Hafiza Mamona Nazir

Hafiza Mamona Nazir

Auto-generated profile, based on publicly available information

Hafiza Mamona Nazir

Department of Statistics, University of Sargodha, Sargodha, Pakistan

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 January 2019 in Complexity
Verified
Journal Article
Open access
PDF file
HNHafiza Mamona Nazir
Hafiza Mamona Nazir
IHIjaz Hussain
Ijaz Hussain
MFMuhammad Faisal
Muhammad Faisal
ASAlaa Mohamd Shoukry
Alaa Mohamd Shoukry
SGShowkat Hussain Gani
Showkat Hussain Gani
IAIshfaq Ahmad
Ishfaq Ahmad
Accurate prediction of hydrological processes is key for optimal allocation of water resources. In this study, two novel hybrid models are developed to improve the prediction precision of hydrological time series data based on the principal of three stages as denoising, decomposition, and decomposed component prediction and summation. The proposed architecture is applied on daily rivers inflow time series data of Indus Basin System. The performances of the proposed models are compared with traditional single‐stage model (without denoised and decomposed), the hybrid two‐stage model (with denoised), and existing three‐stage hybrid model (with denoised and decomposition). Three evaluation measures are used to assess the prediction accuracy of all models such as Mean Relative Error (MRE), Mean Absolute Error (MAE), and Mean Square Error (MSE). The proposed, three‐stage hybrid models have shown improvement in prediction accuracy with minimum MRE, MAE, and MSE for all case studies as compared to other existing one‐stage and two‐stage models. In summary, the accuracy of prediction is improved by reducing the complexity of hydrological time series data by incorporating the denoising and decomposition.

29 April 2026 in IEEE Access
Verified
Journal Article
Open access
MNMalaika Noor
Malaika Noor
HNHafiza Mamona Nazir
Hafiza Mamona Nazir
MQMoiz Qureshi
Moiz Qureshi
HIHasnain Iftikhar
Hasnain Iftikhar
Paulo Canas RodriguesPaulo Canas Rodrigues
Paulo Canas Rodrigues
AHAtef F. Hashem
Atef F. Hashem
Accurate modeling of Wind Speed (WS) data is essential in renewable energy, optimization of agriculture, and disaster risk reduction. However, in the presence of nonlinearity, non-stationarity, and noise in WS, the prediction becomes less efficient. A hybrid framework that combines denoising and decomposition has been developed using WS data from Lahore from 1986 to 2016. Four single-stage preprocessing techniques, namely Moving Average (MA), Singular Spectrum Analysis (SSA), Wavelet Transform (WAV), and Empirical Mode Decomposition (EMD), and a dual-stage denoising pipeline involving Variational Mode Decomposition (VMD) and Empirical Wavelet Transform (EWT) were considered. To model the denoised WS data, statistical models, i.e., Autoregressive Moving Average (ARIMA), machine learning models, including Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), Random Forest (RF), and Support Vector Machine (SVM); and deep learning models, namely Long Short Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and Convolutional Neural Network-LSTM (CNN-LSTM) have been implemented. Furthermore, the data were split 80-20 to assess the performance of the proposed models. Different evaluation metrics, i.e., Mean Absolute Error (MAE), Mean Bias Error (MBE), and Nash Sutcliffe Efficiency (NSE), as well as the Taylor diagram and error prediction box plot, have been used to compare the performance of the proposed methodology. Results confirmed that a simple model, such as ARIMA, with all preprocessing methods, is not suitable for WS prediction. However, SSA along with LSTM (complex model) achieved the best accuracy with least MAE and NSE i.e. 0.0627 and 0.9947 respectively. The application of a dual-stage VMD-EWT somewhat improved the performance of the weaker preprocessing methods, but it did not outperform single-stage denoising methods. Single-stage SSA, when combined with deep learning, yields the most reliable WS predictions, whereas dual-stage denoising improves stability and performance under weaker preprocessing methods.

30 May 2026 in Visnyk of V. N. Karazin Kharkiv National University. Series Geology. Geography. Ecology
Verified
Journal Article
Open access
PDF file
AAAyesha Amir
Ayesha Amir
SASyed Ali
Syed Ali
HNHafiza Mamona Nazir
Hafiza Mamona Nazir
AAAhmed Abd El Aal
Ahmed Abd El Aal
YBYasir Bashir
Yasir Bashir
NRNaveed Rehman
Naveed Rehman
Problem statement. Wargal Limestone plays a significant role due to its substantial reservoir potential. A more in-depth analysis of this formation can be helpful in identifying elusive geobodies. Wargal limestone is a reservoir rock of Permian age in Pakistan. It has proven the reservoir in the Dhurnal field that proves its reservoir facies. In Pakistan, the Western Salt Range presents a wonderful opportunity to study this Permian limestone. Purpose. This study aims to examine the facies and depositional environment of late Permian limestone in Upper Indus Basin by characterizing the geological, mineralogical proxies of rock samples collected from various localities. The geobodies and sedimentary architecture of Wargal Limestone was deciphered using field observation. Methods. Facies analysis, stratigraphic logging, and lithological correlation were used to interpret outcrop data and reconstruct the depositional environments. On the basis of fossils present in lithofacies, its depositional environment is elaborated. Sedimentology was studied using section measurement along with high resolution field photographs were collected using the facies heterogeneity and architecture. The facies were then plotted on high resolution field photographs. The dataset generated includes sedimentological logs, samples, and field photographs to enable reproducibility and future comparative stratigraphic studies. Results. The common facies are carbonaceous shale, sandy limestone, argillaceous limestone, fossiliferous limestone, and dolomitic limestone. Fossils such as bryozoans, brachiopods, bivalves, gastropods, and trilobites date it to the mid-late Permian Five sections of Wargal limestone have been taken to correlate. Lateral facies changes of these facies shows stratigraphic continuity but with some obvious changes in lithology and thickness. We observed the geobodies in the formation, whether it has reefs, sheets, lenses, or channels. There is a chance to locate zoned reservoirs for potential hydrocarbon extraction because of stratigraphic variations. The diverse depositional environments of this formation, which range from lagoonal to reefal settings, are widely known for representing a highly active and diverse paleotethys system. The paper supports Sustainable Development Goal (SDG) 9 by advancing geological knowledge relevant to resource assessment and sedimentary basin analysis, SDG 13: through improved understanding of climatic and earliest environmental settings. Conclusions. As the formation is made up of different types of rock, including fossiliferous limestone and porous dolomite, and since it is situated in a Salt Range, it has good potential for oil and gas exploration in the adjacent Potwar, Kohat and Bannu basins. The findings of this study are recommended for analogues as a guide in developing a framework for future studies of comparable carbonate rock successions in the study region.

1 January 2026 in SCOPUA Journal of Applied Statistical Research
Verified
Journal Article
Open access
PDF file
ASAqsa Saleem
Aqsa Saleem
HNHafiza Mamona Nazir
Hafiza Mamona Nazir
MZMuhammad Zubair
Muhammad Zubair
Accurate river water-level forecasting supports flood risk mitigation and water-resource management in the Indus Basin System, despite increased hydrological variability. This study proposes a hybrid multi-tier ensemble framework for daily river-level forecasting that incorporates 13 heterogeneous machine learning (ML) and deep learning (DL) models from the tree-based, kernel-based, neural network, and deep learning families, which are combined using Simple Averaging, Inverse-RMSE, Adaptive Inverse-RMSE weighting, and stacking-based strategies. Lag-based feature engineering and chronological TimeSeriesSplit validation maintained temporal relationships while preventing information leakage. Support Vector Regression (SVR) and Extra Trees (ET) outperformed ML models, while GRU and LSTM dominated DL architectures. The proposed hybrid ensemble outperformed the persistence benchmark, with an RMSE of 1.7621 m and R^2 of 0.3012. Diebold-Mariano statistical testing verified the comparative forecasting ability with a rigorous statistical significance assessment. Furthermore, the findings show that predicting accuracy increased only up to an ideal ensemble size, after which additional models yielded declining benefits. Overall, the findings show that model diversity and complementary learning characteristics have a greater impact on forecasting performance than increasing ensemble size or using increasingly complex weighting strategies, resulting in a strong framework for operational river water-level forecasting in data-scarce hydrological environments.