General and regional climate models (GCMs/RCMs) exhibit many systematic biases, which affects the simulation accuracy of the real precipitation patterns and consequently the associated hydrological changes. The common approach used to reduce errors in the climate model output is to apply bias correction methods (BCMs) which attempt to adjust the climate simulation by its observation counterpart. In this study, we applied BCM on simulated rainfall over the Bouregrag basin (Morocco) from CanRCM4 CORDEX RCM using linear SCALING method (SCALING), gamma quantile mapping (GQM) and empirical quantile mapping (EQM). Owing to its high performance compared to the others, we applied the EQM method on climate projections under the RCP4.5 scenario. Results showed a decrease up to -50% in the monthly rainfall for the period 2041-2060 with the exception of August and December exhibiting an increase between +20% and +78%. This study supports the need to bias correct climate data before their use in hydrological models where the bias could be irreversible.

Artificial Intelligence (AI) combined with image processing has shown significant improvements through new techniques such as Machine Learning (ML) models. This paper introduces the key methods and algorithms used for Drone image processing. We discuss the benefits and limitations of using ML models instead of classical techniques. Our goal is to classify, categorize and describe the methods that are used in realistic settings of diverse domains of applications. We conducted a systematic literature review where systems presented in the papers were analysed based on their domain, task, technology, and efficiency. By extensively reviewing the existing literature, we successfully identified key themes and trends that emerged across the various research questions. The overall findings of the research emphasise the potential of AI and drone imagery in numerous fields. However, the review also uncovered several challenges that necessitate attention, such as issues related to data quality and the requirement for more advanced AI algorithms. The paper outlines significant innovations in the field and offers recommendations for future research directions. By highlighting cross-disciplinary insights, it delves into methodological approaches, exploring commonalities in AI algorithms and UAVs technologies.

Determining an impervious surface is one of the most important topics of remote sensing because of its great role in providing information that benefits decision-makers in urban planning, sustainable development goals, and environmental protection. In recent years, a great development in this field has occurred due to the huge improvement in the algorithms and techniques that are used to map impervious surfaces. In this paper, the deep learning technique has been implemented to investigate the extraction of impervious surfaces in Marrakesh city, based on Landsat images. 9000 polygons and 13840 points have been taken to prepare label data by random forest in Google Earth Engine (GEE). In addition, all preprocessing steps for remote sensing images have been implemented in GEE. An artificial neural network (ANN) has been used to determine impervious surfaces. After training and testing the proposed network on Landsat image datasets, precision, accuracy, recall, and F1-score matrix scores were 0.79, 0.98, 0.87, and 0.82, respectively. The experimental results show that this method is efficient and precise for mapping the impervious surfaces of Marrakesh city.

Drought is an extreme event that has hit several countries in the world including Morocco. The aim of this research was to assess drought in Morocco with a view to providing information for planning and management of droughts. For this, three drought indices were chosen: Combined Drought Indicator (CDI), Soil Moisture Agricultural Drought Index (SMADI), and Microwave Integrated Drought Index (MIDI). Drought monitor was done during the growing seasons of 2010-2020 using Earth Observation data and cloud computing with the mapping of the drought indices and their inter-comparing via Pearson correlation. The main drought events were tracked and drought characteristics analyzed. Seven drought years were tracked for regions of cereal production. CDI and MIDI were very well correlated, whereas SMADI showed poor correlation with CDI and MIDI. Validation of results was done by comparing the results with another study for the 2015-2016 drought event and comparing yearly precipitation with the long-term average. An Earth Engine App of the three indices was published to make public drought maps.

Abstract. In recent years, deep convolutional neural networks (CNNs) algorithms have demonstrated outstanding performance in a wide range of remote sensing applications, including image classification, image detection, and image segmentation. Urban development, as defined by urban expansion, mapping impervious surfaces, and built-up areas, is one of these fascinating issues. The goal of this research is to explore at and summarize the deep learning approaches used in urbanization. In addition, several of these methods are highlighted in order to provide a comprehensive overview and comprehension of them, as well as their pros and downsides.

Aims: This study aims to predict poor glycemic control during Ramadan among non-fasting patients with diabetes using machine learning models.Methods: First, we conducted three consultations, before, during, and after Ramadan to assess demographics, diabetes history, caloric intake, anthropometric and metabolic parameters. Second, machine learning techniques (Logistic Regression, Support Vector Machine, Naive Bayes, K-nearest neighbor, Decision Tree, Random Forest, Extra Trees Classifier and Catboost) were trained using the data to predict poor glycemic control among patients. Then, we conducted several simulations with the best performing machine learning model using variables that were found as main predictors of poor glycemic control.Results: The prevalence of poor glycemic control among patients was 52.6%. Extra tree Classifier was the best performing model for glycemic deterioration (accuracy = 0.87, AUC = 0,87). Caloric intake evolution, gender, baseline caloric intake, baseline weight, BMI variation, waist circumference evolution and Total Cholesterol serum level after Ramadan were selected as the most significant for the prediction of poor glycemic control. We determined thresholds for each predicting factor among which this risk is present.Conclusions: The clinical use of our findings may help to improve glycemic control during Ramadan among patients who do not fast by targeting risk factors of poor glycemic control.

Integrating the environmental dimension into socio-economic development strategies has become a requirement for the sustainable development. To reach this goal, it is necessary to monitor the state of the environment in order to identify, manage and supervise the different environmental issues and to determine the appropriate responses for them. This objective cannot be achieved without organizing environmental data in an information system that can be used and updated by all the actors and partners concerned by the environmental questions. The aim of implementing an Environmental Information System (EIS) is to develop an integrated framework for the storage, production, management and exchange of environmental information in a decision-making perspective. However, availability and scarcity of data remain a major obstacle for realizing such systems, especially in developing countries. This study aims to discuss the issues and processes of EIS implementation. It details the environmental monitoring models and the challenges related to building the environmental databases. An overview on the EIS implementation in developed countries was carried out, and an example of EIS in MENA region (Morocco) was presented to extract syntheses and recommendations in order to define a guideline to implement a successful EIS especially in data-scarce countries.

By 2030, almost 23.6 million people will die from cardiovascular heart diseases (CVDs). In Morocco, deaths by CVDs represented 38% in 2018. Using machine learning and tracking patient health indicators can reduce this mortality. Indeed, the aim of this study is developing an application that collects and process a stream of geolocation and heart rate data, stores the data and predicts on cardiovascular heart diseases risk. We first construct the machine learning model, define the architecture then we developed and tested the data pipeline. Samsung smartwatch was used to collect heart rate and location, Kafka and Spark were used to collect the streamed data received from the Smartwatch, the Data was then stored in MongoDB. This work produced a development of a complete real time data pipeline from data production to alerts and reports generation using big data and machine learning technologies.

Soil moisture (SM) is a key parameter in the hydrological cycle. Its lack or abundance can cause extreme events such as floods and drought. To track drought, we have chosen the Microwave Integrated Drought Index (MIDI). In order to calculate MIDI, we used SM, precipitation and Land Surface Temperature (LST). The study concerns the 2010-2019 growing seasons. Our study area is composed of some agricultural sites in Morocco: Settat and Meknes areas. MIDI was calculated using two approaches (Al and A2). Then, each approach used two types of classifications (Cl and C2) with five drought classes in order to estimate severity. Four types of MIDI were found: MIDII (approach Al and classification Cl), MIDI2 (approach Al and classification C2), MIDI3 (approach A2 and classification Cl), and MIDI4 (approach A2 and classification C2),. Drought index calculation and drought maps generation was done using Google Earth Engine (GEE). Then, a comparison with a previous study for the 2015-2016 drought event was led in order to validate the results of drought maps. MIDI4 was found to show the best performance in the study.

Data availability is a main element in determining the watershed modeling success. This factor becomes more critical in the case of using spatial models that require space-time distributed data. In developing countries, the implementation of such approaches is often hampered by data scarcity. To deal with this situation, the use of global data captured by earth observation satellites is considered as a major issue. This work aims to show the utility of using this type of information in data scarcity areas, especially for spatial modeling of large watersheds. To estimate the global data contribution, an analysis was performed by comparing them with local measured data. The study focuses on data representing the watershed state (morphological properties) and the input variables of hydrological models (climate data). The quality assessment of these data is calculated through statistical indicators. The comparison of global data grids with local observations shows the utility of using some of these grids for watershed modeling and especially for the state parameters such as topography. For climate parameters, the comparison is very appropriate for the minimum and maximum temperature, and moderate for the humidity and solar radiation but low or very low for rainfall and wind speed data. This work reveals that if global data derived from satellites are an alternative and very promising solution to overcome data scarcity in some areas, they still need to be enhanced to make them more efficient and accurate, especially for climate data.

The United Nations has been estimated that two-thirds of the Moroccan population will live in the cities in 2020, which makes the cities face major challenges, such as putting pressure on infrastructure, health, education, and the environment. In order to avoid these challenges and turn them into gains, it is necessary to put the perfect plans by the policymakers. For this reason, it required to mapping the land cover land use (LC/LU) to measure and map built-up expansion in the cites. This study aims to provide a more effective tool for mapping built-up and non-built areas in the city of Marrakesh in Morocco and compare these results with Atlas of urban expansion project. We used both 8069 polygons from cadastral administrative plans and 9269 control points as input sources, 80% of the data as training data and 20% for the validation, three supervised machine learning classifier algorithms have been used; random forest (RF), support vector machine (SVM), and CART. The results showed that the link between satellite imagery and local cadastral administrative data is more effective, especially in the empty lands in the center and around the city. The overall accuracy and Kappa indicators showed that RF is more powerful than SVM, and CART for mapping urban areas.

Data availability is a main element in determining the watershed modeling success. This factor becomes more critical in the case of using spatial models that require space-time distributed data. In developing countries, the implementation of such approaches is often hampered by data scarcity. To deal with this situation, the use of global data captured by earth observation satellites is considered as a major issue. This work aims to show the utility of using this type of information in data scarcity areas, especially for spatial modeling of large watersheds. To estimate the global data contribution, an analysis was performed by comparing them with local measured data. The study focuses on data representing the watershed state (morphological properties) and the input variables of hydrological models (climate data). The quality assessment of these data is calculated through statistical indicators. The comparison of global data grids with local observations shows the utility of using some of these grids for watershed modeling and especially for the state parameters such as topography. For climate parameters, the comparison is very appropriate for the minimum and maximum temperature, and moderate for the humidity and solar radiation but low or very low for rainfall and wind speed data. This work reveals that if global data derived from satellites are an alternative and very promising solution to overcome data scarcity in some areas, they still need to be enhanced to make them more efficient and accurate, especially for climate data.

Since the advent of the social media in the wake of the Web 2.0 phenomenon, the Volunteered Geographic Information produced by the users of such platforms as Twitter, Facebook, Instagram… have become increasingly used as a major source of data combined with the more traditional authoritative data. The review of a number of case studies in the literature illustrates the pertinence of the use of geo-referenced crowd sourced data to improve the efficiency of risk management approaches particularly in poor mapped area or/and where real-time and up-to-date data is needed. The nature of crowd-sourced data implies that data quality of Volunteered Geographic Information have to be precisely assessed before combining it with authoritative data as shown in the literature review.

Soil moisture is an important parameter among the fifty "Essential Climate Variables" according to Global Climate Observing System (GCOS). It allows to perform several applications in different fields, especially hydrological, meteorological and agricultural ones. There are several methods for measuring this parameter in two main categories: in-situ methods and remote sensing. In this sense, two well-known satellites are invested, namely Soil Moisture and Ocean Salinity (SMOS) from European Space Agency (ESA) launched in 2009 and Soil Moisture Active Passive (SMAP) from National Aeronautics and Space Administration (NASA) launched in 2015. This work will be dedicated to the state of the art of soil moisture downscaling and applications across various regions of the world, including Canada, USA and Spain to take advantage of these studies for a future effective exploitation of soil moisture mapping in a Moroccan context.

The Data, as result of the digital revolution, offers several opportunities in the field of health. Actually, appliances and applications permanently connected to humans and the global digitization of medical documents produce a vast health data: Big Health Data. This data is the subject of several projects in the world given the opportunities offered to optimize this area. The aim of this study is to explore how the combination of Health Data with geolocation can be more opportunistic and how it can be applied in Morocco. In order to achieve this objective, Health Data is introduced by describing projects and global experiences in exploiting this data from different countries, opportunities of the combination with geolocation are highlighted by citing different concrete examples, and a projection of this study on Morocco is detailed.

The Big Data, a result of the digital revolution, offers several opportunities in the field of health. Indeed, appliances and applications permanently connected to humans and the global digitalization of medical documents produce a vast health data: "Big Health Data". This data is the subject of several projects in the world given the opportunities offered to optimize this area. This paper focuses on quantifying the production of scientific articles about Big Health Data research and the most investigated Big Health Data topics. It also presents a mapping of countries producing articles about this subject. In remote sensing using real time categories, we aimed to quantify articles dealing with "big data architectures", technologies and data sources used. A systematic mapping study was conducted with a set of seven research questions by investigating articles from two digital libraries: Scopus and Springer. The study concern articles published in 2017 and the first half of 2018. The results are illustrated by diagrams answering each question from which a set of recommendations are concluded in this area of research. The study shows that this Data is used the most in studies of oncology. Statistics show that while remote sensing and monitoring is a hot topic, real-time use is not as interesting. It was found that there's a lack in studies interested in big data technologies used in real time remote sensing in the field of health. In conclusion, we recommend more focus on research area treating architecture in remote sensing real time Big Health Data systems combined with geolocation.

Remote sensing and image fusion have recognized many important improvements throughout the recent years, especially fusion of optical and synthetic aperture radar (SAR), there are so many published papers that worked on fusing optical and SAR data which used in many application fields in remote sensing such as Land use Mapping and monitoring. The goal of this survey paper is to summarize and synthesize the published articles from 2013 to 2018 which focused on the fusion of Optical and synthetic aperture radar (SAR) remote sensing data in a systematic literature review (SLR), based on the pre-published articles on indexed database related to this subject and outlining the latest techniques as well as the most used methods. In addition this paper highlights the most popular image fusion methods in this blending type. After conducting many researches in the indexed databases by using different key words related to the topic "fusion Optical and SAR in remote sensing", among 705 articles, chosen 83 articles, which match our inclusion criteria and research questions as results,all the systematic study ' questions have been answered and discussed.

With the rapid rate of population growth and economic development, cities face enormous challenges that require both optimal and integrated solutions to meet the needs of growth and to protect the environment and sustainable development. These urban dynamics, which change over time, extend not only horizontally and upward, but also downward. Thus, underground space has been utilized increasingly to relieve the urban surface and to ensure the exploitation of underground resources. The purpose of this study is to evaluate the possibilities of using this space in Casablanca as part of urban land-use planning and, consequently, to suggest an integrated model of exploitation of this space that is adapted to the specificities of the study area. Thus, an analysis of the use of underground spaces in a set of European cities has been performed. The study of the characteristics of this space in Casablanca has been realized according to the levels of geology and hydrogeology and two underground infrastructure projects. This work has led to the implementation of a prototype model named “Sub-Urban Information Modeling”. The model’s objective is to gather all the data and knowledge related to the relevant underground space in an integrated platform that can be shared and updated in order to facilitate the understanding of this environment and its interaction with the surface and to ensure the rational and efficient use of its resources.