Dr Komei Abdi, Dr Hematolah Roradeh,
Volume 8, Issue 4 (1-2021)
Abstract
Objective: Floods are among the most significant natural disasters in Mazandaran Province, particularly in Sari County, where they cause widespread economic, social, and environmental damages each year. The main objective of this research is to identify and map flood hazard zones using machine learning algorithms, namely Random Forest (RF) and Support Vector Machine (SVM), and to apply an ensemble approach in order to enhance prediction accuracy and reduce model uncertainty.
Method: In this study, a set of spatial datasets including a Digital Elevation Model (DEM), land use/land cover derived from satellite imagery, geomorphological indices (slope, aspect, and drainage density), geological data, distance from roads and streams, vegetation index (NDVI), and climatic variables (precipitation and temperature) were collected. These datasets were processed using GIS and RS techniques and prepared for model training and validation. The models’ performance was assessed using evaluation metrics such as Accuracy, F1-score, AUC, and ROC curve analysis.
Findings: The results indicated that both RF and SVM demonstrated high performance in flood hazard mapping, as reflected by strong evaluation metrics. Moreover, the ensemble approach improved prediction reliability and reduced errors compared to single-model predictions. The generated maps revealed that a significant portion of Sari County falls within high and very high hazard zones, which overlap with are::as char::acterized by intense rainfall, high drainage density, and steep slopes.
Conclusion: This research highlights that machine learning algorithms, particularly when applied in an ensemble framework, are powerful tools for identifying flood-prone areas. The findings can serve as a scientific basis for urban planning, disaster management, and flood risk reduction strategies in Sari County and other comparable regions.
Arastoo Yari, Mehdi Feyzolahpour, Neda Kanani,
Volume 10, Issue 4 (12-2023)
Abstract
Earth surface temperature provides important information on the role of land use and land cover on energy balance processes. Therefore, the purpose of this research is to evaluate the LST patterns due to changes in land use (LULC). The studied area is located in Talesh region with an area of 300.6 square kilometers. For this purpose, Landsat images were downloaded in dry and wet seasons from 1365 to 1401. Four user classes were identified by maximum likelihood classification (MLC) and support vector machine (SVM) in 36-year intervals. The Kappa coefficient values for the SVM model were equal to 0.7802 and for the MLC model it was equal to 0.5328. NDVI, NDSI, and NDWI spectral indices were calculated for vegetation, barren soil, and water and were compared with LST in the above years. Changes in land use during the years 1365 to 1401 were an important factor in changes in the temperature of the earth's surface, which averaged from 13.7 degrees Celsius to 39.5 degrees Celsius in the wet season and -0.37 to 41.07 degrees Celsius in the dry season has been variable. Water areas and vegetation have the lowest and barren soil have the highest LST values. The highest negative correlation of -0.74 belongs to the NDVI index in 1365 and the highest positive correlation of 0.79 belongs to the NDSI index in 1365. The area of the forest area has decreased by 20.3% and agricultural land has increased by 217% in 36 years. Barren lands have changed the most and decreased from 2.68 square kilometers to 12 square kilometers. In general, LST has increased due to the increase of human activities such as the expansion of agricultural land and deforestation in the studied period.
Omid Ashkriz, Fatemeh Falahati, Amir Garakani,
Volume 11, Issue 3 (12-2024)
Abstract
The growth of settlements and the increase of human activities in the floodplains, especially the banks of rivers and flood-prone places, have increased the amount of capital caused by this risk. Therefore, it is very important to determine the extent of the watershed in order to increase risk reduction planning, preparedness and response and reopening of this risk. The present study uses the common pattern of the machine and the classification of Sentinel 2 images to produce land cover maps, in order to construct sandy areas and determine land issues affected by the flood of March 2018 in Aqqla city. Also, in order to check and increase the accuracy of the algorithms, three software indices of vegetation cover (NDVI), water areas (MNDWI) and built-up land (NDBI) were used using images. The different sets of setting of each algorithm were evaluated by cross-validation method in order to determine their effect on the accuracy of the results and prevent the optimistic acquisition of spatial correlation from the training and test samples. The results show that the combination of different indices in order to increase the overall accuracy of the algorithms and to produce land cover maps, the forest algorithm is used with an accuracy of 83.08% due to the use of the collection method of higher accuracy and generalizability than compared to. Other algorithms of support vector machine and neural network with accuracy of 79.11% and 75.44% of attention respectively. After determining the most accurate algorithm, the map of flood zones was produced using the forest algorithm in two classes of irrigated and non-irrigated lands, and the overall accuracy of the algorithm in the most optimal models and by combining vegetation indices (MNDWI) was 93.40%. Then, with overlapping maps of land cover and flood plains, the surface of built-up land, agricultural land and green space covered by flood was 4.2008 and 41.0772 square kilometers, respectively.
