HS4.2

Droughts can seriously challenge water management if they have large spatial extents. These extents may change in a warming climate along with changes in underlying hydro-meteorological drivers. Therefore, we ask (1) how streamflow drought spatial extent has changed over the period 1981-2018 in the United States, (2) which physical drivers govern drought spatial extent, and (3) whether/how the importance of these drivers has changed over time. We analyze temporal changes in streamflow drought extents and their drivers using drought events extracted for 671 catchments in the conterminous United States using a variable threshold-level approach. Drought spatial extents are determined as the percentage of catchments affected by drought during a certain month. Then, important drivers are identified by determining the spatial percentage overlap of the area under streamflow drought with precipitation droughts, temperature anomalies, snow-water-equivalent deficits, and soil moisture deficits. Finally, the spatial extent and overlap time series are used in a trend analysis to determine changes in drought spatial extent and to identify changes in the importance of different variables as drivers of drought spatial extent. Our analyses show that (1) drought spatial extents have increased, mainly because of increases in the extent of small droughts; (2) drought extents overall substantially overlap with soil moisture deficits and the relationship of drought to precipitation and temperature varies seasonally; (3) the importance of temperature as a driver of drought extent has increased over time. We therefore conclude that continued global warming may further increase the probability of spatially compounding drought events, which requires adaptation of regional drought management strategies.

How to cite: Brunner, M. I., Swain, D. L., Gilleland, E., and Wood, A. W.: Spatial extent of hydrological drought in the United States: changes and hydro-meteorological drivers, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-374, https://doi.org/10.5194/egusphere-egu21-374, 2021.

Seasonal water supply predictions offer critical information to aid in planning and mitigation of drought impacts. In many northern and montane systems, spring snow information has been shown to be the most important predictor of seasonal drought, since in these systems snow water storage can exceed that of man-made reservoirs. However, a warmer future portends for less precipitation falling in the form of snow, which challenges current prediction methods. This presentation focuses on evaluating physical and statistical techniques for seasonal water supply prediction in snow-fed systems under both historical and future climate conditions with the goal of identifying regions and methods where predictions are likely to remain skillful under future warming. Initial results using downscaled hydrologic projections over the western U.S. indicate that snow information contributes less predictive skill to drought forecasts over roughly two thirds of snow dominated regions by the middle of this century. Significantly greater resilience to warming is seen higher elevation zones (p<0.01) and for prediction methods that include non-snow predictors such as soil moisture. To understand the impact of non-stationary snow conditions on future drought predictions, we conduct a series of idealized experiments to evaluate the relative importance of secular trends versus changing variability of both snow and seasonal climate conditions. This presentation is part of a larger research effort seeking to identify alternatives to snow-based streamflow predictions to advance future drought predictability.

How to cite: Livneh, B., Modi, P., Kasprzyk, J., Ely, B., and Duncan, B.: Evaluating seasonal drought prediction in snow-fed systems past, present, and future: towards identifying resilient prediction techniques, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-1786, https://doi.org/10.5194/egusphere-egu21-1786, 2021.

Regional climate models (RCMs) are commonly used for assessing, at proper spatial resolutions, future impacts of climate change on hydrological events. In this study, we propose a statistical methodological framework to assess the quality of the EURO-CORDEX RCMs concerning their ability to simulate historic observed climate (temperature and precipitation). We specifically focus on the models’ performance in reproducing drought characteristics (duration, accumulated deficit, intensity, and return period) determined by the theory of runs at seasonal and annual timescales, by comparison with high-density and high-quality ground-based observational datasets. In particular, the proposed methodology is applied to the Sicily and Calabria regions (Southern Italy), where long historical precipitation and temperature series were recorded by the ground-based monitoring networks operated by the former Regional Hydrographic Offices. The density of the measurements is considerably greater than observational gridded datasets available at the European level, such as E-OBS or CRU-TS. Results show that among the models based on the combination of the HadGEM2 global circulation model (GCM) with the CLM-Community RCMs are the most skillful in reproducing precipitation and temperature variability as well as drought characteristics. Nevertheless, the ranking of the models may slightly change depending on the specific variable analysed, as well as the temporal and spatial scale of interest. From this point of view, the proposed methodology highlights the skills and weaknesses of the different configurations, aiding on the selection of the most suitable climate model for assessing climate change impacts on drought processes and the underlying variables.

How to cite: Peres, D. J., Senatore, A., Nanni, P., Cancelliere, A., Mendicino, G., and Bonaccorso, B.: A statistical framework for evaluating EURO-CORDEX simulations and derived drought characteristics, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-2330, https://doi.org/10.5194/egusphere-egu21-2330, 2021.

During the last few decades, the frequency of drought has significantly increased in Morocco especially for arid and semi-arid regions, leading to a rising of several environmental and economic issues. In this work, we analyse the spatial and temporal relationship between vegetation activity and drought severity at different moments of the year, across an arid area in the western Haouz plain in Morocco. Our approach is based on the use of a set of more than thirty satellite Landsat images data acquired for the period from 2008 to 2017, combined with the Standardized Precipitation Index (SPI) at different time scales and Standardized water-level Index (SWI). The Mann-Kendall and Sen’s slopes methods were used to estimate SPI trends and the Pearson correlation between NDVI and SPI were calculated to assess the sensitivity of vegetation types to drought. Results demonstrated that SPI experienced an overall upward trend in the Chichaoua-Mejjate region, except for 3-months time scale SPI in summer. The vegetation activity is largely controlled by the drought with clear differences between seasons and timesclaes at which drought is assessed. Positives correlations between the NDVI and SPI are dominant across the entire study area except in June when almost half of correlations is negative. More than 80% of the study domain exhibit a correlation exceeding 0.4 for SPI3 and SPI6 in March. Importantly, this study stresses that the irrigation status of land can introduce some uncertainties on the remote sensing drought monitoring. A weak correlation between the SPI and the SWI was observed at different time-scale. The fluctuations of the piezometric levels are strongly affected by the anthropogenic overexploitation of aquifers and proliferation of irrigated plots.

How to cite: Abdessamad, H., Mohamed El Mehdi, S., and Abdelghani, B.: Monitoring multiscale drought using remote sensing in a Mediterranean arid region., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3389, https://doi.org/10.5194/egusphere-egu21-3389, 2021.

How to cite: Romanowicz, R., Karamuz, E., Napiorkowski, J., and Senbeta, T.: Effects of climate change and human interactions on water balance dynamics in the River Vistula basin, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-5757, https://doi.org/10.5194/egusphere-egu21-5757, 2021.

Soil Moisture (SM) is one of the key observable variables of the hydrological cycle and therefore of high importance for many disciplines, from meteorology to agriculture. This contribution presents a comparison of different products for the mapping of SM. The aim was to identify the best available solution for the operational monitoring of SM as a drought indicator for the entire area of the European Alps, to be applied in the context of the Interreg Alpine Space project, the Alpine Drought Observatory.

The following datasets were considered: Soil Water Index (SWI) of the Copernicus Global Land Service [1]; ERA5 [2]; ERA5-Land [3]; UERRA MESCAN-SURFEX land-surface component [4]. All four datasets offer a different set of advantages and disadvantages related to their spatial resolution, update frequency and latency. As a reference, modelled SM time-series for 307 catchments in Switzerland were used [5]. Switzerland is well suited as a test case for the Alps, due to its different landscapes, from lowlands to high mountain.

The intercomparison was based on a correlation analysis of daily absolute SM values and the daily anomalies. Furthermore, the probability to detect certain events, such as persistent dry conditions, was evaluated for each of the SM datasets. First results showed that the temporal dynamics (both in terms of absolute values as well as anomalies) of the re-analysis datasets show a high correlation to the reference. A clear gradient, from the lowlands in the north to the high mountains in the south, with decreasing correlation is evident. The SWI data showed weak correlations to the temporal dynamics of the reference in general. Especially, during spring and the first part of the summer SM is significantly underestimated. This might be related to the influence of snowmelt, which is not taken into account in the two-layer water balance model used to model SM for deeper soil layers. Low coverage in the high mountain areas hampered a thorough comparison with the reference in these areas.

The results presented here are the foundation for selecting a suitable source for the operational mapping of SM for the Alpine Drought Observatory. The next steps will be to test the potential of MESCAN-SURFEX and ERA5-Land for the downscaling of ERA5 to take advantage of the low latency of ERA5 and the improved spatial detail of the other two datasets.  

Literature:

[1]  B. Bauer-marschallinger et al., “Sentinel-1 : Harnessing Assets and Overcoming Obstacles,” IEEE Trans. Geosci. Remote Sens., vol. 57, no. 1, pp. 520–539, 2019, doi: 10.1109/TGRS.2018.2858004.

[2]  H. Hersbach et al., “ERA5 hourly data on single levels from 1979 to present.” Copernicus Climate Change Service (C3S) Climate Data Store (CDS), 2018.

[3]  Copernicus Climate Change Service, “ERA5-Land hourly data from 2001 to present.” ECMWF, 2019, doi: 10.24381/CDS.E2161BAC.

[4]  E. Bazile, et al., “MESCAN-SURFEX Surface Analysis. Deliverable D2.8 of the UERRA Project,” 2017. Accessed: Jan. 11, 2020. [Online]. Available: http://www.uerra.eu/publications/deliverable-reports.html.

[5]  Brunner, et al.: Extremeness of recent drought events in    Switzerland: dependence on variable and return period choice, Nat. Hazards Earth Syst. Sci., 19, 2311–2323, https://doi.org/10.5194/nhess-19-2311-2019, 2019.

How to cite: Greifeneder, F., Haslinger, K., Seyerl, G., Notarnicola, C., Zappa, M., and Zebisch, M.: Assessing the options for the operational mapping of the soil moisture content in the European Alps, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7445, https://doi.org/10.5194/egusphere-egu21-7445, 2021.

Global warming and anthropogenic activities have significantly altered the hydrological cycle and amplified the extreme events (floods and droughts) in many regions of the world, with associated environmental, economic, and social losses. For effective hydro extremes hazards management, it is significant to understand how climate change influences the occurrence, duration, and severity of the regional dryness/wetness conditions (droughts/floods). The present study was carried out over Upper Jhelum Basin (UJB) in Pakistan which lies in the western Himalaya, a most effected mountainous range by Climate Change. Firstly, a suitable gridded precipitation dataset was selected/chosen among various datasets (APHRODITE, CHIRPS, ERA5, PGMFD, MSWEP) through spatio-temporal comparison against in situ data at monthly, seasonal, and annual scale. Secondly, selected gridded data was adjusted for biases using linear (Linear scaling-LS, Local intensity scaling-LOCI) and nonlinear (Power transformation-PT and Distribution mapping-DM) statistical methods. Finally, standardized precipitation index (SPI) at multiple time scale was used to analyses dryness/wetness conditions in the Upper Jhelum Basin over a 35-year period (1981–2015). Results show the higher capability of ERA5 data to represent the UJB precipitation patterns with correlation coefficient (r=0.79) and normalized standard deviation (nSD=1.1), despite of overestimation especially during peak months. Regarding precipitation bias adjustment, all methods were able to correct the mean values while LOCI and DM take advantage over other two methods to correct wet-day probability and precipitation intensity. The SPI analysis at different time scales showed that wet periods occurred more in the first half of the study period, but at later years, drying periods ranging from moderate to severe continue to be seen with increasing frequency. A strong change in dry/wet conditions was observed around years 1997/1998. This change may be the result of the strongest El Nino event (1997-98) occurred in the history. However, further studies are still needed to check whether there is only a large multi-decadal variation or dry conditions will prevail in future. Overall, these findings would assist to better understand the changing pattern of extreme events with climate variability and help water resources managers to develop basin wide appropriate mitigation and adaptation measures to combat climate change and its consequences. 

How to cite: Ansari, R. and Grossi, G.: Evolution of Dryness/Wetness conditions and their characteristics (duration and severity) across Upper Jhelum Basin, Pakistan, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-12550, https://doi.org/10.5194/egusphere-egu21-12550, 2021.

Snow plays a significant role in surface runoff, groundwater resources, and as an important temporary reservoir for winter precipitation. On the other hand, extreme floods can arise when high melt rates in catchments zone combine with torrential rain at the same time, therefore snowmelt quantities are important for the management of lakes.

This study aims at investigating snow drought characteristics in the catchment area of Lake Urmia, which has recently been faced with the issue of drought and declining water levels. To provide an overview of drought intensity for the last 40 years 1981-2020, the Standardized Snow Melt and Rain Index (SMRI), which accounts for rain and snowmelt deficits, was applied and spatial variations of snow drought were assessed. The index was used in drought analysis based on the ERA5 dataset for the whole study area under three-time scales including the 3-, 6- and 12-month. After determining the dry and wet periods, historical characteristics of droughts were identified, and spatial distribution maps of droughts were plotted.

Results show that during the last years, snow drought events were more frequent, severe, and affected a larger area which shows a spatial spread of drought events. According to the snow index results, most extreme events have happened in the Zarine Rud and Simine Rud sub-catchments which play a key role in increasing the groundwater resources of the Basin. With proper management, these resources can be properly used for lake revitalization.

Keywords: Snow droughts, SMRI, Drought characteristics, Urmia lake, Climate extreme

How to cite: Habibi, M., Schöner, W., and Babaeian, I.: Spatial and temporal assessment of Snow drought characteristic for 1981 to 2020 on Urmia lake catchment scale, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15208, https://doi.org/10.5194/egusphere-egu21-15208, 2021.

How to cite: Poppe Teran, C., Naz, B., Baatz, R., Hendricks-Franssen, H.-J., Nikolaidis, N., and Vereecken, H.: The Effect of Droughts on Ecosystem Water-Use-Efficiency in Europe, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-9426, https://doi.org/10.5194/egusphere-egu21-9426, 2021.

Droughts are considered as one of the most catastrophic natural disasters that affect humans and their surroundings at a larger spatial scale compared to other disasters. Rajasthan, one of India's semiarid states, is drought inclined and has experienced many drought events in the past. In this study, we evaluated different preprocessing and Machine Learning (ML) approaches for drought predictions in Rajasthan for a lead-time of up to 6 months. The Standardized Precipitation Index (SPI) was used as the drought quantifying measure to identify the drought events. SPI was calculated for 3, 6, and 12-month timescales over the last 115-year using monthly rainfall data at 119 grid stations.  ML techniques, namely Artificial Neural Network (ANN), Support Vector Regression (SVR), and Linear Regression (LR), were used to evaluate their accuracy in drought forecasting over different lead times. Furthermore, two data processing methods, namely the Wavelet Packet Transform (WPT) and Discrete Wavelet Transform (DWT), have also been used to enhance the aforementioned ML models' predictability. At the outset, the preprocessed SPI data from both the methods were used as inputs for LR, SVR, and ANN to form a hybrid model. The hybrid models' drought predictability for a different lead-time was evaluated and compared with the standalone ML models. The forecasting performance of all the models for all 119 grid points was assessed with three statistical indices: Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Nash-Sutcliffe Efficiency (NSE). RMSE was used to select the optimal model parameters, such as the number of hidden neurons and the number of inputs in ANN, and the level of decomposition and mother wavelet in wavelet analysis.  Based on these measures, the coupled model showed better forecasting performance than the standalone ML models. The coupled WPT-ANN model shows superior predictability for most of the grid points than other coupled models and standalone models.  All models' performance improved as the timescale increased from 3 to 12 months for all the lead times. However, the model performance decreased as the lead time increased.  These findings indicate the necessity of processing the data before the application of any machine learning technique. The hybrid model's prediction performance also shows that it can be used for drought early warning systems in the state.

How to cite: Hinge, G. and Sharma, A.: Comparison of wavelet and machine learning methods for regional drought prediction, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-218, https://doi.org/10.5194/egusphere-egu21-218, 2021.

West Africa (WA) is prone to food insecurity due to climate-, economic-, conflict-related shocks, as well as high population growth and lack of proper adaptation strategies. As per the USAID’s Famine Early Warning Systems Network, which uses Integrated Phase Classification to classify acute food insecurity (AFI), between 2011 and early 2020, several parts of WA reported the “Stressed” phase of AFI >30% of the time. Food security and livelihood in the region relies substantially on rainfed farming and small-scale water holes. Droughts lead to water deficits resulting in adverse impacts on food production, human and livestock health and agricultural labor opportunities, leading to or worsening of food insecurity. Thus far, the focus of climate, drought outlooks and their impacts, to support food insecurity early warning in this region has mainly been on the seasonal scale (i.e., 3-6 months in future) forecasts whereas use of subseasonal scale (2-4 weeks in future) forecasts has been negligible. Recent advances in routine production (i.e. weekly) and open access to subseasonal forecasts provide an unprecedented opportunity to improve the existing climate services in the region by focusing on the impacts of subseasonal climate characteristics on food insecurity in the region. Here we report on an ongoing project with the AGRiculture HYdrology and METeorology Regional Centre (SERVIR’s WA Hub) that aims to develop a subseasonal water deficit forecasting system to support food insecurity early warning in the region. The presentation will describe  (i) the results of an ongoing analysis examining the influence of subseasonal climate characteristics (e.g. monthly climate variability, length of dry or wet spell) on food insecurity, as measures by different food insecurity indicators (such as vegetation index, food insecurity reports and household level health and malnutrition reports) and (ii) the major accomplishments towards implementation of the water deficit forecasting system, including development and evaluation of prototype products, (iii) capacity building and stakeholder engagement activities with National Meteorological and Hydrological Services across the region. 

How to cite: Shukla, S., Grace, K., Ali, A., McEvoy, D., Turnet, W., Alkhalil, A., Tinni, S., Lona, I., Sanda, I., Cherrington, E. A., Muench, R., and Husak, G.: Forecasting agropastoral water deficits in West Africa to support food insecurity early warning, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3255, https://doi.org/10.5194/egusphere-egu21-3255, 2021.

Drought is one of the most critical hydrometeorological phenomena in terms of impacts to society because it affects soil water content, and consequently, crop production and human diets, in some cases under critical conditions, drought produces starving and people migration. Although Colombia is a tropical country, there are areas of the territory that have periods of drought that cause important economic damages such as fires, death loss in cattle, reduction of the capacity to supply water to persons, impacts to agriculture and fish farming.

Due to recent advances in terms of spatial and temporal resolutions of remote sensing and Artificial Intelligence techniques, it is possible to develop Automatic Learning Models supported on historic information. In this research was built a classifier  Random Forest (RF) and Bagged Decision Tree Classifier (DTC) model to predict, spatial and temporal drought occurrence in Colombia, using remote sensing data as land surface temperature, precipitation, soil water contentl, and evapotranspiration, and macro climatic variables information as ONI, MEI and SOI.  It was used the Standardized Precipitation Index (SPI) with 3-month time scale, that allows identifying agricultural drought events. The results showed that Random Forest provides the best outcomes. In terms of recall and precision, RF produced 0.84 and 0.59 and DTC brought a 0.8 and 0.33, respectively, to predict drought. The above, evidence that models could overestimate the number of times where drought occurs, in contrast with normal or humid conditions. On the other hand, False Positive and False Negative rates are important facts for measuring the development of models. In this case, the FP and FN rates are 7.5% and 2% for RF and 21% and 2.5% for DTC respectively, that means that both models made fewer mistakes predicted real drought events, but had more errors forecasting real normal or humid condition, especially, DTC model. RF can provide a better performance predicting drought and normal/humid conditions in contrast with DTC. The implementation of the developed model can allow governmental entities assessment and monitor agricultural drought over time. Taking, in consequence, actions to mitigate the impacts of droughts in their territories.

How to cite: Herrera, D. and Aristizábal, E.: Artificial Intelligence and machine learning model for spatial and temporal prediction of drought in the Colombia Caribbean region., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3599, https://doi.org/10.5194/egusphere-egu21-3599, 2021.

Streamflow drought forecasting is a key element of contemporary Drought Early Warning Systems (DEWS). The term streamflow drought forecasting, rather than streamflow forecasting, however, has created confusion within the scientific hydro-meteorological community, as well as in operational weather and water management services. Streamflow drought forecasting requires an additional step, which is the application of a drought identification method to the forecasted streamflow time series. The way, how streamflow drought is defined, is the main reason for this misperception. The purpose of this study, therefore, is to provide a comprehensive overview of the application of different drought identification approaches to forecast streamflow drought, incl. its characteristics, such as drought occurrence, timing, duration, and deficit volume, across the pan-European river network and for the Rhine River in more detail. In this study, the implications of different approaches for forecasting streamflow drought are elaborated using the extreme 2003 drought in Europe, as an example. The forecasted 25 ensemble streamflow data with 7-month lead time (LT) were obtained from the LISFLOOD hydrological model fed with seasonal meteorological forecasts from the European Centre for Medium-range Weather Forecasts system 5 (ECMWF SEAS 5). Streamflow droughts were analyzed using the daily and monthly Variable Threshold methods (VTD and VTM), daily and monthly Fixed Threshold methods (FTD and FTM), and the Standardized Streamflow Index with 1-month accumulation period (SSI-1). Our results clearly show that streamflow drought characteristics derived with different approaches deviate, which are partly associated with different climate regions across Europe. Using the forecasts initiated in July 2003 for LT=7-month, first, the daily drought approaches forecast more drought events than the monthly approaches. Second, the VT droughts (VTD and VTM), incl. SSI-1 forecast a lower number of drought occurrences than the FT droughts (FTD and FTM), which highlights the importance of taking seasonality into account. Overall, the FT approaches predict a longer drought duration, earlier drought timing, and higher drought deficit volume in many European rivers than the VT approaches. The characteristics of SSI-1 drought, in general, are close to what is being identified by the VTM approach. A detailed analysis of the drought forecasts for the Rhine River indicates that the number of drought events derived from the median of ensemble members can be predicted relatively well, but with lower skill for other drought characteristics. The use of monthly-aggregated forecasted flow data (e.g. VTM, FTM, and SSI) seems to be the best practice for seasonal drought forecasts because it will alleviate the drought forecast skill. The monthly drought threshold approaches, however, will forecast higher drought duration and deficit volume than using daily datasets. The choice of the drought identification method when forecasting streamflow drought, ultimately rests with the end-users and we need to realize that there is no one drought identification approach that fits all needs.

How to cite: Sutanto, S. J. and Van Lanen, H. A. J.: Application of multi drought definitions to forecast streamflow drought across Europe, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3952, https://doi.org/10.5194/egusphere-egu21-3952, 2021.

Drought propagation processes interlink closely with the water cycle, which has so far been mostly investigated without tracking temporal propagation across multiple types of drought. Central Asia, including Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan, and Xinjiang (China) areas, is one of the most drought-prone areas in the world and extremely vulnerable to water scarcity. Understanding the multiple pathways of drought propagation over time in Central Asia is necessary for food security, human health, and poverty alleviation. In this study, we quantify the propagation time and track the details of temporal propagation processes of drought across the atmosphere, the geosphere, and the hydrosphere. The standardized indices are calculated using variables directly related to each type of drought: precipitation (SPI), evapotranspiration (SEDI), soil moisture (SSI), and runoff (SRI)^{1, 2}. The drought propagation processes are divided into the development stage and recovery stage. The propagation time at different stages between multiple types of drought is calculated by the Pearson correlation coefficient (p<0.05)³ and run theory method⁴. Besides, the potential influencing factors on the temporal propagation are explored from the meteorological, land cover, and water management aspects. As the main results, the propagation time in winter is longer than in summer. And topography has a significant impact on drought propagation time. These key findings could further benefit the early warning of drought and facilitate the drought mitigation-adaptation in both Central Asia and other continents.

Keywords: Temporal propagation, Central Asia, Influencing factors, Land cover

1. Gevaert, A. I.; Veldkamp, T. I. E.; Ward, P. J., The effect of climate type on timescales of drought propagation in an ensemble of global hydrological models. Hydrol. Earth Syst. Sci. 2018, 22, (9), 4649-4665.
2. Barella-Ortiz, A.; Quintana-Segui, P., Evaluation of drought representation and propagation in regional climate model simulations across Spain. Hydrol. Earth Syst. Sci. 2019, 23, (12), 5111-5131.
3. Barker, L. J.; Hannaford, J.; Chiverton, A.; Svensson, C., From meteorological to hydrological drought using standardised indicators. Hydrol. Earth Syst. Sci. 2016, 20, (6), 2483-2505.
4. Wu, J. F.; Chen, X. H.; Yao, H. X.; Liu, Z. Y.; Zhang, D. J., Hydrological Drought Instantaneous Propagation Speed Based on the Variable Motion Relationship of Speed-Time Process. Water Resour. Res. 2018, 54, (11), 9549-9565.

How to cite: Tian, L., Ho, S., Disse, M., and Tuo, Y.: The Temporal Propagation Processes of Multiple Types of Drought in Central Asia, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-6499, https://doi.org/10.5194/egusphere-egu21-6499, 2021.

An increasing number of regions and countries are confronted with droughts as well as an increase in water demand. Inevitably, this leads to an increasing pressure on the available water resources and associated risks and economic impact for the water dependent sectors. In order to prevent big drought impacts, such as agricultural damage and food insecurity, timely and focused drought mitigation measures need to be carried out. To enable this, the detection of drought and its sector-specific risks at early stages needs to be improved. One of the main challenges is to develop compound and impact-oriented drought indices, that make optimal use of innovative techniques, satellite products, local data and other big data sets.

Here, we present the development of a Next Generation Drought Index (NGDI) that combines multiple freely available global data sources (eg. ERA5, MODIS, PCR-GLOBWB) to calculate a range of relevant drought hazard indices related to meteorological, hydrological, soil moisture and agricultural drought (eg. SPI, SPEI, SRI, SGI, VCI). The drought hazard indices are aggregated at district level, while considering the percentage area exposure of the drought impacted sector (exposure). In addition, the indices are enriched with local and national scale drought impact information (eg. online news items, social media data, EM-DAT database, GDO Drought news, national drought reports). Results are presented at sub-national scales in interactive spatial and temporal views, showing the combined drought indices and impact data.

The NGDI approach is being tested for the agricultural sector in Mali, a country with a vulnerable population and economy that faces frequent dry spells which heavily impact the functioning of the important agricultural activities that sustain a large part of the population. The computed drought indices are compared with local drought data and an analysis is made of the cross-correlations between the indices within the NGDI and collected impact data.

We aim at providing the NGDI information to a broad audience as well as co-creation of further NGDI developments. Hence, we would like to reach out to interested parties and identify collaboration opportunities.

How to cite: Hendriks, D., Hazenberg, P., Gotte, J., Trambauer, P., Haag, A., Donchyts, G., and Sperna Weiland, F.: Development of a Next Generation Drought Index: combining multiple global and local data sources to enable detection of sector-specific drought impacts, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10886, https://doi.org/10.5194/egusphere-egu21-10886, 2021.

Artificial reservoirs are one of the main water supply resources in the Mediterranean areas; their management can be strongly affected by the problems of drought and water scarcity. The reservoir water level is the result of the hydrological processes occurring in the upstream catchment, which, in turn, depend on meteorological variables, such as rainfall and temperature. It follows that a reliable forecast model of the meteorological forcing, along with a reliable water balance model, could enhance the correct management of a reservoir. With regard to the rainfall/temperature forecast model, the use of forecast climate data in the mid-term may provide further support for the future water level estimation of reservoirs.

From the perspective of the water balance model, instead, among the approaches used to predict the water levels for the next future, those based on data-driven methods have been demonstrated to be particularly capable of correctly reproducing the correlation between a dependent variable (e.g., water level, volume) and some covariates (e.g., temperature, precipitation).

This study describes the preliminary results of a novel application that exploits the Seasonal Forecast (SF) data, produced at the European Centre for Medium-Range Weather Forecasting (ECMWF), within a data-driven model aimed to predict the reservoir water volume at mid-term scale, up to 6 months ahead in four reservoirs of the Sicily (Italy) here considered as a case study. For each case, a NARX (Nonlinear AutoRegressive network with eXogenous inputs) neural network is calibrated to reproduce the monthly stored water volume starting from the monthly precipitation and mean monthly air temperature variables.

Preliminary results showed that the NARXs have the capability to reproduce the water levels in the investigated period (January 2017 - April 2020), including the variations during more or less dry periods. All this despite the SF data have not been previously treated with downscaling and/or bias correction techniques.

How to cite: Francipane, A., Arnone, E., and Noto, L. V.: Combining a data-driven approach with seasonal forecasts data to predicting reservoir water volume in the Mediterranean area., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-14019, https://doi.org/10.5194/egusphere-egu21-14019, 2021.

Hydrological droughts can trigger socioeconomic disruptions which can cause large impacts on societies. Therefore, their forecast is extremely important for early warning and disaster mitigation. Several modelling approaches have been developed for this purpose, and models based on machine learning have become popular during the last decades. In this study, we evaluated the performance of five commonly used models in drought forecasting: Autoregressive Integrated Moving Average (ARIMA), multiple linear (MLM), Random Forest (RF), K-nearest Neighbours (KN) and Artificial Networks (ANN) models.
The study site was the Paute River basin (4816.56 km2) located in the mid-high and eastern part of the Ecuadorian Andes. The "Daniel Palacios" dam, the oldest of 4 strategic dams and which alone generates 35% of Ecuador's national energy demand, was considered the lowest point in the basin. The basin is susceptible to hydrological droughts, which have led to power shortages in the past (e.g., in 1995, 1999 and 2009).
As hydrological drought predictand we used the monthly Standardized Streamflow Index (SSI) accumulated at 1-, 3-, 6- and 12-months, while the predictor variables were precipitation, temperature, river flow and the climatic indices associated with the El Niño-Southern Oscillation, ENSO (El Niño 1+2, El Niño 3, El Niño 4, and El Niño 3.4, which are strongly related to droughts in Ecuador). Data were obtained from the National Institute of Meteorology and Hydrology (INAMHI) and the Climate Prediction Centre of NOAA.
The models were evaluated for lead times of 1, 3, 6 and 12 months through the determination coefficient (R2), the Nash-Sutcliff efficiency (NSE) and the Root Mean Square Error (RMSE). The models' capability to distinguish between occurrence and non-occurrence of droughts (here defined as SSI≤-1) was assessed with a receiver operating characteristic (ROC) diagram. Models were validated using the expanding window (walk forward approach) as a back testing strategy starting as calibration and validation periods Aug/1984-Dec/2010 and Jan/2011-Dec/2019 (75 and 25% of the data, respectively).
As expected, SSI at longer aggregations can be predicted more accurately than at shorter ones with all models. This fact is explained because the former ones more effectively reduce the noise than the latter due to the increase in filter length. The greater the lead time, the less reliable the prediction. Considering a lead time of 1 month, the best model was the ANN for SSI-1 and SSI-3 (R2=0.64, ROC=0.67; and R2=0.78, ROC=1.00 respectively), and the MLM model for SSI-6 and SSI-12 (R2=0.86, ROC=0.66, and R2=0.96, ROC=0.99 respectively). However, very similar performances were obtained in the latter cases by the ANN model (R2=0.86, ROC=0.66 R2=0.91 and ROC=0.75 respectively) and the ARIMA model (R2=0.83, ROC=0.98 R2=0.93 and ROC=0.99 respectively). ARIMA models showed large errors for lead times longer than 1 month, so an ANN model is recommended. However, to maximize their potential, further research could explore modifications of the ANN architecture or the input data. Results indicate that these models can be used to forecast hydrological droughts in the Paute river basin and can be used to support reservoir operation decisions.

How to cite: Araneda-Cabrera, R. J., Bermudez, M., and Puertas, J.: Short-term hydrological drought forecasting in the Paute river basin, Ecuadorian Andes, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-16318, https://doi.org/10.5194/egusphere-egu21-16318, 2021.