land hydrology monitoring. One of the largest uncertainties in GRACE measurements comes from correlated errors in the form of north-south stripes. The complex nature of these errors along with their large magnitude that can completely obscure hydrology signals necessitate new filtering methods. As part of my research, I designed innovative filtering techniques for GRACE measurements based on artificial intelligence and rigorous mathematical models in order to reduce the deficiencies of the currently used methods. As an application, I developed hydrology models and studies the variability of terrestrial and ground water in North America. To assist users from a wide spectrum of geosciences in using these new techniques, I also developed free, open-access software and toolboxes. Reducing the uncertainties of GRACE satellite products would result in more accurate hydrology models that will also improve the prediction of future hydrology states, leading to models that can serve as early warning systems for extreme events, such as floods and droughts. The more reliable monitoring of terrestrial water and groundwater variations is also of great importance for the estimation of available freshwater deposits and the development of optimized water management systems. A state-of-the-art land hydrology model can also be applicable to various users, such as glaciologist for studying the evolution of glaciers and ice sheets, and geophysicists for monitoring ground deformations due to the Earth’s crust movement.

Funding sources: Natural Sciences and Engineering Research Council (NSERC) of Canada, Werner Graupe International Fellowship in Engineering.

Collaborators: None.

Figures:

Figure

Figure 1 – Removal of correlated errors from GRACE monthly models. The water mass variations are in terms of equivalent water height.

 Figure 2 – Monthly terrestrial water storage variations in North America in terms of equivalent water height. The data are estimated from a combination of GRACE and land hydrology models.

 Publications:

  1. Piretzidis D (2020) Land Hydrology Studies in North America Using GRACE and Hydrology Models. Ph.D. Thesis, University of Calgary, url: http://hdl.handle.net/1880/111902
  2. Piretzidis D, Sideris MG, Tsoulis D (2019) Comparison of criteria for the identification of correlated orders in GRACE spherical harmonic coefficients. International Association of Geodesy Symposia, doi: https://doi.org/10.1007/1345_2019_83
  3. Piretzidis D, Sideris MG (2019) Stable recurrent calculation of isotropic Gaussian filter coefficients. Computers & Geosciences, doi: https://doi.org/10.1016/j.cageo.2019.07.007
  4. Kuczynska-Siehien J, Piretzidis D, Sideris MG, Olszak T, Szabo V (2019) Monitoring of extreme land hydrology events in central Poland using GRACE, land surface models and absolute gravity data. Journal of Applied Geodesy, doi: https://doi.org/10.1515/jag-2019-0003
  5. Wang H, Xiang L, Steffen H, Wu P, Jiang L, Shen Q, Piretzidis D, Sideris MG, Hayashi M, Jia L (2019) Converse trends of the terrestrial and ground water storage changes in Canada and the United States. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W13, 1793-1796, doi: https://doi.org/10.5194/isprs-archives-XLII-2-W13-1793-2019
  6. Piretzidis D, Sra G, Karantaidis G, Sideris MG, Kabirzadeh H (2018) Identifying presence of correlated errors using machine learning algorithms for the selective de-correlation of GRACE harmonic coefficients. Geophysical Journal International, doi: https://doi.org/10.1093/gji/ggy272
  7. Piretzidis D, Sideris MG (2018) SHADE: A MATLAB toolbox and graphical user interface for the empirical de-correlation of GRACE monthly solutions. Computers & Geosciences, doi: https://doi.org/10.1016/j.cageo.2018.06.012

Software:

The software developed in this work is available for download at https://github.com/dimitriospiretzidis. Additional routines and resources can be found in the supplementary material of the author’s publications, following the doi links.