A novel multivariate multiscale index for drought-flood abrupt alternations: Considering precipitation, evapotranspiration, and soil moisture

Drought-flood abrupt alternation (DFAA), or the rapid shift between droughts and floods, is a complex phenomenon that cannot be fully captured by individual meteorological or hydrological factors alone. However, previous studies have rarely taken into account multiple factors simultaneously, and have identified DFAA events using either precipitation or runoff. This study proposed a novel index, the Multi-factor Standardized DFAA Index (MSDFI), which integrated multiple factors (i.e., precipitation, evapotranspiration, and soil moisture) and multiple timescales using copula functions to detect DFAA events. The proposed index was applied to the Pearl River Basin (PRB), containing the largest river in southern China, to examine the applicability of the index and explore the spatial and temporal evolution patterns of DFAA from 1979 to 2018. The potential driving mechanisms behind these patterns were also explored. The results demonstrated that the proposed MSDFI, considering multiple variables and time scales, can effectively identify DFAA events. The identified drought to flood (DTF) and flood to drought (FTD) events aligned with the hydro-meteorological transition months in the PRB. There were more than 32 DTF and FTD events during 1979–2018 in the PRB and these events were uniformly distributed spatially. The DTF events were mainly observed during spring and winter, while FTD events mainly occurred in summer and autumn. The DTF intensity ranged from 0.97 to 1.75, whereas the FTD intensity ranged from 0.87 to 1.75. Both DTF and FTD showed spatial patterns with slightly higher intensities in the west than that in the east. The potential drivers of DTF and FTD differed, with precipitation, surface pressure, and NDVI being key factors for DTF, and precipitation and evapotranspiration being key factors for FTD. These findings suggested that the proposed index and approach are effective and can be applied to any region globally.