Zanco Journal of Pure and Applied Sciences

Countries in the Middle East region are classified as arid and semi-arid regions in the world. Because of its geographical location, it is characterized by fluctuating rainfall, which is exposed to frequent floods and causes human and material losses, on the other hand there is a lack of short-duration interval rainfall data. Therefore, to achieve hydrological studies of floods and planning for flood mitigation structure, it is necessary to derive the Rainfall Intensity- Duration- Frequency (IDF) curve, as there is a traditional method for calculating IDF curves, which is based only on two statistical distributions and only one statistical test. In this research, Soil conservation service (SCS) Type II for 24-hour rainfall distribution suitable for the arid and semi-arid regions was adopted to derive the IDF curve for maximum daily rainfall data in Erbil city. The method used has few computational steps than the traditional methods, the number of parameters is less, the number of used statistical distributions is 6, and the number of statistical tests is 4, which increases reliability. The optimal distribution was chosen for Erbil city, and the IDF curves were derived. The maximum daily rainfall was used during a specified return period from 2 to 200 years. The results showed that the maximum difference between the IDF curves obtained from the SCS theory and the traditional method is about 11%. Also, empirical relationships for representation IDF Talbot, Bernard and Ishiguro were applied, and through the test, Bernard was found to be the best

Urbanization leads to increase lands cost which resulted in the construction of high towering vertically toward the sky and use underground for construction. Foundations of these structures need deep excavation. Anti-slide piles are a suitable option for excavation stability, which involves bored cast-in-place piles constructed along the line of the excavation as a retaining structure using closely bored piles. There is a gap between two individual piles. For this purpose, an experimental study on anti-sliding piles was investigated, For further investigation of contiguous pile system, and their behavior under different soil types, and to study the effect of embedment ratio and relative density on lateral displacement of the contiguous pile.
Physical scale model tests were performed to investigate the effect of the fine-grained soils, coarse-grained soils, various embedment ratios, and density on the lateral displacement. The test results showed that lean clay (CL) soil is more stable to resist lateral displacement induced by inter-particle cohesion. Increasing the embedding ratio significantly reduced the lateral displacement, and with increasing backfill density of sandy soil for high embedment ratio the lateral displacement noticeably decreased, while for the low value of embedment ratio the lateral displacement increase with increase backfill density.

Building footprint extraction from UAV (Unmanned Aerial Vehicle) imagery data has been an active research topic in the photogrammetric and remote sensing communities in the past two decades. The traditional methods for building extraction from high-resolution imagery data are time-consuming and may not provide desired results. Recently, effective high-level approaches have been developed for buildings footprint extraction. However, their efficiency must be balanced by reducing the processing time required to obtain acceptable results using high-resolution imagery. This paper introduces an automatic method to extract building footprints from UAV imagery using deep learning algorithms. A Mask R-CNN (Region Based Convolutional Neural Network) model has been applied to building footprint extraction. For the building extraction, three experiments have been achieved. For the algorithm testing, two study areas have been selected. An orthophoto has been produced for each study area using photogrammetric software based on UAV imagery. Three experiments have been achieved for building extraction from the study area. The first experiment was based on using the pre-trained model only. In the second trial, the raw model was trained based on the study area only was used. While in the third trial, the model was trained based on fine-tuning (based on satellite imagery) and the pre-trained model with UAV training data. The analysis showed that the highest accuracy rate of the building footprint extraction increased to around 95% through using fine-tuning and more data sets in the model training, specifically with similar data sets to the study area.

This paper presents a framework for the electrification of an isolated rural residential area through a decentralized hybrid renewable plus diesel energy system. Solar/wind/diesel power units have been combined to supply continuous electrical power to a small isolated rural community in the Kurdistan Region of Iraq. The selection of the unit rates for an optimum cost-effective solution is evaluated. The framework activities include wind/solar speed/ Radiation data collection via monitoring instruments on the test location and a time-based load forecasting calculation method. Based on the available data for the wind/solar attributes a control system is designed to manage the hourly synchronization of the three units. The system is tested by the MATLAB simulator with satisfactory results.

Erbil Basin is one of the most important groundwater resources in the Middle East, it is located between both Greater Zab and Lesser Zab Rivers. Due to the importance of the basin locations and its major source for water supply, various researchers studied the hydrological, hydrogeological characteristics of the Erbil basin. It also divided into three sub-basins that are (Kapran, Central, and Bash-tepa). This paper struggle to present the actual conditions of the selected basin and determine the development process of the groundwater management in the area. Thus, the article evaluates and compare different hydrological and hydrogeological studies related to the Erbil basin, and shows different measurements of the basin boundary by used the researchers. As well as, summarized the characteristics and properties of the aquifer system. In addition, the paper estimated change in head using different catchment area for each basin. At the end of this study, it is concluded that the Erbil basin require more investigations and need to be managed by authorities in order to keep this vital source for future generations.

Iraq is a country well known for having a lot of problems and open conflicts. After a century of wars and growing political rivalry, there are still a huge number of refugees and internally displaced people (IDPs) in Iraq. According to the United Nations (UN) guiding principles, IDPs are people who are forced to leave their places of habitual residence and move to another place. They didn’t cross internationally recognized borders but still can’t return to their homes. After the crisis in 2014, the number of IDPs in Iraq rapidly increased, reaching 6 million, with the loss of their livelihoods, employment, and source of income from their place of origin. The issue we are addressing in this paper is predicting the next changes in the IDP population in each governorate among different shelter types. To have an advance and clear vision of the next changes regarding humanitarian aid, deliver humanitarian aid in advance, analyze return barriers faced by IDPs, distribute the limited funding available by UN organizations, and define which areas of emergence need more attention. This study aims to find a suitable model for predicting a future change in the IDP population while a limited amount of data is available. As a result, a new model called recurrent neural network distance (RNN_Distance) was proposed, consisting of a combination of traditional recurrent neural network (RNN) and Euclidean distance techniques with 99% accuracy. Also, we compared the result of the proposed algorithm with various existing machine learning models such as decision trees, random forests, linear regression, autoregressive integrated moving average (Arima), and Facebook Prophet (FB Prophet), then discussed the outputs. RNN_Distance was proposed, consisting of a combination of traditional recurrent neural networks (RNN) and Euclidean distance techniques with 99% accuracy. Also, we compared the result of the proposed algorithm with various existing machine learning models such as decision trees, random forests, linear regression, autoregressive integrated moving average (Arima), and Facebook Prophet (FB Prophet), then discussed the outputs.