Future Hydrological Drought Risk Assessment Based on Nonstationary Joint Drought Management Index

Abstract

As the environment changes, the stationarity assumption in hydrological analysis has become questionable. If nonstationarity of an observed time series is not fully considered when handling climate change scenarios, the outcomes of statistical analyses would be invalid in practice. This study established bivariate time-varying copula models for risk analysis based on the generalized additive models in location, scale, and shape (GAMLSS) theory to develop the nonstationary joint drought management index (JDMI). Two kinds of daily streamflow data from the Soyang River basin were used; one is that observed during 1976-2005, and the other is that simulated for the period 2011-2099 from 26 climate change scenarios. The JDMI quantified the multi-index of reliability and vulnerability of hydrological drought, both of which cause damage to the hydrosystem. Hydrological drought was defined as the low-flow events that occur when streamflow is equal to or less than Q80 calculated from observed data, allowing future drought risk to be assessed and compared with the past. Then, reliability and vulnerability were estimated based on the duration and magnitude of the events, respectively. As a result, the JDMI provided the expected duration and magnitude quantities of drought or water deficit.The GAMLSS framework has been applied to hydrological frequency analysis to provide the design criteria for managing future drought risk [11,12]. Bivariate frequency analysis based on the GAMLSS-copula model was also conducted and proved useful in hydrological prediction [13][14][15]. GAMLSS can be useful to estimate nonstationary index. For example, Wang et al. [16] proposed the time-dependent standardized precipitation index and Bazrafshan and Hejabi [17] suggested using the nonstationary reconnaissance drought index.This study performed statistical analyses with 26 climate change scenarios for the Soyang River basin in South Korea to construct the bivariate time-varying copula models based on GAMLSS theory. Copula functions were used to combine the drought information estimated from low-flow events, which were reliability and vulnerability. Then a new drought index, joint drought management index (JDMI), was developed based on the nonstationary copula model. The potential drought or water supply failure events were quantified as durations and magnitudes by JDMI-based risk assessment. Study Area and Data

Drought Risk Analysis, Forecasting and Assessment under Climate Change

“…This Special Issue's ten articles advance our understanding of drought's complex phenomena and its interaction with climate change and human activity. The newly proposed drought indices [36,37] will serve as effective tools for drought monitoring under changing environmental conditions. The indices can incorporate multiple inputs for drought calculation, which is more realistic than traditional methods.…”

“…This Special Issue includes 10 peer-reviewed articles covering a wide range of research topics related to drought monitoring, drought forecasting and drought risk analysis in a changing climate. Specific issues include the development of a modified composite drought index (MCDI) and a non-stationary joint drought management index (JDMI) [36,37], climate change influences on drought patterns and crop yields [38,39], meteorological and hydrological drought risk under future climate change predictions [40,41], extreme drought assessment and its relationship with the Indian Ocean dipole (IOD) mode [42], severe drought prediction using atmospheric teleconnection patterns (ATPs) [43], drought forecasting using stochastic models [44] and hydrological drought risk estimations based on changing climate conditions and human activities [45]. These studies use statistical approaches, field measurements and mathematical methodologies.…”

“…MCDI is a reliable drought monitoring index that represents integrated meteorological and agricultural drought knowledge. In another study, Yu et al [37] formulated a non-stationary joint drought management index (JDMI) for hydrological drought risk assessments. They formulated a bivariate time-varying copula model using generalized additive models for location, scale and shape (GAMLSS) and determined future low-flow drought using simulated data for South Korea's Soyang River basin.…”

Nonstationary Analyses of the Maximum and Minimum Streamflow in Tamsui River Basin, Taiwan

“…A widely used approach to model nonstationary hydro-climatic series is the timevarying moments method, indicated by Khaleq et al [42], which incorporates time-varying parameters into probability models with the same form of stationary condition. The GAMLSS is a popular tool to achieve this purpose in hydrology and dynamically detects evolution of probability distributions with time or other covariates [43][44][45][46].…”

Nonstationary Flood Hazard Analysis in Response to Climate Change and Population Growth

“…Otherwise, the nonstationary models with only time covariates are regarded as lacking a physical mechanism. Thus, researchers have introduced some physical covariates such as rainfall, temperature, climate indices, reservoir index, crop area, irrigation area, and impervious surface to inform nonstationary models [13][14][15][16][17][18][19][20][21][22][23], and conduct flood hazard analysis [24][25][26][27][28][29][30][31][32]. In changing environments, to guarantee that the flood hazard analysis is closely related to the operation of the hydrologic projects and the early flood warning system, we must associate flood hazards with the design lifespan of hydrological projects due to a flood hazard of a certain flood event exceeding a given flood quantile over the project's lifespan being significantly different from that over other time periods of the same length due to the nonstationarity/evolution of flood distributions.…”