How to cite: Fatima, S. and Udmale, P. D.: Drought (trigger-1) assessment in Maharashtra at Sub-district Level in the 21st century using multiple SPI and Dry Spell combinations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-721, https://doi.org/10.5194/egusphere-egu26-721, 2026.

EGU26-721

Download Close

Drought is one of the most widespread hydroclimatic hazards, characterised by slow onset, long duration, and complex propagation. While Markov chains have recently gained attention for drought prediction, their potential to characterise changing drought-state dynamics has not yet been fully explored. This study proposes a multi-tier Markov chain (MC) framework to evaluate shifts in drought transition behaviour across 133 catchments in Great Britain under observed and future climate conditions.

Using SPI- and SSI-based drought classifications defined over seven discrete categories, 7×7 MC matrices were constructed for each catchment. The analysis employs the eFLaG dataset derived from the UKCP18 regional climate projections, combining simulations from 12 regional climate models and four hydrological models (G2G, GR6J, GR4J, PDM). Three time periods were assessed: the observed baseline (1989-2018), the near future (2020-2049), and the far future (2050-2079), yielding three MC transition matrices per catchment.

The first tier of the framework applies a non-parametric permutation test to determine whether differences between transition matrices across time periods represent statistically significant shifts rather than sampling variability. For catchments exhibiting significant changes, the second tier decomposes each matrix into interpretable components- such as persistence (matrix trace), upward and downward mobility, and direction-specific transitions (Wet to Wet, Dry to Dry, Wet to Dry, Dry to Wet). This approach identifies which transition pathways drive observed temporal changes and whether future climates are associated with increased persistence, greater drying tendencies, or altered recovery patterns.

The proposed multi-tier MC framework provides a systematic means to detect, localise, and interpret evolving drought-state dynamics, offering insights relevant for water-resource planning and climate-adaptation strategies. The results will contribute to an improved understanding of potential future changes in spatio-temporal drought behaviour across Great Britain and demonstrate the broader utility of Markov chains for drought-risk assessment beyond purely predictive applications.

How to cite: Gaur, N., Medina-Lopez, E., and Beevers, L.: Analysing drought-state transition dynamics across Great Britain using a multi-tier Markov framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-875, https://doi.org/10.5194/egusphere-egu26-875, 2026.

EGU26-875

Download Close

Drought is a complex and multi-dimensional natural hazard including hydro-climatic driven and socio-economic aspects. The impacts of drought are generally shaped by spatial variability, duration and its persistence as well. Therefore, monitoring and forecasting drought are challenging task in many folds: i) Traditional drought indices such as Standardized Precipitation Index (SPI) or its variant Standardized Precipitation-Evapotranspiration Index (SPEI) are highly accepted but such indices often focus on the deviations from normal conditions within a particular time scale, which limits their ability to capture comprehensive assessment of a given region. ii) These indices require a statistical distribution describing variable of climatic factors in concern, which is extremely difficult to obtain a unique distribution that may fit to basin characteristic entirely. iii) Those are not capable of assessing drought severity and persistence over a basin. To overcome these limitations, Successive Coincidence Deficit Index (SCDI) was previously introduced in order to establish drought severity, persistence, and spatial characteristics. This study offers a new variant of SCDI, referred to as Moving Coincidence Index (MCI) based on the idea that identifies drought events triggered by simultaneous occurrence of precipitation deficits and temperature anomalies, without relying on probability distribution fitting or data normalization. The proposed MCI was applied to the Upper Tigris River Catchment (UTRC), Türkiye, which is one of important trans-boundary catchments in the Middle East. Historical analyses were conducted using long-term gauge-based precipitation and temperature observations for the period 1972–2011. The propose methodology was extended to investigate future drought behavior by using bias-corrected CMIP6 climate projections under SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios. Drought characteristics were evaluated across multiple temporal windows (1-, 3-, 6-, and 12-month) to represent meteorological, agricultural, and hydrological drought processes. Results from the historical period indicate that MCI effectively captures prolonged and successive drought conditions and provides consistent spatial patterns when compared with commonly used drought indices. Shorter time scales reveal highly localized drought behavior, while longer accumulation periods highlight persistent and basin-wide drought structures. Future projections show a pronounced increase in drought persistence and spatial coherence, particularly under higher emission scenarios. The application of MCI for CMIP6 projections enables the identification of potential changes in the spatial distribution and seasonal characteristics of coincident hot–dry conditions across the basin. As a conclusion, the integration of MCI with CMIP6 projections provides a robust and flexible framework for assessing present and future drought dynamics. The findings suggest critical insights for climate adaptation strategies, reservoir operation, and sustainable water resource management in drought-prone and transboundary river basins.

How to cite: Demir, C., Şahin, A. U., and Özkaya, A.: Monitoring Spatiotemporal Drought Events by Moving Coincidence Index Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2755, https://doi.org/10.5194/egusphere-egu26-2755, 2026.

EGU26-2755

Download Close

Effective drought monitoring in agricultural systems requires accurate estimation of root zone soil moisture to assess crop water stress and optimize irrigation decisions, yet translating continuous satellite-derived surface soil moisture into root zone dynamics remains a significant challenge.

This study presents muSEC (multilayer Surface Evaporative Capacitor), a physically based model developed from the recently proposed Surface Evaporative Capacitor (SEC) framework. muSEC links surface observations to deeper soil layers during drydown periods through a two-stage evaporation formulation and simplified vertical redistribution scheme, maintaining physical parameters across different soil types.

Spatial variability was assessed by evaluating the model across sites with contrasting soil textures and land uses, combining Time Domain Reflectometry and Cosmic Rays in situ measurements with NASA SMAP satellite retrievals. The latter provide high temporal resolution and show strong correlation with ground observations. When compared against models of varying complexity, muSEC demonstrated robust performance in reproducing soil moisture dynamics at multiple depths, thereby confirming its potential to predict agricultural water availability and drought conditions from satellite-derived surface observations.

This model framework enables deeper root-zone drought forecasting from readily available satellite surface observations, thus supporting the development of effective early warning systems and improved irrigation management in water-scarce agricultural regions.

How to cite: Olivero, A., Martini, T., Gentile, A., Gisolo, D., Canone, D., and Ferraris, S.: From Near-Surface to Root-Zone Soil Water Losses: A Physically Based Model for Drought Monitoring , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7941, https://doi.org/10.5194/egusphere-egu26-7941, 2026.

EGU26-7941

Download Close

The Standardized Precipitation Index (SPI) is computed over CONUS using daily precipitation estimates from the NOAA Daily U.S. Climate Gridded Dataset (NClimGrid-Daily). From the NClimGrid-SPI (1951-present; 0.05°x0.05°), we derive historical hydro-climatological conditions and drought information from the Drought Severity and Coverage Index (DSCI) which combines drought levels into a single areal value (from 0 to 500). One of our objective is to better understand drought dynamic and particularly how drought episodes evolve from short term rainfall deficit (i.e., less than three months) to persistent drought condition (i.e., beyond nine months). To investigate how those cascading effect work, we use a Machine Learning (ML) approach to identify spatio-temporal patterns of drought episodes over CONUS. Several unsupervised ML clustering algorithms are tested using an ensemble of features including drought duration, rainfall accumulation, drought severity (maximum DSCI, time of maximum DSCI), seasonality (drought beginning and end dates), location (latitude, longitude). Results show that the most severe drought events (i.e., DSCI > 350) are those that have the longest durations and for which drought relief is associated with higher rainfall accumulation regardless of the location considered. Furthermore, there is an apparent consistency across accumulation scales and the number of parameters selected with an optimum number of clusters around four. The Euclidian distance ML models tested seems to be able to define spatiotemporal areas of similar drought patterns. Differences between models are observed in terms of spatial definitions and predominance  of a cluster at a given location. The strongest prevalence of a given cluster has allowed to isolate areas of coherence such as the Pacific Northwest, the PNW, the Eastern Seaboard and the Southeast, and the Southwest area along the MX-US border. Domain delineations are weaker for areas such as the Rockies, the Midwest, and the Great Plains. While the SPI algorithm assumes a Gamma (McKee et al., 1993) or a Pearson III (Guttman, 1998) distribution for monthly rainfall accumulation periods, results show that this assumption might not be optimal depending on the domain considered and the accumulation period when applied to daily drought monitoring.

How to cite: Prat, O., Eldho, I., Coates, D., Nelson, B., Shaw, M., and Ansari, S.:  Determination of Spatial and Temporal Drought Patterns Over CONUS Using Unsupervised Machine Learning Clustering Algorithms , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8330, https://doi.org/10.5194/egusphere-egu26-8330, 2026.

EGU26-8330

Download Close

We address a need for hydrometeorological early warning and information systems (EWIS) in Southern Africa. In the project 'Co-Design of Hydrometeorological Information system for Sustainable Water Resource Management in Southern Africa' (Co-HYDIM-SA) we want to enhance water security in the two transboundary regions: Cuvelai-Cunene and Notwane (Namibia and Angola; Botswana and South Africa.

The Global Precipitation Climatology Centre (GPCC) has many years of experience in hosting an operational and publicly available global drought monitoring service, by combining the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI). For the SPI (SPEI) the 'gamma' ('log-logistic') distribution is fitted to the cumulative distribution function of the precipitation (climatic water balance) data. The main challenge of drought monitoring in the focus region is data scarcity. Therefore, we opted for the well-known and widely used SPI and SPEI, because these rely solely on precipitation and temperature data when calculating the potential evapotranspiration following Thornthwaite (1948). Nevertheless, observation data is difficult to acquire. Only few parameters are available and gaps in time series from stations are often present. That’s why, we work on a flexible data input in the operational system, where we can decide which data source should be used. For precipitation we mainly rely on the gridded GPCC dataset based on station data, whereas for temperature the gridded dataset from the Climate Prediction Center (CPC) is used. Furthermore, we plan to include satellite products (GIRAFE, CHIRPS, GPCP) and reanalysis (ERA5-Land) datasets. For the data acquisition and the implementation of the product, the collaboration with stakeholders in the focus region is essential. Therefore, they are included in the decision making and informed about our progress. The ‘co-design’ approach is an essential part of the project and is achieved by a close partnership with local Universities and a regular contact to the stakeholders.

At the EGU26 I want to present the Co-HYDIM-SA project, my findings and challenges we have encountered. Until today, we have calculated time series for the two Drought Indices (SPI, SPEI) and compared them with specific drought events. In general, the indices are consistent with the described droughts. One disadvantage of the SPI is that it has limitations during the dry season, especially for short term data aggregation. Whereas, the SPEI is characterized by its all-year round usability, due to the integration of potential evapotranspiration in addition to the precipitation data. As a next step, we will compare the grid data to station time series and evaluate the results by calculating skill scores.

References:

• Thornthwaite, C. W. (1948). An Approach toward a Rational Classification of Climate. Geographical Review, 38(1), 55–94. https://doi.org/10.2307/210739

How to cite: Appenheimer, J., Rustemeier, E., Ziese, M., and Finger, P.: Meteorological Drought Monitor for two transboundary regions in Southern Africa, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10126, https://doi.org/10.5194/egusphere-egu26-10126, 2026.

EGU26-10126

Download Close

Soil moisture dynamics are governed by the balance between water supply from precipitation and atmospheric demand driving evapotranspiration. Thereby, the relative roles of precipitation (P) deficits and enhanced evapotranspiration (ET) for inducing dry soils are unclear, including their variation across regions and drought phases. However, this is crucial because heat-driven drying and rainfall deficits imply distinct drought evolution patterns, different drought responses to global change, and require different water management strategies. 

In this study, we identify anomalously low surface soil moisture events and compare concurrent precipitation deficits with actual ET anomalies in a consistent framework. More specifically, we separate each dry event into two development and two recovery phases, and classify each phase into P-dominated, ET-dominated, Compound-dominated (P deficit with increased ET), or non-dominant regimes. We use gridded observation-based datasets over Europe at a daily resolution covering the study period 2001–2021.

Across Europe, the drought development phase is mostly characterized as Compound-dominated in humid to transitional climate regions in central and northern Europe. By contrast, in more arid Mediterranean regions, we find P-dominated regimes toward which become more frequent as drought development progresses. The weaker role of ET in southern Europe has to do with less amount of vegetation and more vegetation water limitation which constrains transpiration as a main contributor of ET, while atmospheric water demand is actually high in these regions. 

For the drought recovery phase we find mostly compound-dominated regimes. This indicates that rainfall events contribute to overcoming the peak dry soil moisture anomalies while this is supported by reduced ET. The latter may be relatively cloudy and colder-than-usual weather associated with precipitation as well as drought legacy effects limiting vegetation functioning and hence transpiration beyond the actual water deficit period.

While the overall results are robust, regional patterns depend on the choice of datasets and thresholds used in the identification of dry events. Overall, our analysis provides a physically interpretable typology of soil drought evolution that can support drought diagnosis and early-warning systems.

How to cite: Ochse, H., Ruiz-Vásquez, M., and Orth, R.: Role of precipitation deficits versus increased evapotranspiration for dry soils in Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12556, https://doi.org/10.5194/egusphere-egu26-12556, 2026.

EGU26-12556

Download Close

Droughts are systemic hazards with far-reaching consequences for food security, economic stability, and the environment. While traditionally characterised by deviations from normal conditions over static spatial areas or point-based time series, droughts are increasingly recognised as dynamic continuous processes with large memory effects that propagate through interlinked hydrological, ecological, and socio-economic systems (i.e. “drought as a continuum”). Despite this conceptual shift, gaps remain in capturing the evolving nature of droughts as they move across space and persist through time. This study presents a novel object-based tracking framework based on a three-dimensional Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for identifying and characterising droughts as explicit spatiotemporal entities at the global scale. The proposed methodology integrates in a novel way the Standardized Precipitation-Evapotranspiration Index (SPEI) at two complementary scales: SPEI-01 to capture rapid onset and SPEI-03 to monitor evolving persistence. The spatiotemporal identification of drought events is achieved through a two-stage clustering process: first, a 2D DBSCAN identifies spatial clusters from instantaneous intensity values; second, these entities are integrated into a 3D DBSCAN framework to establish connectivity across the temporal dimension, defining cohesive drought events globally. Additionally, we introduce a novel Drought Event Index, a composite metric synthesising an event’s duration, pace, extent, and intensity into a single metric that enables direct comparison of drought events across diverse geographical locations and historical periods. Methods are applied to the ERA5 reanalysis dataset for the period 1940-2025. Results indicate a marked increase in the frequency and intensity of drought events in recent decades compared to the period 1950-1990, while accurately identifying the spatiotemporal dynamics of recent significant events around the globe, such as the 2018 and 2022 drought events in Europe, and the unprecedented 2019-2025 multi-year droughts in South America. The proposed methodological framework evaluates dynamics often unaccounted for by static analysis, thus enabling the quantitative assessment of droughts as a continuum at the global scale and across different timescales.

How to cite: Hrast Essenfelder, A., Toreti, A., Cammalleri, C., and Vicente-Serrano, S.: Drought as a Continuum: quantifying global Spatiotemporal Connectivity of Drought Events, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12986, https://doi.org/10.5194/egusphere-egu26-12986, 2026.

EGU26-12986

Download Close

Europe is undergoing increasingly extreme events, especially droughts that have become more frequent and severe. The observed conditions lead to water scarcity, agricultural impacts, and river flow issues, with projections indicating worsening conditions despite some regional variability. It is therefore crucial to find methods that can monitor drought conditions such as intensity, categories, and patterns, and that can assess the pacing of those changes over Europe. In recent years, there is an increasing application of geodetic techniques such as the Gravity Recovery and Climate Experiment (GRACE) and the Global Positioning System (GPS) in hydroclimatic research that enable monitoring of the continental water storage and Earth's displacement by observing the gravity field variations or the changes in the position of permanent stations, respectively. The recalculation of these changes into Drought Severity Index (DSI) provides a successful method for studying drought characteristics. However, limitations of both techniques, such as GRACE signal leakage and systematic errors of GPS, do not allow for an unambiguous assessment of drought. Thus, in our study, we overcome the limitations of both geodetic techniques by calculating a Multivariate DSI (MDSI) based on a combination of time series using the Frank copulas concept. We focus on emphasizing the potential of MDSI in describing drought characteristics compared to GRACE-DSI and GPS-DSI, as well as the sensitivity of DSIs to regional and local hydroclimatic and hydrometeorological changes recorded in Europe. In view of the sensitivity of both techniques to different temporal signals, we also take a step further by defining a new modified MDSI (mMDSI), which is the next step in climate change research. We divide GRACE-derived and GPS-observed displacement series into three temporal scales, i.e., short-term, seasonal, and long-term, which we then convert to DSI. The total mMDSI is defined as a combination of various temporal signals of GRACE-DSI and GPS-DSI. We perform spatial and temporal analyses to identify patterns of climate change, e.g., wetting/drying hotspots, and assess the reliability of mMDSI/MDSI by comparison with various meteorological and hydrological datasets. We prove that MDSI and mMDSI are key methods for decision-makers that may be applied in establishing preventive strategies to mitigate the effects of droughts in regions indicating ‘warning’ conditions.

How to cite: Lenczuk, A., Ndehedehe, C., Klos, A., and Bogusz, J.: Assessment of the potential of combined geodetic-based drought indices for studying climate change in Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21651, https://doi.org/10.5194/egusphere-egu26-21651, 2026.

EGU26-21651

Download Close

Effective drought and water-resource management is a fundamental challenge worldwide. In recent decades, the intensification of drought has become a serious challenge in northwestern Iran, particularly in the Lake Urmia basin, where rising temperatures and declining heavy rainfall have accelerated water scarcity. Therefore, monitoring drought and studying its trends is crucial.

This study evaluates drought patterns at seven meteorological stations using the Standardized Precipitation Index (SPI) and the Standardized Precipitation-Evapotranspiration Index (SPEI) at 3, 12, and 24-month time scales. The Innovative Trend Analysis (ITA) method, supported by the Seasonal Kendall test, was used to identify and assess drought behavior.

The ITA method clearly showed drought trends, whereas the Seasonal Kendall test often failed to detect any trends in short-term data. The results showed that the stations of Tabriz and Urmia have more dry and normal periods, while wet periods have reduced, indicating a reduction in overall moisture. Mahabad, Saqqez, Maragheh, and Sarab had a decrease in all categories (dry, normal, and wet), which demonstrates severe and persistent drought. SPEI also identified short-term droughts in Mahabad and Tekab, which SPI was unable to capture.

Frequency analysis using McKee’s classification showed that most months fall within the normal range; however, ITA trends indicated that the intensity and persistence of normal periods are decreasing in many stations. These results indicate that ITA trends can identify which stations enter drought rapidly, retain moisture stability, and is critical for water storage planning and early warning systems.

Overall, the integration of SPI and SPEI with statistical and trend methods provides a comprehensive framework for drought monitoring in semi-arid regions. The findings suggest that the use of ITA is highly effective for water resource management, long-term change prediction, and strengthening adaptation strategies in the sensitive and critical Lake Urmia basin.

How to cite: Ziafati, N., Khalili, K., Rezaie, H., Fathollahzadeh Attar, N., Rodrigues Pereira da Franca, M. J., and Pourzangbar, A.: Innovative trend analysis method for drought indicators and pattern detection of the Urmia lake basin, Iran, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5676, https://doi.org/10.5194/egusphere-egu26-5676, 2026.

EGU26-5676

Download Close

Drought poses a growing risk to society and ecosystems in Sweden, creating major challenges for water supply, agriculture, and emergency response. Although the country has long been regarded as water-rich, recent drought events have exposed significant vulnerabilities and highlighted the need to improve national preparedness. Within this context, the ACT4Drought project, funded by the Swedish Research Council (FORMAS), aims to co-develop an actionable service for drought and water scarcity at sub-seasonal to seasonal (S2S) timescales. We use bias-adjusted seasonal meteorological forecasts from the ECMWF SEAS5 prediction system, which provides ensemble forecasts up to seven months ahead. These forecasts are used to drive the Swedish national hydrological model (S-HYPE) and generate forecasts of soil moisture, discharge and related drought indicators. We evaluate the seasonal predictability of droughts across meteorological, agricultural and hydrological aspects, using the Standardized Precipitation Index (SPI), Standardized Precipitation and Evapotranspiration Index (SPEI), Standardized Soil Moisture Index (SSMI), and Standardized Streamflow Index (SSI) at 1 to 3-month aggregations, and assess their forecast skill across initialization times, lead times and spatial domains. By identifying where and when seasonal forecasts reliably capture drought conditions, this work provides a foundation for more robust operational drought early warnings and advances Sweden’s capacity for drought preparedness.

How to cite: Du, Y., Canedo Rosso, C., Bertolatus, S., and Pechlivanidis, I. G.: Seasonal Predictability of Hydrometeorological Drought in Sweden, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3343, https://doi.org/10.5194/egusphere-egu26-3343, 2026.

EGU26-3343

Download Close

Skillful sub-seasonal to seasonal (S2S) hydrologic forecasts are essential for proactive, risk-based water management, yet the practical boundary of their usefulness - the predictability limit - remains poorly quantified for high-resolution drought indicators. Here, we use the operational High-resolution Sub-seasonal Hydroclimatic Forecasting System, HS2S (https://www.ufz.de/HS2SForcasts4Germany), providing daily ensemble soil-moisture forecasts for Germany since 2020, and quantify predictability limits with CRPS (Continuous Ranked Probability Score), a strictly proper scoring rule for probabilistic forecasts.

HS2S couples the mesoscale Hydrologic Model (mHM; https://mhm-ufz.org) with ECMWF extended-range ensemble meteorological forecasts. In the latest version of the forecasting system (Hs2S v0.2), 51 atmospheric ensemble forecasts are interpolated from 10~km to 1~km using external drift kriging and subsequently bias-corrected, enabling near-real-time hydrologic forecasting and uncertainty estimates.

We quantify predictability limits for recent drought conditions in Germany, focusing on the persistent multi-year drought of 2018--2022 and the acute drought conditions observed in 2025. Using the Soil Moisture Index (SMI; total soil column), we diagnose how forecast skill decays with lead time (up to 42~days) and how this decay varies across space. To contextualize the added value of meteorological forcing versus hydrologic persistence, we benchmark HS2S against (i) an Ensemble Streamflow Prediction (ESP)-style reference that propagates initial hydrologic conditions with historical meteorological sequences and (ii) a purely statistical ARIMA baseline. We further isolate the contribution of initial hydrologic conditions, derived from high-density German Weather Service (DWD) station observations, and show how land-surface "memory'' can extend useful predictability beyond that provided by meteorological forcing alone. The results provide a benchmark for further impact-based drought early warning studies and identify actionable windows of opportunity in which high-resolution forecasts add decision-relevant value.

How to cite: Najafi, H., Shrestha, P. K., Boeing, F., Kelbling, M., Thober, S., Rakovec, O., and Samaniego, L.: Quantifying the Sub-seasonal Predictability Limit of 1-km Soil Moisture Drought in Germany, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11980, https://doi.org/10.5194/egusphere-egu26-11980, 2026.

EGU26-11980

Download Close

In the Netherlands, flood forecasting and early warning systems are well established and operationally embedded. However, despite an increasing frequency of drought events and impacts over the past decades, drought early warning systems remain comparatively less developed. This gap is critical, as growing climate variability is expected to intensify agricultural, ecological, and hydrological stress even in temperate regions. Standardized drought indices such as the Standardized Precipitation Index (SPI) and Standardized Streamflow Index (SSI) provide an established framework for drought monitoring and forecasting, but they strongly depend on the underlying probability distributions used to represent hydroclimatic variability and extremes. Poor distribution choices can distort index values and reduce forecast reliability, especially for moderate to extreme drought events.

In this study, we develop an enhanced drought early warning approach for the Netherlands using SPI (1-, 3-, 6-, and 12-month) and SSI (1- and 3-month) accumulation periods. Forecasts are derived from the operational European Flood Awareness System (EFAS) and ECMWF SEAS5 seasonal predictions. Reference indices are computed from historical precipitation and streamflow using ERA5-Land and EFAS datasets. For each grid cell, candidate distributions are fitted to accumulated monthly variables, and the dominant distribution is selected for standardization. To ensure the selected distributions remain valid under forecast conditions, we evaluate distribution performance using ECMWF hindcasts, applying a lead-month climatology framework (fitting and testing distributions per initialization month and lead time). Forecast indices are then evaluated against reference indices.

The use of correct distributions is expected to improve SPI/SSI forecast performance and enhance skill in predicting moderate to extreme drought events, particularly at short to medium lead times. This work supports operational integration of drought early warning into the Dutch forecasting center.

How to cite: Odongo, R. A., Sutanto, S. J., Biemans, H., and Paparrizos, S.: Evaluating probabilistic distributions for drought forecasting system in the Netherlands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20109, https://doi.org/10.5194/egusphere-egu26-20109, 2026.

EGU26-20109

Download Close

Earth System Science Interdisciplinary Center, University of Maryland

Terrestrial Water Storage (TWS) integrates information from various important sources of moisture, each with distinct temporal and spatial dynamics, including groundwater, soil moisture, and surface water storage. TWS anomalies, hence, can serve as an indicator of drought, and are being used operationally, such as by the U.S. Drought Monitor. TWS can be simulated by land surface models and observed from satellites like GRACE/GRACE-FO, providing extensive spatial and temporal coverage in near-real time, which is particularly attractive in data-sparse regions that are also food insecurity hot spots. FLDAS (Famine Early Warning Systems Network Land Data Assimilation System)-Forecasts provide TWS forecasts at the subseasonal to seasonal scale (S2S). While past research has found the TWS forecasts to be a skillful predictor of Leaf Area Index (used as a surrogate of vegetative productivity) at 3 months lead time, further research is needed to facilitate operational application of TWS forecasts in supporting food insecurity early warning. This presentation summarizes recent research that (i) evaluates the skill of TWS forecasts from the FLDAS-forecasts system relative to GRACE/GRACE-FO observations and highlights the inter-model differences that lead to differences in TWS forecasts, (ii) investigates the role that each of the TWS components plays in the predictability of TWS at the S2S scale, and highlights the role of rootzone soil moisture in TWS predictability. Together, these analyses provide insights into both the promise and limitations of producing S2S forecasts of TWS using either land surface models or statistical models. We focus our analysis on data-sparse, food-insecure regions in Africa where data limitations are widespread and any improvement in forecast skill can be translated into improved early warnings of agricultural drought.

How to cite: Shukla, S., Anderson, W., Li, B., Cook, B., Hazra, A., Slinski, K., and McNally, A.: Investigating the Predictability of Terrestrial Water Storage at Subseasonal to Seasonal Scale to support drought and food insecurity early warning in Data-Sparse Regions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14348, https://doi.org/10.5194/egusphere-egu26-14348, 2026.

EGU26-14348

Download Close

Droughts have become increasingly frequent in recent years across Central Asia, posing significant challenges to water management, agriculture, and socio-economic stability in the region. Accurate forecasting of droughts is crucial for mitigating their impacts, yet effective prediction relies on comprehensive and high-quality datasets from the source areas. In Central Asia, however, such datasets are often sparse or incomplete, limiting traditional monitoring and forecasting approaches. To address this challenge, we employ remote sensing datasets to forecast potential drought occurrence across the region. By leveraging satellite-derived indicators of snow cover, vegetation index (NDVI), and precipitation anomalies, we develop predictive models capable of identifying areas at risk of drought even under limited ground-based observations. The results demonstrate the potential for remote sensing approaches to fill critical data gaps, providing timely and actionable information for decision-makers. Implementation of these forecasts at the policy level can support proactive drought management, resource allocation, and adaptation strategies, ultimately enhancing regional resilience to increasing drought frequency.

We have integrated the developed methodology of drought forecasting into MODSNOW-Tool as an additional functionality of forecasting droughts. 

How to cite: Gafurov, A., Weiss, T., and Akhundzada, N.: Drought forecasting for improved water management and , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14790, https://doi.org/10.5194/egusphere-egu26-14790, 2026.

EGU26-14790

Download Close

Drought is defined in a variety of different ways. One method is through the use of the SPEI (Standardised Precipitation and Evapotranspiration Index) derived from the HadUK4 by the UK CEH (Centre for Ecology and Hydrology) and made available at 1km for the UK only and globally through a new product, the Global Multi-Index Drought (GMID) at 0.1º . After an exhaustive test of various dependent variables, three variables were chosen for deep learning training, 1km LST from the VIIRS instrument onboard the NASA-NOAA satellites is combined with Precipitation and Soil-evapotranspiration monthly datasets then downscaled to 1km from 0.1º to train a deep learning model to forecast 1 km SPEI. These forecasts can then be compared with the aforementioned 1km (UK only) and 0.1º (UK, Ireland and France) SPEI and GMID datasets respectively. Examples will be shown of the UK , Ireland and France regions. Farmers, NGOs, government scientists and policy makers require drought forecasts at near human scale and as far ahead as possible for water conservation planning. These 1km results need to be downscaled to human level at 10m.

An unique processing system for generating 10m spectral and broadband albedo which is part of the Copernicus Global Land Monitoring Service called S2GM (Sentinel-2 Global Mosaic) has been employed to generate 10m products [1,2]. From these spectral albedos, simple vegetation indices such as NDVI can be derived over a monthly time period and NDVI can be employed to downscale the 1km forecasts up to 10m. This application of a composited product eliminates the problems of cloud cover at mid-latitudes which Sentinel-2 sampling every 5-daily has. Examples from the 2022 and 2025 droughts will be shown for the UK, Ireland and France (UKIF). The monitoring of drought through the water extent of reservoirs using S2GM monthly composite spectral albedos will also be shown as an independent method of drought assessment.

The GTIF-UKIF drought capability results will be shown in the context of crop-type and vegetation productivity at the 10m level using an unique webGIS system developed for all the Green Transition Information Factory (GTIF) capabilities (gtif-uk-ireland-france.net). These results indicate that this drought monitoring and forecasting method may have the potential to be rolled out across the rest of Europe and southwards across Africa to provide forecasts 3-6 months ahead of a drought.

The authors would like to thank ESA for contract no. 4000144118/24/I-NS, Burak Bulut of UK CEH for the SPEI and GMID datasets and Gillian Watson for the NDVI processing and downscaling of the SPEI.

Cited references
[1] Muller J.P., Song R. Brockley D., Whillock M., 2023. Sentinel-2 Global Mosaic HR-Albedo Algorithm Theoretical Basis Document S2GM-UCL-ATBD-v3.1 https://s2gm.land.copernicus.eu/help/documentation

[2] Muller, J-P., Song, R., Griffiths, P., 2025. Bi-facial PV solar power systems for mixed use of arable and grassland, an evaluation over GB and Ireland taking into account environmental exclusion areas. DOI: 10.5194/egusphere- EGU25-18951 

How to cite: Muller, J.-P., Song, R., and Griffiths, P.: Forecasting drought using SPEI at the 10m level with ERA5 and Sentinel-2 spectral albedo products as part of ESA-GTIF project for the UK, Ireland and France., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20482, https://doi.org/10.5194/egusphere-egu26-20482, 2026.

EGU26-20482

Download Close

Among other water resources, surface water, subsurface water, groundwater, and water supply are all adversely impacted by drought, a natural occurrence. As a type of hydrological drought, groundwater drought reflects both the unique features of the aquifer and human caused disturbances to the hydrological system. It is evident that human activity has both direct and indirect effects on the worsening of groundwater drought. Groundwater withdrawals are frequently used to meet water needs during hydrological and agricultural droughts because groundwater storage offers resilience. As a result, excessive groundwater extraction may make drought more severe. Quantitatively characterizing groundwater drought is extremely difficult due to the complex nature of groundwater flow systems and the difficulties in obtaining field observations pertaining to aquifers. By offering early warnings, long-term drought forecasting is essential to reducing drought risks.

Accurate long-term drought forecasting has long been of interest to researchers, but it is difficult because accuracy typically declines with forecasting period. This study's main goal is to present a novel hybrid deep learning model, Deep Feedforward Natural Networks (DFFNN), improved by War Strategy Optimization (WSO), for high accuracy long lead time drought forecasting. One of the vital aquifers in Odisha (Keonjhar district) was monitored for groundwater drought using the Standardized Groundwater Level Index (SGI), and forecasts were made for a range of lead times, including 1, 3, 6, 9, 12, and 24 months. For this study, monthly groundwater level data from 10 observation wells over a 25-year period (1996–2021) were collected. The observation wells were chosen based on their uniform distribution within the aquifer area and the completeness of their data records. The WSO algorithm was used to optimize important DFFNN parameters, such as the number of neurons and layers, learning rate, training function, and weight initialization. Two well known optimizers, Particle Swarm Optimization (PSO) and Grey Wolf Optimization, were used to validate the model's performance. 

Outcomes revealed that DFFNN-WSO model attained superior performance for SGI 24 (t + 12) with a coefficient of determination (r²) of 0.9847, Root Mean Square Error (RMSE) of 0.1035, willmott index of agreement (IoA) of 0.9812; for SGI 24 (t + 9) with r² = 0.8965, IoA = 0.8906 and RMSE = 0.1942; for SGI 12 (t + 6) with r² = 0.8473, IoA = 0.8352 and RMSE = 0.2315; for SGI 24 (t + 3) with r² = 0.7915, IoA = 0.7846 and RMSE = 0.2693; and for SGI 24 (t + 1) with r² = 0.7725, IoA = 0.7642 and RMSE = 0.3187 at the W5 station. Analysis of results indicated that DFFNN-WSO model outperformed all applied models consistently at all locations, and it considerably enhanced drought forecasting accurateness, with highest improvements for SGI 24 (t + 12) and moderate gains for SGI 24 (t + 1). The suggested model is a useful tool for real time drought monitoring and management since it offers precise and timely drought predictions, allowing for well-informed decision making to lessen the effects of drought.  

Keywords: Deep Feedforward Natural Networks (DFFNN); War Strategy Optimization; Standardized Groundwater Level Index (SGI); Water scarcity; Keonjhar

How to cite: Samantaray, S., Sahoo, A., and Satapathy, D. P.: Improving Long-term Drought Forecasting with a novel Hybrid Deep Learning model based on Standardized Groundwater Level Index, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22079, https://doi.org/10.5194/egusphere-egu26-22079, 2026.

EGU26-22079

Download Close

The drought monitoring and forecasting are essential for effective water resources management and decreasing climate risks because of increasing climatic variability. In order to simulate the 12-month Standardized Precipitation Evapotranspiration Index (SPEI-12), this paper evaluates the appropriateness and the comparative performance of gradient boosting-based machine learning models namely; Gradient Boosting Regressor, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost). A rigorous evaluation methodology is adopted to ensure scientific accuracy and applicability of the operation where statistical goodness of fit measures, hydrological efficiency measures, diagnostics of errors, bias measures, test of significance, and accuracy of threshold-based drought classification are all undertaken. According to the results, the learning capacity of all gradient boosting models is high in the course of the training, and R² and NSE values are between 0.98 and 0.99, which suggests that the variability of SPEI-12 is depicted well. LightGBM and CatBoost outperformed the other approaches in both R², NSE, and KGE values and lower RMSE and bias in the testing stage, therefore, the models were the most predictable and applicable. It is interesting to note that LightGBM is generally accurate and efficient, whilst CatBoost is more resistant to outliers, which is demonstrated by lower average relative error. LightGBM is the most superior approach when compared to other model with evaluation metrics (R² of 0.87, NSE of 0.86, KGE of 0.83, and the lowest RMSE of 0.37). Evaluation using the threshold indicates the operational strength of the proposed framework, and all models were highly accurate in detecting moderate and severe situations of drought. In 67.23% of the test cases the model correctly forecasted an event of drought at a tolerance of 10% which rose to 90.64% at a tolerance of 100 percent which is corroborated by the fact that it is a realistic model that can be useful in an operational drought early warning system. Models were most effective under intense drought conditions with a high degree of accuracy of over 90 percent at the 100 percent mark, which means that it is reliably applicable in detecting severe drought conditions that are necessary in emergency response planning. The model performance was strongly validated by means of the rigorous statistical analysis using various statistical metrics which included: R² NSE, KGE, RMSE, P-Bias, and F-statistics. This multimeric method ensured comprehensive evaluation that can be used in operation in different climatic regions. In general, the findings indicate that machine learning models based on the gradient boosting are a valid and useful approach to predict the drought index over the long run. This paper demonstrates the unique advantages of boosting techniques in the long-term drought index (SPEI-12) modelling and the importance of selecting and validating the model with numerous statistical measures. The proposed approach holds tremendous potential in improving risk assessment for drought monitoring.

How to cite: Panigrahi, S. and Kumar Vidyarthi, V.: Scale-Dependent Gradient Boosting Algorithms for SPEI Drought Prediction, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16440, https://doi.org/10.5194/egusphere-egu26-16440, 2026.

EGU26-16440

Download Close

Groundwater is a critical resource in semi-arid regions, particularly in the Haouz plain of central Morocco, where climatic variability and growing anthropogenic pressures are causing increased stress on aquifer systems. This study aims to assess future groundwater drought in the Haouz aquifer under conditions of data scarcity by integrating regional climate projections from the Med-CORDEX initiative with advanced machine learning techniques. The research is driven by the need for reliable, spatially resolved forecasts in regions where hydrological and groundwater data are limited or unavailable. The core methodology involves the use of meteorological drought indices to quantify drought events based on climate variables. These indices were calculated using historical and projected climate data derived from Med-CORDEX simulations under two Representative Concentration Pathways: RCP 4.5 and RCP 8.5. In the absence of dense ground-based monitoring networks, the study relies on ERA5 reanalysis data and virtual station datasets to create an input matrix suitable for predictive modeling. Machine learning models were trained to estimate groundwater drought conditions using climate predictors and geographical variables. Among the models tested, Random Forest exhibited superior performance, capturing non-linear interactions and delivering high predictive accuracy (R² > 0.9). The results reveal a significant intensification of drought conditions over time, particularly in the long term under the RCP 8.5 scenario, with increased occurrence and severity of extreme drought events projected in the latter half of the 21st century. The western part of the aquifer is identified as highly vulnerable, experiencing the most pronounced drought intensification. In contrast, the eastern portion shows a degree of resilience, maintaining near-normal drought conditions even under severe climate scenarios. This spatial variability underscores the importance of localized groundwater management strategies. The study concludes that coupling regional climate projections with machine learning offers a promising approach for groundwater drought forecasting in data-scarce environments. The modeling framework developed is scalable and adaptable to similar hydrological systems facing data limitations. 

How to cite: El Bouazzaoui, I., Ait El Baz, A., Ait Brahim, Y., Machay, H., and Bougadir, B.: Forecasting groundwater drought in data-scarce regions using a machine learning approach and Med-CORDEX climate projections: the case of the Haouz aquifer (Morocco), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2342, https://doi.org/10.5194/egusphere-egu26-2342, 2026.

EGU26-2342

Download Close

Machine learning is now widely used for environmental forecasting. Although predictive skill often varies only modestly across architectures, interpretability remains a persistent challenge, reducing transparency and limiting stakeholders’ ability to understand model behavior, build trust, and apply forecasts in practice. Balancing accuracy and interpretability are therefore essential for scientific credibility and real-world decision-making. In this context, Graph Attention Network (GAT) is particularly promising. Graph representations encode spatial dependencies and capture complex non-Euclidean relationships, such as upstream–downstream hydrological connectivity or large-scale teleconnections, that conventional grid-based models often struggle to represent. Attention mechanisms then adaptively weight information from different neighbors, helping the model focus on the most informative signals while offering a transparent view of which connections drive each prediction. Here, we evaluate the transferability and representational capacity of GAT for soil-moisture drought forecasting by modeling hydrological response units (HRUs) as nodes in a soil-moisture interdependence graph that preserves connectivity between locations. Beyond predictive accuracy, our analyses show that the model learns stable, physically meaningful relationships and yields interpretable hydrological insights. Feature-importance results reveal consistent links between key predictors and drought dynamics across both space and time. Attention diagnostics indicate pronounced seasonality: weights respond to the relative variability of source-node inputs, producing alternating dominance of high- and low-elevation sources between winter and summer. Spatially, the model consistently prioritizes same-elevation connections, suggesting that it internalizes distinct hydrological regimes in its learned representation. We also highlight three ongoing efforts: 1) extending evaluation to additional climatic regions to test transferability; 2) exploring hybrid GAT–sequence architectures to better capture temporal dynamics, while carefully assessing potential trade-offs in systematic, physically meaningful interpretability; and 3) developing an easy-to-use, open-source codebase to support broader use and reproducibility.

How to cite: Tuo, Y., Wirthensohn, M., Zhu, X., Peng, J., and Disse, M.: On the Value of Graph Attention Network for Interpretable Drought Forecasting, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19634, https://doi.org/10.5194/egusphere-egu26-19634, 2026.

EGU26-19634

Download Close

Recent improvements of the hydrological, open source (OS) LISFLOOD model aimed to support both flood- and drought-related applications. The latest model upgrades are very promising for drought monitoring use cases, for which the sources of improvements can be grouped into four main areas: (i) updated meteorological forcings improving the quality of the gridded model inputs; (ii) revised static maps providing an improved representation of catchment morphology and soil properties; (iii) structural model revisions that enhance the physical consistency of simulated water fluxes; and (iv) the adoption of a new calibration objective function, the Joint Divergence Kling–Gupta Efficiency (JDKGE), which improves low-flow performance while maintaining or improving accuracy for high flows compared to the previous calibration using the Kling–Gupta Efficiency.

In this study, we evaluate the cumulative effect of these developments with a focus on drought monitoring and forecasting applications. Using multi-source observational data and different benchmarking strategies, we evaluate the accuracy and physical consistency of the new operational LISFLOOD model setup of the European and Global Flood Awareness Systems (EFAS version 6 and GloFAS version 5) of the the Copernicus Emergency Management Service (CEMS). The evaluation focuses on two key hydrological variables for drought monitoring, namely river flows and soil moisture, at the European and global scale. Beyond the two raw variables, we examine the performance of drought indicators, including the Low Flow Index and Soil Moisture Index from the European and Global Drought Observatories (EDO and GDO), and assess their ability in detecting drought events, using both hazard observations and impact data as reference. Results from long-term simulations show substantial improvements in drought detection thanks to the new developments in OS LISFLOOD and associated CEMS setups. Similar improvements in drought forecasting skill are also anticipated and will be investigated in further work.

How to cite: Ficchì, A., Bavera, D., Grimaldi, S., Moschini, F., Pistocchi, A., Russo, C., Mazzetti, C., Wortmann, M., Prudhomme, C., Salamon, P., and Toreti, A.: Advances in drought monitoring using an operational hydrological model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20246, https://doi.org/10.5194/egusphere-egu26-20246, 2026.

EGU26-20246

Download Close

This study analyzes trade-offs in water supply during shortages and droughts, focusing on sustainable measures at various demand nodes. It introduces drought-specific methods and optimal strategies. The methodology includes: (1) deriving an analytical solution for the water shortage index in a cross-watershed, network-flow, diverse supply system; (2) drought low-flow frequency analysis; (3) designing low-flow events using the Alternating Block Method; (4) a multi-objective simulation optimization model for resource allocation; (5) pattern analysis of water supply for industrial, livelihood, and agricultural needs; and (6-7) trade-off analysis of fallow strategies with cross-watershed diversion using recycled and hyporheic water. To address extreme events, the model's objective shifts from a yearly water shortage index to a ten-day modified shortage index (MSI), aiming to reduce tap and irrigation shortages. Decision variables include dam releases, tap and irrigation water supply, and regional diversion, with constraints on flow continuity and physical limits. The cross-watershed reservoir network-flow allocation model in Taiwan is developed using GAMS. Without agricultural fallow during the 2020 drought, tap water shortages would reach 29.43%, 18.13%, and 12.58% in Hsinchu, Taoyuan, and Banxin. Opening the Taoyuan-Hsinchu support pipeline reduces shortages by 4.16%-5.58% under non-fallow and fallow scenarios. Optimal fallow can cut shortages in Shimen and Taoyuan by 35.39% and 28.41%, respectively. During 200-year drought scenarios, shortages only occur in Hsinchu by 13.81%-15.32%, and pipeline operation reduces shortages to below 0.11%. To bring shortages below 3%, fallows are necessary across all areas during long-lasting, high-return droughts, where shortages maximum rise to 95.67%. Recycled water further helps reduce shortages in Shimen and Taoyuan by up to 9.18%.

How to cite: Huang, C.-L. and Hsu, N.-S.: Analytical trade-off simulation-optimization of drought-resistant water supply allocation strategies under various demands using a multi-objective cross-watersheds network-flow model , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2158, https://doi.org/10.5194/egusphere-egu26-2158, 2026.

EGU26-2158

Download Close

Agricultural Drought Risk constitutes one of the most significant and long-term damaging impacts of climate change, primarily contributing to food insecurity. Despite the large number of previous research activities on drought risk management, some countries remain excluded from the global drought studies while vulnerable communities are still exposed to famine and livelihood loss. This critical gap prove that the applied drought assessment techniques faced successive refinements over time including dataset and methodologies but exhibit notable limitation regarding both spatial assessment and theoretical consistency.

The present study examines a comparative analysis of agricultural drought risk across two Tunisian watersheds, Medjerda and Merguellil, which are characterized by distinct climatological conditions. The analytical framework integrated the three core components of agricultural drought risk: hazard, vulnerability, and exposure while adopting a resource nexus perspective to capture the interdependencies among the selected indicators of each component.  

Drought indicators were collected from remotely sensed data over the period 2016-2024 considered as the latest drought period in Tunisia. The hazard indicators were represented by Precipitation condition index (PCI), Temperature condition index (TCI), Vegetation condition index (VCI) and Soil moisture condition index (SMCI). The vulnerability indicators included Runoff, Ground Water (GW), Primary Productivity (NPP) and Nighttime Light (NL). The exposure indicators were cropping area and population density. All indicators were normalized to ensure integration within drought analysis framework. This study employed two temporal lags initially addressing the short-term dynamics of drought hazard on a monthly scale followed by yearly assessment of drought risk components. The combination process of drought indicators was conducted by three objective weighting techniques: Principal Component Analysis (PCA), Gaussian Mixture Model (GMM) and Entropy to create time series of drought risk maps.

The spatial structure of obtained drought risk maps was analyzed using spatial pattern indices, including the Gini Index, along with four landscape metrics: Number of Patches (NP), Landscape Shape Index (LSI), Shannon’s Diversity Index (SHDI), and Contagion Index (CONTAG). These indices were considered as objective functions within multiple Pareto optimization scenarios to identify the most relevant spatial configuration of drought risk maps.  

The optimization results provided robust evidence indicating that the entropy-based approach was the most effective method in drought risk monitoring. The Medjerda watershed, which is characterized by sub-humid regime, faced strong drought variability with a severe drought period recorded in 2023, while drought risk trend remained gradual in the semi-arid watershed, Merguellil, showing slight change in 2022 and 2023.

The drought assessment determined the contribution of drought indicators in creating each component, the highest weight was assigned to VCI within monthly and yearly hazard component. Considering the vulnerability component, NPP exhibited the highest contribution followed by GW in the case of Medjerda and NL in the case of Merguellil. The cropping area had highest weight within exposure component. The results offer an objective and reliable assessment of the temporal drought risk variability and quantitatively reveal the climate–water–food nexus shaping drought risk. Overall, the study confirms the viability of using integrated risk assessment for sustainable water-use in agriculture. 

How to cite: Hammami, H., Chapagain, S. K., Zarei, A., and Schütze, N.: Comprehensive Management of Agricultural Drought Risk: Integrating the Climate-Water-Food Nexus , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2254, https://doi.org/10.5194/egusphere-egu26-2254, 2026.

EGU26-2254

Download Close

Groundwater dynamics shape how drought is experienced in landscapes: they regulate the persistence of streamflow, controlling the duration and magnitude of ecological stress linked to low flows, and govern recovery trajectories long after rainfall deficits ease. Despite this, groundwater is often weakly represented in routine drought characterization, largely because piezometric records are sparse, discontinuous, and unevenly distributed, and because groundwater responses are filtered through storage, geology, and time-lagged recharge processes that obscure simple attribution to atmospheric anomalies. Robust and comprehensive drought diagnostics and early warning need methods that link meteorological forcing to interpretable indicators of groundwater storage and release.

The analysis is conducted in near natural headwater catchments in southern Spain, thereby reducing the influence of pumping. We propose a triangulation approach to characterize drought propagation using three complementary components: meteorological drought forcing measured with Standardized Precipitation- Evapotranspiration Index (SPEI), groundwater drought state obtained from piezometric data, measured with the Standardized Groundwater Index (SGI), and groundwater-controlled discharge behaviour captured through a simple baseflow proxy extracted from gauged streamflow and Terraclimate modelled runoff data (Abatzoglou et al., 2018).

Meteorological drought is represented by the SPEI evaluated across accumulation windows from 1 to 48 months. Observed groundwater head series are quality-controlled, filled in and regularised using transfer-function noise timeseries modelling with the Pastas software (Collenteur et al., 2019) to obtain continuous records, from which SGI is computed using a month wise non-parametric standardisation. Baseflow is derived from observed discharge and runoff data using a consistent separation approach, and standardised to enable direct comparison with SGI as a second, catchment-integrated representation of groundwater state.

We explore drought propagation by mapping correlations and response lags between SPEI and both groundwater anomaly indicators, SGI and standardized baseflow, identifying the dominant memory windows and seasonality of sensitivity. Predictive performance is then assessed using regressions for interpretable relationships between groundwater response and the most informative SPEI scales, and Random Forest regression to capture further interactions. We stratify and interpret these relationships by lithology, aquifer properties and catchment size. We further test whether SPEI–groundwater relationships exhibit structural changes over time, via moving-window correlations, wavelet analysis and segmented analyses across sub-periods and seasons.

Across sites, the triangulation reveals coherent but aquifer-dependent propagation patterns, which are presented with narratives and diagrams of drought propagation pathways. SGI and baseflow-based state indicators consistently align with SPEI at intermediate to long accumulation windows, reflecting nuanced modulation in storage and recession dynamics. Importantly, baseflow proxies complement SGI by providing a continuous, integrated signal of groundwater release that can support and strengthen monitoring, especially where piezometric data are sparse. The combined framework delivers operationally relevant SPEI trigger windows and predictive models for anticipating groundwater-related anomalies in Mediterranean environments.

References
Abatzoglou, J. T., Dobrowski, S. Z., Parks, S. A., & Hegewisch, K. C. (2018). TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958–2015. Scientific data, 5(1), 1-12.
Collenteur, R. A., Bakker, M., Caljé, R., Klop, S. A., & Schaars, F. (2019). Pastas: Open source software for the analysis of groundwater time series. Groundwater, 57(6), 877-885.

How to cite: Serrano-Acebedo, P. and Limones, N.: Two groundwater stories, one drought: Standardized Groundwater Index and baseflow proxies under climatic forcing in near-natural aquifers in southern Spain, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6912, https://doi.org/10.5194/egusphere-egu26-6912, 2026.

EGU26-6912

Download Close

Conventional drought monitoring in Greece has largely relied on in-situ measurements (rain gauges, reservoir records) to infer meteorological and hydrological indices. Despite the fact that the gauge measurements are valuable, most of the mountains basins lack of them. Moreover, reservoir information is not always frequent or openly available and meteorological indicators alone do not always reflect the evolving situation in major water-supply reservoirs. For this reason, Satellite Earth observation in combination with reanalysis data provide a strong complement. Satellite imagery allows reservoir water extent to be mapped directly and repeatedly, while meteorological data capture the spatial variability across entire river basins, supporting both situational awareness and longer-term analysis.

In this study, an open-data, long-term monitoring pipeline was implemented in Google Earth Engine, combining freely available satellite and reanalysis datasets. Monthly reservoir surface-water extent (2017–2025) was derived from Sentinel-2 optical imagery using multiple water indices (NDWI, MNDWI, AWEI) with consistent cloud/shadow masking and monthly compositing. A key element for “long-memory” drought assessment was added through the JRC Global Surface Water Monthly Recurrence dataset (1984–2021) from post-processed satellite retrievals, which provided an historical baseline. ERA5-Land reanalysis data were used to characterize climate conditions, including precipitation for the calculation of Precipitation Index SPI (3/6/12 months), temperature anomalies, a heat-ratio metric (share of days with daily Tmax above the historical 90^th percentile) and snow cover fraction for relevant mountainous headwaters.

The above methodology was applied for two water-supply systems under clear “emergency” pressure: the Attica system, where Mornos is the main source and Evinos supports it via transfer, and the Aposelemis system in Crete, which also depends on inflows linked to the Lasithi area. During 2024-2025 Attica experienced persistently low reservoir levels, with 2025 being among the lowest conditions since the Evinos reservoir was integrated and broadly comparable to the 2007–2008 major drought. In 2025, the Mornos reservoir declined from ~65% of its historical maximum extent in May to ~51% by September, marking the lowest levels recorded in the past two decades, despite limited meteorological relief during winter 2024/25. Evinos showed stronger monthly fluctuations, with values in the most stressed months commonly around ~60% of seasonal maxima. In Crete, Aposelemis shifted from high reservoir capacity during 2019–2022 (often ~80–90% of maximum extent) to a prolonged decline after 2023, reaching approximately one-third of maximum reservoir coverageduring 2025. This evolution is consistent with persistent precipitation deficits and increased heat stress across the region.

The integrated EO–reanalysis assessment showed that drops in reservoir levels often follow meteorological drought indicators with a delay of months to even years, highlighting the need for continuous monitoring. Using Google Earth Engine and open satellite and reanalysis data, a scalable open-data pipeline was developed for near-real-time drought tracking and water-resource awareness, supporting proactive drought management in Greece and other Mediterranean basins.

How to cite: Konis, A., Askitopoulos, A., Pagana, V., and Kontoes, C. (.: Diachronic drought assessment of Greek water-supply reservoirs using open EO and reanalysis data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7136, https://doi.org/10.5194/egusphere-egu26-7136, 2026.

EGU26-7136

Download Close

With droughts increasing in frequency and severity worldwide, reliable monitoring and forecasting systems, along with transparent accuracy assessment, are crucial for effective drought management and decision-making. Here, we evaluate the performance of three drought hazard indicators (DHIs) provided by the global, multi-sectoral drought hazard monitoring and forecasting system that has been developed within the OUTLAST project and is available via the WMO’s HydroSOS website. In OUTLAST, a consistent framework is applied to produce sector-specific DHIs for global monitoring and seasonal forecasts of droughts. To do so, climate data from ERA5 (for monitoring) and bias-corrected SEAS5 (for seasonal forecasts) are used to calculate meteorological DHIs as well as to force the Global Crop Water Model and the global hydrological model WaterGAPto derive agricultural and hydrological DHIs, respectively.

This study aims to assess the performance of three DHIs from multiple sectors, including (1) the standard precipitation index (SPI), (2) the rainfed crop drought hazard indicator (RFCDI), and (3) the empirical percentiles of streamflow (Q-EP), in an informative and user-friendly way. This is done by (a) a comprehensive comparison of OUTLAST DHIs against the same DHIs calculated with independent, preferably observation-based data, such as (1) remote sensing-based precipitation, (2) remote sensing-based actual and potential evapotranspiration, and (3) in-situ observed streamflow of large river basins, all for the historic period 1981-2020; and (b) a detailed evaluation of the capability of two example seasonal forecasts, issued in March 2018 and March 2022, to predict Northern Hemisphere spring and summer droughts across sectors. For each DHI, four drought classes are defined, with drought conditions being identified by a return period of at least five years.

For the historic period, the derived drought classes agree in about 50% of drought months globally (Q-EP: 49%, RFCDI: 51%), with higher agreement in the case of SPI (59%). The agreement is in general highest in temperate and cold climate zones, except for RFCDI, which performs best in arid regions (61%), where Q-EP only has a small agreement with in-situ streamflow droughts (36%). SPI has the lowest agreement in tropical regions (44%), where the agreement of RFCDI and Q-EP is slightly higher (46% resp. 47%). This low agreement of OUTLAST-SPI with remote sensing-based SPI reflects the known high uncertainties of ERA5 precipitation (which is used in OUTLAST) in the tropics, that partly propagates to modelled RFDCI and Q-EP. Differences between different DHIs and climate zones reflect the uncertainties and limitations of both the individual models used to compute the OUTLAST DHIs and the independent data sets used for comparison. At the same time, the consistent framework to produce multi-sectoral DHIs allows to analyze the effect of drought- and error-propagation in the hydrological cycle on the ability to capture observed drought conditions by model-based DHIs.

The results of these comparisons will be provided to the users of the OUTLAST drought hazard monitoring and forecasting system, and by that support informed drought management and decision-making across multiple sectors worldwide.

How to cite: Trautmann, T., Abbasi, N., Weber, J., Vu, T., Dietrich, S., Döll, P., Kunstmann, H., Lorenz, C., and Siebert, S.: Assessing the accuracy of multi-sectoral drought hazard indicators from the OUTLAST drought monitoring and seasonal forecasting system at the global scale, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11664, https://doi.org/10.5194/egusphere-egu26-11664, 2026.

EGU26-11664

Download Close

This study presents a data-driven framework for identifying drought-affected paddy rice fields associated with actual agricultural drought events in South Korea. The proposed approach examines spatiotemporal patterns of multiple vegetation- and moisture-related indices derived from high-resolution satellite observations to distinguish paddy fields experiencing water stress from normal growing conditions. Spectral–temporal characteristics of paddy fields and barren land are analyzed to detect paddy pixels exhibiting barren-like behavior during drought periods. The framework is demonstrated over Chungcheongnam-do, a major agricultural region where severe water shortages in paddy fields were reported during recent drought events. A Long Short-Term Memory (LSTM) model is employed to capture temporal dependencies in vegetation dynamics. Satellite observations from non-drought years are used for model training and validation, and the trained model is subsequently applied to drought years to identify anomalous paddy field responses. Drought-affected paddy areas are delineated based on the persistence and duration of barren-like conditions relative to the crop phenological cycle. To enhance interpretability, permutation-based feature importance analysis is conducted to assess the contribution of individual indices and to identify those most effective in distinguishing drought-affected conditions. By establishing quantitative criteria for delineating previously ambiguous drought-impacted paddy areas, the proposed framework provides a basis for improved assessment of agricultural drought impacts and supports more robust monitoring of crop stress under variable hydroclimatic conditions.

Keywords: Agricultural Drought, Paddy Rice Fields, Vegetation Dynamics, Satellite Remote Sensing, Data-driven Framework

Acknowledgement

This work was supported by the Korea Environmental Industry & Technology Institute (KEITI) through Water Management Program for Drought, funded by the Korea Ministry of Climate, Energy and Environment (MCEE). (RS-2022-KE002032) and was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (RS-2024-00334564). Also, This research was supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education(RS-2024-00356439) and was supported by the National Research Foundation of Korea (NRF) (RS-2021-NR060085) funded by the Korea government (MSIT).

How to cite: Kim, H., Baik, J., Cha, H., Park, K., Ku, S., and Jun, C.: Identifying drought-affected paddy rice fields using satellite-based temporal vegetation dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15101, https://doi.org/10.5194/egusphere-egu26-15101, 2026.

EGU26-15101

Download Close

Drought monitoring systems often rely on multiple standardized indices computed at fixed time scales, leaving end users with fragmented information and weak links to actual impacts. Here we present Drought Scan (DS), an operational drought monitoring and forecasting system designed to provide a synoptic, impact-oriented view of drought at the river-basin scale.

DS is entirely driven by basin-aggregated monthly precipitation and builds on a continuous multi-scale representation of standardized precipitation anomalies (SPI from 1 to 36 months). The core of the system is a synthetic indicator, D_(SPI), obtained through a weighted aggregation of multi-scale SPI values. The weighting scheme is optimized against observed river discharge, maximizing the correlation between D_(SPI) and standardized monthly streamflow (SQI1). As a result, unlike conventional indices, D_(SPI) acts as a proxy of hydrological stress, despite being derived solely from precipitation. This makes the indicator explicitly impact-oriented and directly interpretable in terms of water availability.

The system integrates three complementary components: (i) a multi-scale SPI heatmap that reveals drought triggers, persistence, and propagation across temporal scales; (ii) the D_(SPI) indicator, which condenses this information into a single, basin-specific drought signal calibrated on hydrological response; and (iii) the cumulative deviation from normal (CDN), which captures the long-term memory of wet and dry phases and contextualizes drought severity within multi-year precipitation regimes.

By construction, DS bridges the meteorological–hydrological continuum without relying on hydrological modeling or extensive ancillary data. Once an impact-oriented indicator is defined from precipitation alone, the system naturally lends itself to be applied into forecast estimates at sub-seasonal and seasonal scales: projected precipitation can be propagated through the same framework to obtain forecasts of D_(SPI), i.e. forecasts of drought conditions expressed in terms of expected hydrological stress. Different forecasting approaches can be adopted (numerical such as those provided by Copernicus Climate Change Service or those estimated by machine learning algorithms), but the emphasis remains on the indicator and its interpretability rather than on the predictive technique itself. To facilitate this interpretation, forecasts are coupled with probabilistic scenarios that also can allow the quantification of rainfall needed to recover from drought phases.

DS is conceived as a climate service tool developed within the Drought Central framework (www.droughtcentral.it), suitable for monitoring, early warning, and scenario exploration, and designed to translate complex drought dynamics into information that is robust, transparent, and operationally meaningful for water management and decision-making.

How to cite: Di Paola, A., Magno, R., Di Giuseppe, E., Quaresima, S., Rocchi, L., and Pasqui, M.: Drought Scan: an impact-oriented drought monitoring system bridging precipitation and hydrological response, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17298, https://doi.org/10.5194/egusphere-egu26-17298, 2026.

EGU26-17298

Download Close

Vapour pressure deficit (VPD) is a key indicator of atmospheric dryness, plant water stress, stomatal conductance, and crop productivity. Under climate change, rising air temperatures increase the capacity of the atmosphere to hold water vapor, leading to higher VPD even in regions where precipitation has not declined.  Atmospheric drought is therefore an important but still underrepresented component of drought risk assessments, which have traditionally focused on precipitation and soil moisture alone. In Central Europe, recent heatwaves and drought events have caused substantial agricultural and ecological impacts, but the long-term behaviour of VPD and its interaction with soil moisture remain not fully clarified.

The objective of this study is to assess long-term changes in atmospheric drought, evaluating the reliability of reanalysis-based VPD, and quantifying the coupling between atmospheric conditions, soil moisture, and agricultural productivity in Czechia. The results will support improved drought monitoring and impact assessment in the context of ongoing climate change.

This study analyses VPD from station observations and reanalysis data in Czechia for 50 years (1975-2024), together with soil moisture data from reanalysis and annual crop yield data. The performance of reanalysis-based VPD is evaluated against station observations using bias, root-mean-square error, correlation, and their ability to reproduce observed extreme VPD events. This comparison assesses whether reanalysis data are suitable for studying atmospheric drought and extremes at regional scale. Long-term changes in VPD and soil moisture are evaluated using non-parametric trend methods. Analyses are performed for annual and growing-season means as well as for drought-relevant metrics, including maximum VPD and the annual number of extreme VPD days. The relationship between atmospheric and soil drought is investigated across daily to monthly time scales. Impacts on agriculture are assessed by relating annual crop yields to growing-season VPD and soil moisture.

How to cite: Kalmár, T. and Beranová, R.: Atmospheric Drought under Climate Change: Vapour Pressure Deficit Trends and Impacts in Czechia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17850, https://doi.org/10.5194/egusphere-egu26-17850, 2026.

EGU26-17850

Download Close