Investigating the flood-affected areas of Khuzestan in the period from 7 March 2019 to 24 Avril 2019 using NDVI, NDBaI and NDWI indices and analyzing the degradation process of Hourol Azim wetland from 2000 to 2023 using Random Forest Model (RTC)

The Hourol Azim wetland is one of the border water resources of Iran and Iraq, which has witnessed severe floods and a severe decrease in water level in recent years, and has experienced environmental crises in the form of drought, destruction of farms, and dust storms. In this research, the random forest algorithm was used to investigate the change process of Hourol Azim wetland from 2000 to 2023, and in addition, to investigate the expansion of water areas caused by the flood of April 2019 from the spectral indices NDWI, NDVI and NDBaI in the period of March 7. It was used on April 8 and April 24, 2019. With the reduction of the wetland level, the temperature of 77 degrees Celsius was recorded in some parts of the northeast of the wetland in May 2023. The LST index showed the highest positive correlation with the NDBaI index at the rate of 0.72 on March 7, 2019, and the highest negative correlation between the NDVI index and LST was obtained at the rate of -0.73 in 2000. In 2000, the surface of Hourol Azim wetland was equal to 256 square kilometers, and this amount increased to 780 square kilometers in 2023. However, at the time of the flood in April 2018, at the maximum intensity of the flood, the water level of the lagoon increased to 3200 square kilometers along with the outflow of the Karkhe River. At the time of the flood, vegetation experienced its lowest level and the largest area of 11,843 square kilometers was allocated to barren lands. The results show that the random forest model has detected different types of land use with high accuracy.