Drought Assessment Research Articles

A Bio-inspired and Deep Learning Based Hybrid Model for Agricultural Drought Assessment

Agricultural droughts can cause many serious hazards. Drought monitoring indices, namely Normalized Difference Vegetation Index (NDVI), Atmospherically Resistant Vegetation Index (ARVI), Soil Adjusted Vegetation Index (SAVI), and Enhanced Vegetation Index (EVI) have been used for an agricultural drought assessment. Satellite images from the Kolar region of Karnataka are used to calculate these indices. This paper proposes an integration model based on Convolutional Neural Networks (CNN) and a bio-inspired algorithm (Sparrow Search Algorithm (SSA) and Barnacles Mating Optimizer (BMO)) considering the indices as population. Performance is compared with the standalone CNN model in terms of efficiency. For the CNN, the accuracy, time taken for Epoch1, and time taken for Epoch2 is 91%, 16s (3s/step), and 2s (2s/step), respectively. For the CNN integrated with SSA, it is 94%, 3s (3s/step) and 0s (43ms/step), respectively. For the CNN integrated with BMO, it is 94%, 3s (2s/step) and 0s (46ms/step) respectively.

Flash Drought Typologies and Societal Impacts: A Worldwide Review of Occurrence, Nomenclature, and Experiences of Local Populations

Abstract Flash droughts, characterized by rapid onset and intensification, are increasingly occurring as a consequence of climate change and rising temperatures. However, existing hydrometeorological definitions fail to encompass the full range of flash droughts, many of which have distinct local physical attributes. Consequently, these events often go undetected or unforecast in generic global flash drought assessments and are underrepresented in research. To address this gap, we conducted a comprehensive survey to gather information on local nomenclature, characteristics, and impacts of flash droughts worldwide. The survey revealed the widespread occurrence of these phenomena, highlighting their underresearched nature. By analyzing case studies, through literature review often in local languages to unearth elusive studies, we identified five different types of flash droughts based on their specific characteristics. Our study aims to increase awareness about the complexity and diverse impacts of flash droughts, emphasizing the importance of considering regional contexts and the vulnerability of affected populations. The reported impacts underscore the need for better integration of all flash drought types in drought research, monitoring, and management. Monitoring a combination of indicators is crucial for timely detection and response to this emerging and escalating threat. Significance Statement This study aims to better understand flash droughts worldwide and their varying characteristics and impacts. We surveyed the experiences of people affected by flash droughts and then examined a wide range of literature, including non-English and nonacademic sources. This helped us understand how flash droughts can differ from those commonly studied in the United States and China. We identified and described five types of flash droughts, some of which may not be detected by current drought measurement methods. It is crucial to include all types of flash droughts in drought monitoring systems and management plans, as they are expected to become more common due to global warming. We can then better prepare for and reduce the impacts of this growing threat.

Assessment of Drought and Desertification Outlook in China Using Meteorological Drought Indices

Changing land use patterns in China may result in adverse impacts on the climate in Northeast Asia. In particular, desertification of the Chinese mainland can also have a significant influence on the climate in South Korea, such as the occurrence of yellow dust storms. This study aimed to forecast the desertification in China through a comprehensive drought analysis. Three drought indices namely, SPI, GEVI, and SPEI, were determined for eight cities in China namely, Beijing, Guangzhou, Shanghai, Xianyang, Urumqi, Chongqing, Wuhan, and Weihai. There was a strong correlation between SPI and SPEI, while GEVI showed a slightly lower correlation. The long-term droughts that can trigger desertification exhibited high correlations among the considered indices. To assess the impact of climate change, periods of severe drought occurrence were compared in ten-year intervals since 1970. It was observed that severe droughts occurred throughout China in the 1970s and 1990s. However, there was no apparent trend in increasing drought days or worsening drought conditions. Desertification in China can be influenced not only by the rainfall but also by the changes in land use pattern and exploitation of water resources. Therefore, additional research is deemed necessary to evaluate desertification in China.

Meteorological Drought Assessment for Jhelum River Basin, Jammu and Kashmir\'s Sangam and Ram Munshi Bagh Stations

Abstract: Drought is a severe natural calamity that harms the global ecology. To examine drought conditions including intensity and duration, a variety of drought indicators (DIs) are commonly employed. This study assesses the occurrences of drought in Kashmir's Jhelum River Basin (JRB) at the Sangam and Ram Munshi Bagh stations. Several DI approaches, such as Deciles, Percent of Normal (PN), and Standardized Precipitation Index (Normal-, Log-, and Gamma-SPI), are used to analyse 20 years of monthly recorded precipitation data from 2002 to 2023. Every technique is used on the annual long-term precipitation data set. The findings indicate that nearly identical outcomes are obtained for the stations by the DI approaches. While the normalSPI indicates rainy and less droughty conditions, the log-SPI and gamma-SPI forecast moderate dry conditions. The results emphasize that the PN method predicts more moderate drought years in comparison with SPI method, however, Deciles method shows longer period of extreme and severe drought than other methods. As a result, the five methods indicate various drought intensities in 2017, 2018,2001, and 2002 with a peak moderate drought in 2017 in the Sangam station. The Ram Munshi Bagh station area experienced the less drought conditions from 2002 to 2023 with a moderate drought in 2017 and 2018 as well. Therefore, moderate drought conditions happened in 2017 -2018 in both stations confirm to the recorded drought reports for the same region.

Spatiotemporal consistency and inconsistency of meteorological and agricultural drought identification: A case study of India

Drought identification is challenging because the severity, intensity and frequency of extreme drought events are heterogenous. Incorporating multiple datasets is to generate standardized drought indices for effective drought assessment, monitoring, and develop early warning systems. This study aims to explore the spatio–temporal relationship among drought indices of precipitation, soil moisture, vegetation condition and health based on standardizing multiple drought indices with trend detection. The indices were derived through standardized drought analysis from 2001 to 2020 over India and regional level. Moreover, time series decomposition was utilized to remove seasonal and random components for trend detection. This study enhanced the comprehension of meteorological and agricultural drought conditions in India by standardizing and analyzing trends in multiple drought indices. The results revealed that the most severe drought periods for subregions were consistent for all indices with trend analysis. All indices also reported a significant decline in trend for the 2009 drought event across the subregions in India. These findings contributed to a better understanding of drought dynamics in the subregion. However, vegetation-based indices were not always consistent to the indices of precipitation and soil moisture due to ecosystem resilience or heat stress. This study offered policymakers diverse information, including correlations and variations of multiple drought indices to assist in detecting the most severe periods of drought at a regional scale.

A new non-stationary standardised streamflow index using the climate indices and the optimal anthropogenic indices as covariates in the Wei River Basin, China

Study regionCatchment area above the Huaxian station along the Wei River Basin, China. Study focusThis study attempts to construct a new Non-stationary Standardized Streamflow Index (NSSI) applicable to the variable streamflow sequence of the Wei River Basin based on the climate index and the optimal anthropogenic index, and analyse the drought characteristics of the basin. The climate index is used to quantify climate change factors and three anthropogenic indices are used to quantify the factor of human activities, including the reservoir index, the human-induced index calculated based on the Variable Infiltration Capacity (VIC) hydrological model and the Long Short-Term Memory (LSTM) model machine learning approach, respectively. New hydrological insights for the regionThe human-induced index based on the LSTM model is more suitable for quantifying anthropogenic factors in the Wei River Basin. The NSSI performs better than the SSI in drought identification. The NSSI based on the LSTM model can capture more frequent severe drought and extreme drought events. The frequency of severe drought and extreme drought is higher in summer and autumn than in the others. The NSSI can better characterize the hydrological drought processes under a non-stationary condition, thus it can provide a more effective reference for regional drought assessment and related policy-making from the perspective of a changing environment.

Drought characterization: A systematic literature review

<p>This study examined the worsening severity of global droughts caused by climate change. However, the multiple definitions and varied range of drought indices pose challenges in effectively monitoring and assessing the prevalence and severity of droughts. This study aims to give a comprehensive overview of the various drought definitions found in the literature and how they have evolved based on their applications. Specifically, the focus was to shed light on the dynamic nature of drought characterization and offer insights into the factors that shaped its conceptualization over time. Within this context, this study explored three primary categories of drought indices: climatic, remote sensing, and composite. Each category was discussed in relation to its utility in specific fields, such as meteorological, agricultural, and hydrological drought assessments, along with an analysis of their strengths and limitations. Furthermore, this study presents modified meteorological drought indices that have been adapted to better monitor agricultural droughts. Additionally, the authors used geographic information systems to create a map showing the distribution of drought-related publications globally over the past decade. The findings showed that countries with arid and semi-arid climates are more actively involved in drought research, highlighting their particular interest and concern regarding the subject matter. The implications of this study emphasize the urgent need for immediate and coordinated efforts to address the escalating issue of droughts caused by climate change. By improving monitoring and assessment methods and focusing on tailored strategies in vulnerable regions, it is possible to mitigate the far-reaching consequences of drought and to build more resilient communities and ecosystems.</p>

Spatiotemporal Changes and Hazard Assessment of Hydrological Drought in China Using Big Data

The intensification of the regional water cycle resulting from climate change, coupled with the influence of human activities, has brought about alterations in the frequency, scale, and intensity of droughts. In this study, based on hydrological big data and the standardized runoff drought index (SRI), the multi-scale spatiotemporal evolution of hydrological drought in China from 1948 to 2014 was analyzed using the run-length theory and gravity center model. Meanwhile, the hydrological drought hazard index was constructed to analyze the distribution of the hazard levels of drought in China. The results showed that, during 1948~2014, there was an opposite spatial distribution between the average intensity and the average coverage–duration–frequency of drought in the Yellow River Basin, Haihe River Basin and southeastern river basins. The drought situation in most river basins in China has shown an aggravating trend, among which the southeastern river basins, Haihe River Basin, Songliao River Basin and Pearl River Basin have generally shown an aggravating trend. The drought situation in China was severe in the 1950s and 1960s, gradually reduced in the 1970s, 1980s, and 1990s, and the drought situation was the mildest in the 1990s. After entering the 21st century, the drought situation began to worsen sharply. Meanwhile, from 1948 to 2014, the hazard level of drought generally presented a pattern of high in the west and north, and low in the east and south. The hazard levels of drought in the northwest and northeast were generally higher, and those in the southwest and southeast regions were generally lower. In general, the hazard levels of drought were relatively high in most areas of China.

Detecting drought-prone regions through drought indices

Abstract Climate change has led to heightened variability in global rainfall patterns, resulting in greater unpredictability and inconsistency, and it has led to the origin of meteorological drought situation. This has amplified the frequency of droughts or drought-like conditions worldwide. India, being primarily agrarian, faces significant challenges due to drought, affecting various regions intermittently. Given the urgency of addressing recurring drought issues, it is crucial to determine specific ‘drought-prone’ areas through the analysis of historical and current meteorological data. It is still a challenge to quantitatively understand where and to what extent the impact of rainfall patterns could lead the drought. Whether any region likely comes under drought-prone area or not? Can we help policy makers to apply their knowledge effectively? It will help to undertake the long-term mitigation measures for drought assessment and management which encompasses early warning, monitoring, and relief toward the good health of the society. The present study is a further step in the same direction in which Akola district in Maharashtra, India has been assessed for drought-prone declaration using two drought measuring indices; seasonality index (SI) and aridity index (AI). For this, the measured meteorological data precipitation, and potential evapotranspiration, from 1952 to 2002 is made available from India Water Portal.

Assessment of Drought and Zinc Stress Tolerance of Novel Miscanthus Hybrids and Arundo donax Clones Using Physiological, Biochemical, and Morphological Traits.

High-yield potential perennial crops, such as Miscanthus spp. and Arundo donax are amongst the most promising sources of sustainable biomass for bioproducts and bioenergy. Although several studies assessed the agronomic performance of these species on diverse marginal lands, research to date on drought and zinc (Zn) resistance is scarce. Thus, the objective of this study was to investigate the drought and Zn stress tolerance of seven novel Miscanthus hybrids and seven Arundo clones originating from different parts of Italy. We subjected both species to severe drought (less than 30%), and Zn stress (400 mg/kg-1 of ZnSO4) separately, after one month of growth. All plants were harvested after 28 days of stress, and the relative drought and Zn stress tolerance were determined by using a set of morpho-physio-biochemical and biomass attributes in relation to stress tolerance indices (STI). Principal component analysis (PCA), hierarchical clustering analysis (HCA) and stress tolerance indices (STI) were performed for each morpho-physio-biochemical and biomass parameters and showed significant relative differences among the seven genotypes of both crops. Heatmaps of these indices showed how the different genotypes clustered into four groups. Considering PCA ranking value, Miscanthus hybrid GRC10 (8.11) and Arundo clone PC1 (11.34) had the highest-ranking value under both stresses indicating these hybrids and clones are the most tolerant to drought and Zn stress. In contrast, hybrid GRC3 (-3.33 lowest ranking value) and clone CT2 (-5.84) were found to be the most sensitive to both drought and Zn stress.

Assessing meteorological and agricultural drought characteristics and drought propagation in Guangdong, China

Study regionGuangdong, China Study focusDroughts have resulted in significant crop failure in Guangdong province, where annual crop yield exceed 12 million tons. However, there is a lack of a comprehensive assessment of both meteorological drought and agricultural drought in Guangdong, particularly regarding the influencing factors and drought propagation. This study aims to address this gap by utilizing the standardized precipitation evapotranspiration index (SPEI) and soil water deficit index (SWDI) to represent meteorological drought and agricultural drought, respectively. To examine the characteristics of drought in the past 50 years (1973–2022), the year is divided into flood and non-flood season based on the temporal distribution of precipitation. New hydrological insights for the regionOver the last five decades, the number of meteorological and agricultural drought days decreases from coastal to inland areas. Notably, agricultural droughts have become more pronounced in the last two decades. More agricultural drought days appears during the non-flood season than the flood season. Precipitation emerges as the most influential factor for meteorological drought, while potential evapotranspiration, actual evapotranspiration and temperature are identified as the primary factors for agricultural drought. The drought propagation time ranges from 10 to 45 days which is shorter in central Guangdong than other regions, and it is longer in non-flood season than in flood season.

Remote sensing data for drought stress and croplands productivity assessment in Kherson region

Formulation of the problem. Remote sensing data might be used for indirect assessment of croplands conditions and drought stress through the calculation of specific vegetation indices, such as vegetation health index (VHI), agriculture stress index (ASI), and drought intensity or weighted mean vegetation health index (WMVHI). However, the accuracy of these indices is not clear for some territories. For example, the South of Ukraine is a zone of risky agriculture, because of low natural moisture supply and high evapotranspiration. Moisture supply is the main limiting factor for sustainable crop production in this region. The goals of this study were: 1) to assess the reliability of the mentioned vegetation indices in drought assessment through the direct comparison with the UNEP aridity index; 2) to find out whether remote sensing drought indicators could be used for the yield prediction of major crops on the regional scale. Methods. The study was conducted for Kherson region of Ukraine, as it is one of the most arid regions of the country with very high drought risks. The data on average weighted annual VHI, ASI, and WMVHI for the period 1984-2022 (Season 1) were collected and generalized from the FAO Earth Observation services. UNEP aridity index was calculated using the data from Kherson regional hydrometeorological center. Correlation and linear regression analysis were performed using common statistical methodology. Results. As a result, it was found that 1) all the studied remote sensing drought indicators demonstrate poor correlation with the aridity index, therefore, they should not be used to determine meteorological drought in the region; 2) all the studied remote sensing indices, especially VHI, demonstrate moderate-to-strong correlation with the yields of certain crops, cultivated in Kherson region (R=0.54-0.86), and could be used for the yield prediction; 3) the aridity index have poor relation to the yields of major crops, cultivated in the studied area; 4) VHI-based linear regression models for the crops’ yields prediction are reliable and reasonable for scientific and practical use just for cereal crops, and are much less accurate for grain corn and sunflower; 5) based on the study findings, it could be concluded that aridity index provides pure climatological characteristics of the region, while the studied vegetation indices are mainly focused on the level of drought stress that impacts crops during the growing season. Scientific novelty and practical significance. The article provides novel insights on the implementation of remote sensing data in drought risks assessment in crop production, and their utilization for the purpose of croplands productivity prediction. The study has theoretical and practical importance for current agriculture, and the findings could be used both in scientific, educational, and practical purposes.

Revolutionizing Agriculture with Satellite Technology for Farmers: A Review

Farmers in India have benefited considerably from the combination of satellite-enabled services and data obtained on the ground. The India Meteorological Department, Ministry of Earth Sciences, provides weather forecasting, agro-advisory services, agromet services, soil moisture monitoring, and agricultural extension initiatives to encourage agricultural operations in India. Indian Space Research Organization?s (ISRO) also partners with the Ministry of Agriculture and Farmers Welfare on several satellite data and geographic information systems-based agricultural applications. ISRO, in collaboration with the Ministry of Agriculture and Farmers Welfare, has developed applications including horticultural crop inventory and site suitability for expansion in unutilized places; crop assessment using medium- and high-resolution satellite data; field information gathering with field photos using a smartphone application; and crop cutting experiments based on satellite-derived crop vigor information. ISRO has provided technology for FASAL and the National Agricultural Drought Assessment and Surveillance System to the Department of Agriculture Cooperation and Farmer Welfare. ISRO has also incorporated the Central Water Commission?s monitoring of irrigation systems. Overall, satellite-enabled services have transformed agricultural operations in India by providing farmers with precise and timely data that enable them to make educated decisions about their crops, resulting in increased crop yields and financial returns.. KEYWORDS :Agriculture, Farming, Satellite technology, Meteorology, Space organization

Assessment of meteorological drought using Standard Precipitation Index model (SPI) in Herat, Afghanistan

Assessment of meteorological drought using Standard Precipitation Index model (SPI) in Herat, Afghanistan

Vegetation Stress Monitor—Assessment of Drought and Temperature-Related Effects on Vegetation in Germany Analyzing MODIS Time Series over 23 Years

Over the past two decades, and particularly since 2018, Central Europe has experienced several droughts with strong impacts on ecosystems and food production. It is expected that under accelerating climate change, droughts and resulting vegetation and ecosystem stress will further increase. Against this background, there is a need for techniques and datasets that allow for monitoring of the timing, extent and effects of droughts. Vegetation indices (VIs) based on satellite Earth observation (EO) can be used to directly assess vegetation stress over large areas. Here, we use a MODIS Enhanced Vegetation Index (EVI) time series to analyze and characterize the vegetation stress on Germany’s croplands and grasslands that has occurred since 2000. A special focus is put on the years from 2018 to 2022, an extraordinary 5-year period characterized by a high frequency of droughts and heat waves. The study reveals strong variations in agricultural drought patterns during the past major drought years in Germany (such as 2003 or 2018), as well as large regional differences in climate-related vegetation stress. The northern parts of Germany showed a higher tendency to be affected by drought effects, particularly after 2018. Further, correlation analyses showed a strong relationship between annual yields of maize, potatoes and winter wheat and previous vegetation stress, where the timing of strongest relationships could be related to crop-specific development stages. Our results support the potential of VI time series for robustly monitoring and predicting effects of climate-related vegetation development and agricultural yields.

Spatio-temporal variation and dynamic risk assessment of drought and flood disaster (DFD) in China

It is important to understand the variations of Drought-Flood-Disaster (DFD) from the past to future for disaster management decision-making under the backdrop of global warming. This study explored the spatio-temporal variations of DFD risks in China from the past (2000–2020) to future (2020–2100) based on the natural and socio-economic datasets. The main findings include: (1) The spatial pattern of disaster vulnerability decreased from 2000 to 2100. The overall vulnerability distribution showed a spatial pattern of high east and low west, and its eastern distribution presented high north and low south. (2) The drought hazard danger degree showed a trend of decreasing first and then increasing with obvious spatial heterogeneity. The flood hazard danger degree had a decreasing spatial distribution from southeast to northwest, with an overall gradually increasing trend. (3) The DFD risks showed a spatial pattern of high east and low west, and were more closely related to disaster vulnerability. Overall, the flood and drought risks had good consistency at the highest level (VI), mainly distributed in the North China Plain, Northeast Plain, eastern Sichuan, and southern Guangdong. This level means that the most severe damage to life and property will occur in these regions. (4) The DFD showed the direction of “north(east)-south(west)”, and the main axis of the ellipse moved north eventually. The DFD hot spots were mainly distributed in North China and the middle and lower reaches of the Yangtze River. The DFD cold spots were mainly distributed in the southwest, Qinghai-Tibet region, and Gansu province.

Performance of potential evapotranspiration models across different climatic stations in New South Wales, Australia

Study regionNew South Wales, southeast Australia Study focusEstimating potential evapotranspiration (ETp) rates, detecting its temporal trends and analysing its interannual oscillation are critical for long-term assessment of water availability and regional drought. This study aimed to evaluate the comprehensive performance of 12 simplified models in characterising ETp against the benchmark Penman model across different climate sites in southeast Australia. This study used Taylor skill score (S), normalised root mean square error (nRMSE) and relative mean bias error (rMBE) to estimate models’ capability in estimating ETp rates. Then, this study adopted Mann-Kendall test and continuous wavelet transform (CWT) to test temporal trends and periodicity of ETp estimated by all models. New hydrology insights for the regionJensen-Haise model was capable to produce fair (nRMSE ≤ 30%) estimates of daily ETp across all stations. Models except Mak1 were generally able to produce reasonable estimates of ETp at larger time scale. In addition, we found that the 12 alternative ETp models generally agreed with the Penman model on the primary (9.6–12.4-year) and quasi (2.6–3.9-year) periods of ETp, but they did not show matchable ability in detecting ETp temporal trends. The comprehensive investigation on models’ performance will shed light on models’ selection in estimation of drought and hydrological cycle.

Drought Assessment in Greece Using SPI and ERA5 Climate Reanalysis Data

The present work aims to assess the spatial variability and the trends of the annual rainfall and meteorological drought in the entire territory of Greece utilising the ERA5 reanalysis precipitation dataset of the European Centre for Medium-Range Weather Forecasts (ECMWF), which spans from January 1940 to December 2022 (an 83-year period). Drought assessment took place based on the Standardized Precipitation Index (SPI) for timescales ranging from 1 month to 12 months. Evaluation was carried out by calculating SPI using observed rainfall data from five meteorological stations. The annual rainfall and drought severity trends for timescales of 1 (SPI-1), 3 (SPI-3), 6 (SPI-6) and 12 (SPI-12) months were analysed using the Theil–Sen slope method and the Mann–Kendall trend test. The results indicate significant, both increasing and decreasing, annual precipitation trends at the 95% significance level for the Aegean Islands, western Crete and western mainland of Greece. The results also indicate significant drought trends for SPI-12 for the Aegean Islands and western Peloponnese. Trend analysis for SPI-1, SPI-3 and SPI-6 indicate a mixture of non-significantly increasing wetting trends and increasing drought trends at the national scale. In conclusion, the ERA5 dataset seems to be a valuable tool for drought monitoring at the spatial scale.

Meteorological drought assessment in northern Bangladesh: A machine learning-based approach considering remote sensing indices

Meteorological drought, driven by inadequate precipitation, has significant repercussions for water resources, agriculture, and human well-being. This study conducted an extensive assessment of meteorological drought in northern Bangladesh, employing remote sensing indices and machine learning techniques. The main aim was to evaluate meteorological drought occurrences in northern Bangladesh from 2010 to 2019, utilizing seven drought parameters and a machine learning model. Utilizing a Random Forest (RF) model, this study employed the Standardized Precipitation Index (SPI) as the dependent variable and seven remote sensing indices as independent variables. Through this methodology, the study assessed the significance of these indices generated by the model and integrated them, culminating in the creation of a meteorological drought distribution map spanning 2010 to 2019. This approach offers novel insights by probing the interplay and collective impacts of these indices, shedding light on previously unexplored aspects of regional drought patterns of northern Bangladesh. The major findings showed that precipitation strongly influenced both short-term and long-term meteorological drought episodes. Moreover, land surface-related indices, such as Evapotranspiration (ET) and Normalized Difference Water Index (NDWI), exhibited a more pronounced impact on short-term drought occurrences, while vegetation-related indices like Normalized Multi-band Drought Index (NMDI) and Normalized Difference Vegetation Index (NDVI) demonstrated greater influence over long-term drought events. During this timeframe, the Rajshahi division experienced frequent extreme and severe drought events. Moderate droughts and abnormally dry conditions were widespread. The Barind tract area consistently faced moderate to extreme droughts, with exceptions in 2011, 2014, and 2019. On average, over 5% of the region had extreme droughts, while more than 12% experienced severe droughts during this decade. Long-term drought indicators (SPI 6 and SPI 9) consistently showed higher frequencies of extreme and severe droughts compared to short-term indicators (SPI 1 and SPI 3), emphasizing the influence of prolonged rainfall deficits on extreme droughts and the relevance of longer time frames for severe drought dynamics. The RF model demonstrated strong performance with accuracy ranging from 81% to 95%. Low prediction errors (RMSE 6% to 31%) and high out-of-bag (OOB) accuracy ranging from 76% to 98% highlighted its accuracy. The F1 score consistently exceeded 76%, indicating high precision and recall. Cross-validation values ranged from 78% to 94%, affirming reliable generalization to new data. Incorporating the main findings, this study contributes valuable insights for the formulation of targeted drought mitigation strategies in northern Bangladesh. It is imperative to note that the scope of this study is confined to the northern region of Bangladesh, and generalizing these findings to other regions should be exercised with caution. Nevertheless, the research methodology and approach can serve as a model for future studies in related fields, advancing knowledge of how to assess droughts using remote sensing and machine learning methods.

A review of widely used drought indices and the challenges of drought assessment under climate change.

Under climate change, drought assessment, which can address nonstationarity in drought indicators and anthropogenic implications, is required to mitigate drought impacts. However, the development of drought indices for a reliable drought assessment is a challenging task in the warming climate. Thus, this study discusses factors that should be considered in developing drought indices in changing climate. Inconsistent drought assessment can be obtained, depending on the baseline period defined in developing drought indices. Therefore, the baseline period should represent the contemporary climate but should also correspond to long enough observations for stable parameter estimation. The importance of accurate potential evapotranspiration (PET) for drought indices becomes higher under a warming climate. Although the Penman-Monteith method yields accurate PET values, depending on the climate and vegetation cover, other suitable PET formulas, such as the Hargreaves method, with fewer hydrometeorological data can be used. Since a single drought index is not enough to properly monitor drought evolution, a method that can objectively combine multiple drought indices is required. Besides, quantifying anthropogenic impacts, which can add more uncertainty, on drought assessment is also important to adapt to the changing drought conditions and minimize human-induced drought. Drought is expected to occur more frequently with more severe, longer, and larger areal extent under global warming, since a more arid background, which climate change will provide, intensifies land-atmosphere feedback, leading to the desiccation of land and drying atmosphere. Thus, an accurate drought assessment, based on robust drought indices, is required.