Elsevier

Journal of Hydrology

Volume 591, December 2020, 125549
Journal of Hydrology

Research papers
Prognostic and diagnostic assessment of hydrological drought using water and energy budget-based indices

https://doi.org/10.1016/j.jhydrol.2020.125549Get rights and content

Highlights

  • Novel approach was proposed to assess the hydrological drought using multi-sensored remote sensing data.

  • Hydrological drought was assessed using prognostic and diagnostic approach.

  • EWDI and WBDI was developed using remote sending data.

  • Evapotranspiration is a one of major factor affecting hydrological drought.

Abstract

Long-term meteorological drought can lead to hydrological drought by restricting water resources required by humans, such as streamflow and reservoir storage. We developed two new indices for monitoring hydrologic drought-based satellite-derived evapotranspiration, a major factor in hydrological drought occurrence. The Water Budget-based Drought Index (WBDI) estimates potential runoff by the differences between precipitation and evapotranspiration using water budget analysis while the Energy-based Water Deficit Index (EWDI) is an index of available water capacity through evapotranspiration and soil moisture based on solar radiation using energy budget analysis. We used these, along with the existing Standardized Precipitation-Evapotranspiration Index (SPEI, based on precipitation, atmospheric temperature, and evapotranspiration), to map the spatiotemporal patterns of drought on the Korean Peninsula from 2001 to 2014. For validation against actual drought conditions, we compared the results with streamflow data from five gauging stations in South Korea. EWDI—the diagnostic approach—performed best when assessing current hydrological drought conditions, while WBDI and SPEI—prognostic approach—best captured drought conditions after 2–3 months of lag time. Our results confirmed that evapotranspiration is a major factor affecting hydrological drought; the new methods demonstrated here make it possible to evaluate drought through diagnostic and prognostic perspectives depending on the situation, thereby improving scientific drought evaluation capacity.

Introduction

In the last two decades, prolonged droughts (lasting for over a year) have frequently occurred around the world (Keyantash and Dracup, 2002, Nalbantis and Tsakiris, 2009, Mu et al., 2013). Many can be classified as hydrological droughts that cause a shortage of available water resources needed for human life, such as river discharge or reservoir storage (Van Loon, 2015, Zhang et al., 2015). The basic concept of analyzing hydrological drought is the hydrologic cycle. Precipitation and evapotranspiration are essential factors to consider when evaluating hydrological drought due to their close relationship with available water resources in the hydrologic cycle (Zhang et al., 2019). Many studies have been conducted to analyze hydrological drought through changes in precipitation and evapotranspiration (Doesken and Garen, 1991, Anderson et al., 2007, Anderson et al., 2011, Anderson et al., 2013, Leng et al., 2015, Huang et al., 2017, Liu et al., 2017, Vicente-Serrano et al., 2010, Wu et al., 2018, Guo et al., 2020). Anderson et al., 2007, Anderson et al., 2011, Anderson et al., 2013) developed the Evaporative Stress Index (ESI), a drought index, using the ratio of potential to actual evapotranspiration and produced a drought map of the United States for comparison with the existing Standardized Precipitation Index (SPI) and the Palmer Drought Severity Index (PDSI). Vicente-Serrano et al. (2010) developed a Standardized Precipitation Evapotranspiration Index (SPEI) to assess drought based on the probabilistic distribution of precipitation and potential evapotranspiration and compared its hydrological monitoring performance with the SPI. Liu et al. (2017) developed a new drought index, the standardized wetness index (SWI), based on the ratio of SPEI to actual evapotranspiration. Huang et al. (2017) examined the propagation time from meteorological to hydrological drought through seasonal characteristics. Wu et al. (2018) analyzed relationship between hydrological and meteorological drought for reservoir operation perspectives. Doesken and Garen (1991) proposed a hydrological drought index, Surface Water Supply Index (SWSI), to identify drought conditions associated with hydrological fluctuations using streamflow, reservoir storage, and precipitation. However, these drought indices have several limitations. In the case of the SPEI, the requirement for a serially complete dataset for both temperature and precipitation may limit its use due to insufficient data being available. Being a monthly index, rapidly developing drought situations may not be identified rapidly (Vicente-Serrano et al., 2010, WMO, 2016). For the SWSI, as data sources change or additional data are included, the entire index has to undergo recalculation to account for these changes in the inputs, making it difficult to construct a homogeneous time series. As calculations may vary between basins, it is difficult to compare basins or homogeneous regions (Doesken and Garen, 1991, WMO, 2016). The conventional drought indices have been considered based on the hydrometeorological variables measured by the network of in situ data tools, whereas using satellite data is robust substituted by providing decisive hydrometeorological variables for drought analysis at the enormously higher spatial scale than the capability of in situ network devices (Sur et al., 2019). Because satellite-based datasets have a fixed spatio-temporal resolution, there may be limitations depending on the research objectives. For example, in a study that performs a drought analysis of a subcatchment scale, if a passive microwave sensor output is used, accurate analysis cannot be performed because the spatial resolution is too coarse. However, since the goal of this study is to analyze the monthly drought on a regional scale, it is essential to use satellite data in this study. In addition, there is another advantage to using satellite data. Conventional drought indices cannot be calculated in ungauged areas, while satellite data can be used to overcome these shortcomings. Notably, the Korean peninsula, which is a study area, includes North Korea, and it is challenging to obtain ground observation data for North Korea due to political problems. However, because this study used satellite data, drought analysis of the entire Korean peninsula, including North Korea, could be performed. Details of the satellite data utilized are given in section 2.2.

Other previous studies analyzing hydrological droughts have developed many drought indices. A common feature of the indices developed is the application of the circulation of water and energy. Among them, Zhang et al. (2019) developed a standardized moisture anomaly index (SZI) using potential evapotranspiration and analyzed the global scale in different climatic zones. Particularly, the SZI was estimated using the water-energy balance approach. In the basic water budget equation, the difference between precipitation and evapotranspiration was assumed as runoff, and the value of evapotranspiration was calculated by applying the energy balance concept. The method of standardizing the difference between rainfall and evapotranspiration proposed by Zhang et al. (2019) was used to analyze the hydrological drought in this study. The drought index using the difference between rainfall and evapotranspiration is termed as the Water Budget based Drought Index (WBDI) in this study. The WBDI is advantageous in its ability to predict potentially available water content for the near future, using the water balance equation. Details of the newly developed WBDI are given in section 3.1.1. In addition, for comparison, we applied the Energy-based Water Deficit Index (EWDI, proposed by Sur et al., 2015a, Sur et al., 2015b) formulated by combining actual and potential evapotranspiration and soil moisture through the energy budget algorithm.

The calculated WBDI and EWDI were compared by estimating these along with the SPEI (based on precipitation and temperature fluctuations) for the Korean Peninsula from 2001 to 2014. The results were verified by comparison with actual drought conditions drawn from observed streamflow data. The year 2001 and 2014, when the most severe drought occurred on the Korean Peninsula, were set as the testing period. Both WBDI and EWDI are revised indices based on the drought index developed in previous studies. The main difference between the two indices is the physical concept of drought analysis for water- and energy budget perspectives. A study that comparatively analyzes and verifies hydrological phenomena according to two concepts has been conducted before, but few studies have been conducted on the Korean Peninsula region, which is a study area. Through comparison and verification in a specific area, it is possible to analyze whether hydrological phenomena are interpreted according to physical concepts.

Section snippets

Study area

The Korean Peninsula is located in northeast Asia between 33 and 43° latitude and 124–132° longitude, with a total area of 219,020 km2 and altitude ranges of 0–2,744 m (Hwang et al., 2013). The region has a monsoon climate with an average annual rainfall of 1,100 mm that latitudinally varies from North Korea (919.7 mm) to South Korea (1,307.7 mm). Land cover is categorized by the IGBP classification using MODIS global landcover product (MCD12Q1) (Fig. 1).

Data

To estimate drought indices, various

Hydrological drought indices

In this study, hydrological droughts were compared in terms of the water and energy balance equation. Evapotranspiration is the most important variable to be considered when applying the water and energy balance equation (Yilmaz et al., 2014, Farhadi et al., 2014, Farhadi et al., 2016). Both equations use evapotranspiration as an input variable, wherein the concept of hydrological drought index is determined by the physical meaning of each evapotranspiration. First, in the case of EWDI based on

Temporal patterns of hydrological drought indices

As per the statistical analysis of past hydrometeorological records, since the year 2000, severe droughts have occurred in the Korean Peninsula in 2001, 2007 to 2008, and 2014 to 2015 (Kwon et al., 2016, Jang et al., 2016; Park et al., 2018a, Park et al., 2018b). Jang et al. (2016) selected the most severe periods of drought by analyzing outliers of monthly rainfall over the last 40 years. Among them, severe spring droughts in the southern part of South Korea were reported in 2001. Kwon et al.

Conclusions

We proposed and tested two new satellite-based hydrologic drought indices based on the water budget (WBDI) and energy budget (EWDI) and applied these, along with the existing SPEI-3 index (based on precipitation, atmospheric temperature, and evapotranspiration), to the Korean Peninsula for the period 2001 to 2014. We then compared the spatial results for the severe drought years of 2001 and 2014 and verified the accuracy of all three methods by comparison with streamflow data from five gauging

CRediT authorship contribution statement

Chanyang Sur: . Seo-Yeon Park: Methodology, Investigation. Jong-Suk Kim: . Joo-Heon Lee: Conceptualization.

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.

Acknowledgement

This research was supported by Korea Environment Industry & Technology Institute (KEITI) though Water Management Research Program, funded by Korea Ministry of Environment (MOE) (79616) and this work was supported by the faculty research fund of Sejong University in 2020.

References (41)

  • Bae, S., Lee, S.-H., Yoo, S.-H., Lim, T. 2018. Analysis of Drought Intensity and Trends Using the Modified SPEI in...
  • Doesken, N.J., Garen, D., 1991. Drought monitoring in the Western United States using a surface water supply index....
  • L. Farhadi et al.

    Estimation of land surface water and energy balance parameters using conditional sampling of surface states

    Water Resour. Res.

    (2014)
  • L. Farhadi et al.

    Mapping land water and energy balance relations through conditional sampling of remote sensing estimates of atmospheric forcing and surface states: mapping water and energy balance relations

    Water Resour. Res.

    (2016)
  • G.H. Hargreaves et al.

    Reference crop evapotranspiration from temperature

    Appl. Eng. Agric.

    (1985)
  • K. Hwang et al.

    Estimation of instantaneous and daily net radiation from MODIS data under clear sky conditions: a case study in East Asia

    Irrig. Sci.

    (2013)
  • Jang, S., Rhee, J., Yoon, S., Lee, T., Park, K., 2017. Evaluation of GPM IMERG Applicability using SPI based Satellite...
  • Jang, H.W., Cho, H.W., Kim, T.W., Lee, J.H. 2016. Quantitative characterization of historical drought events in Korea...
  • J. Keyantash et al.

    The quantification of drought: an evaluation of drought indices

    Bull. Amer. Meteor. Soc.

    (2002)
  • H. Kim et al.

    Validation of MODIS16 global terrestrial ET product in various climate and land cover types in Asia

    KSCE J. Civil Eng.

    (2012)
  • Cited by (11)

    • Propagation from meteorological to hydrological drought and its application to drought prediction in the Xijiang River basin, South China

      2023, Journal of Hydrology
      Citation Excerpt :

      Currently, research on drought propagation is mainly investigating the statistical relation between meteorological and hydrological drought indices by correlation analysis, wavelet transform and mutual entropy, etc. (Junqueira et al., 2020; Yu et al., 2020; Ding et al., 2021; Zhu et al., 2021; Rose and Chithra, 2022). These researchers pointed out that drought propagation has the typical characteristics of pooling, lag, and lengthening (Shin et al., 2018; Sur et al., 2020; Lohpaisankrit and Techamahasaranont, 2021; Wu et al., 2022). The findings have improved our understanding of the relationship between long-term drought index series.

    • Evaluation of drought propagations with multiple indices in the Yangtze River basin

      2022, Journal of Environmental Management
      Citation Excerpt :

      This study adopted SPEI to evaluate the meteorological drought index because SPEI reflected the hydro-climatic water balance for the evaluation of difference between precipitation and potential evaporation. Hydrological drought is characterized as a considerable shortage of water in the land surface of the water balance (Khatiwada and Pandey, 2019; Sun et al., 2018; Sur et al., 2020; Yang et al., 2020). Various indices to evaluate hydrological droughts including surface water supply index (SWSI), Palmer hydrological drought index (PHDI), and standardized runoff index (SRI) have been applied (Esfahanian et al., 2017; Guttman, 1991; Shafer and Dezman, 1982; Shukla and Wood, 2008; Zhu et al., 2016).

    • Drought vulnerability assessment: Solution for risk alleviation and drought management among Iranian farmers

      2022, International Journal of Disaster Risk Reduction
      Citation Excerpt :

      Hydrological drought usually happens later than meteorological and agricultural droughts because the effects of precipitation decline, including soil moisture, the river flows, and the level of groundwater and water resources, typically appear with some delay in hydrological systems [40]. Hydrological drought happens when the balance level of surface and underground water reserves is lower than their typical level [41]. Socioeconomic drought, addressed in this research, differs from other types of droughts as it relates human activities to the elements of climatic, agricultural, and hydrological droughts [42].

    View all citing articles on Scopus
    View full text