Impact of land-use and climate change on future extreme flows: a study for three dam watersheds in Alborz and Tehran provinces of Iran

Abstract

This study assesses the impact of land-use and climate change on hydrological regime of three dam watersheds (Karaj, Latian and Mamlu) in Alborz and Tehran Provinces of Iran. Daily precipitation and temperature data from CMIP6 are transiently downscaled to ten climatic stations using the LARS-WG under global warming scenarios SSP1-1.9, SSP2-4.5, and SSP5-8.5. The Cellular Automata-Markov-chain machine learning is used to simulate future land-use maps (2021–2080) by training and testing its multilayer perceptron neural network with observed land-use change during the period 1995–2015. The study area will experience warming by 0.78, 2.1, and 2.4 °C under SSP1-1.9, SSP2-4.5, and SSP5-8.5, respectively, and precipitation anomalies by +129.3, −95.6, and −54.2 mm, respectively, compared to the baseline period 1991–2014. Extreme precipitation depth will increase at all stations under SSP1-1.9. However, precipitation change depends on the storm’s return period, station, and scenario under warmer scenarios. The SWAT-predicted river flow over three watersheds will increase, compared to the baseline period, under SSP1-1.9 but decrease under SSP2-4.5 and SSP5-8.5. Among combinations of land-use and climate change scenarios, land-use scenario High under SSP1-1.9 leads to greatest annual streamflow, while no change in land-use under SSP2-4.5 results in maximum reduction in streamflow over three watersheds. Extreme flows over Karaj and Latian watersheds show no sensitivity to different land-use scenarios due to negligible land-use development in future but they still are sensitive to climate change scenarios. Meanwhile, extreme flows over Mamlu watershed show significant sensitivity to both land-use and climate change scenarios due to significant land-use change in future.