Glavesh Ahmed Shamsaddin started her PhD research in 2024 on the thesis entitled "Climate Reconstruction Using Annual Diameter Increment of Stem Analysis and Carbon Sequestration of Natural Pinus brutia Ten. in Duhok Governorate." Combining dendrochronology with dendroclimatology in combination with carbon modeling, it deals with the most critical environmental problems within semi-arid regions of Kurdistan-Iraq. The research is framed on three main objectives: (1) reconstructing past climate trends by investigating the relationship between tree-ring growth and precipitation/temperature; (2) forecasting past precipitation and temperature patterns from tree-ring data to fill the gaps left by the limited number of available meteorological records; and (3) developing regression models for estimating the carbon sequestration potential of Pinus brutia based on diameter and height measurements. Glavesh uses up-to-date statistical tools and specialized software, such as CooRecorder, to extract and analyze data from tree cores and discs at five microsites in Duhok. Her findings will help improve our understanding of the history of climate change and the role forests play in emissions mitigation, potentially offering practical insights into sustainable forest management. This research is under the guidance of Dr. Yaseen Taha Mustafa and Dr. Tariq K. Salih and will surely contribute substantially to climate science and environmental conservation.
Milat Hasan Abdullah began his research project in 2024, titled ‘Integrative Machine Learning and Remote Sensing Approaches for Forest Conservation in Duhok, Kurdistan Region of Iraq: From Monitoring to Management.’ This ongoing project focuses on tackling critical challenges in forest conservation through the integration of advanced machine learning (ML) techniques with remote sensing (RS) and GIS technologies. The study is designed to address four primary objectives: (1) monitoring deforestation and forest degradation over time using time-series satellite data; (2) analyzing the impact of climate change on forest health and vulnerability; (3) optimizing reforestation efforts by identifying high-priority degraded areas through cluster analysis; and (4) developing a dynamic Forest Recovery Index to evaluate and enhance reforestation success and adaptive management strategies. Milat employs a comprehensive methodology, incorporating diverse datasets such as forest inventories, meteorological data, and satellite imagery, combined with cutting-edge ML algorithms. The findings are anticipated to provide dynamic, actionable insights for policymakers and practitioners aiming for sustainable forest management and conservation practices. This work is being conducted under my guidance and is expected to yield significant contributions in the next one to two years.
Yaseen Taha Mustafa, PhD
Professor of Applied Remote Sensing & GIS
University of Zakho,
P.O. Box 12, Zakho, Duhok, Kurdistan Region - Iraq
Mobile: +9647504922553
Email: yaseen.Mustafa@uoz.edu.krd
Abdulqadeer I. A. Rash completed his PhD research between 2021 and 2025, focusing on the assessment, modeling, and simulation of Land Use/Land Cover (LULC) dynamics in selected districts in the northeast of Erbil Province, Kurdistan Region, Iraq. His research aimed to evaluate the effectiveness of hybrid modeling approaches for predicting future land transformations, identify optimal methods for LULC change detection, and examine the influence of such changes on land surface temperature. The study further investigated the key biophysical and anthropogenic drivers shaping land change processes in the region. Abdulqadeer employed advanced machine learning algorithms, integrating multi-temporal remote sensing data and field-based observations to ensure robust analytical outcomes. His work also extended to forecasting future LULC patterns for 2040 under various Shared Socioeconomic Pathway (SSP) climate scenarios using Random Forest modeling. This comprehensive research was carried out in close collaboration with Asst. Prof. Dr. Rahel Othman Hamad, whose scientific contributions were instrumental in guiding the study’s analytical depth and contextual relevance.
Hendaf N. Habeeb conducted his PhD research from 2021 to 2024, focusing on spatiotemporal analysis of forest cover change using advanced machine learning algorithms in Duhok and Amedy districts, Kurdistan Region of Iraq. His research aimed to estimate forest vegetation cover utilizing deep learning models, specifically Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), and Ensemble approaches, evaluating the impacts of various geoenvironmental factors, including anthropogenic influences, climate, soil properties, and topography. The study also employed Random Forest algorithms to investigate the spatiotemporal dynamics of forest vegetation cover in the Duhok district from 2013 to 2023, identifying significant transitions in dominant ecological drivers, such as shifts from altitude and rainfall towards soil moisture and groundwater levels. Additionally, the research explored future forest vegetation scenarios for the Duhok district using Ensemble Deep Learning (EDL) models integrated with Land Use and Land Cover (LULC) data and climate projections, predicting significant reductions in forest cover by 2060 under severe climate scenarios. The research leveraged data from multiple sources, including Landsat and Sentinel-2 satellite images, meteorological, geological, and field validation data. Hendaf Nasruldeen Habeeb's PhD work resulted in several peer-reviewed publications. Professor Yaseen T. Mustafa supervised this comprehensive research project.
Abdullah A. Abdullah dedicated the years from 2021 to 2024 to his doctoral research on managing uncertainty in deep learning models through a Bayesian approach. His work explored the adaptation of a non-Bayesian deep learning framework to incorporate Bayesian deep learning (BDL) by utilizing variational inference as an approximation technique. The research aimed to assess the confidence levels in predictions generated by deep learning models using BDL, with a particular focus on quantifying uncertainty in MLP-mixer architectures rather than merely achieving optimal performance. This comprehensive research effort resulted in the publication of four high-impact papers in prestigious journals, contributing significantly to the field of uncertainty quantification in deep learning. The project also benefited from a productive collaboration with Asst. Prof. Dr. Masoud M. Hassan, whose contributions enhanced the depth and impact of the research.
Kaiwan K. Fatah pursued his PhD research (2015–2019) on the integration of GIS and remote sensing to investigate the interplay between climate, land use/land cover (LULC), surface/groundwater sustainability, and natural disasters in the Akre District, Kurdistan Region, Iraq. His dissertation focused on achieving several key objectives: mapping and delineating flood-susceptible areas using remote sensing and GIS-AHP approaches; identifying landslide vulnerability zones and conducting risk assessments through GIS-based multi-criteria evaluation and frequency ratio analysis; mapping groundwater potential zones with the aid of machine learning algorithms applied to remotely sensed data; and exploring the relationships among LULC, climatic factors, and groundwater level fluctuations using GIS, remote sensing techniques, and deep learning approaches. The research culminated in the publication of four high-impact papers in prestigious, ranked journals, contributing significantly to the understanding of natural hazards and resource sustainability in the Akre District. This work was conducted in collaboration with Asst. Prof. Dr. Imaddadin O. Hassan, whose expertise greatly enriched the research outcomes.
Islam Sabah Khurshed successfully concluded his Master of Science project in the year 2024, with a focus on elucidating the pivotal role played by remote sensing technology in the evaluation and comprehension of the spatial and temporal fluctuations in surface soil moisture (SSM) and vegetation cover. Conducted over a twenty-year period from 2001 to 2021 within the Batifa region in the northern part of the Kurdistan Region of Iraq, his research leveraged Landsat satellite imagery to categorize land use and land cover (LULC) by employing a support vector machine (SVM) algorithm, and it systematically extracted soil classifications on an annual basis. The study was distinguished by its exceptional accuracy in LULC mapping, as demonstrated by the high values of overall accuracy and kappa coefficient. Furthermore, Khurshed innovated a random forest (RF) modeling technique for the estimation and mapping of SSM, which showcased considerable predictive precision. His research elucidated the interconnections between SSM and vegetation cover, revealing the influence of human activities on the ecological conditions in Batifa. The project, which benefited from the active collaboration of Assistant Professor Dr. Mohammed A. Fayyahd, highlighted the critical importance of remote sensing technologies in contributing to informed strategies for sustainable environmental management and the conservation of soil resources.
Halmat Sarbast Khalaf accomplished her MSc project in 2023, focusing on the crucial yet understudied domain of Soil Organic Matter (SOM) in Batifa, a district of Zakho in the Kurdistan region of Iraq. Her research utilized advanced machine learning algorithms, notably Random Forest (RF) and Extreme Gradient Boosting (XGBoost), to predict SOM content, leveraging a unique dataset that combined remotely sensed data from Landsat 8 imagery with field observations, including spectral bands and a Digital Elevation Model (DEM). Khalaf's study employed indices such as NDVI and SAVI, among others, to train and test the predictive models. Her findings highlighted the superior accuracy of the XGBoost algorithm in SOM prediction, with significant variables like SAVI and DEM playing a crucial role. This pioneering work, conducted in collaboration with Asst. Prof. Dr. Mohammed A. Fayyadh, offers invaluable insights for agricultural stakeholders in Batifa and similar regions, promoting sustainable practices and informed decision-making for soil conservation and agricultural productivity.
Wassfi H. Sulaiman completed his MSc in 2023, focusing on the critical evaluation of groundwater (GW) availability in the Zakho Basin-Iraqi Kurdistan Region. It is challenged by political complexities, environmental changes, and urban expansion. Utilizing the Analytic Hierarchy Process (AHP) alongside geospatial techniques, Sulaiman's study identified potential GW sites by integrating various criteria such as slope, flow accumulation, and rainfall, among others, into a composite index. The accuracy of his model was validated through significant statistical measures, highlighting areas with varying GW potential. His research culminates in the creation of a Groundwater Potential Index (GWPZ) map, offering invaluable insights for sustainable GW management. This framework not only aids local stakeholders in enhancing water resource utilization but also sets a precedent for addressing GW conservation in other drought-affected regions.
Bushra Shaban Yousif completed her MSc in 2023, exploring soil texture's impact on agriculture and land management in the Batifa, Zakho, Kurdistan Region of Iraq. Addressing the scarcity of digital soil maps, she utilized Landsat 8 imagery and a Digital Elevation Model (DEM) to estimate soil textures using machine learning models: Random Forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoost). Among these, XGBoost showed superior accuracy in predicting soil textures, particularly for clay, silt, and sand. Her analysis, based on 96 soil samples, revealed a predominance of loamy textures within the study area, contributing to a better understanding of soil heterogeneity. This research aids in creating digital soil maps, enhancing agricultural planning, and promoting sustainable soil management in regions with varied soil textures. This project was a collaborative effort with Asst. Prof. Dr. Mohammed A. Fayyadh.
Rebar T. Ali completed his PhD research from 2019 to 2022, focusing on the utilization of geoinformatics techniques for analyzing the environmental impact in Duhok Province, Kurdistan Region, Iraq. The primary objectives of his research included assessing changes in vegetation cover within the four districts of Duhok governorate from 2000 to 2019, using the modified soil-adjusted vegetation index (MSAVI2) in relation to elevation, slope, and aspect. Additionally, he investigated the relationship between spatiotemporal vegetation distribution and climatic factors. Furthermore, he assessed the vulnerability of land degradation (LD) in the Duhok district by integrating remote sensing, GIS, and the Analytic Hierarchy Process (AHP) approach. The research also aimed to determine the extent of LD in the area and identify its spatial distribution within the Duhok district. Moreover, the research aimed to spatially identify the distribution of vegetation and land surface temperature (LST) in the Duhok district and analyze the spatiotemporal relationship between LST and vegetation, as well as the relationship between LST and air temperature. Various data sources were employed in the research project, including satellite images from Landsat and Sentinel-2, meteorological data, geological data, and ground truthing data. As a result of his PhD research, Rebar T. Ali produced three publications. Professor Abdulla A. Omar actively participated in this collaborative project.
Mohammed H. Obeyed completed his PhD between 2016 and 2020, focusing on the establishment of allometric equations for determining the wet weight and dry weight of individual trees and tree stands. Throughout his research, he conducted various field surveys and engaged in laboratory work. Additionally, he utilized high-resolution satellite imagery (WorldView-2) to identify and delineate single trees or groups of trees. The primary outcomes of his study include the estimation of aboveground biomass and carbon sequestration for natural stands of Quercus Aegilops in Duhok Province. Furthermore, he successfully estimated and mapped the aboveground biomass of natural Quercus Aegilops using WorldView-3 imagery. We have been working on this together with Prof. Zaki M. Akrawee.
Shaheen Abdulkareem pursued her PhD research (2015–2019) on the intersection of spatial agent-based modeling (ABM) and artificial intelligence (AI). In her doctoral thesis, she concentrated on investigating the potential of AI-based learning techniques to enhance the decision-making capabilities of agents within ABMs. Shaheen conducted a comprehensive examination of the effects of various learning algorithms employed to augment social, spatial, and institutional intelligence, both at the individual and group levels. She applied these methodologies to two distinct spatial ABMs: a cholera diffusion disease model aimed at analyzing individual protective behavior, and a flood risk housing market model focused on studying expectation formation. Collaboratively, the research was conducted in conjunction with Prof. Tatiana Filatova and Assoc. Prof. Dr. E-W Augustijn.