Quantifying the relative importance of potential evapotranspiration and timescale selection in assessing extreme drought frequency in conterminous China
Introduction
Extreme drought is defined according to precipitation deficit and evapotranspiration within a specific timeframe relative to normal ranges (Hao et al., 2018; Um et al., 2020). The timeframe used for measuring extreme drought (referred to as timescale) can range from days to months (Zhao et al., 2017; Zhang et al., 2017). As one of the biggest threats to social, economic and environmental sustainability (Kelley et al., 2015; Xu et al., 2015), extreme drought is predicted to increase in the next several decades (IPCC, 2014; Smirnov et al., 2016). Assessing extreme drought frequency (EDF) is the basis for designing proactive, risk-based response strategies to reduce the negative impacts of droughts (Joiner et al., 2018; Wang and Asefa, 2019).
A common measure for quantifying EDF is a range of drought indices, which use single numeric values estimated from various hydroclimatic variables (Wu et al., 2015; Mukherjee et al., 2018). Among various drought indices, Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) have been commonly used due to their applicability to multiple time scales, simplicity, and standardization (Lorenzo-Lacruz et al., 2010; Wu and Chen, 2019). Xia et al. (2018) used SPI, which only considers precipitation for identifying the EDF and is based on precipitation at 1-, 3- and 6-month timescales. Zhao et al. (2017) constructed SPEI based on precipitation and actual evapotranspiration data for the evaluation of extreme and severe drought in China at multiple timescales ranging from 1 to 48 months.
Previous observations and investigations suggested the major sources of uncertainty in assessing EDF arise from the calculation of potential evapotranspiration (PET) and the choice of timescale (Zhao et al., 2017; Liu et al., 2019). Specifically, conflicting results for estimated EDF depend on whether PET is incorporated in the computation of drought indices (Dai and Zhao, 2017) since the influence of temperature on extreme drought can be quantified and described by PET (Valipour et al., 2017). Wang et al. (2015) investigated extreme drought change in China using the SPI for 1961–2012 and found no increase in EDF throughout China. Chen and Sun (2015) detected changes in extreme drought in China using the SPEI and reported that extreme drought had become more frequent across China. Different timescales identify different time lags for the start of water shortage (Zhang et al., 2017). Values of drought indices at different timescales are substantially different, thus leading to discrepancies in the evaluation of EDF among different studies (Fluixásanmartín et al., 2018). Ayantobo et al. (2017) applied the SPI and SPEI to assess the EDF in conterminous China at multiple timescales ranging from 1-week to 24-months and found that extreme drought occurred with lower frequency as the timescale increased. Although previous studies have verified the significant impacts of PET and timescale separately, how PET and timescale combine and interact with each other to shape the spatial and temporal patterns of EDF remains unknown.
The sensitivity of EDF to PET and timescale selection varies temporally and spatially, from monthly to annual timescales and from regional to national spatial scales (Zhang and Shen, 2019). The spatial heterogeneity of climatic conditions determines the causes and consequences of extreme drought (Zhang et al., 2017; Um et al., 2020). In the arid region, characterized by low precipitation and intense PET, water deficits could be intensified by high temperature (Luo et al., 2017; Um et al., 2020) and have a strong cumulative effect for a given timescale (Peng et al., 2019). Cumulative water deficits can enhance the magnitude of drought to increase the EDF assessed by drought indices (Liu et al., 2016; Pastor-Guzman et al., 2018). In humid regions, however, extreme droughts mostly originate from precipitation deficit (Luo et al., 2017; Um et al., 2020) with a weak cumulative effect (Peng et al., 2019). Apart from the spatial variability, the temporal extent of extreme drought is another critical aspect because the magnitude of water deficit is determined by various hydroclimatic variables which are subject to the variations in timescale (Dey and Mishra, 2017, Ning et al., 2019). For instance, the intensity and spatial pattern of precipitation greatly influence the monthly PET but may have little influence on the mean annual PET (Zhang et al., 2020). Therefore, the differences of PET can be safely ignored on mean annual EDF but not at the monthly scale (Greve et al., 2016). To consider the differences in the temporal and spatial patterns of EDF, the contribution of PET and timescale should be independently investigated for monthly and annual extreme drought events in regions with different climate conditions.
Earlier studies have focused on the uncertainties in the driving climate variables (Ahmed et al., 2018; Mukherjee et al., 2018), whereas the direct contributions of PET and timescale to EDF have been overlooked (Tirivarombo et al., 2018). Therefore, the objective of this study is to analyze the temporal and spatial patterns of EDF using drought indices and to determine the individual and interactive contributions of PET and timescale. We used the monitoring data from 695 meteorological stations operating in conterminous China from 1970 to 2010 with the SPI and SPEI at 3- and 12-month timescales to calculate the EDF. Then, we modelled the monthly and annual effects of PET and timescale at the regional and national scales. We employed sensitivity analysis to assess the response of PET to changes in driving climate variables (i.e. precipitation and temperature), as well as to better understand the relative importance of driving climate variables affecting the impact of PET on EDF. The analysis framework used in this study was shown useful for improving understanding of the relationship between drought indices and extreme drought, and it could prove helpful in providing appropriate drought indices to estimate EDF and model other datasets which consider extreme drought.
Section snippets
Study area and meteorological data
Our study area is the entirety of the terrestrial land of China (Fig. 1), which has diverse patterns of precipitation and evapotranspiration. This complexity increases uncertainties when attempting to quantify the frequency of extreme drought across the entire country (Ayantobo et al., 2017; Yao et al., 2018). The aridity index (AI), defined as the ratio of annual evapotranspiration and precipitation, (Ahmed et al., 2018) is suitable for classifying the type of climate and disentangling some of
Annual variations at the national scale
Substantial differences in annual EDF were observed among SPI and SPEI at 3- and 12-month timescales (Fig. 2). Significant variations and differences among drought indices were found in 1970–1980 and 1995–2010. During 1980–1995, annual EDF from SPI and SPEI showed relatively small variations and discrepancies (Fig. 2a). The average number of extreme drought events during 1970–2010 for SPI3, SPI12, SPEI3, and SPEI12 were 66, 92, 99, and 103, respectively (Fig. 2b). Drought indices with 3-month
Discussion
Different conclusions were obtained in publications concerning EDF in China (Table 2). Most research indicated that EDF had an increasing trend over the last decade due to decreasing precipitation and increasing PET (Sun et al., 2017). This was also confirmed by the current research (Fig. 2, Fig. 7, Fig. 8). Our results indicated that the interactive effects of PET and timescale selection on EDF varied with the temporal and spatial scaling of the extreme drought occurrence. For conterminous
Conclusions
For conterminous China, PET exerted a greater effect than timescale on EDF estimates, most of the time. This pattern was more obvious in the arid and semi-arid regions, where the relative importance of the PET effect was 25% greater than the timescale effect. The roles of PET and timescale on EDF variation were different. PET was the primary contributor to the temporal variability of EDF, whereas timescale selection had a stronger effect on the spatial variability of EDF. Additionally, the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research was funded by the National Key Research and Development Program of China (Grant No. 2016YFA060084) and National Natural Science Foundation of China (Grant No. 31961133027, 31971486).
References (80)
- et al.
Impacts of climate variability and change on seasonal drought characteristics of Pakistan
Atmos. Res.
(2018) - et al.
Spatial comparability of drought characteristics and related return periods in mainland China over 1961–2013
J. Hydrol.
(2017) A new uncertainty importance measure
Reliab. Eng. Syst. Saf.
(2007)- et al.
Separating the impacts of climate change and human activities on streamflow: a review of methodologies and critical assumptions
J. Hydol.
(2017) - et al.
A global quantitation of factors affecting evapotranspiration variability
J. Hydrol.
(2020) - et al.
A modified Mann-Kendall trend test for autocorrelated data
J. Hydrol.
(1998) - et al.
Global relationships among traditional reflectance vegetation indices (NDVI and NDII), evapotranspiration (ET), and soil moisture variability on weekly timescales
Remote Sens. Environ.
(2018) - et al.
Inter-comparison of satellite-retrieved and global land data assimilation system-simulated soil moisture datasets for global drought analysis
Remote Sens. Environ.
(2019) - et al.
The impact of droughts and water management on various hydrological systems in the headwaters of the Tagus River (Central Spain)
J. Hydrol.
(2010) - et al.
Spatial and temporal analysis of drought variability at several time scales in Syria during 1961–2012
Atmos. Res.
(2018)
Interaction of vegetation, climate and topography on evapotranspiration modelling at different time scales within the Budyko framework
Agric. For. Meteorol.
Remote sensing of mangrove forest phenology and its environmental drivers
Remote Sens. Environ.
Drought monitoring and analysis: standardised precipitation evapotranspiration index (SPEI) and standardised precipitation index (SPI)
Phys. Chem. Earth
Variability of the aridity index and related drought parameters in Greece using climatological data over the last century (1900–1997)
Atmos. Res.
Impacts of potential evapotranspiration on drought phenomena in different regions and climate zones
Sci. Total Environ.
Selecting the best model to estimate potential evapotranspiration with respect to climate change and magnitudes of extreme events
Agric. Water Manag.
Spatiotemporal trends of dryness/wetness duration and severity: the respective contribution of precipitation and temperature
Atmos. Res.
Spatio-temporal variation of drought in China during 1961–2012: a climatic perspective
J. Hydrol.
Drought evolution, severity and trends in mainland China over 1961–2013
Sci. Total Environ.
Comparison of the climate indices based on the relationship between yield loss of rain-fed winter wheat and changes of climate indices using gee model
Sci. Total Environ.
Response of vegetation to different time-scales drought across China: spatiotemporal patterns, causes and implications
Glob. Planet. Chang.
A unified framework of water balance models for monthly, annual, and mean annual timescales
J. Hydrol.
Timescale differences between South-PDSI and SPEI for drought monitoring in China
Phys. Chem. Earth
Crop Evapotranspiration-Guidelines for Computing Crop Water Requirements-FAO Irrigation and Drainage Paper 56
SPEI: Calculation of the Standardized Precipitation-Evapotranspiration Index. R Package Version 1.17.0
A review of statistical methods for determination of relative importance of correlated predictors and identification of drivers of consumer liking
J. Sens. Stud.
Changes in drought characteristics over China using the standardized precipitation evapotranspiration index
J. Clim.
A methodology for flash drought identification: application of flash drought frequency across the United States
J. Hydrometeorol.
Global warming and 21st century drying
Clim. Dyn.
R: A language and environment for statistical computing
Uncertainties in historical changes and future projections of drought. Part i: estimates of historical drought changes
Clim. Chang.
Searching for the optimal drought index and timescale combination to detect drought: a case study from the lower Jinsha River basin, China
Hydol. Earth Syst. Sci.
A two-parameter Budyko function to represent conditions under which evapotranspiration exceeds precipitation
Hydrol. Earth Syst. Sci.
Relative importance for linear regression in R: the package relaimpo
J. Stat. Softw.
Seasonal drought prediction: advances, challenges, and future prospects
Rev. Geophys.
Monitoring the 1996 drought using the standardized precipitation index
Bull. Am. Meteorol. Soc.
Statistical Methods in Water Resources
Dry-wet variations and cause analysis in Northeast China at multi-time scales
Theor. Appl. Climatol.
Impacts, Adaptation, and Vulnerability: Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
Cited by (9)
A novel approach for the identification of conservation priority areas in mountainous regions based on balancing multiple ecosystem services – A case study in the Hengduan Mountain region
2022, Global Ecology and ConservationCitation Excerpt :Moreover, the inclusion of more ES types involved in the process of determining the optimal CPAs makes it more difficult to consider multiple services at the same time (Li et al., 2021). Therefore, an accurate identification of typical regional ESs is crucial for the accuracy and scientific nature of the results (Ma et al., 2021a, 2021b, 2021c). In addition, the use of more risk values results in more conservation scenarios being obtained, making it more likely to obtain the precise location of the optimal CPAs (Zhang et al., 2014).
Quantitative driving analysis of climate on potential evapotranspiration in Loess Plateau incorporating synergistic effects
2022, Ecological IndicatorsCitation Excerpt :ET0 as the theoretical upper limit of evaporation, is the amount of water fully supplied from the lower cushion surface to the air (Li et al., 2017; Zhou et al., 2020). It is an important index for assessing regional dry-wet conditions and water resource deficit, and also a necessary condition for calculating crop water demand and actual evapotranspiration (Qin et al., 2021; Singer et al., 2021; Ma et al., 2021a,b). Accurate estimation of ET0 is very important in irrigation scheduling, water use optimization and other hydrological work (Mohammadi and Mehdizadeh, 2020; Hadadi et al., 2022).
Can ecological landscape pattern influence dry-wet dynamics? A national scale assessment in China from 1980 to 2018
2022, Science of the Total EnvironmentCitation Excerpt :The formation and occurrence mechanism of drought is complex, associating with various factors concurrently (Yao et al., 2020). Many ecological processes occurred on land surface such as vegetation evapotranspiration and runoff would interact with regional wet-dry environment (Aguilos et al., 2021; Kyatengerwa et al., 2020; Ma et al., 2021; Wang et al., 2019). In recent years, in addition to the direct effects of climatic factors on drought (Vicente-Serrano et al., 2015; Yang et al., 2020), some indirect effects associated with human activities, especially land use/land cover change, have gradually attracted worldwide attention (Fan et al., 2015; Júnior et al., 2015; Peng et al., 2019).