Abstract
A flash drought (FD) event is a relatively new form of severe occurrence, described by the quick onset and intensification of drought situations with serious consequences. This paper aims to understand the wide variety of studies and serve as a basis for future research potentials of FDs. An effort has been made to understand the advantages and limitations of the existing indices used for FD identification. A study in India was carried out for the identification of FDs using daily gridded data of soil moisture for eight days (octad). The results reveal that the Indus basin faced around 82% and the Cauvery basin faced about 88% of severe FDs during the monsoon and non-monsoon seasons, respectively. Additionally, the results show that the Indus basin faces FD in the monsoon season since the basin has mostly barren areas and wasteland. This review also highlights the role of humidity, wind and radiation, soil characteristics, climate oscillations, and the relation between FDs and heavy-rainfall-induced flash floods. Furthermore, the paper has depicted some specific research-needs to monitor, forecast, plan, and respond with crucial points.
HIGHLIGHTS
This study discusses the most extensive indices used to identify FDs.
A study is carried out in India to identify FDs using daily gridded soil moisture data for eight days (octad).
The study briefly explains the factors that need to be included in identifying and monitoring FDs.
The study also depicts the crucial points for future research-needs to monitor, forecast, plan, and respond to FDs.
INTRODUCTION
Droughts are frequently seen as slow-moving natural hazards. However, some severe droughts have such a rapid onset or intensification that it appears as though they occur in a ‘flash’, making them and their consequences challenging to forecast and plan for (NIDIS 2021). The term ‘flash droughts (FDs)’ is commonly used to describe droughts that develop quickly over a short time (Osman et al. 2021). FDs have been a research topic since 2002 (Svoboda et al. 2002), and scientific research on this concern is currently increasing, with a significant increase in publications beginning from 2013 until the present (Lisonbee et al. 2021). The publications increased considerably after the FD of 2012 in the central USA, and the research is continually rising every year (Liang & Yuan 2021). The research on FD has been getting a lot of attention in recent years (Hu et al. 2021; Jiao et al. 2021; Liang & Yuan 2021; Lisonbee et al. 2021; Mishra et al. 2021; Osman et al. 2021) in scientific fields. A recent study indicates that the intensification rates of drought have accelerated over sub-seasonal timescales (Pendergrass et al. 2020). Furthermore, the study highlights a shift toward a greater prevalence of flash droughts in 74% of the global regions that were identified in the Intergovernmental Panel on Climate Change (IPCC) Special Report on extreme events during the past 64 years (IPCC 2021; Yuan et al. 2023). Areas with a high incidence of flash droughts include Brazil, the Sahel, the Great Rift Valley, and India. Additionally, there are noteworthy local hotspots in the central United States, southwestern Russia, and northeastern China (Christian et al. 2021). Another study results indicate that regions experiencing flash droughts are typically those with humid or semi-humid climates. These regions include Southeast Asia, East Asia, the Amazon Basin, eastern North America, and southern South America (Qing et al. 2022). However, Sreeparvathy & Srinivas (2022) identified that areas with the highest incidence of meteorological flash droughts are primarily located in arid and semi-arid regions. FDs occur over sub-seasonal-to-seasonal timelines (‘weeks’ or ‘weeks to months’), posing a novel challenge to the growing interest in enhancing their identification and forecast climate change (Pendergrass et al. 2020). FDs are caused by prolonged periods of soil moisture depletion (Ford & Labosier 2017; Yuan et al. 2019; Liu et al. 2020; Osman et al. 2021), less precipitation (Mo & Lettenmaier 2016), exceptionally high temperature (e.g., heatwaves), and strong wind (Chen et al. 2019), all of which result in abnormally high evaporation and evapotranspiration rates (Christian et al. 2019a; Li et al. 2020a). FD events appear to occur more frequently than is commonly believed and can inflict significant agricultural, water, and economic losses that intensify in the absence of timely forecast and identification (NIDIS 2021). For instance, some of the impacts of past FD events in different countries are presented:
India: A recent study shows that FDs affected around 10%–15% of rice crop and maize crop cultivation lands in the country between 1951 and 2018, and about 40% of India faced severe FDs in 1979 (Mahto & Mishra 2020).
China: During the 2013 summer, an FD hit 13 provinces in southern China, destroying almost 2 million hectares of crops in Guizhou and Hunan alone (Yuan et al. 2015).
Russia: FD that hit western parts of Russia in the summer of 2010 was more rapid and intensified, which affected the wheat yields that reduced by 70% and led to a decline in wheat production in Russia (Christian et al. 2020). The 2010 FD across western Russia had at least these two immediate impacts: (1) the severe loss of life (about 11,000 deaths) caused by the heatwave along with FD and (2) the loss of approximately 20 million MT of wheat (34% decline) compared with Russia's production in the preceding two years (Hunt et al. 2021).
Central US: The 2012 FD had a significant and extensive impact on vegetation, agriculture, and the economy of the United States, with more than $30 billion in agricultural losses (Otkin et al. 2016; Liang & Yuan 2021; Lesinger & Tian 2022).
US Northern Plains: The resultant impact of the 2017 Northern Plains FD in the United States was about $2.6 billion in agricultural losses (NIDIS 2021). He et al. (2019) found that the FD led to a 25% decline in cropland ET, a 6% reduction in crop yield over the area between April and September 2017.
The objective of this paper is to combine the concepts related to FD, bring attention to the aspects of FD that are already understood and those that are not, recognize the inadequacies in the present indices used to identify and define FD, and depict some specific points for research needs. To achieve this, the paper will address the following questions:
What are the advantages and limitations of the existing indices used for identifying or defining FDs?
What are the things that the researchers miss and how to include them in improving identification and prediction?
What are the research-needs to monitor, forecast, plan, and respond to FDs?
DROUGHT INDICES USED TO IDENTIFY FDS
Soil moisture indices
Soil Moisture Index (SMI): The SMI was developed by the High Plains Regional Climate Center to detect the quick start of a drought event by analyzing the reported aridness of the soil compared with the capability of vegetation to obtain water. The SMI ranges from 5 to −5, where 5 denotes field capacity and −5 denotes the wilting point as it is a soil–moisture-based drought index (Sridhar et al. 2008; Hunt et al. 2009). The SMI is based on a range of ±5, with negative numbers signifying times of water shortage. A drought event is defined as a period with negative SMI values, and the average value for the 0–0.5 m layer soil profile is used to determine the daily SMI. The SMI has been used to identify the onset of FDs based on the rapid decline in soil moisture in a three-week period (Mozny et al. 2012; Hunt et al. 2021).
Advantages: The advantage of using SMI is that it is feasible to acquire data from the remote sensing system and to evaluate the frequent measurements of satellite observations as well as to validate the accuracy. Also, SMI has a benefit over soil reflectance as it reflects soil–water conditions better.
Limitations: SMI is calculated as the average value for the 0–0.5 m soil layer. However, it does not include SMI values for multiple soil layers. In fact, during the spring season, the upper layer of soil (sandy soils) may dry out and obstruct the growth of crops. The zero value of SMI indicates no drought. However, it can lead to drought conditions or maybe recovery from drought.
Soil Moisture Volatility Index (SMVI):Osman et al. (2021) proposed a definition using SMVI, which is constructed based on average root zone soil moisture (RZSM). In this definition, an FD is said to occur when (1) the one-pentad (five-day) running RZSM falls below the four-pentad (20-day) running average for a period of at least four pentads; and (2) by the end of the period, RZSM drops below the 20th percentile for that time of the year. This index is sensitive to disruptions during the beginning of drought, though rain episodes can change it (Osman et al. 2021). This index demonstrates an ability to identify the start of FDs irrespective of the vegetation or wetness conditions of a particular area similar to the detected impacts on vegetation (Osman et al. 2022). The SMVI is used to estimate the incidence rate, spatial distribution, and seasonality of FD across the contiguous United States (Osman et al. 2021).
Advantages: The advantage of the SMVI approach lies in its capability to detect rapid changes regarding a slower drying trend.
Limitations: SMVI shows an ability to describe the onset of FD events. However, it is restricted to reporting the severity and intensity of FD events. The index is only applicable for intervals in drought onset, and the index must be reset with rainfall events.
Evapotranspiration indices
Evaporative Demand Drought Index (EDDI): The EDDI is an experimental tool that uses the anomalies of evaporative demand to indicate the spatial coverage and severity of FDs (Hobbins et al. 2016). The potential evapotranspiration (PET) detects and monitors FD onset and captures FDs triggered by climate variables (Won & Kim 2020; Noguera et al. 2021). The EDDI has significant strength in capturing the early indications of water stress from timescales of one to 12 weeks since 1979, making it a powerful system like USDM for FD preparedness (McEvoy et al. 2016). The EDDI can verify the occurrences of historical FD events to provide an early-warning system. Therefore, this index is used in many studies (Yao et al. 2018; Christian et al. 2019a; Nguyen et al. 2019; Li et al. 2020b; Parker et al. 2021).
Advantages: The main strength of EDDI is its early warning abilities and its efficiency in the attribution of drought changing aspects. The EDDI is a powerful tool for drought preparedness as it acquires the signals of water stress at weekly to monthly timescales. The EDDI can also be deconstructed to discern the impact of individual evaporative drivers on the onset and continuity of drought, including factors like radiation and advection components. The EDDI can report a diversity of frequent droughts without precipitation data.
Limitations: EDDI performs only with evaporative demand and with no other variables (e.g., precipitation). Therefore, it may not be reasonable to interpret FDs using only evaporative demand. EDDI is limited to only the contiguous United States.
Evaporative Stress Index (ESI): The ESI represents the ratio of actual evapotranspiration to PET, i.e., ESI = ET/PET (Anderson et al. 2016; McEvoy et al. 2016). The ESI represents these anomalies of the actual-to-potential ET ratio taken through the energy equilibrium using remote sensing parameters such as land surface temperature (LST) and leaf area index (LAI) as inputs (AghaKouchak et al. 2015; Anderson et al. 2016; Lee et al. 2021; Ahmad et al. 2022). The negative ESI value indicates that standardized ET is less than the average for given time intervals, usually, two, four, eight, or 12 weeks, progressing at a seven-day interval, which shows the decreased soil moisture in addition to vegetative stress (Hunt et al. 2021). In many studies, the ESI has precisely identified the FD events and given an early warning in a couple of weeks (Nguyen et al. 2021). If the ESI reduces rapidly and this is sustained for at least two weeks, it indicates the onset of FD (Nguyen et al. 2019).
Advantages: The ESI generates very high-resolution data, and it does not need precipitation data. The ESI has a special significance in regions with minimum in situ precipitation observation.
Limitations: ESI is derived from remote sensing assessment, which is limited to snow-free periods. ESI cannot competently capture the soil moisture from the adjacent fields due to its spatial constraint.
Standardized Evapotranspiration Deficit Index (SEDI): The SEDI is an FD index that integrates climate and ecosystem systems by measuring the difference between actual and PET (Zhang et al. 2019; Li et al. 2020b). The SEDI attempts to identify both heat-wave-driven and water-deficit-driven FDs. It is mostly suitable for monitoring and evaluating FD hazards globally (Kim et al. 2019). Li et al. (2020b) used SEDI to identify FD based on these criteria: (i) the extent of the drought is more than five pentads but less than 12 pentads, (ii) the cumulative SEDI instantaneous intensification rate is at or lower than 25% of the cumulative incidence rate of the variation in the cumulative SEDI value throughout FD growth, and (iii) during the FD growth phase, the average rapid intensification rate is at or less than 40% of the cumulative distribution rate of the change in the cumulative SEDI (Li et al. 2020b).
Advantages: The SEDI has an exceptional application in describing the biological transformations of ecosystems in reaction to the subtleties of drought intensity as compared with other indices related to precipitation and temperature.
Limitations: SEDI can be used for drought assessment, which is related to crops and vegetation, but its use is limited to drought impacts such as river discharge, reservoir storage, or groundwater. SEDI has a tendency for large uncertainty in humid and semi-humid regions. Thus, it is more suitable for arid and semi-arid regions.
Precipitation and evaporation indices
Standardized Precipitation Index (SPI): The SPI has been developed to measure precipitation deficiency over various timescales. It is used as a drought index that considers the relevance of temporal scales in the study of water availability and usage (Guttman 1998). It is used to show the efficiency of detecting FD vs field observations (Hunt et al. 2014). It is significant to know that SPI alone is not likely to lead to the occurrence of FDs since precipitation deficiency is only a single variable of numerous factors that will lead to quick intensification (Otkin et al. 2013). Precipitation is an essential component and a driver of FDs which gives multiple forms of information for calibration and validation of drought conditions (Goyal et al. 2022). In addition, precipitation is not related to evaporative demand, which is specifically significant for FDs (Ford 2022). However, SPI can be used with other indices or parameters, for instance, the SPI and SPEI (Standardized Precipitation Evaporation Index) were utilized in the study of ‘rapid onset drought’, which proved that both SPI and SPEI for one month are relatively sensitive to the onset of FD events and also closely tracked the decrease in soil water (Hunt et al. 2014; Noguera et al. 2021).
Advantages: The SPI has an advantage in computation and usage for any geographical location and for any number of intervals as the SPI values are in units of standard deviation from the long-term mean. The SPI is easy to compute compared with other drought indices as it is based on only precipitation (single input parameter).
Limitations: SPI uses only precipitation, which makes it less linked with the ground conditions. Long-term precipitation data along with climate conditions are required for the accurate identification of FDs.
Standardized Precipitation Evaporation Index (SPEI): The SPEI is a drought index based on SPI, measured by precipitation and PET (Vicente-Serrano et al. 2010), but the only change is that the temperature parameter is included in this index to capture climatic water stability (Vicente-Serrano et al. 2010; Faiz et al. 2020). It is used to detect FD events based on a short time scale of one month and weekly high-frequency data and anomalous decline in SPEI index values in four weeks of a short time period focusing on the development stage of an event (Noguera et al. 2020). The SPEI can simultaneously detect variations in rainfall and temperature, which are constructed using weekly rainfall and PET (Hu et al. 2021). A previous study stated that drought indices such as SPI and SPEI probably cannot identify/monitor an FD as they are precipitation-based (Kim et al. 2019). However, the SPEI comprises an indicator of heat-waves, indicating the signs of heat wave-driven droughts (Hari et al. 2020).
Advantages: The SPEI links multi-timescale properties of the SPI with evapotranspiration data, which makes it more suitable for climate change analyses. The main advantage of the SPEI over the SPI is that it counts the role of temperature through its effect on PET.
Limitations: SPEI is sensitive to the method of computing PET and it requires more data compared with the precipitation SPI. As the SPEI is a monthly index, rapidly developing drought conditions may not be identified quickly.
Flash-drought-specific indices
Flash Drought Intensity Index (FDII): The FDII describes both the quick intensification rate and subsequent severity of FD events (Otkin et al. 2021). It is computed using two components: extreme rate of intensification (FD_INT) and average drought severity (DRO_SEV). Otkin et al. (2021) used the least intensification rate equal to a 15-percentile reduction in soil moisture through a four-pentad period to identify FD. A study of the 2012 FD in the United States found that the FDII description of extreme drought situations corresponded more directly to locations having bad crop situations and high production losses than the intensification rate component (FD_INT) individually.
Advantages: The FDII offers a more extensive measure of FD severity than other existing identification approaches that emphasize only on the rate of intensification. The FDII is also able to depict severe drought situations associated with crop conditions and yield damages based on the intensification rate alone.
Limitations: FDII is derived using the soil moisture output from the Noah land surface model, which is limited to the 0–40 cm layer. However, various challenges in using soil moisture from remotely sensed data limit their use for FD events. FDII is limited to only the contiguous United States.
Flash Drought Stress Index (FDSI): The FDSI is a composite index for monitoring FDs using thresholds of soil-hydrologic and land–atmospheric regimes (Sehgal et al. 2021). The Soil Moisture Stress (SMS; shows drought severity) and Relative Rate of Drydown (RRD; shows drought intensification rate) are combined to provide FDSI. The FDSI has been extensively proved globally across daily, weekly, and monthly timescales using a suitable vegetation and meteorology drought index. The assessment of FDs using FDSI is done at the global level (e.g., the regions of South Africa, Australia, and Northern Great Plains).
Advantages: The FDSI is a composite index used for global FD monitoring, and it can be extensively used across daily, weekly, and monthly timescales. The FDSI is generated on a daily basis and the data are available for the most recent years from 2015 to the present.
Limitations: FDSI is sensitive to changing land-surface heterogeneity, land–atmospheric interactions, and evolving meteorological anomalies. FDSI is generated using SMAP observations from 2015 to the present, which shows the period of record for data is short.
Rapid Change Index (RCI): The RCI has been developed to capture moisture stress changes over multiple weeks. It shows the areas experiencing rapid changes in moisture stress as being conditional on weekly fluctuations in the ESI. Then the onset of a drought event is likely when the RCI value is negative (Otkin et al. 2014, 2015). When variations in the ESI during a two-week interval are less than the 20th percentile, the RCI at each time step is used to determine the intensification phase of the FD event (Nguyen et al. 2021). The RCI provides significant drought early-detection abilities and real-time predictions (Chen et al. 2020), which might be applied to report an elevated risk of drought onset in sub-seasonal time frames (Otkin et al. 2018).
Advantages: The RCI provides an early warning capability to notify stakeholders about the increase in the risk for rapidly developing drought events. The potential advantage of the RCI is that it can be used for both rapidly increasing and rapidly decreasing moisture stress. The RCI may be used to predict future increases in drought severity as depicted by USDM.
Limitations: Drought is likely to develop when RCI is negative. Since RCI changes with time, it is difficult to capture FD development by monitoring RCI maps. RCI is limited to only the contiguous United States.
Quick Drought Response Index (QuickDRI): The QuickDRI was developed to identify quickly and for early detection of developing drought events. The multiple input parameters, namely, precipitation (as SPI), soil moisture, evapotranspiration (as ESI), and vegetation health (as Standardized Vegetation Index; SVI), are used to show anomalous drought conditions over the previous four weeks for a one-standard deviation threshold (Chen et al. 2019; Jiao et al. 2021; Osman et al. 2021).
Advantages: The purpose of QuickDRI is to provide an alert to identify the regions of developing drought and the regions of intensifying drought situations indicating a wet signal. The QuickDRI can employ SPEI to integrate evaporation calculation using both temperature and precipitation, instead of only precipitation.
Limitations: QuickDRI may not be meaningful for dormant vegetation, snow cover, and frozen soils, since it could lead to a false indication that is not related to a change in the drought condition. QuickDRI is limited to only the continental United States.
A STUDY ON FREQUENCY AND MEAN DURATION OF FDS OVER INDIA
Study area and data used
The daily precipitation and temperature datasets for 24 river basins were collected from the India Meteorological Department (IMD), and the soil moisture datasets were obtained from five soil moisture products (ESA-CCI, ERA-Interim, MERRA-2, GLDAS-2-Noah, and IMDAA). All the datasets were re-gridded to a resolution of 0.25° × 0.25° using the IDW (inverse distance weighing) method. Please refer to Table 1 for all datasets and their sources.
Data . | Period . | Source . | Original resolution . | Final resolution . |
---|---|---|---|---|
Precipitation | 1981–2014 | Indian Meteorological Department (IMD) | 0.25°×0.25° | 0.25°×0.25° |
Temperature | 1981–2014 | Indian Meteorological Department (IMD) | 1°×1° | 0.25°×0.25° |
Soil moisture | 1981–2014 | ESA-CCI SM COMBINED datasets (https://www.esa-soilmoisture-cci.org/node/145) | 0.25°×0.25° | 0.25°×0.25° |
Global Land Data Assimilation System version 2 Noah (https://ldas.gsfc.nasa.gov/gldas/) | 0.25°×0.25° | |||
ERA-Interim soil moisture reanalysis product (https://apps.ecmwf.int/datasets/data/interimfulldaily/levtype=sfc/) | 0.75°×0.75° | |||
Modern-Era Retrospective analysis for Research and Applications, Version 2 (https://gmao.gsfc.nasa.gov/reanalysis/MERRA-2/data_access/) | 0.5°×0.67° | |||
IMDAA soil moisture data (https://rds.ncmrwf.gov.in/) | 0.12°×0.12° |
Data . | Period . | Source . | Original resolution . | Final resolution . |
---|---|---|---|---|
Precipitation | 1981–2014 | Indian Meteorological Department (IMD) | 0.25°×0.25° | 0.25°×0.25° |
Temperature | 1981–2014 | Indian Meteorological Department (IMD) | 1°×1° | 0.25°×0.25° |
Soil moisture | 1981–2014 | ESA-CCI SM COMBINED datasets (https://www.esa-soilmoisture-cci.org/node/145) | 0.25°×0.25° | 0.25°×0.25° |
Global Land Data Assimilation System version 2 Noah (https://ldas.gsfc.nasa.gov/gldas/) | 0.25°×0.25° | |||
ERA-Interim soil moisture reanalysis product (https://apps.ecmwf.int/datasets/data/interimfulldaily/levtype=sfc/) | 0.75°×0.75° | |||
Modern-Era Retrospective analysis for Research and Applications, Version 2 (https://gmao.gsfc.nasa.gov/reanalysis/MERRA-2/data_access/) | 0.5°×0.67° | |||
IMDAA soil moisture data (https://rds.ncmrwf.gov.in/) | 0.12°×0.12° |
Methodology
In this study, soil moisture (SM) is used to detect FD events in India between 1981 and 2014. The daily gridded data of SM were converted to an octad (SM averaged for eight days). As a result, each year got a total of 46 octads, with 16 octads for the rainy season (i.e., the 19th–34th octads) and the residual octads for the non-rainy (non-monsoon) season (i.e., from first to 18th and 35th to 46th octads). The definition used for the FD event is as ‘when the mean SM percentile for eight decreases to 20th percentile or less from above 40th percentile and an average rate of drop not <5%iles for every eight days’, the drop from the 40th to the 20th percentile indicating the beginning of FD (Zhang & Yuan 2020). The retrieval time of the FD event was recognized after the SM began to either fall slowly or rise, and when the SM percentile increases to 20 (or higher), the drought is over. Furthermore, for FDs, a minimum period of 24 days (three time-steps) was measured to eliminate short dry spells that significantly less affect the environment. The FD threshold was set at the 20th percentile (Yuan et al. 2019) to eliminate the conventional droughts from the study.
Results and discussion
THE FACTORS THAT NEED TO BE INCLUDED IN IDENTIFYING AND MONITORING FDS
Role of humidity
The relationship between humidity and FD is such that increased humidity can mitigate drought conditions by reducing transpiration, the process by which plants lose water through their leaves (Georgii et al. 2017). However, high humidity can also make the outdoor environment feel more unpleasant, even if the relative humidity is lower than that of a cooler day with higher relative humidity. In general, while humidity can have a certain effect on drought conditions, it is only one of several factors that contribute to FD and its level of severity (Means 2021). The measure of the amount of moisture present in the atmosphere is referred to as humidity. The onset of FDs occurs suddenly when drought conditions arise during high-temperature periods accompanied by low humidity levels (Gamelin et al. 2022; Gong et al. 2022). Humidity plays a critical role in determining whether a region is at risk from FDs. When there is low atmospheric moisture available, it becomes increasingly hard for vegetation and soil surfaces to retain enough water, resulting in aridness which leads to a quick depletion of soil moisture content – elevating susceptibility to wildfires especially within habitats vulnerable to frequent drought incidents. Understanding how varying meteorological factors such as climate change exacerbate existing risks needs to be better researched by experts, policymakers, and drought management agencies, thus to come up with more effective strategies in preventing the consequences of these extreme events.
Role of wind and radiation
A study indicated that sudden rises in evapotranspiration rates, induced by variations in heat, wind, and radiation, preceded all the FDs analyzed (Sreeparvathy & Srinivas 2022). FD occurs when there is a decline in the amount of precipitation, which is further intensified by unusual high temperatures, winds, and radiation. These joint climate conditions can cause swift and substantial changes to the local climate (NIDIS 2022). The soil moisture was abundant before the FD, but it declined as the evapotranspiration rates increased. Therefore, the researchers inferred that variations in evapotranspiration rates are a reliable predictor of FDs (Chen et al. 2020). The strong winds can hasten the process of evapotranspiration and cause soil moisture to deplete rapidly, which is a key factor contributing to the occurrence of FDs. FDs are characterized by a sudden onset of drought conditions caused by a combination of low soil moisture, high temperatures, and dry soil, and they are frequently associated with atmospheric anomalies such as low humidity, strong winds, extreme temperatures, and scant precipitation. The reason for this is that winds carry thermal energy to a region and take away the evaporated moisture, which can accelerate the drying of the soil.
Role of soil characteristics
Soil characteristics, including type, depth, and water-holding capacity factors, can affect the onset and severity of FDs and play an important role in soil moisture and the rate at which water is absorbed and released (Chen et al. 2019; Christian et al. 2019b; Qing et al. 2022). Understanding these factors and their interaction is crucial in forecasting and mitigating the consequences of FDs. Efficient handling of water resources, such as soil preservation, land utilization practices, and irrigation, can help in reducing the effects of FDs (Subramoniam et al. 2022). The existence of different vegetation and soil types in various regions can also play a role in the emergence of FDs (Chen et al. 2019). Crops can experience moisture stress at a faster rate by extracting water from deeper layers of soil, which can contribute to the rapid onset of FDs (Qing et al. 2022). Soils that are shallow, which have limited water retention capacity and are vulnerable to rapid evaporation, can increase the possibility and intensity of FDs. This comprises soil types like sandy soils, loamy sands, and those with a high proportion of gravel or rock fragments. Additionally, low organic matter content in soils can also play a role in triggering FDs. This is because soils with less organic matter are less capable of holding moisture and supporting healthy plant growth, leading to quicker soil moisture reduction during times of high temperature or low precipitation (Qing et al. 2022).
Influences of climate oscillations
Climate oscillations are variations in oceanic and atmospheric conditions that happen over a span of several years to decades. These oscillations can have an influence on weather patterns and moisture availability, which can impact the occurrence and severity of FDs. The intensity and severity of FDs can be strongly influenced by climate oscillations, which can occur over large distances through teleconnections (Lesinger & Tian 2022). An intensification of climate warming and an increase in the frequency of El Niño events can intensify the occurrence of FDs in multiple major croplands at the same time, leading to significant consequences for food production (Mahto & Mishra 2023). Climate oscillations are large-scale variations in oceanic and atmospheric conditions that can affect weather patterns and moisture availability, thus influencing the development of FDs (Yuan et al. 2023). El Niño–Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), and Indian Ocean Dipole (IOD) are examples of climate oscillations that can impact FDs. ENSO is a pattern of cyclical warming and cooling in the Pacific Ocean that can disturb global weather patterns. NAO refers to the variation of atmospheric pressure patterns over the North Atlantic Ocean, and IOD refers to the variability of sea surface temperatures in the Indian Ocean. During positive IOD events, there tends to be a decrease in rainfall over parts of Southeast Asia and Australia, increasing the risk of FDs. These climate oscillations can have a significant impact on the occurrence of FDs in different regions globally. Thus, understanding the mechanisms by which climate oscillations affect FDs is vital in addressing the problems related to the rapid intensification of drought and its consequences for agriculture, water resources, ecosystems, and the human environment (Christian et al. 2021).
Connection between FDs and heavy-rainfall-induced flash floods
Flash droughts and flash floods induced by heavy rainfall are connected through their impact on the ground. Flash floods can be more destructive in arid regions due to the limited ability of dry soils to absorb water, leading to a higher likelihood of surface runoff and flooding, whereas in wetter regions, the soil can more readily absorb water, resulting in less severe flooding (Gupta et al. 2023; Yin et al. 2023). FDs are more common in humid regions with lower aridity compared with slow droughts, with a frequency that is two to three times higher (Yuan et al. 2023). When an FD occurs, the soil can become dry and compacted, which can decrease its permeability and ability to absorb rainwater. This can lead to rapid runoff of water from the surface of the ground and dead vegetation rather than water being absorbed into the soil during heavy rainfall, resulting in flash flooding. The IPCC Report emphasized with high confidence that the process of aridification and the occurrence of heavy rainfall will lead to severe flooding (IPCC 2022). Consequently, flash floods can become more likely in certain situations due to the occurrence of FDs. In addition, the safety of dams becomes a serious concern when there is a possibility of larger floods. This problem is compounded by the occurrence of flash floods during drought intervals, which creates a misleading sense of safety (Mediero et al. 2007).
CONCLUDING REMARKS
The study describes an understanding of the level to which FD characteristics and indicators are susceptible to the use of drought indices, as these are the basis for evaluating the areas most vulnerable to FDs. The study in India reveals that the Indus basin area faced around 82% of severe FD events during the rainy season and the Cauvery basin area faced about 88% of severe FD events during the non-rainy season. In this study, FD is identified using a rapid intensification approach and soil moisture is used as an input variable/driver to define the octad-scale (eight-day) temporal scale using soil moisture percentiles. Reanalysis data (GLDAS) used in the study provide better soil moisture datasets despite satellite-based soil moisture datasets (Chen et al. 2013). Also, adopting a consistent definition and index for what characterizes an FD provides helpful information to administrators, researchers, scientists, and farmers to understand these extreme events of the climate system. However, the standard guidelines should be decided to differentiate FD episodes from conventional droughts. Given the worldwide climatology of the occurrence of FD events and the quick land-surface dryness associated with fast onset, drought occurrences have the potential to have major consequences beyond agricultural production loss and economic loss. Furthermore, it is well acknowledged that current monitoring and forecasting tools do not give enough early warning for FDs (NIDIS 2021; Mukherjee & Mishra 2022a); therefore, certain research needs mentioned below are required in future work.
The monitoring should consist of the suitability of indicators like soil moisture, evapotranspiration, and climatic variables. Furthermore, systematic approaches are required to monitor the development of FD accurately beyond the rapid onset and intensification period. Future research is needed to advance the data sources from in situ observations (soil moisture, vegetation, crop health, etc.), satellite data (remote sensing), and climate models' output data along with meteorological and drought impact datasets, which have a significant role in monitoring FDs accurately.
Forecasting methods are needed for identifying novel machine-learning algorithms and data sources. Also, researchers should have a better understanding of the land–atmosphere connections and physical systems triggering FD. Also, their validation is required to know their functioning and suitability to forecast FDs using existing and upcoming models. Likewise, to develop a forecast method or model, researchers should have weekly predictions at a higher resolution and use better data integration.
For planning, the research needs to be carried out on FDs during every season of the year. Planning should be done of given impacts on water resources, ecosystem, well-being of agriculture and economy for decision-making, and impact evaluation. The research plan is essential to study how the event impacts are effectively reported by the organizational programs, decision-making systems, and policies to expand relief for FDs. Furthermore, a detailed social science study is also necessary for public awareness and participation in effective interaction.
During the response, government bodies at different levels (regional, state, local, specific groups, etc.) must evaluate the technical drought data and share them with scientific and non-scientific persons for monitoring and early warning of FDs. The response approaches must be efficient to communicate with the end-users of FD impacts by developing dashboards, interactive maps, apps, etc. Communication supports understanding of practical monitoring and forecast information. The research should also develop appropriate indicators and factors for drought response actions and decision-making.
ACKNOWLEDGEMENTS
S.R. would like to acknowledge the support of the Department of Science and Technology (DST), Government of India, New Delhi, for his INSPIRE Fellowship. We present this work as a tribute to Prof. Manish Kumar Goyal on their birthday, acknowledging their steadfast support, invaluable guidance, and constant encouragement.
CONSENT TO PARTICIPATE
The authors declare that they are aware of and consent to their participation on this paper.
CONSENT TO PUBLISH
The authors declare that they consent to the publication of this paper.
AUTHOR CONTRIBUTIONS
S.R.: Conceptualization, Software, Visualization, Writing – original draft, and Writing – review and editing; V.P.: Methodology, Formal analysis, Visualization; M.K.G.: Supervision, Validation.
FUNDING
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.