Predicting the distribution pattern of economically important benthic waters of Hormozgan province based on geographic information system (GIS) using neural networks

Number of pages: 139 File Format: Not Specified File Code: 29347
Year: Not Specified University Degree: Not Specified Category: IT Information Technology Engineering
  • Part of the Content
  • Contents & Resources
  • Summary of Predicting the distribution pattern of economically important benthic waters of Hormozgan province based on geographic information system (GIS) using neural networks

    Master's thesis in the field of management

    Information technology (M.Sc)

    Academic year 1389-1390

     

    Abstract

    In order to check and predict the distribution pattern of economically important benthic fish, the catch data of 10 species including:  1- Saurida tumbil 2- White halva (Pampus argenteus)  3- Black halva (Parastromateus niger) 4- Rays 5- Common stonefish (Pomadasys kaakan) 6- Shourideh (Otolithes ruber)  7- Sharks, 8- Big catfish (Arius thalassinus), 9- String-tailed guazim (Nemipterus japonicus) and 10- Big-headed horse mane (Trichiurus lepturus)  related to the years 1387 and 1388 were analyzed in the waters of Hormozgan province.

    Using Spreadsheet software, initial analysis was done, and with geographic information system (GIS) software, spatial distribution maps of the mentioned species were prepared based on Catch Per Unit of Area (CPUA). After that, in order to predict the pattern of distribution,  Maps of the physical and chemical parameters of the water in the region including: temperature, turbidity, salinity, density, dissolved oxygen, pH, chlorophylla, electrical conductivity, depth, distance from the shore, fishing time and longitude and latitude were prepared. After conversion, the maps were used as inputs to Artificial Neural Networks (ANNs) software, with a percentage of information for training, a percentage for validation, and another percentage for testing the performance of the artificial neural network. were used and the best artificial neural network model with  High efficiency percentage was chosen as a model for prediction. By using the model on the limited information of hydrology and fishing in the region, it is possible to predict the distribution pattern of the desired fish, and by using the distribution pattern, it is possible to guide the fishing fleet and precisely determine the fishing areas according to geographical coordinates. Human societies perform Optimum exploitation can help in maintaining these God-given reserves. On the one hand, overfishing, destructive environmental factors, destruction of habitats, vulnerability of aquatic communities, limited ability to restore reserves, and on the other hand, the needs of human societies, endanger their population. Therefore, in order to properly exploit and sustainably develop, it is necessary to always monitor the process of changes in aquatic populations. One of the available ways to achieve this goal is to conduct regular research in order to determine any possible changes in different aquatic populations. Surveys conducted in the Persian Gulf showed that the environmental conditions in this region are always affected by strong annual changes. The presence of extreme fluctuations in environmental conditions, including changes in temperature and water salinity throughout the year in the Persian Gulf, causes disturbance in the marine environment and affects the biological and behavioral characteristics of aquatic animals. Therefore, high concentration of salt and high temperature of water in this region along with severe climate changes are among the effective factors in the ecosystem of this region. The exchange and movement of water through the Strait of Hormuz is one of the other effective factors in the changes in the environmental conditions of this region, as a result of which waters with less salinity and more oxygen and nutrients enter the Persian Gulf (Nikoyan et al., 2014).

    Although it seems that the fishes of the Persian Gulf are able to tolerate high levels of temperature and salinity, severe changes in environmental conditions in the surface waters have effects. It has more on the surface deposits of the Persian Gulf compared to the Gulf, because temperature and salinity changes are often more noticeable in the surface layers of water. This has caused severe seasonal fluctuations in the spread and distribution of surface species compared to benthic species. The abundance of small and large surface fish species, especially in the cold months of the year in the Persian Gulf, confirms these changes.Sardine fish, which is considered a surface species, is found in coastal areas and in areas where the water has a lower temperature. But when the surface water temperature reaches its maximum in summer, these fishes migrate to the depth of water and near the seabed, which has a lower temperature (Nikoyan et al., 2004).

    In general, the life cycle of different aquatic species largely depends on their living conditions. The better aquatic resources and their environment are available, and it is by understanding these relationships that it is possible to legitimize fishing situation forecasting systems (Vali Elahi, 1374). The biological diversity of fish species, the presence of mangrove forests on the Iranian coasts of the Persian Gulf, the existence of numerous and strategic islands in this water area, the extraction and export of oil from this area, the evaporation of water, are among the most important factors that are effective in explaining these special conditions (Ghaffari Cherati, 1375).

    The advancement of marine and coastal technology can put a lot of environmental pressure on the Persian Gulf. These pressures can be caused by shipping, ports, oil and petrochemical industries, mines, salting, fishing, renovation and agriculture on the marine environment (Sheppard et al., 1992; Price 1993).

    In order to investigate and monitor the aquatic population and determine the season and catch rate of any aquatic species, the Iranian Fisheries Research Institute conducts regular research patrols in the Persian Gulf and the Sea every year. Oman does. To do this important thing, expensive equipment such as a vessel equipped with research tools and equipment, skilled human resources, imposes a lot of cost on fisheries research, which is unavoidable. Considering the scope of the work and the size of the investigated area and the great diversity of species, by choosing a single management, the entire waters of the Persian Gulf into the water basins of the three provinces of Khuzestan. Bushehr and Hormozgan were divided, and the waters of each province were divided into a number of sub-regions according to the area covered by it (Parasamanesh, 1373 related to the waters of Khuzestan; Niamimandi and Soharyan, 1373 related to the waters of Bushehr province; Vali Nasab et al., 1373 related to the waters of Hormozgan province). Research patrols were carried out in each province on a seasonal basis and then a joint report related to the entire waters of the Persian Gulf was compiled (Kambozia et al., 1375). It was also planned again from 1381 to re-evaluate the bottom aquatic resources of the Persian Gulf and the Sea of ??Oman under a single management and careful coordination (Dehghani et al., 1383). Daryanbard et al., 2013; Vali Nasab et al., 2014).

    The important economic fishes whose data were studied and analyzed in this research include common hasun (Saurida tumbil),  White halva (Pampus argenteus),  Black halva (Parastromateus niger),  Shield fish (Rays),  Common stonecrop (Pomadasys kaakan),  Otolithes ruber,  Sharks, big catfish (Arius thalassinus),  The string-tailed guazim (Nemipterus japonicus) and the big-headed horse mane (Trichiurus lepturus) were related to the years 1387 and 1388 in the waters of Hormozgan province.

    This research is looking for a solution that, by using modern information technologies and using different software, can determine the distribution pattern of economic benthic species using the information system. Geographical, fishing areas, species distribution changes during the years 2017 and 2018, CPUA relative abundance index (catch rate per unit area) and the distribution pattern of benthic economic species using artificial neural network software to predict and reduce part of the costs. In this method, by reducing the amount of fishing areas (sampling stations), we can achieve the same results as before.

    Using geographic information system software, they provide us with the possibility of surface distribution with fishing points, and our neural network software helps in predicting the distribution of an aquatic species.

  • Contents & References of Predicting the distribution pattern of economically important benthic waters of Hormozgan province based on geographic information system (GIS) using neural networks

    Abstract..

    Chapter 1 - General

    Introduction..

    Outline of the topic.

    Ecology of the Persian Gulf.

    Necessity of conducting research.

    Research area.

    1-5-1- Spatial range of sampling of aquatic animals.

    1-5-2- The investigated area and Sampling stations for physical and chemical sampling of water. Time range of aquatic and hydrological sampling. 1-7- Questions and hypotheses. 1-7-1- Research questions. 1-7-2- Research hypotheses.

    Definitions of terms.

    Chapter Two - Background of the research, frameworks and foundations and documentation

    2-1- Background of the research.

    2-2- Introduction of ten species of demersal fish.

    2-2-1- Common halva Saurida tumbil.

    2-2-2- White halva Pampus argenteus.

    2-2-3- Black halva Parastromateus niger.

    2-2-4- Shield fish (Rays) 2-2-9- Nemipterus japonicus. 2-2-10- Bighead horse mane Trichiurus lepturus. 2-3- Artificial neural network. 2-3-1- History of artificial intelligence. 2-3-2- Multilayer perceptron.

    Chapter 3: Materials and methods and method of conducting research

    3-1- Tools and methods.

    3-1-1- Tools and equipment.

    3-1-1-1- Equipment on board.

    3-2- Work method.

    321 investigated areas and determination of sampling stations.

    3-3- Sampling method.

    3-3-1 Calculation method of CPUA and living mass.

    3-3-2- Method of measuring physical and chemical parameters of water.

    3-4- Information analysis method.

    Chapter four: Analysis and expression of research results

    4-1- Depth of stations Water sampling. 4-2- Vertical and horizontal distribution of physical and chemical parameters of water. 4-2-1- Water temperature. 4-2-2- Electrical conductivity. 4-2-3- Salinity. 4-2-4- Density. 4-2-5- Dissolved oxygen. 4-2-6- pH.

    4-2-7- Chlorophylla.

    4-2-8- Turbidity.

    4-3- Amount of CPUA and total living mass of bottom trawl fish in Persian Gulf and Sea of ??Oman.

    4-3-1- Amount of CPUA in 1387.

    4-3-2- Amount of living mass in 1387.

    4-3-3- Amount of CPUA in 1388.

    4-3-4- Amount of live mass in 1388.

    4-4- Live mass, CPUA and distribution of aquatic animals by species, years 1387 and 1388.

    4-4-1- Hassoun Common Saurida tumbil.

    4-4-1-1- Neural network analysis.

    4-4-2- White halva Pampus argenteus.

    4-4-2-1- Neural network analysis.

    4-4-3- Black halva Parastromateus niger.

    4-4-3-1- Neural network analysis.

    4-4-4- Rays shield fish.

    4-4-4-1- Neural network analysis.

    4-4-5- Common stonefish Pomadasys kaakan.

    4-4-5-1- Neural network analysis.

    4-4-6- Otolithes ruber.


    4-4-6-1- Network analysis Nervous.

    4-4-7- Sharks.

    4-4-7-1- Neural network analysis.

    4-4-8- Big catfish Arius thalassinus.

    4-4-8-1- Neural network analysis.

    4-4-9- Nemipterus japonicas.

    4-4-9-1- Neural network analysis.

    4-4-10- Bighead horse mane Trichiurus lepturus.

    4-4-10-1- Neural network analysis.

    Chapter five: discussion and conclusion

    5-1- Physical and chemical parameters of water.

    5-1-1- Water temperature 5-1-2-electrical conductivity .

    5-2- Live mass, CPUA and distribution of aquatic animals by species, years 1387 and 1388.

    5-2-1- Common halva Saurida tumbil.

    5-2-2- White halva Pampus argenteus.

    5-2-3- Black halva Parastromateus niger.

    5-2-4- Shield fish Rays.

    5-2-5- Common stonefish Pomadasys kaakan.

    5-2-6- Otolithes ruber.

    5-2-7- Sharks.

    5-2-8- Big catfish Arius thalassinus.

    5-2-9- Nemipterus japonicas. 5-2-10- Trichiurus lepturus mane.

Predicting the distribution pattern of economically important benthic waters of Hormozgan province based on geographic information system (GIS) using neural networks