Floods are devastating natural disasters, causing significant economic losses and casualties globally. The accelerating water cycle due to global warming increases the frequency and intensity of extreme hydrological events like floods. Current flood risk projections often rely on hydrological models using CMIP5 data, but CMIP6 offers improved precipitation simulations. China, experiencing significant flood losses, requires updated risk assessments utilizing CMIP6 data. Rapid urbanization in China further amplifies flood exposure; from 1992 to 2015, urban land in floodplains increased by 542.21%. While some studies have examined flood risk with urban growth or climate change, the relative importance of both factors remains unclear. Previous research either neglected climate change, only considered asset densification within existing urban areas, or used limited urban expansion data outside the SSP framework. This study addresses these gaps using high-resolution data and advanced modeling techniques to project flood risk in seven major Chinese urban agglomerations, representing a significant portion of China's population and GDP. These agglomerations, situated in major floodplains and with diverse urban expansion patterns, provide valuable case studies for investigating the impact of urban expansion on flood risk.

Existing literature highlights the destructive power of floods and the impact of climate change on increasing flood frequency and intensity. Studies using CMIP5 data have projected global flood risk increases, but the availability of CMIP6 data with enhanced precipitation simulations necessitates updated projections, especially for regions like China significantly impacted by floods. The interplay between socio-economic development and flood risk is also explored, with a focus on rapid urbanization in developing countries like China. Previous research has shown the effects of urbanization and climate change on flood risk independently and in combination, however, the resolution of urbanization data and its incorporation with climate change projections has been limited. Moreover, the impact of the spatial pattern of urban expansion on flood risk has been largely unexplored. The use of ISIMIP3b bias-corrected CMIP6 outputs, combined with high-resolution urban expansion data under the SSP framework, distinguishes this study from previous work.

This study employed a model cascade combining the Variable Infiltration Capacity (VIC) hydrological model and the CaMa-Flood hydrodynamic model. Five bias-corrected CMIP6 forcings (from ISIMIP3b) representing different warming levels (1.5°C, 2.0°C, 2.5°C, and 3.0°C) and socio-economic scenarios (SSP1-2.6, SSP3-7.0, and SSP5-8.5) were used to drive the VIC model to generate daily runoff. The runoff data was then input into CaMa-Flood to simulate daily river discharge and flood extent. High-resolution (approximately 500m) urban land expansion data from Chen et al. (2020) were overlaid onto the inundation maps to determine inundated urban land areas, serving as the indicator for flood risk. The study focuses on seven major urban agglomerations in China (Beijing-Tianjin-Hebei, Central Plains, Central Shanxi Plain, Yangtze River Delta, Triangle of Central China, Chengdu-Chongqing, and Pearl River Delta), accounting for a significant portion of China's population and GDP. To assess the contributions of climate change and urban expansion, three scenarios were considered: only socio-economic development (SSP), only climate change (RCP), and both climate change and urban expansion (SSP-RCP). The 100-year flood protection levels were considered to assess flood control effectiveness. The Gumbel distribution was used to fit the historical annual maximum daily river water storage to estimate the magnitude of 100-year floods. The relative contributions of climate change and urban expansion to changes in flood risk were quantitatively analyzed using defined equations (6-9). The China Meteorological Forcing Dataset (CMFD) was used for model validation and to establish flood depth-recurrence relationships.

The study projects a substantial increase in flood risk in the seven urban agglomerations. Under warming levels of 1.5°C to 3°C, the 30-year average annual inundated urban land areas are projected to be 4.7 to 19 times greater than historical levels, across different scenarios. Southern China (PRD, CC, and TCC agglomerations) faces the most significant increases. The inundated proportion of urban land will increase dramatically (5 to 16 times historical levels). Urban land is more likely to be inundated than non-urban land, and newly developed urban land is more easily inundated than historically developed urban land. Climate change is the dominant driver, contributing more than 50% in most cases and nearly 100% in southern China. However, urban expansion plays a more significant role in northern China (BTH and CSP agglomerations) at lower warming levels. The edge expansion pattern of urban land increases inundation risk as newly-developed areas are often closer to rivers. When both climate change and urban expansion are considered, flood risk is 10-50% higher than considering climate change alone. The inundated extent of newly-developed urban land is about 10% greater than original urban land, except for the YRD agglomeration due to its flat terrain and large historical floodplains. The 100-year flood protection levels, effective historically, are insufficient for future flood control demands.

These findings highlight the significant threat of increased flood risk in China's major urban agglomerations due to the combined effects of climate change and urban expansion. The substantial projected increase in inundated urban land underscores the urgency for integrated flood risk management strategies. Climate change mitigation is crucial to limit warming and reduce flood hazard, and urban planning should prioritize sustainable development and avoid high-risk zones. Strengthening flood protection levels is also essential, although it may involve substantial investments. The edge-expansion pattern of urbanization warrants attention for its contribution to increased flood vulnerability. The study's findings are relevant not only to China but also to other rapidly urbanizing regions worldwide. Current flood risk assessments that neglect future urban expansion may significantly underestimate future losses.

This study demonstrates a substantial future increase in fluvial flood risk in seven major Chinese urban agglomerations, driven primarily by climate change but significantly exacerbated by urban expansion, especially through edge-expansion patterns. Limiting global warming to below 2°C, or even 1.5°C, is crucial. Integrated strategies are necessary, combining climate change mitigation, responsible urban planning, and enhanced flood protection levels to minimize future flood losses. Further research should incorporate models that simulate urban land change while considering flood risk and investigate the effects of human activities such as reservoir construction and drainage systems on flood risk.

The study's projections might overestimate inundated urban land areas due to the exclusion of human activities like reservoir construction, water withdrawal, and urban drainage systems. The urban land expansion data does not explicitly account for flood risk. While the study found that even without climate change, urban expansion alone leads to increased inundation, the projections might still overestimate the risk. Further research employing models that consider flood risk in urban planning scenarios would enhance accuracy. The study focuses on seven agglomerations and may not fully capture the risks faced by other urban areas in China. Changes in geology and hydrology due to urbanization were not considered.