Drought–flood abrupt alternations (DFAAs) have a greater impact on ecosystems and socioeconomic environments than lone droughts or floods. Despite the significant impact of DFAAs, research has paid little attention to their evolutionary characteristics, particularly in relation to vegetation growth in the Heilongjiang River Basin. Therefore, this study focuses on the Heilongjiang River Basin and employs the DFAA Index to identify and analyze abrupt alternation events from 1970 to 2019. It also examines the annual and interannual distributions of vegetation growth changes from 2000 to 2019, based on the Normalized Difference Vegetation Index. Lastly, it utilizes correlation analysis to investigate the responsive relationship between vegetation growth and DFAA events. The results indicate the following: (1) Within the Heilongjiang River Basin, the number of drought-to-flood events increased over time, whereas the number of flood-to-drought events decreased over time. The frequency of mutation was relatively high in the northern region, low in the eastern region, elevated in spring and summer, and reduced in winter. (2) The Normalized Difference Vegetation Index was lowest in January, highest in July, and approximately 0 during the winter. The vegetation coverage reached its peak during the summer. (3) Vegetation changes in response to DFAAs exhibited a significant time lag. Vegetation changes in spring–summer lagged behind DFAA events by 3–4 months, while in summer–autumn, the lag was approximately 3 months. These results are of great significance for the early warning and prevention of DFAAs in the Heilongjiang River Basin.

Songhua River Basin, northeast China, has seen significant changes due to climate change and human activities from 1990 to 2000, when forests were largely reclaimed and agricultural land was taken up to change the terrestrial water cycle drastically. This paper investigates hydrological changes in three basins: the main stream basin of the Songhua River, the Second Songhua River Basin, and the Nenjiang River Basin. Machine learning and signal processing techniques have been applied to reconstruct historical river records with high accuracy, achieving determination coefficients exceeding 0.97. The physically based WEP model effectively simulates both natural hydrological patterns and human-induced hydrological processes in the northern Nenjiang region. Climate projections indicate clear temperature increases across all scenarios. The most significant warming is observed under the SSP5-8.5 scenario, where runoff increases by 8.52% to 12.02%t, with precipitation driving 62% to 78% of the changes. Summer runoff shows the most significant increase, while autumn runoff decreases, particularly in the Nenjiang Basin, where permafrost loss alters spring melt patterns. This change elevates flood risk in summer, with the rate of increase strongly dependent on the scenario. Water resources show strong scenario dependence, with the average growth rate of SSP5-8.5 being 4 times that of SSP1-2.6. A critical threshold is reached at a 2.5 °C increase in temperature, triggering system instability. These results emphasize the need for adaptation to spatial differences to address emerging water security challenges in rapidly changing northern regions, including nonlinear hydroclimatic responses, infrastructure resilience to flow changes, and cross-basin coordination.

Climate change has exacerbated the occurrence of complex extreme hydrological events in high-latitude cold regions, among which drought–flood abrupt events (DFAAEs) threaten food and water security, and accurately predicting their future evolution remains a key challenge. This study used the Community Water Model (CWatM) hydrological model, combined with five CMIP6 climate models, to simulate runoff datasets for historical periods (1985–2014) and future shared socioeconomic pathways (SSPs: SSP126, SSP370, SSP585: 2015–2100). We calculated the DFAA index (DFAAI), analyzed the spatiotemporal distribution patterns and predicted future trends of DFAAEs in the Heilongjiang River Basin, and explored their climatic driving mechanisms. The main conclusions are as follows: (1) Under SSPs, precipitation and evaporation increase from northwest to southeast, and temperature increases from north to south; hotspots expand inland. By 2100, annual precipitation will reach 655, 700, and 720 mm; mean air temperature will rise by 3, 6, and 7 °C; and annual evapotranspiration will reach 460, 515, and 521 mm. (2) Relative to the historical period, DFAAEs increase from 5.9 to 6.6, 7.1, and 7.5 events per year (SSP126/370/585). Coverage rises from 10.6% to 12.7%, 17.1%, and 19.0%, while mean intensity remains 1.8–2.0. Across both the historical period and SSPs, the shares of light (69–74%), moderate (20–24%), and severe (6–8%) events are stable. (3) Principal Component 1 (PC1,62.9%) reflects a precipitation-dominated wetting mode with synchronous increases in evapotranspiration and is the primary driver of DFAAI variability. PC2 (20.3%) captures an energy-related mode governed mainly by evapotranspiration and indirectly modulated by air temperature, providing a secondary contribution. These results clarify DFAA mechanisms and inform water-resources security planning in the Heilongjiang River Basin.