Hoa Thi Pham
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hoa Thi Pham has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
Joseph Awange
4 shared publicationsSten Claessens
4 shared publications- Joseph Langat Awange
Department of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong Kong, China
2 shared publications Michael Kuhn
Curtin Univ
2 shared publications- Michael Kuhn
Curtin University
2 shared publications - Binh Van Nguyen
Geology Faculty, Hanoi University of Natural Resources and Environment, Hanoi 100000, Vietnam
1 shared publication
Latest Publications
- Open Access
- 22 August 2026
- Published by Springer Nature in Natural Resources Research
- Vol. 35 (5) , 3187-3220
A Framework for Characterizing Component-Wise Nonstationarity in GRACE/GRACE-FO-Derived Terrestrial Water Storage Anomalies
Understanding the nonstationarity of GRACE/GRACE-FO-derived terrestrial water storage anomalies (TWSA) is crucial for interpreting component-scale water storage variability, improving TWSA modeling, and supporting evaluation of land-surface and global hydrological models. However, existing studies lack methods for exploring TWSA nonstationarity in sufficient detail. To address this gap, this study introduces a framework comprising four steps: (i) filling TWSA gaps using two independent observation-only methods under an uncertainty-referenced dual-reconstruction strategy; (ii) decomposing TWSA into semi-annual, annual, long-term, and residual components while reducing and diagnosing signal mixing, with the long-term component further separated into trend and interannual–decadal variability (IA-D); (iii) evaluating the extracted components using complementary spectral and time–frequency methods; and (iv) estimating time-varying and average semi-annual amplitude (SAA) and annual amplitude (ANA), while characterizing IA-D in the time and time–frequency domains. The framework integrates non-wavelet-based techniques (Singular Spectrum Analysis, Fast Fourier Transform) and wavelet-based methods (Continuous Wavelet Transform, Multiresolution Analysis of the Maximal Overlap Discrete Wavelet Transform), along with local least-squares fitting, change-point detection, and local-extrema analysis. The Vietnam application demonstrates its ability to characterize component-wise TWSA nonstationarity across diverse hydroclimatic and geographical conditions. SAA fluctuates in all regions, ANA generally weakens with regional exceptions, and IA-D exhibits regionally varying high- and low-value episodes and evolving 2-year and 5-year periodicities. Compared with conventional least-squares fitting, the framework remains broadly consistent with full-record average seasonal amplitude estimates, while revealing time-varying SAA and ANA and providing more reliable average SAA estimates where IA-D variability is much larger than the semi-annual component.
- 22 August 2026
- Published by Springer Nature in Natural Resources Research
- Vol. 35 (5) , 3187-3220
Open AccessA Framework for Characterizing Component-Wise Nonstationarity in GRACE/GRACE-FO-Derived Terrestrial Water Storage Anomalies
Understanding the nonstationarity of GRACE/GRACE-FO-derived terrestrial water storage anomalies (TWSA) is crucial for interpreting component-scale water storage variability, improving TWSA modeling, and supporting evaluation of land-surface and global hydrological models. However, existing studies lack methods for exploring TWSA nonstationarity in sufficient detail. To address this gap, this study introduces a framework comprising four steps: (i) filling TWSA gaps using two independent observation-only methods under an uncertainty-referenced dual-reconstruction strategy; (ii) decomposing TWSA into semi-annual, annual, long-term, and residual components while reducing and diagnosing signal mixing, with the long-term component further separated into trend and interannual–decadal variability (IA-D); (iii) evaluating the extracted components using complementary spectral and time–frequency methods; and (iv) estimating time-varying and average semi-annual amplitude (SAA) and annual amplitude (ANA), while characterizing IA-D in the time and time–frequency domains. The framework integrates non-wavelet-based techniques (Singular Spectrum Analysis, Fast Fourier Transform) and wavelet-based methods (Continuous Wavelet Transform, Multiresolution Analysis of the Maximal Overlap Discrete Wavelet Transform), along with local least-squares fitting, change-point detection, and local-extrema analysis. The Vietnam application demonstrates its ability to characterize component-wise TWSA nonstationarity across diverse hydroclimatic and geographical conditions. SAA fluctuates in all regions, ANA generally weakens with regional exceptions, and IA-D exhibits regionally varying high- and low-value episodes and evolving 2-year and 5-year periodicities. Compared with conventional least-squares fitting, the framework remains broadly consistent with full-record average seasonal amplitude estimates, while revealing time-varying SAA and ANA and providing more reliable average SAA estimates where IA-D variability is much larger than the semi-annual component.
- Open Access
- 25 March 2026
- Published by Wiley in Jawra Journal of the American Water Resources Association
- Vol. 62 (2) , e70101
Stationary and Nonstationary Influences of Teleconnections on Terrestrial Water Storage Anomalies in Vietnam
Previous studies of teleconnection (TC) impacts on Terrestrial Water Storage Anomalies (TWSA) rarely focus on Vietnam or on the nonstationary nature of TC–TWSA relationships. This study addresses these gaps by examining both stationary and nonstationary TC influences on TWSA using correlation analysis (Pearson and cross‐spectral methods) and TC‐related TWSA extraction. The extracted TC‐related TWSA is consistent with correlation patterns: absolute zero‐lag (lagged) correlations are 0.18–0.51 (0.18–0.62), with TCs leading TWSA by 0–8 months and explaining 20%–64% of TWSA variance. Dominant TCs differ by region and are generally stronger in the south. Nonstationarity is evident from (i) large correlation shifts between 2003–2012 and 2013–2022 (e.g., from none to 0.80 or from −0.19 to 0.76) and (ii) time–frequency variability in TC–TWSA strength. Separating TCs into interannual and decadal components improves TWSA modeling, especially when distinct TCs are used for each scale; decadal influences dominate in the north, interannual in the south. These results highlight the value of timescale‐specific TC integration for TWSA reconstruction and forecasting and caution against transferring nonstationary TC–TWSA links between periods.
- 25 March 2026
- Published by Wiley in Jawra Journal of the American Water Resources Association
- Vol. 62 (2) , e70101
Open AccessStationary and Nonstationary Influences of Teleconnections on Terrestrial Water Storage Anomalies in Vietnam
Previous studies of teleconnection (TC) impacts on Terrestrial Water Storage Anomalies (TWSA) rarely focus on Vietnam or on the nonstationary nature of TC–TWSA relationships. This study addresses these gaps by examining both stationary and nonstationary TC influences on TWSA using correlation analysis (Pearson and cross‐spectral methods) and TC‐related TWSA extraction. The extracted TC‐related TWSA is consistent with correlation patterns: absolute zero‐lag (lagged) correlations are 0.18–0.51 (0.18–0.62), with TCs leading TWSA by 0–8 months and explaining 20%–64% of TWSA variance. Dominant TCs differ by region and are generally stronger in the south. Nonstationarity is evident from (i) large correlation shifts between 2003–2012 and 2013–2022 (e.g., from none to 0.80 or from −0.19 to 0.76) and (ii) time–frequency variability in TC–TWSA strength. Separating TCs into interannual and decadal components improves TWSA modeling, especially when distinct TCs are used for each scale; decadal influences dominate in the north, interannual in the south. These results highlight the value of timescale‐specific TC integration for TWSA reconstruction and forecasting and caution against transferring nonstationary TC–TWSA links between periods.
- 1 February 2026
- Published by American Society of Civil Engineers (ASCE) in Journal of Surveying Engineering
- Vol. 152 (1)
Location-Specific Ensemble Kriging: New Approach for Local Hybrid Geoid–Quasigeoid Estimation
- 1 February 2026
- Published by American Society of Civil Engineers (ASCE) in Journal of Surveying Engineering
- Vol. 152 (1)
Location-Specific Ensemble Kriging: New Approach for Local Hybrid Geoid–Quasigeoid Estimation
- Open Access
- 22 August 2025
- Published by Taylor & Francis in Survey Review
- Vol. 58 (409) , 297-329
Local hybrid geoid/quasigeoid development using machine learning with consideration of neighbour spatial characteristics
- 22 August 2025
- Published by Taylor & Francis in Survey Review
- Vol. 58 (409) , 297-329
Open AccessLocal hybrid geoid/quasigeoid development using machine learning with consideration of neighbour spatial characteristics
- Open Access
Evaluation of Three Feature Dimension Reduction Techniques for Machine Learning-Based Crop Yield Prediction Models
Machine learning (ML) has been widely used worldwide to develop crop yield forecasting models. However, it is still challenging to identify the most critical features from a dataset. Although either feature selection (FS) or feature extraction (FX) techniques have been employed, no research compares their performances and, more importantly, the benefits of combining both methods. Therefore, this paper proposes a framework that uses non-feature reduction (All-F) as a baseline to investigate the performance of FS, FX, and a combination of both (FSX). The case study employs the vegetation condition index (VCI)/temperature condition index (TCI) to develop 21 rice yield forecasting models for eight sub-regions in Vietnam based on ML methods, namely linear, support vector machine (SVM), decision tree (Tree), artificial neural network (ANN), and Ensemble. The results reveal that FSX takes full advantage of the FS and FX, leading FSX-based models to perform the best in 18 out of 21 models, while 2 (1) for FS-based (FX-based) models. These FXS-, FS-, and FX-based models improve All-F-based models at an average level of 21% and up to 60% in terms of RMSE. Furthermore, 21 of the best models are developed based on Ensemble (13 models), Tree (6 models), linear (1 model), and ANN (1 model). These findings highlight the significant role of FS, FX, and specially FSX coupled with a wide range of ML algorithms (especially Ensemble) for enhancing the accuracy of predicting crop yield.
Open AccessEvaluation of Three Feature Dimension Reduction Techniques for Machine Learning-Based Crop Yield Prediction Models
Machine learning (ML) has been widely used worldwide to develop crop yield forecasting models. However, it is still challenging to identify the most critical features from a dataset. Although either feature selection (FS) or feature extraction (FX) techniques have been employed, no research compares their performances and, more importantly, the benefits of combining both methods. Therefore, this paper proposes a framework that uses non-feature reduction (All-F) as a baseline to investigate the performance of FS, FX, and a combination of both (FSX). The case study employs the vegetation condition index (VCI)/temperature condition index (TCI) to develop 21 rice yield forecasting models for eight sub-regions in Vietnam based on ML methods, namely linear, support vector machine (SVM), decision tree (Tree), artificial neural network (ANN), and Ensemble. The results reveal that FSX takes full advantage of the FS and FX, leading FSX-based models to perform the best in 18 out of 21 models, while 2 (1) for FS-based (FX-based) models. These FXS-, FS-, and FX-based models improve All-F-based models at an average level of 21% and up to 60% in terms of RMSE. Furthermore, 21 of the best models are developed based on Ensemble (13 models), Tree (6 models), linear (1 model), and ANN (1 model). These findings highlight the significant role of FS, FX, and specially FSX coupled with a wide range of ML algorithms (especially Ensemble) for enhancing the accuracy of predicting crop yield.
- Open Access
Enhancing Crop Yield Prediction Utilizing Machine Learning on Satellite-Based Vegetation Health Indices
Accurate crop yield forecasting is essential in the food industry’s decision-making process, where vegetation condition index (VCI) and thermal condition index (TCI) coupled with machine learning (ML) algorithms play crucial roles. The drawback, however, is that a one-fits-all prediction model is often employed over an entire region without considering subregional VCI and TCI’s spatial variability resulting from environmental and climatic factors. Furthermore, when using nonlinear ML, redundant VCI/TCI data present additional challenges that adversely affect the models’ output. This study proposes a framework that (i) employs higher-order spatial independent component analysis (sICA), and (ii), exploits a combination of the principal component analysis (PCA) and ML (i.e., PCA-ML combination) to deal with the two challenges in order to enhance crop yield prediction accuracy. The proposed framework consolidates common VCI/TCI spatial variability into their respective subregions, using Vietnam as an example. Compared to the one-fits-all approach, subregional rice yield forecasting models over Vietnam improved by an average level of 20% up to 60%. PCA-ML combination outperformed ML-only by an average of 18.5% up to 45%. The framework generates rice yield predictions 1 to 2 months ahead of the harvest with an average of 5% error, displaying its reliability.
Open AccessEnhancing Crop Yield Prediction Utilizing Machine Learning on Satellite-Based Vegetation Health Indices
Accurate crop yield forecasting is essential in the food industry’s decision-making process, where vegetation condition index (VCI) and thermal condition index (TCI) coupled with machine learning (ML) algorithms play crucial roles. The drawback, however, is that a one-fits-all prediction model is often employed over an entire region without considering subregional VCI and TCI’s spatial variability resulting from environmental and climatic factors. Furthermore, when using nonlinear ML, redundant VCI/TCI data present additional challenges that adversely affect the models’ output. This study proposes a framework that (i) employs higher-order spatial independent component analysis (sICA), and (ii), exploits a combination of the principal component analysis (PCA) and ML (i.e., PCA-ML combination) to deal with the two challenges in order to enhance crop yield prediction accuracy. The proposed framework consolidates common VCI/TCI spatial variability into their respective subregions, using Vietnam as an example. Compared to the one-fits-all approach, subregional rice yield forecasting models over Vietnam improved by an average level of 20% up to 60%. PCA-ML combination outperformed ML-only by an average of 18.5% up to 45%. The framework generates rice yield predictions 1 to 2 months ahead of the harvest with an average of 5% error, displaying its reliability.