This study aims to estimate the biomass and carbon sequestration potential of naturally occurring Calabrian pine (Pinus brutia Ten.) in the trunk and branches across three distinct sites in the Duhok Governorate, Kurdistan Region of Iraq. At each site, a total of 43–60 trees were measured to develop robust allometric models for estimating tree biomass and carbon storage, using diameter at breast height (DBH) and total tree height as key predictors. Site-specific variables such as elevation, slope, and soil characteristics, which can significantly influence biomass accumulation and carbon dynamics, were also taken into account. Tree selection encompassed a range of size classes and age groups to ensure a representative sampling of the natural forest structure and growth variability. Carbon content was estimated by applying a widely accepted carbon conversion factor of 0.5 to the calculated dry biomass. The findings reveal that trees in Zawita store substantially more carbon, with an average of 326.7 kg for a tree with a 40 cm DBH, compared to 123.6 kg in Atrosh and 157.9 kg in Belkef. The highest overall carbon stock in trunk and branches was recorded in Zawita, reaching up to 577.6 kg for a tree with 50 cm DBH, indicating superior sequestration capacity and higher biomass productivity relative to the other sites. These results carry important implications for forest management, carbon accounting, and climate change mitigation strategies in the region. Moreover, the study provides a valuable baseline for long-term ecological monitoring, sustainable forestry planning, and carbon stock assessments. By emphasizing the ecological and climate-related importance of natural pine forests in semi-arid mountainous environments, the research supports conservation initiatives and underscores the role of Pinus brutia as a vital carbon sink contributing to both regional sustainability and global climate goals.

This study explores the spatiotemporal patterns and drivers of deforestation and forest degradation along the politically sensitive Iraq–Turkey border within the Duhok Governorate between 2015 and 2024. Utilizing paired remote sensing (RS) and high-end machine learning (ML) methods, forest dynamics were simulated from Sentinel-2 imagery, climate datasets, and topographic variables. Seven ML models were evaluated, and XGBoost consistently outperformed the others, yielding predictive accuracies (R2) of 0.903 (2015), 0.910 (2019), and 0.950 (2024), and a low RMSE (≤0.035). Model interpretability was further improved through the application of SHapley Additive exPlanations (SHAP) to estimate variable contributions and a Generalized Additive Model (GAM) to elucidate complex nonlinear interactions. The results showed distinct temporal shifts; climatic factors (rainfall and temperature) primarily influenced vegetation cover in 2015, whereas anthropogenic drivers such as forest fires (NBR), road construction (RI), and soil exposure (BSI) intensified by 2024, accounting for up to 12% of the observed forest loss. Forest canopy cover decreased significantly, from approximately 630 km2 in 2015 to 577 km2 in 2024, mainly due to illegal deforestation, road network expansion, and conflict-induced fires. This study highlights the effectiveness of an ML-driven RS analysis for geoinformation needs in geopolitically complex and data-scarce regions. These findings underscore the urgent need for robust, evidence-based conservation policies and demonstrate the utility of interpretable ML techniques for forest management policy optimization, providing a reproducible methodological blueprint for future ecological assessment.

Changes in climate and land use/cover (LULC) become serious challenges and have a major impact on groundwater level (GWL) depletion in various regions worldwide. This research aimed to assess the impact of climate variations and LULC shifts on GWL in the Akre district, Kurdistan region of Iraq, through a novel Geo-informatics and machine learning (XGBoost) framework spanning 1990 to 2022. Data sources included hydro-climatic measurements from TerraClimate, Landsat 5 and 8 images for LULC mapping, and GWL records were considered. Spatial analysis of hydro-climatic variables was done using simple Kriging within a GIS framework, while LULC changes over time were analyzed with a random forest classifier. The research observed significant alterations in land classes, with increases in built-up, soil, rocks areas and decreases in vegetative, agriculture and water bodies, particularly in the northern region. The climate analysis indicated rising temperatures and evapotranspiration rates alongside falling precipitation, soil moisture, and runoff levels. The XGBoost model identified annual temperature as a critical factor in GWL reduction, although climate changes had a more substantial impact than LULC alterations. The model effectively predicted GWL trends, as evidenced by high accuracy (R2 of 0.83) and AUC-ROC of 0.87) and low error metrics (RMSE of 11.7 and MAE of 14.28). The GWL for the years 2010 and 2000 were predicted based on the GWL map generated for 2022. Throughout the study period, GWL exhibited a significant decline, with an average reduction exceeding 35 m. Specifically, depletion ranged from 10 to 15 m between 2000 and 2010, followed by a more pronounced decline of 25–30 m from 2010 to 2022. This depletion was most severe in the eastern and northern areas and was primarily driven by unfavorable climatic conditions and LULC changes. These insights are vital for local policymakers and stakeholders in managing groundwater resources sustainably and formulating appropriate water extraction guidelines.

This study examines the spatiotemporal dynamics of forest cover in Duhok district, Iraq, where urbanization, industrial activities, and agricultural expansion have significantly impacted the environment. Utilizing an Ensemble Deep Learning approach, this study integrated Land Use and Land Cover data and environmental factors (climate, topography, and geology) to model forest cover changes from 2000 to 2060. Land Use and Land Cover maps were generated using the Random Forest classifier, whereas future climate scenarios were based on Coupled Model Intercomparison Project Phase 6 projections. The findings of this study highlight the critical role of environmental factors in shaping forest cover distribution and demonstrate the efficacy of advanced deep learning techniques in environmental modeling. The analysis reveals a projected decline in forest cover from 364.17 km² in 2000 to 240.50 km² in 2022, with a further decrease to 156.93 km² by 2060 under the most severe climate scenario. These insights highlight the profound impact of environmental factors on forest cover and underscore the transformative potential of advanced deep learning techniques for predicting and understanding environmental changes. These findings offer a crucial roadmap for policymakers, equipping them with the tools needed to devise strategies that can mitigate the looming threats posed by climate change and human activities to these vital ecosystems.

Understanding the dynamic relationships between geoenvironmental factors and forest vegetation cover is crucial for effective forest management and planning. This study investigates the spatiotemporal dynamics of forest cover in the Duhok District in the Kurdistan Region of Iraq over a decade (2013–2023), emphasizing the impact of geoenvironmental factors via Random Forest algorithms and Landsat data. This research integrates datasets including fractional vegetation cover (FVC), groundwater levels, climate data, topography, and soil moisture data, offering a comprehensive analysis of the factors influencing forest cover. The results show that in 2013, altitude and rainfall were the primary factors influencing FVC, with areas of higher altitudes and adequate rainfall exhibiting up to 30% denser forest cover. By 2023, soil moisture and groundwater levels had emerged as the dominant factors, with soil moisture levels accounting for 25% of the variation in FVC. This shift underscores the increasing importance of water management strategies to maintain forest health. The Random Forest model demonstrated high predictive accuracy, achieving an R2 value of 0.918 (RMSE of 0.016 and MAE of 0.013) for 2013 and 0.916 (RMSE of 0.018 and MAE of 0.014) for 2023, underscoring the model’s robustness in handling nonlinear ecological processes. This study’s insights are crucial for guiding sustainable forest management practices and assisting decision-makers in formulating strategies for resource management, environmental preservation, and future planning. This study underscores the necessity of adaptive management strategies that consider evolving climatic and hydrological conditions, emphasizing continuous monitoring and advanced technologies to ensure the resilience of forest ecosystems.

Groundwater (GW) is a crucial and increasingly scarce natural resource, that is affected by climate change and mismanagement. To manage GW resources effectively, it is crucial to accurately identify GW potential zones (GWPZs) using modern techniques. This study aimed to employ and assess geoinformatics-based machine learning (ML) models to delineate GWPZs in the Akre district, Kurdistan region of Iraq. Six nonparametric ML models were used: a support vector machine (SVM), k-nearest neighbours (KNN), decision tree (DT), random forest (RF), gradient boost DT (GBDT), and extreme gradient boosting (XGBoost). These models were trained on diverse GWPZ-favourable influencing factors, encompassing topographic, hydrological, geological, and environmental aspects. The findings of this study revealed that the XGBoost model outperformed the other nonparametric models in terms of best-fit performance and accuracy in generating a GW potential map (GWPM), achieving a R2 of 0.88, a root mean square error (RMSE) of 11.348, and a mean absolute error (MAE) of 6.623. Notably, over half of the study area (53%) was categorised as having high or very high GWPZs, primarily in the low-lying Rovia Plain. The study identified rainfall, elevation, lineament density (LD), drainage density (Dd), topographic wetness index (TWI), and slope as the most significant factors influencing GWPZ modelling. This study provides a comprehensive framework for GW resource management, ecological conservation, and urban development planning. These insights are crucial for stakeholders, policymakers, and local authorities in strategic resource planning and environmental stewardship.

Global flood events, intensified by an intricate interplay of natural dynamics and human activities, pose severe environmental and safety threats, driving the demand for advanced vulnerability assessments. This research innovates with a holistic geo-informatics methodology, pinpointing flood-prone regions in Erbil City, Kurdistan Region, Iraq. It harnesses the Analytic Hierarchy Process and Geographic Information System synergies, enriched with insights from multispectral satellite observations, to conduct an exhaustive analysis of key flood susceptibility indicators. These encompass geological, hydrological, meteorological, and anthropogenic parameters, integrated into a robust Geographic Information System analytical matrix. Through strategic factor analysis and prioritization, the study distills a detailed flood susceptibility cartogram, classifying regional vulnerabilities into distinct risk echelons: high (encompassing 25%), moderate (38.1%), low (28.4%), and very low (8.5%). The final flood map highlights an acute susceptibility epicenter in the northeast, contrasting with the minimal threat southwest. This demarcation reaffirms the Analytic Hierarchy Process-Geographic Information System fusion's precision in translating multifaceted data into strategic intelligence, proving instrumental for scholars, administrative frameworks, and developmental strategists. The insights garnered are pivotal in sculpting informed urban development and flood counteractive blueprints, instrumental in mitigating flood repercussions and preserving ecological and human integrities.

Accurate estimation of forest vegetation cover and understanding of ecosystem factors are essential for effective forest management and planning. This study enhances forest vegetation cover precision in Duhok and Amedy districts, Kurdistan Region, Iraq, through deep learning (DL) models, including Convolutional Neural Networks, Deep Neural Networks (DNN), and an Ensemble approach. This study analyzed 12 ecosystem factors, utilizing 11 as inputs to predict forest cover, which served as the output variable. The Ensemble model was the most accurate, with coefficient of determination (R²) values of 0.790 and 0.740, root mean square error (RMSE) values of 0.037 and 0.057, and mean absolute error (MAE) values of 0.028 and 0.044 for the Duhok and Amedy districts, respectively. It outperformed the other models, highlighting the benefits of integrating multiple DL strategies. The results identified significant influences of anthropogenic factors, climate, and rock composition in Duhok and soil, climate, and topography in Amedy. This study highlights the superior capability of the Ensemble model in discerning key ecosystem drivers and underscores the importance of tailored environmental conservation strategies for regional sustainability. This study advocates a refined model-based approach in ecosystem management and planning.

Bayesian deep learning (BDL) has emerged as a powerful technique for quantifying uncertainty in classification tasks, surpassing the effectiveness of traditional models by aligning with the probabilistic nature of real-world data. This alignment allows for informed decision-making by not only identifying the most likely outcome but also quantifying the surrounding uncertainty. Such capabilities hold great significance in fields like medical diagnoses and autonomous driving, where the consequences of misclassification are substantial. To further improve uncertainty quantification, the research community has introduced Bayesian model ensembles, which combines multiple Bayesian models to enhance predictive accuracy and uncertainty quantification. These ensembles have exhibited superior performance compared to individual Bayesian models and even non-Bayesian counterparts. In this study, we propose a novel approach that leverages the power of Bayesian ensembles for enhanced uncertainty quantification. The proposed method exploits the disparity between predicted positive and negative classes and employes it as a ranking metric for model selection. For each instance or sample, the ensemble's output for each class is determined by selecting the top ‘k' models based on this ranking. Experimental results on different medical image classifications demonstrate that the proposed method consistently outperforms or achieves comparable performance to conventional Bayesian ensemble. This investigation highlights the practical application of Bayesian ensemble techniques in refining predictive performance and enhancing uncertainty evaluation in image classification tasks.

The accurate quantification of surface soil moisture (SSM) and vegetation cover using remote sensing techniques is essential for effective environmental management. This study investigated the spatial variations in SSM and vegetation cover in the Batifa region of the Kurdistan Region of Iraq. Landsat-8 images of the study area were classified using a support vector machine (SVM), and the soil land type was subsequently extracted. A random forest (RF) algorithm was developed to retrieve SSM using Landsat data in conjunction with in situ measurements. The results demonstrated that the RF algorithm achieved a high coefficient of determination ( R2=0.80 ) for the SSM retrieval. The study area exhibited distinct distributions of SSM and normalized difference vegetation index (NDVI) values across different ranges. The low range of SSM (2.21%–3.34%) and NDVI (−0.020–0.172) values occupied approximately 25% of the soil area, whereas the moderate range of SSM (3.34%–4.05%) and NDVI (0.172–0.238) values covered approximately 50% of the soil area. A high range of SSM (4.05%–6.49%) and NDVI (0.238–0.935) values was found in approximately 25% of the region. The southern part of Batifa experienced drought conditions, whereas the northern part exhibited higher SSM levels. Anthropogenic resources caused a decrease in vegetation and SSM in Batifa. These findings have significant implications for sustainable management of water and soil resources in the Batifa area.

The identification of land use/land cover (LULC) changes is important for monitoring, evaluating, and preserving natural resources. In the Kurdistan region, the utilization of remotely sensed data to assess the effectiveness of machine learning algorithms (MLAs) for LULC classification and change detection analysis has been limited. This study monitors and analyzes LULC changes in the study area from 1991 to 2021 using a quantitative approach with multi-temporal Landsat imagery. Five MLAs were applied: Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), and Extreme Gradient Boosting (XGBoost). The results showed that the RF algorithm produced the most accurate maps of the three-decade study period, accompanied by a high kappa coefficient (0.93–0.97) compared with the SVM (0.91–0.95), ANN (0.91–0.96), KNN (0.92–0.96), and XGBoost (0.92–0.95) algorithms. Consequently, the RF classifier was implemented to categorize all obtainable satellite images. Socioeconomic changes throughout these transition periods were revealed by the change detection results. Rangeland and barren land areas decreased by 11.33 % (−402.03 km2) and 6.68 % (−236.8 km2), respectively. The transmission increases of 13.54 % (480.18 km2), 3.43 % (151.74 km2), and 0.71 % (25.22 km2) occurred in agricultural land, forest, and built-up areas, respectively. The outcomes of this study contribute significantly to LULC monitoring in developing regions, guiding stakeholders to identify vulnerable areas for better land use planning and sustainable environmental protection.

Soil organic matter (SOM) is an essential component of soil fertility that plays a vital role in the preservation of healthy ecosystems. This study aimed to produce an SOM-level map of the Batifa region in northern Iraq. Random forest (RF) and extreme gradient boosting (XGBoost) models were used to predict the SOM spatial distribution. A total of 96 soil samples were collected from the surface layer (0–30 cm) of both cropland and soil areas in Batifa. In addition, remote sensing data were obtained from Landsat 8, including bands 1–7, 10, and 11. Supplementary variables such as the normalized difference vegetation index (NDVI), soil-adjusted vegetation index (SAVI), brightness index (BI), and digital elevation model (DEM) were employed as tools to predict SOM levels across the region. To evaluate the accuracy of the RF and XGBoost models in predicting SOM levels, statistical metrics, including mean absolute error (MAE), root mean square error (RMSE), and determination coefficient (R2), were used, with 80% of the data used for prediction and 20% for validation. The findings of this study revealed that the XGBoost model exhibited higher accuracy (MAE = 0.41, RMSE = 0.62, and R2 = 0.92) in predicting SOM than the RF model (MAE = 0.65, RMSE = 0.96, R2 = 0.79). Band 10, DEM, SAVI, and NDVI were identified as the most important predictors for both the models. The methodology employed in this study, which utilizes machine learning models, has the potential to map SOM in similar settings. Furthermore, the results offer significant insights for the stakeholders involved in soil management, thereby facilitating the enhancement of agricultural techniques.

Groundwater availability in the Zakho Basin faces significant challenges due to political issues, border stream control, climate change, urbanization, land use changes, and poor administration, leading to declining groundwater quantity and quality. To address these issues, this study utilized the Analytic Hierarchy Process (AHP) and geospatial techniques to identify potential groundwater sites in Zakho. The study assigned weights normalized through the AHP eigenvector and created a final index using the weighted overlay method and specific criteria such as slope, flow accumulation, drainage density, lineament density, geology, well data, rainfall, and soil type. Validation through the receiver operating characteristic (ROC) curve (AUC = 0.849) and coefficient of determination (R2 = 0.81) demonstrated the model’s accuracy. The results showed that 17% of the area had the highest potential as a reliable groundwater source, 46% represented high-to-moderate potential zones, and 37% had low potential. Flat areas between rivers and high mountains displayed the greatest potential for groundwater development. Identifying these potential sites can aid farmers, regional planners, and local governments in making precise decisions about installing hand pumps and tube wells for a regular water supply. Additionally, the findings contribute to the development of a sustainable groundwater management plan, focusing on improving water usage and protecting water-related ecosystems in the region. Identification of the optimum influencing factors, arrangement of the factors in a hierarchy, and creation of a GWPI map will allow further planning for groundwater preservation and sustainability. This project can be conducted in other areas facing droughts.

Soil texture is a key physical property that has a large effect on several other soil properties that are important for managing and planning agriculture. In the Batifa region (Iraq), there is currently no digital soil map on a moderate scale available. In this regard, remotely sensed data can aid in mapping soil-texture fractions. The purpose of this study was to evaluate the performance of three different machine-learning (ML) models (random forest [RF], support vector regression [SVR], and extreme gradient boosting [XGBoost]) to spatially estimate soil-texture classes using Landsat 8 and a digital elevation model (DEM). To this end, 96 soil samples with a surface layer 0–30-cm deep were collected to estimate their soil-texture frac-tions using 19 variables. These comprised Landsat Spectral Bands 1–7, 10, and 11, the Normalized Difference Soil Index, Enhanced Vegetation Index, Simple Soil Ratio Clay Index, Brightness Index, Grain Size Index, Normalized Difference Vegetation Index, Normalized Difference Sand Index, Soil Adjusted Vegetation Index, Landsat Bareness Index, and DEM. High coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE) values were used to select the best ML model. The results of this study showed that the XGBoost model estimated the soil-texture fractions (clay: R2 = 90%, silt: R2 = 85%, and sand: R2 = 91%) better than the RF (clay: R2 = 86%, silt: R2 = 66%, and sand: R2 = 86%) and SVR (clay: R2 = 72%, silt: R2 = 56%, and sand: R2 = 86%) models. The best estimators of soil-texture fractions were the GSI for clay, the DEM for silt, and Band 10 for sand, followed by other variables based on satellite data. In general, however, the DEM is considered a good estimator for all soil-texture fractions. These findings will help support techniques for managing soil in places where the surface soil has different textures.

Convolutional neural networks (CNNs) have become a popular choice for various image classification applications. However, the multi-layer perceptron mixer (MLP-Mixer) architecture has been proposed as a promising alternative, particularly for large datasets. Despite its advantages in handling large datasets and models, MLP-Mixer models have limitations when dealing with small datasets. This study aimed to quantify and evaluate the uncertainty associated with MLP-Mixer models for small datasets using Bayesian deep learning (BDL) methods to quantify uncertainty and compare the results to existing CNN models. In particular, we examined the use of variational inference and Monte Carlo dropout methods. The results indicated that BDL can improve the performance of MLP-Mixer models by 9.2 to 17.4% in term of accuracy across different mixer models. On the other hand, the results suggest that CNN models tend to have limited improvement or even decreased performance in some cases when using BDL. These findings suggest that BDL is a promising approach to improve the performance of MLP-Mixer models, especially for small datasets.

Landslides are among the most critical geo-environmental disasters for both humans and the environment. This study intended to create a landslide susceptibility map (LSM) for the Akre District of Kurdistan Region, Northern Iraq. In this paper, 15 landslide causative criteria including elevation, slope, curvature, aspect, topographic wetness index, topographic roughness index, steam power index, lithology, lineament density, soil types, land use/cover, normalized difference vegetation index, distance to roads, rainfall, and distance to streams were analysed. For this purpose, frequency ratio (FR), analytic hierarchy process (AHP), and ensemble FR-AHP models based on geoinformatics techniques were applied. The LSMs were then assigned into five categories based on their susceptibility levels: very low, low, medium, high, and very high. The results showed that the high and very high landslide susceptibility categories for the FR, AHP, and hybrid FR-AHP models were 27.75% (732,01 km2), 28.44% (750,36 km2), and 28.84% (761 km2), respectively. The results showed that the majority of historical landslide incidents occurred in mountainous terrain in the northern parts of the study area, which are classified as high and very high susceptibility zones. The predicted rate curves for the FR, AHP, and hybrid FR-AHP models had areas under curve of the receiver operating characteristic curve (AUC-ROC) values of 93.4%, 89.3%, and 93.8%, respectively, which indicate that the ensemble FR-AHP model provides more reliable and accurate results for LSM. The LSM generated via the hybrid FR-AHP model can be utilised by local authorities, managers, and decision-makers in further land use/cove planning to mitigate the devastating influences of landslides in the area.

The temperature rise has become a serious environmental concern affected by both human and natural factors. Worldwide, rising land surface temperatures have emerged as the most pressing issue facing the twenty-first century. In the last two decades, a curious change was realized in temperature in the Duhok district of Iraq. Hence, this study examined the spatiotemporal land surface temperature (LST) distribution and Modified Soil Adjusted Vegetation Index (MSAVI2) and the correlation between them in the Duhok district in three different years 2001, 2011, and 2021 using Landsat satellite images. Air temperature data from seven weather stations were used to validate the LST results. The study's findings revealed that the Duhok district’s LST has risen during the study period. In general, the average LST has been increasing at a rate of 0.15 °C per year. Other findings showed that the vegetation cover of the Duhok district has changed dynamically. In all three years of study, the regression analysis results indicated that there was a negative correlation between LST and MSAVI2. This method of evaluation will be useful in guiding future urban management work and local government strategies.

In recent decades, floods have been the most common, complex, and destructive natural calamities worldwide. Hence, for inclusive flood risk assessment, creating flood susceptibility mapping to demarcate flood-vulnerable zones is fundamental for decision makers. To assess flood-prone locations in the Akre, Iraqi Kurdistan Region, fundamental susceptibility mapping was undertaken using geographic information systems, remote sensing, and an analytic hierarchy process model. To assess flood susceptibility, the geographic information systems framework used 15 ideal causative factors for flooding: altitude, slope, distance to streams, flow accumulation, drainage density, rainfall, soil type, lithology, curvature, topographic wetness index (TWI), topographic roughness index, stream power index, stream transport index, land use/land cover, and normalized difference vegetation index. The factors contributing to flooding were optimally weighted with respect to the proposed model. The final flood susceptibility map was reclassified into five different classes of susceptibility to flooding: very low (16.64% of the study area); low (19.53%); moderate (38.92%); high (17.83%); and very high (7.08%). The area under the curve values for the predicted rate and success rate of the AHP model were 0.956 and 0.971, respectively. Therefore, the results were accurate and reliable. The AHP model is a powerful method for fundamental susceptibility mapping to mitigate the serious impacts of flooding and assist scholars, local governments, and policymakers in future master planning.

The resection in 3D space is a common problem in surveying engineering and photogrammetry based on observed distances, angles, and coordinates. This resection problem is nonlinear and comprises redundant observations which is normally solved using the least-squares method in an iterative approach. In this paper, we introduce a vigorous angular based resection method that converges to the global minimum even with very challenging starting values of the unknowns. The method is based on deriving oblique angles from the measured horizontal and vertical angles by solving spherical triangles. The derived oblique angles tightly connected the rays enclosed between the resection point and the reference points. Both techniques of the nonlinear least square adjustment either using the Gauss-Newton or Levenberg – Marquardt are applied in two 3D resection experiments. In both numerical methods, the results converged steadily to the global minimum using the proposed angular resection even with improper starting values. However, applying the Levenberg – Marquardt method proved to reach the global minimum solution in all the challenging situations and outperformed the Gauss-Newton method.

In the last decade, Deep Learning (DL) has revolutionized the use of artificial intelligence, and it has been deployed in different fields of healthcare applications such as image processing, natural language processing, and signal processing. DL models have also been intensely used in different tasks of healthcare such as disease diagnostics and treatments. Deep learning techniques have surpassed other machine learning algorithms and proved to be the ultimate tools for many state-of-the-art applications. Despite all that success, classical deep learning has limitations and their models tend to be very confident about their predicted decisions because it does not know when it makes mistake. For the healthcare field, this limitation can have a negative impact on models predictions since almost all decisions regarding patients and diseases are sensitive. Therefore, Bayesian deep learning (BDL) has been developed to overcome these limitations. Unlike classical DL, BDL uses probability distributions for the model parameters, which makes it possible to estimate the whole uncertainties associated with the predicted outputs. In this regard, BDL offers a rigorous framework to quantify all sources of uncertainties in the model. This study reviews popular techniques of using Bayesian deep learning with their benefits and limitations. It also reviewed recent deep learning architecture such as Convolutional Neural Networks and Recurrent Neural Networks. In particular, the applications of Bayesian deep learning in healthcare have been discussed such as its use in medical imaging tasks, clinical signal processing, medical natural language processing, and electronic health records. Furthermore, this paper has covered the deployment of Bayesian deep learning for some of the widespread diseases. This paper has also discussed the fundamental research challenges and highlighted some research gaps in both the Bayesian deep learning and healthcare perspective.

Scientific and academic researches and studies trying to present a multi-range of techniques and methods focusing on groundwater pollution, potentials, assessment, and prediction, Groundwater is the most important resource of fresh water now and many researchers trying to cover all about this resource to get sustainable development. This review aims to create an overview of groundwater analysis and forecasting methods. The study is based on the need to select and group research papers into best-defined methodological categories. The article gives an overview of recent advancements in groundwater potential zone analysis approaches, as well as ongoing research objectives based on that overview. This review has overviewed papers and researches been published last decade 2010 -2020 have been done depending on the data sources from the global online database, which could obtain many papers and research studying the groundwater potential zones and other aspects related to groundwater. The aim of reviewing multiple types of research and papers on determining groundwater potential zones by applying the best techniques and selecting the most suitable factors that affect groundwater potential zones.

Land degradation is a complex process and significant environmental problem affected by both natural and anthropogenic driving factors. Globally, the prevention of land degradation has become one of the most significant challenges of the twenty-first century. Over the last decade, a marked change was observed in land degradation in the Duhok district of Iraq. Geographic Information System (GIS), Multispectral Remote Sensing satellite image, and Analytic Hierarchy Process are efficient tools for modelling and assessing the risks of land degradation. To analyze land degradation, in this study, various physical and human-induced factors were used in the GIS environment; slope, elevation, drainage, precipitation, geology, vegetation coverage, and land use/land cover. The data were analyzed and weights assigned to each factor using the analytic hierarchy process and mapped using GIS techniques, resulting in a land degradation map. Field observations were carried out to better understand the degree of various factors that contribute to land degradation by using recent technologies and a global positioning system in the study area. The results show that 11.74% and 28.69% of the study area were affected by high and moderate levels of land degradation, respectively, while 34.77% and 24.79% of the study area experienced slight and no degradation, respectively. Slope gradient, rainfall, and distribution of vegetation were identified as the primary causes of land degradation in the area. This method of evaluation is intended to be beneficial to governments and researchers.

This process of industrial zones site selection requires to fulfill requirements and standards, simultaneously reducing environmental and public health costs and maximizing technical and economic benefits. A geographic information system was used in this study to evaluate land suitability to propose suitable sites for industrial zones in Duhok district, Kurdistan Region of Iraq. The defined parameters were classified into two groups, specifically, environmental and socioeconomic criteria. The information layers were prepared and were standardized into Boolean logic. The parameters maps were combined using the Boolean logic technique, and the suitability map was achieved on two classes including unsuitable and suitable. The results showed that only about 10% (104 Km2) of the study area was evaluated as suitable areas for industrial zones, which are mainly found in the north, northwest, and central parts of the study area. The results also illustrated that most of the industries within the study area have been located in inappropriate sites which have had a severe impact on the air quality of Duhok city. This paper would present a solution for industrial site selection in Duhok and it is helpful for regional planning of the country.

The concept of ecosystem service (ES) was originally developed to illustrate the benefits that natural ecosystems generate for society and to raise awareness for biodiversity and ecosystem conservation. In recent years, geographical information system (GIS) has become a powerful tool for mapping (ES) within a landscape, which visualizes spatial and temporal patterns and changes in ecosystems and their services.Mapping (ES) is necessary for the progress of strategies that will guarantee their future supply and to support the policies in a more effective way. The comprehensive literature review were conducted from international databases such as Elsevier, Springer, Wiley, and Google Scholar. We used the key terms including ‘mapping’, ‘maps’, ‘ES or ecosystem service, ‘ecosystem functions’, ‘landscape functions’, ‘evaluation of ES’, and ‘assessment of services’. in order to identify mapping ecosystem services and their challenges and opportunities. In total, 65 research papers were found firstly, which 34 of them were selected for reviewing. The most mportant challenges are insuffieicnet generation of ES in the context of managed systems, need to estimate associations among indicators of (ES) incomplete understanding of the nature of associations among services, and the lack of a general numerical outline to address these relations.

The integration of remote sensing techniques and Geographic Information System has a wide use to quantify the spatial and temporal distribution of vegetation cover. Over the last decade, a remarkable change was noticed in both climate and vegetation cover in Duhok. The Modified Soil Adjusted Vegetation Index (MSAVI2) was extracted from Landsat satellite images over the 20 years (2000 to 2019). For analyzing the vegetation changes, the terrain data including elevation, slope, and aspect and climate data temperature and precipitation are used. The result shows that from 2000–2019, the average mean MSAVI2 is 0.361 and the trend increased in 77.9% of the study area. The northern and northeastern areas of the study area revealed a significant increase in vegetation, while in the low land areas it is decreased. The amount of precipitation and temperature degree affect the spatiotemporal distribution of vegetation cover. The MSAVI2 showed a positive relationship with precipitation and temperature. At elevation less than 2000 m, with increasing elevation the MSAVI2 is increasing, but when the elevation reaches 2000 m, the MSAVI2 is decreasing and negatively related to elevation. The vegetation has a positive relation with slopes less than 45°, and at slopes higher than 45°, the MSAVI2 is decreased. The impact of aspect on the vegetation figured out that the largest MSAVI2 is detected in the shady slope due to relatively less evapotranspiration.

The rapidly and wide use of remote sensing and accurately obtain information on the spatiotemporal distribution of large-scale vegetation is of great significance for improving and managing the Environment. To assess and analyze the spatiotemporal variation of vegetation status in Kurdistan Region of Iraq (KRGI), we used time series NDVI-based vegetation that are extracted from MOD13Q1 MODIS product over 20 years (2000 - 2019). The results showed that vegetation was mainly distributed in the north-east to south-east of the KRGI, while west region has less distributed and almost no vegetation. This is clearly remarkable in the south-west part of the region (Garmian administration). While, the most dominated vegetation province was Duhok province in KRGI during study period. There is a noticeable temporal variation in vegetation over a period of 20-year in the KRGI. The lower vegetated cover area is observed in the years 2000, 2008, and 2009. The increase/decrease of vegetated cover area is not only effected by climate conditions. The anthropogenic resource is also one of the main resources that has a major influence on the increase/decrease of vegetation.

The study was aimed to develop above ground biomass(AGB) and its component models for an individual tree and stand of Quercus aegilops The benefits of this study are to know the amount of forest biomass to help us to estimate the amount of the lost or emitted carbon during deforestation and will give a clear idea of the forest capacity in capturing and storing carbon C in the forest ecosystem.. The study was conducted in six locations in the northeastern Duhok province in Kurdistan region of Iraq. Twenty-one trees were selected according to their diameter classes and felled to measure fresh weight (FW) and dry weight (DW) from different organs, including (stem, branches, leaves, and whole tree). In addition to, all trees with diameter at breast height (D ≥5 cm) in 89 plots of 0.04 ha, each was measured. Allometric equations of individual trees were used for estimating AGB and its component depending on D only. The DW of AGB and its components were converted into C by multiplying it on a half. The AGB estimated for FW and DW of the entire study area are 157.5285, 115.1153 (Mg ha-1) respectively. The results showed that Csequestration in the stands for stems, branches, leaves and the whole tree are 146.005, 73.9333, 31.38121and 249.9924 Mg ha-1 respectively.

Modern societies are exposed to a myriad of risks ranging from disease to natural hazards and technological disruptions. Exploring how the awareness of risk spreads and how it triggers a diffusion of coping strategies is prominent in the research agenda of various domains. It requires a deep understanding of how individuals perceive risks and communicate about the effectiveness of protective measures, highlighting learning and social interaction as the core mechanisms driving such processes. Methodological approaches that range from purely physics-based diffusion models to data-driven environmental methods rely on agent-based modeling to accommodate context-dependent learning and social interactions in a diffusion process. Mixing agent-based modeling with data-driven machine learning has become popularity. However, little attention has been paid to the role of intelligent learning in risk appraisal and protective decisions, whether used in an individual or a collective process. The differences between collective learning and individual learning have not been sufficiently explored in diffusion modeling in general and in agent-based models of socio-environmental systems in particular. To address this research gap, we explored the implications of intelligent learning on the gradient from individual to collective learning, using an agent-based model enhanced by machine learning. Our simulation experiments showed that individual intelligent judgement about risks and the selection of coping strategies by groups with majority votes were outperformed by leader-based groups and even individuals deciding alone. Social interactions appeared essential for both individual learning and group learning. The choice of how to represent social learning in an agent-based model could be driven by existing cultural and social norms prevalent in a modeled society.

Machine learning (ML) algorithms steer agent decisions in agent-based models (ABMs), serving as a vehicle for implementing behaviour changes during simulation runs. However, when training an ML algorithm, obtaining large sets of micro-level human behaviour data is often problematic. Information on human behaviour is often collected via surveys of relatively small sample sizes. This paper presents a methodology for training a learning algorithm to guide agent behaviour in a spatial ABM using a limited survey data sample. We apply different implementation strategies using survey data and Bayesian networks (BNs). By being grounded in probabilistic directed graphical models, BNs stand out among other learning algorithms in that they can be based on expert knowledge and/or known datasets. This paper presents four alternative implementations of data-driven BNs to support agent decisions in a spatial ABM. We differentiate between training BNs prior to, or during the simulation runs, using only survey data or a combination of survey data and expert knowledge. The four different implementations are then illustratedusing aspatialABMofcholeradiffusion for Kumasi,Ghana. The results indicate that a balance between expert knowledge and survey data provides the best control over the learning process of the agents and produces the most realistic agent behaviour.

Background Millions of people worldwide are exposed to deadly infectious diseases on a regular basis. Breaking news of the Zika outbreak for instance, made it to the main media titles internationally. Perceiving disease risks motivate people to adapt their behavior toward a safer and more protective lifestyle. Computational science is instrumental in exploring patterns of disease spread emerging from many individual decisions and interactions among agents and their environment by means of agent-based models. Yet, current disease models rarely consider simulating dynamics in risk perception and its impact on the adaptive protective behavior. Social sciences offer insights into individual risk perception and corresponding protective actions, while machine learning provides algorithms and methods to capture these learning processes. This article presents an innovative approach to extend agent-based disease models by capturing behavioral aspects of decision-making in a risky context using machine learning techniques. We illustrate it with a case of cholera in Kumasi, Ghana, accounting for spatial and social risk factors that affect intelligent behavior and corresponding disease incidents. The results of computational experiments comparing intelligent with zero-intelligent representations of agents in a spatial disease agent-based model are discussed. Methods We present a spatial disease agent-based model (ABM) with agents’ behavior grounded in Protection Motivation Theory. Spatial and temporal patterns of disease diffusion among zero-intelligent agents are compared to those produced by a population of intelligent agents. Two Bayesian Networks (BNs) designed and coded using R and are further integrated with the NetLogo-based Cholera ABM. The first is a one-tier BN1 (only risk perception), the second is a two-tier BN2 (risk and coping behavior). Results We run three experiments (zero-intelligent agents, BN1 intelligence and BN2 intelligence) and report the results per experiment in terms of several macro metrics of interest: an epidemic curve, a risk perception curve, and a distribution of different types of coping strategies over time. Conclusions Our results emphasize the importance of integrating behavioral aspects of decision making under risk into spatial disease ABMs using machine learning algorithms. This is especially relevant when studying cumulative impacts of behavioral changes and possible intervention strategies.

Pinus brutiaTen. is the most important coniferous tree species grow naturally in some mountains of Iraqi Kurdistan Region. It is well adapted to both climatically and soil conditions. The diameter at breast height (DBH) is the significant character which has a major effect on both volume, and height of the tree, and indirectly with the site index. The purpose of this study was to utilize remote sensing techniques in estimating the DBH. This is performed by developing models between the collected DBH measurements from the field and the extracted tree crown area from high resolution satellite imagery (WorldView-2). Five models were developed and the selection of the best model was based on coefficient of determination (R 2 ), Root Mean Square Error (RMSE), Bias, and Accuracy criteria. Accordingly, the model DBH= 1.2768 TCA − 14.381 was the most appropriate model to estimate the DBH for Pinus brutia Ten. in this study.

Estimating fire damage in the forest areas is an important task and a main target for the forestry directorate. This paper presents an approach for estimating burned forest areas in Duhok Governorate that caused by Turkish jet attacks on July 2015. The presented approach is based on using Landsat Data Continuity Mission LDCM (Landsat-8) satellite images. The spectral bands of Landsat-8 imagery were analyzed and the most sensitive bands to the burned areas were identified. Moreover, spectral indices: Normalized difference vegetation index (NDVI) and Normalized Burned Ratio (NBR) were used to identify and determine the burned areas. As a result, a map of burned areas was created with 8981 hectares of the total areas. This approach can be applied for the whole Kurdistan region over multi-temporal interval to estimate burned areas that happened either by Turkish jet attack or by other sources.

Forest sustainability requires an effective programme such as monitoring, managing, analyzing and classifying tree species through collecting required information. For such a purpose, field-based assessment provides the needed information. However, this method is costly, time consuming, and therefore assessment frequency is low. This often allows undesirable forest information to develop that do not coincide with management objectives. Satellite remote sensing and its techniques provide an efficient methodology and potentially low-cost alternative to field-based assessment. However, these techniques require the development of methods; such as semi-automatic tree detection and classification. These methods help to easily and accurately extract the required information in a fastest period. In this study, we investigated the potential of the newly developed high resolution satellite sensor, 8-band WorldView-2 (WV-2) imagery for identifying and mapping tree species in the Zawita sub-district, Duhok, Kurdistan region-Iraq. We performed object-based classification method to identify and map of Sixteen tree species including: calabrain pine (Pinus brutia), Almand (prunus duclis), Azarole hawthorn (Crataegus azarolus), Judas tree (cercis siliquastrum L), oriental plane (platanus orientalis), white poplar, silver (populus alba), white willow (Salix alba), Valonia Oak (Quercus aegilops), Gall Oak (Quercus Infectoria), common walnut (juglans Nigra), chinaberry (Melia azedarach), Tera Binth (pistacia khinjuk Stocks), syrian ash (fraxinus syriaca), White Mulberry (Morus alba), common fig (ficus carica), oleaster (Elaeagnus angustifolia). The accuracy assessment is achieved based on the random selection of validation samples, which showed an overall classification accuracy of 77% with Kappa coefficient of 69%. Based on the results of this study we concluded that the WV-2 sensor; with high spatial resolution and additional bands (coastal, yellow, red-edge and NIR2), is attributed as a proper satellite imagery for forest trees classification.

Countries in Middle East and North Africa (MENA region) are considered as arid and semi-arid areas that are suffering from water scarcity. They are expected to have more water shortages problem due to climatic change. Iraq is located in the Middle East covering an area of 433,970 square kilometers populated by 31 million inhabitants. One of the solutions suggested to overcome water scarcity is Rainwater Harvesting (RWH). In this study Macro rainwater harvesting technique had been tested for future rainfall data that were predicted by two emission scenarios of climatic change (A2 and B2) for the period 2020-2099 at Sulaimaniyah Governorate north east of Iraq. Future volumes of total runoff that might be harvested for different conditions of maximum, average, and minimum future rainfall seasons under both scenarios (A2 and B2) were calculated. The results indicate that the volumes of average harvested runoff will be reduced when average rainfall seasons are considered due to the effect of climatic change on future rainfall. The reduction reached 10.82 % and 43.0% when scenarios A2 and B2 are considered respectively.

Rainfall is a significant portion of hydrologic data. Rainfall records, however, are often incomplete due to several factors. In this study, the inverse distance weighting (IDW) method integrated with GIS is used to estimate the rainfall distribution in Duhok Governorate. A total of 25 rain fall stations and rainfall data between 2000 and 2010 were used, where 6 rainfall stations were used for cross-validation. In addition, the relationship between interpolation accuracy and two critical parameters of IDW (Power α value, and a radius of influence) was evaluated. Also, the rainfall distribution of Duhok Governorate was classified. As an output of this study and in most cases, the optimal parameters for IDW in interpolating rainfall data must have a radius of influence up to (15 - 60 km). However, the optimal α values varied between 1 and 5. Based on the results of this study, we concluded that the IDW is an appropriate method of spatial interpolation to predict the probable rainfall data in Duhok Governorate using α = 1 and search radius = 105 km for all the 25 rainfall stations.

Kurdistan Region (KR) of Iraq has suffered from the drought period during the seasons 2007- 2008 and 2008-2009 that affected the human and economic activities of the region. Macro rainwater harvesting (Macro RWH) is one of the techniques that can ensure water availability for a region having limited water resources. This technique is based on Soil Conservation ServiceCurve Number (SCS-CN) method and the Watershed Modeling System (WMS) was used to estimate the runoff. Rainfall records of Sulaymaniyah area for the period 2002-2012 were studied and an average season was selected (2010-2011). The results of the application of the WMS model showed that about 10.76 million cubic meters could be harvested. The results also showed that the quantity of the harvested runoff was highly affected by rainfall depth, curve number values, antecedent moisture conditions (AMC) and the area of the basins.

Macro rainwater harvesting techniques (Macro RWH) are getting more popular to overcome the problem of water scarcity in arid and semi-arid areas. Iraq is experiencing serious water shortage problem now despite of the presence of Tigris and Euphrates Rivers. RWH can help to overcome this problem. In this research, RWH was applied in Koya City in its districts, North West Iraq. Twenty-two basins were identified as the catchment area for the application of RWH technique. Watershed modeling system (WMS), based on Soil Conservation Service-curve number (SCS-CN) method, was applied to calculate direct runoff from individual daily rain storm using average annual rainfall records of the area. Two consecutive adjustments for the curve number were considered. The first was for the antecedent moisture condition (AMC) and the second was for the slope. These adjustments increased the total resultant harvested runoff up to 79.402 × 106 m3. The average percentage of increase of harvested runoff volume reached 9.28%. This implies that water allocation is of the order of 2000 cubic meter per capita per year. This quantity of water will definitely help to develop the area.

Canopy leaf area index (LAI) is a quantitative measure of canopy foliar area. LAI values can be derived from Moderate Resolution Imaging Spectroradiometer (MODIS) images. In this paper, MODIS pixels from a heterogeneous forest located in The Netherlands were decomposed using the linear mixture model using class fractions derived from a high-resolution aerial image. Gaussian Bayesian networks (GBNs) were applied to improve the spatio-temporal estimation of LAI by combining the decomposed MODIS images with a spatial version of physiological principles predicting growth (3PG) model output at different moments in time. Results showed that the spatial-temporal output obtained with the GBN was 40% more accurate than the spatial 3PG, with a root-mean-square error below 0.25. We concluded that the GBNs improved the spatial estimation of LAI values of a heterogeneous forest by combining a spatial forest growth model with satellite imagery.

The use of satellite remote sensing (RS) has salient progress in water budget calculation and it performs in watershed management. An RS technique can properly enhance hydrogeologic surveys. Moreover, to have an intimate understanding of the changes in water level fluctuations, it is also important to relate them to the surrounding eomorphic, structural, climatic and geologic factors. This research serves twofold. The first one is to operationalize the use of RS and Geographic Information Systems (GIS) techniques to assess the change of water surface area in Duhok dam located in Duhok city, Kurdistan region-Iraq. The second is to present and interpret the available statistical data on water level fluctuations in Duhok dam. The change of water surface in the Duhok dam is examined over a 11 year period using satellite images taken between 2001 and 2012. Three Landsat Enhanced Thematic Mapper Plus (Landsat 7 ETM+) images acquired on 13 June, 2001, 11 June, 2006 and on 11 June, 2012, respectively, were used. The change is tracked from the images using Band 7 with the help of Normalized Difference Water Index (NDWI). The accuracy assessed by using the Normalized Difference Area index (NDAI), and the change in water surface area analysed by comparing it with the related meteorological data of the dam. Results show that the estimated water surface area by RS matches the one on the ground with small relative error (less than 2.15%). A decrease of slightly more than 23% was observed in the water surface area this 11 year period. In addition, over this time period, climate conditions (rainfall, temperature and evaporation) in the study area have been changed significantly. These changes could have affected the reservoir surface area, but so also could external human interference around the dam.

Macro Rainwater Harvesting (RWH) has been tested at Koysinjaq (Koya) District, Kurdistan region of Iraq, due to its limited source of water.The studied area consists of four basins with total area of 228.96 km2. The estimating volumes of harvested runoff for the four selected basins together for the study period (2002-2011) were calculated using the Watershed Modeling System (WMS) which is based on Soil Conservation Service Curve Number (SCS-CN) method.In this research, a comparison between maximum and minimum rainfall seasons was conducted to give better understanding for the events that is governing the harvested runoff collection.The results show that, the total harvested runoff ranged from14.83 to 80.77(*106 m3) from the four selected basins together. This indicates that the technique of Macro RWH can be considered to provide a new source of water to contribute to reduce the problem of water scarcity.

Biophysical parameter values such as LAI have proved useful in a number of environmental applications. An approach is presented for producing the spatio-temporal estimation of leaf area index (LAI) of a heterogeneous forest using Moderate Resolution Imaging Spectroradiometer (MODIS) satellite images. This is performed by decomposing MODIS LAI for a heterogeneous forest using the Linear Mixture Model (LMM) and the information about the class fraction from an aerial image. Results showed that the decomposed MODIS LAI values were estimated well with maximum and minimum RMSE of 0.37, and 0.17, respectively. We concluded that our approach can be used to decompose MODIS LAI successfully for any heterogeneous forest.

The rapid urban development in the Duhok city since the 1990s has dramatically enhanced the potential impact of human activities. To identify and monitor this urban development effectively, remote sensing provides a viable source of data from which updated land cover information can be extracted efficiently and cheaply. In this study, three satellite datasets, Landsat Thematic Mapper (Landsat TM), and two Landsat Enhanced Thematic Mapper Plus (Landsat 7 ETM+), acquired during 1989, 2001 and 2012, respectively, were used to detect and evaluate Duhok’s urban expansion. Two change detection techniques were tested to detect areas of change. The techniques considered were image differencing, and post-classification comparison. The land use/land cover (LULC) maps of the years 1989, 2001 and 2012 were produced and the changes were determined with significant accuracies. The simplicity of our methods and the minimal investment of time and money make incorporation of remotely sensed data into urban growth a potentially powerful tool for the urban planner/manager. Moreover, the understanding of the spatial and temporal dynamics of Duhok’s urban expansion is the cornerstone for formulating a view about the future urban uses and for making the best use of the limited resources that are available.

The leaf area index (LAI) is a biophysical variable related to atmosphere-biosphere exchange of CO2. One way to obtain LAI value is by the Moderate Resolution Imaging Spectroradiometer (MODIS) biophysical products. In this paper, we use this product to improve the physiological principles predicting growth model within a Gaussian Bayesian network (GBN) setup. The MODIS time series, however, contains gaps caused by persistent clouds, cloud contamination, and other technique problems. We used the Expectation Maximization (EM) algorithm to estimate these missing values. During a period of 26 successive months, the EM algorithm is applied to four different cases: successively and not successively missing values during two different winter seasons, successively and not successively missing values during one spring season, and not successively missing values during the full study. Results show that the maximum value of the averaged absolute error between the original values and those estimated equals 0.16. This low value indicates that the estimated values well represent the original values. Moreover, the root mean square error of the GBN output reduces from 1.57 to 1.49 when performing the EM algorithm to estimate the not successively missing values. We conclude that the EM algorithm within a GBN can adequately handle missing MODIS LAI values and improves the estimation of the LAI.

Estimating the contribution of forests to carbon sequestration is commonly done by applying forest growth models. Such models inherently use field observations, such as leaf area index ( LAI ), whereas relevant information is also available from remotely sensed images. The purpose of this study is to improve the LAI estimated from the Physiological Principles Predicting Growth ( 3 -PG ) model by combining its output with LAI derived from Advanced Spaceborne Thermal Emission and Reflection Radiometer ( ASTER ) satellite imagery. A Bayesian network ( BN ) approach is proposed to take care of the different structure of the inaccuracies in the two data sources. It addresses the bias in the 3 -PG model and the noise of the ASTER images. Moreover, the EM algorithm is introduced into BN to estimate missing the LAI ASTER data, since they are not available for long time series due to the atmospheric conditions. This paper shows that the outputs obtained with the BN were more accurate than the 3 -PG estimate, as the root mean square error reduces to 0.46, and the relative error to 5.86 percent. We conclude that the EM -algorithm within a BN can adequately handle missing LAI ASTER values, and BN s can improve the estimation of LAI values. Ultimately, this method may be used as a predicting model of LAI values, and handling the missing data of ASTER images time series.

Leaf area index (LAI) is a biophysical variable that is related to atmosphere-biosphere exchange of CO 2 . One way to obtain LAI value is by the Moderate Resolution Imaging Spectroradiometer (MODIS) biophysical products (LAI MODIS). The LAI MODIS has been used to improve the physiological principles predicting growth (3-PG) model within a Bayesian Network (BN) set-up. The MODIS time series, however, contains gaps caused by persistent clouds, cloud contamination, and other retrieval problems. We therefore formulated the EM-algorithm to estimate the missing MODIS LAI values. The EM-algorithm is applied to three different cases: successive and not successive two winter seasons, and not successive missing MODIS LAI during the time study of 26 successive months at which the performance of the BN is assessed. Results show that the MODIS LAI is estimated such that the maximum value of the mean absolute error between the original MODIS LAI and the estimated MODIS LAI by EM-algorithm is 0.16. This is a low value, and shows the success of our approach. Moreover, the BN output improves when the EM-algorithm is carried out to estimate the inconsecutive missing MODIS LAI such that the root mean square error reduces from 1.57 to 1.49. We conclude that the EM-algorithm within a BN can handle the missing MODIS LAI values and that it improves estimation of the LAI.

Estimating the contribution of the forests to carbon sequestration is commonly done by applying forest growth models. Such models inherently use field observations such as leaf area index (LAI), whereas a relevant information is also available from remotely sensed images. This paper aims to improve the LAI estimated from the forest growth model [physiological principals predicting growth (3-PG)] by combining these values with the LAI derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery. A Bayesian networks (BNs) approach addresses the bias in the 3-PG model and the noise of the MODIS images. A novel inference strategy within the BN has been developed in this paper to take care of the different structures of the inaccuracies in the two data sources. The BN is applied to the Speulderbos forest in The Netherlands, where the detailed data were available. This paper shows that the outputs obtained with the BN were more accurate than either the 3-PG or the MODIS estimate. It was also found that the BN is more sensitive to the variation of the LAI derived from MODIS than to the variation of the LAI 3-PG values. In this paper, we conclude that the BNs can improve the estimation of the LAI values by combining a forest growth model with satellite imagery.

Estimating the contribution of the forests to carbon sequestration is commonly done by applying forest growth models. Such models inherently use field observations such as leaf area index (LAI), whereas a relevant information is also available from remotely sensed images. This paper aims to improve the LAI estimated from the forest growth model [physiological principals predicting growth (3-PG)] by combining these values with the LAI derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery. A Bayesian networks (BNs) approach addresses the bias in the 3-PG model and the noise of the MODIS images. A novel inference strategy within the BN has been developed in this paper to take care of the different structures of the inaccuracies in the two data sources. The BN is applied to the Speulderbos forest in The Netherlands, where the detailed data were available. This paper shows that the outputs obtained with the BN were more accurate than either the 3-PG or the MODIS estimate. It was also found that the BN is more sensitive to the variation of the LAI derived from MODIS than to the variation of the LAI 3-PG values. In this paper, we conclude that the BNs can improve the estimation of the LAI values by combining a forest growth model with satellite imagery.

In this paper, we provide a solution to the system of non-integer differential equation of order 0 < q < 1, by the technique of Laplace transformation and with interest to property of Mittag-Leffler function, with the help of the programming technique of Maple.