Multiple causes of nonstationarity in the Weihe  annual low-flow series

Xiong, Bijing; Xiong, Lihua; Chen, Jie; Xu, Chong‐Yu; Li, Lingqi

doi:10.5194/hess-22-1525-2018

Cited by 28 publications

(23 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, there is no such restriction in GAMLSS, allowing time series modeling for observations with any underlying distribution (Rigby & Stasinopoulos, ). Because of its ability to relate time series variables with external covariates, GAMLSS has been employed by many studies for nonstationary modeling and attribution (Jiang et al, ; López & Francés, ; Tan & Gan, ; Villarini, Smith, et al, ; Villarini et al, , ; Xiong et al, ; Zhang et al, , ). GAMLSS has also been used for stochastic simulation of climate variables and in weather generators (Chun et al, ; Serinaldi & Kilsby, , ; Tye et al, ; Verdin et al, ).…”

Section: Methodsmentioning

confidence: 99%

Evaluation of the Stationarity Assumption for Meteorological Drought Risk Estimation at the Multidecadal Scale in Contiguous United States

Apurv

Cai

2019

Water Resources Research

View full text Add to dashboard Cite

In this study, we analyze the nonstationarity in meteorological droughts at the multidecadal scale in different parts of the contiguous United States during 1901-2017. We develop metrics to compare the drought risk calculated under the assumptions of stationarity and nonstationarity and identify their spatial and temporal patterns. By analyzing the variability of drought risk in the past and exploring its ongoing patterns, we evaluate in which regions of the contiguous United States the assumption of stationarity can be safely used for drought risk planning and management. We find statistically significant interdecadal changes in the probability distribution functions of drought severity in parts of the Northwest, upper Midwest, the Northeast, eastern parts of Great Plains and in parts of Arizona, New Mexico, Utah, and Nevada in the Southwest. In these regions, the nonstationary risk has been significantly higher than the stationary estimate of risk in the past, which shows that the assumption of stationarity can lead to the underestimation of drought risk in these regions. The multidecadal drought risk shows low variability in California, parts of northern and western Great Plains, Ohio Valley, and in the Southeast, since the statistical properties of droughts have not changed significantly in these regions during 1901-2017. However, the meteorological drought risk has increased in California and the Southeast in the recent decades due to the influence of global warming and hence the assumption of stationarity for risk estimation may lead to underestimation of drought risk in future in these regions if this effect of global warming persists.Plain Language Summary Traditionally, statistical approaches adopted by water resource managers for planning and design of water resource systems and infrastructure are based on the assumption of stationarity; that is, it is assumed that the probabilistic characteristics of the hydrological and meteorological processes do not change with time, and hence, the planning and designs for future can be based on the past observations. In this paper, we have evaluated the validity of the stationarity assumption for meteorological drought risk estimation at the multidecadal scale by comparing drought risk calculated under the assumptions of stationarity and nonstationarity, respectively, in different parts of the continental United States. We find statistically significant nonstationarity in meteorological droughts in the Northwest, upper Midwest, the Northeast, eastern Great Plains and in parts of Nevada, Utah, Arizona, and New Mexico in the Southwest United States, which results in high interdecadal variability of drought risk in these regions. This result demonstrates that the assumption of stationarity can lead to underestimation of drought risk in these regions, thereby exposing water resource systems to failure under severe droughts.

show abstract

Section: Methodsmentioning

confidence: 99%

Evaluation of the Stationarity Assumption for Meteorological Drought Risk Estimation at the Multidecadal Scale in Contiguous United States

Apurv

Cai

2019

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…In the case of GGA distribution, considering that the shape parameter 3  is quite sensitive and difficult to estimate, we assumed it to be constant as other studies did (e.g. : Du et al 2015;López and Francés 2013;Xiong et al 2018;Yan, Xiong, Guo, et al 2017). The Akaike Information Criterion (AIC;…”

Section: Model Selection 215mentioning

confidence: 99%

The influence of global climate and local hydrological variations over streamflow extremes: The tropical-mountain case

Contreras

Mendoza

Pacheco

et al. 2019

Preprint

View full text Add to dashboard Cite

Abstract. Hydrological extremes such as floods and droughts are the most common and threatening natural disasters worldwide. Particularly, tropical Andean headwaters systems are prone to hazards due to their complex climate conditions. However, little is known about the underlying mechanisms triggering such extremes events. In this study, the Generalized Additive Models for Location, Scale and Shape (GAMLSS) were used for investigating the relations between the Annual-Peak-Flows (APF) and Annual-Low-Flows (ALF), respecting to climate and land use/land cover (LULC) changes. Thirty years of daily streamflow data-sets taken from two Andean catchments of southern Ecuador are used for the experimental research. Global climate indices (CI), describing the large-scale climate variability were used as hypothetical drivers explaining the extreme's variations on streamflow measures. Additionally, the Antecedent-Cumulative-Precipitation (AP) and the Standardized-Precipitation-Index (SPI), and LULC percentages were also included as possible direct drivers – synthetizing local climate conditions and localized hydrological changes. The results indicate that AP and SPI clearly explain the extreme streamflow variability. Nonetheless, global variables play a significant role underneath the local climate. For instance, ENSO and CAR exert influence over the APF, while ENSO, TSA, PDO and AMO control ALF. Furthermore, it was found that LULC changes strongly influence both extremes; although this is particularly important for relative more disturbed catchments. These results provide valuable insights for future forecasting of floods and droughts based on precipitation and climate indices, and for the development of mitigation strategies for mountain catchments.

show abstract

“…Unfortunately, the paucity of progress in model development is partly due to structural inadequacy. For example, dynamic components in hydrological models are oversimplified due to a poor understanding of their physical mechanisms (Xiong et al, 2019;Deng et al, 2016Deng et al, , 2018Dakhlaoui et al, 2017;Sarhadi et al, 2016;Pathiraja et al, 2016;Ouyang et al, 2016). Previous studies have demonstrated that the assumption of time-invariant parameters is usually inappropriate.…”

Section: Introductionmentioning

confidence: 99%

Dynamics of hydrological-model parameters: mechanisms, problems and solutions

Lan

Lin

et al. 2020

Hydrol. Earth Syst. Sci.

Self Cite

View full text Add to dashboard Cite

Abstract. It has been demonstrated that the application of time-varying hydrological-model parameters based on dynamic catchment behavior significantly improves the accuracy and robustness of conventional models. However, the fundamental problems for calibrating dynamic parameters still need to be addressed. In this study, five calibration schemes for dynamic parameters in hydrological models were designed to investigate the underlying causes of poor model performance. The five schemes were assessed with respect to the model performance in different flow phases, the transferability of the dynamic parameters to different time periods, the state variables and fluxes time series, and the response of the dynamic parameter set to the dynamic catchment characteristics. Furthermore, the potential reasons for the poor response of the dynamic parameter set to the catchment dynamics were investigated. The results showed that the underlying causes of poor model performance included time-invariant parameters, “compensation” among parameters, high dimensionality and abrupt shifts in the parameters. The recommended calibration scheme exhibited good performance and overcame these problems by characterizing the dynamic behavior of the catchments. The main reason for the poor response of the dynamic parameter set to the catchment dynamics may be the poor convergence performance of the parameters. In addition, the assessment results of the state variables and fluxes and the convergence performance of the parameters provided robust indications of the dominant response modes of the hydrological models in different sub-periods or catchments with distinguishing catchment characteristics.

show abstract

Multiple causes of nonstationarity in the Weihe annual low-flow series

Cited by 28 publications

References 52 publications

Evaluation of the Stationarity Assumption for Meteorological Drought Risk Estimation at the Multidecadal Scale in Contiguous United States

Evaluation of the Stationarity Assumption for Meteorological Drought Risk Estimation at the Multidecadal Scale in Contiguous United States

The influence of global climate and local hydrological variations over streamflow extremes: The tropical-mountain case

Dynamics of hydrological-model parameters: mechanisms, problems and solutions

Contact Info

Product

Resources

About