Abstract

This study describes the method of using the CA-Markov model to predict the distributed land use in the Gorgan River basin in 2040 based on historical evolution.future land use changes for 2040 have been simulated using the integrated CA-Markov model according to the scenario of continuation of the current management process. Considering the phenomenon of climate change, which is the consequence of the increasing growth of human activities, Intensity-Duration-Frequency (IDF) curves in three time periods in the form of historical period (1999-1980), recent period (2000-2020) and future period (2021-2040) under the conditions of climate change, extraction and their changes were investigated. took For future needs, the CAMS-CSM1-0 model from the CMIP6 model set has been used for two optimistic (SSP2-4-5) and pessimistic (SSP-5-8.5) scenarios. Then the HEC-HMS hydrological model was used to investigate the effect of land use change on the hydrological responses of the drainage basin. CA-Markov model prediction results indicate a decrease of 13.89% of forest lands from 2020 to 2040. Urban and agricultural areas have increased by 40.65% and 5.27% respectively in the same period of time. According to the results obtained from the flood simulation in 2040, assuming constant rainfall and taking into account land use changes, the peak discharge will increase by 51.11, compared to 2020, and also the flood volume will be 53.88%. It will increase compared to 2020. The further increase in urbanization and the destruction of forest lands leads to more significant flood peak and volume changes. By examining the flow hydrograph in different return periods, for three rainfall scenarios, it was determined that rainfall and land use changes had an effective role in increasing the peak discharge and flood volume, considering the 24 percent decrease in rainfall in the third scenario (2040-2021) , the peak discharge in this scenario for the 100-year return period has increased by 3.96% compared to the first scenario. The results indicate an increase in flood peak and flood volume and a decrease in pastures, forests and an increase in urban and agricultural areas between 1986 and 2040.

Keywords: Flood, integrated CA-Markov model, hydrological model, Gorgan River basin

Golestan province has suffered countless human and financial losses due to historical and destructive floods. The inflow of surface runoff and flood to the wide and vast lands of the province, the low storage capacity and failure to direct the flood to safe natural and artificial reservoirs, as well as the improper drainage conditions have caused flooding and waterlogging and damage. And there is a need to consider effective management strategies in this regard for the region. Predicting land cover changes is a useful method to achieve a general vision for better management of natural resources and protection of agricultural lands in urban areas and is very effective in carrying out long-term measures (Varga et al. 2019). (A Markov model is a model in which the future state of a system can be predicted based solely on the previous state. The prediction of future changes is obtained by creating the transition probability matrix (LULC) from period one to period two (Hyandieh et al., 2017). The aim of the current research is to evaluate land use changes in the Gorgan River basin with the help of satellite images for the years 1986 to 2020 and to model and predict land changes using the CA-Markov model until 2042 and to analyze its historical evolution from 1986 to 2020 (i.e. the base period ) and predict the observed trend for the future period of 2040-2021 and use HEC-HMS-based hydrological models to investigate the impact of predicted and synthesized land use change on typical historical flood values.

Methods and material

The model needs topographic data, soil type, land use, meteorological and hydrological data. Topographical data was obtained by digital elevation model (DEM) of 30 meters. Soil data and land use data have been extracted from the processing of LandsatTM, Landsat ETM+ and TIRS OLI satellite images for the years 1986, 2006 and 2020, respectively, to classify and investigate land use changes in the Gorgan River basin and have been prepared for input into the model. . Meteorological data including daily precipitation and temperature of the Gorgan River basin from 1983 to 2020 were obtained from the Golestan Regional Water Company, and in the recent and future period, IDF curves have been estimated based on climate change. The daily rainfall data of 9 stations including Golestan and Vashamgir dams in the basin were analyzed in the HEC-HMS model. The hydrological data of daily runoff from 1986 to 2020 were evaluated at the Agh Qola station in the calibration phase of the Gorgan River.

Results and discussion

The daily runoff hydrological data from 1986 to 2020 was evaluated at the Agh Qola station in the calibration phase of the Gorgan River. The results indicate that land use changes, in addition to the increase in peak flow in three 20-year periods, have also caused an increase in the volume of floods, and the increase in volume in 2001 compared to the flood of 1983 has increased by 46% and compared to 2020 was 80 percent. What is clear from the forecast data of the future rainfall, according to the 24% decrease in rainfall in the third scenario (2040-2021), the peak discharge in this scenario for the 100-year return period is 3.96%, compared to the first scenario.



Conclusion

we have had 85.48%, and 83.87% increase in discharge and flood volume in 2020 compared to 1986. With the assumption of constant rainfall, the increase of 51.11% of discharge and 53.88% in volume has been calculated for the year 2040. The results of the return period of 2 to 100 years also show an increase in flow rate and volume for the three rainfall periods. The peak runoff rate in the return period of 2 years for the first period was 106.05 cubic meters per second and in the return period of 100 years. The same period has reached 685.04 cubic meters per second.