HS7.2

Abstract

Precipitation is highly variable in space and time. Ground-based precipitation gauging networks such as those from national weather services are often not able to capture this variability. Weather radars have the potential to capture the spatio-temporal characteristics of rainfall fields but they also suffer from specific errors such as attenuation. The increasing number and availability of opportunistic sensors (OS), such as commercial microwave links (CML) and personal weather stations (PWS), provides new opportunities to improve rainfall estimates based on ground observations.

We have developed a geostatistical interpolation method that allows a combination of different opportunistic sensors and their specific features and geometric properties, e.g., point and line information. In addition, the uncertainty of the different data sets can be considered [1].

The flood event in the western provinces of Germany in July 2021 showed that both, the precipitation interpolations based on rain gauge data from the German National Weather Service and radar-based precipitation products, underestimated precipitation. We show that the additional information of OS data can improve precipitation estimates in terms of areal precipitation amounts and spatial distribution.  

References
[1] Graf, M., El Hachem, A., Eisele, M., Seidel, J., Chwala, C., Kunstmann, H. and Bárdossy, A.: Rainfall estimates from opportunistic sensors in Germany across spatio-temporal scales, https://doi.org/10.1016/j.ejrh.2021.100883

How to cite: Eisele, M., Bárdossy, A., Chwala, C., Demuth, N., El Hachem, A., Graf, M., Kunstmann, H., and Seidel, J.: Improvement of rainfall estimates using opportunistic sensors - the example of the flood in Rhineland-Palatinate in July 2021, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6968, https://doi.org/10.5194/egusphere-egu22-6968, 2022.

The need for high-resolution and low maintenance weather stations is the major factor behind the increasing adoption of Non-Catching Gauges (NCGs) by national weather services and research institutions. Data from such instruments are used for several applications and in numerous research fields, where instrumental biases can have a strong impact. For NCGs, rigorous testing and calibration are more challenging than for catching gauges. Hydrometeor characteristics like particle size, shape, fall velocity and density must be carefully reproduced to provide the reference precipitation, instead of the equivalent water flow used for the calibration of catching gauges. Instrument calibration is usually declared by the manufacturers, using internal procedures developed for the specific technology employed. No standard calibration methodology exists, that encompass all or at least most of the available NCGs (Lanza et al. 2021). The EURAMET project 18NRM03 ‘INCIPIT’ on the ‘Calibration and accuracy of non-catching instruments to measure liquid/solid atmospheric precipitation’, was initiated in 2019 to address such issues.

A calibration device was developed to achieve individual drop generation on demand and in-flight measurement of the released drops. Water drops in the range from 0.5 to 6 mm in diameter are generated to mimic natural raindrops. A high-precision syringe pump is used to form the drop of the desired volume at the tip of a calibrated nozzle. A high-voltage power supply is used to apply a large potential difference between the nozzle and a metallic ring, and the resulting electric field triggers the release of the drop. A precision motorized gantry moves the generator across the horizontal plane, to cover different releasing positions within the instrument sensing area. By either varying the release height or accelerating the drop using compressed air, different fractions of the terminal velocity can be achieved, depending on the drop size. A second gantry, just above the gauge under test, aligns the plane of focus of a high-resolution camera with the fall trajectory of the drop. Three images of the same drop are captured in a single picture, using speedlights triggered at fixed time intervals. Photogrammetric techniques and a photodiode to measure the time between flashes provide the shape, size, speed, and acceleration of the drop. This characterizes each released drop before it reaches the instrument sensing area and, by comparison with the gauge measurement, the instrumental bias is obtained. Laboratory tests are presented to assess the performance of the calibration device.

This work is funded as part of the activities of the EURAMET project 18NRM03 “INCIPIT Calibration and Accuracy of Non-Catching Instruments to measure liquid/solid atmospheric precipitation”. The project INCIPIT has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme.

References:

Lanza L.G. and co-authors, 2021: Calibration of non-catching precipitation measurement instruments: a review. J. Meteorological Applications, 28.3(2021):e2002.

How to cite: Chinchella, E., Stagnaro, M., Cauteruccio, A., and Lanza, L. G.: A precision raindrop generator to calibrate non-catching rain gauges, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7339, https://doi.org/10.5194/egusphere-egu22-7339, 2022.

Adjustments of the wind-induced bias of conventional catching type rain gauges derive from collection efficiency (CE) curves that can be obtained either from field experiments or from numerical simulation (Lanza and Cauteruccio, 2021). The use of numerical simulation allows to overcome the limitations of the experimental installations and monitoring campaigns (e.g., the many influencing variables involved and the variability of the rainfall process) to cover a wide range of wind speed and rainfall intensity (RI) conditions. Also, the accuracy of the measurements taken as a reference is still an issue in field experiments.

A Lagrangian particle tracking (LPT) model, suitably validated in the wind tunnel (see Cauteruccio et al., 2021), is applied to the results of computational fluid dynamic (CFD) simulations of the airflow field surrounding a rain gauge to derive a simple formulation of the collection efficiency curves as a function of wind speed (Cauteruccio and Lanza, 2020). A new parameterization is proposed to highlight the influence of rainfall intensity, based on the typical form of the drop size distribution (DSD) of rainfall events (data from the Italian territory). The methodology is applied to a cylindrical gauge, which has the typical outer shape of most tipping-bucket rain gauges, as a representative specimen of operational measurement instruments.

Using rainfall intensity as a controlling factor for the collection efficiency has solid physical bases in the relationship between RI and the DSD (Colli et al., 2020), and the role of RI can only be quantified using numerical simulations of both the airflow field (using CFD) and the particle motion (via the LPT).

A simple formulation of the adjustment curves is obtained, which can be easily applied in an operational context, since wind velocity is the only ancillary variable required to perform the adjustment. Wind is often measured by operational weather stations together with the precipitation intensity, so the correction adds no relevant burden to the cost of meteo-hydrological networks.

References

Cauteruccio, A. and L.G. Lanza (2020). Parameterization of the collection efficiency of a cylindrical catching-type rain gauge based on rainfall intensity. Water, 12(12), 3431. https://doi.org/10.3390/w12123431.

Cauteruccio, A., Brambilla, E., Stagnaro, M., Lanza, L.G. and D. Rocchi (2021). Wind tunnel validation of a particle tracking model to evaluate the wind-induced bias of precipitation measurements. Water Resour. Res., 57(7), e2020WR028766. https://doi.org/10.1029/2020WR028766.

Colli, M., Stagnaro, M., Lanza, L.G., R. Rasmussen and J.M. Thériault (2020). Adjustments for Wind-Induced Undercatch in Snowfall Measurements Based on Precipitation Intensity, J. Hydrometerol., 21, 1039-1050, https://doi.org/10.1175/JHM-D-19-0222.1.

Lanza, L.G and A. Cauteruccio (2021). Accuracy assessment and intercomparison of precipitation measurement instruments. Chapter 1, p. 3 – 35. In: Michaelides, S. (ed.), Precipitation Science. Elsevier, Amsterdam, Netherlands. ISBN: 978-0-12-822973-6, pp. 833. https://doi.org/10.1016/B978-0-12-822973-6.00007-X.

How to cite: Lanza, L. G. and Cauteruccio, A.: Influence of the drop size distribution on the collection efficiency of catching gauges as a function of rainfall intensity, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7871, https://doi.org/10.5194/egusphere-egu22-7871, 2022.

Vertically pointing radars (VPRs) provide detailed observations of precipitating cloud systems as they pass directly over the radar site. Two VPRs operating side-by-side and at different millimeter wavelengths (mm-wave) will observe the same raindrops but will have different return signals due to wavelength dependent raindrop backscattering and attenuation characteristics. These differences enable the retrieval of raindrop size distributions and vertical air motions. Yet, as the rain rate increases, the attenuation increases. Eventually, at some combination of path length [km] and rain specific attenuation [dB/km], the attenuation [dB] will extinguish high frequency VPR return signals; limiting high frequency VPRs to studying rain processes close to the ground. 

In order to estimate how far VPRs can measure into rain shafts, this study simulated constant rain rate precipitation columns and then estimated the path length needed to produced enough attenuation to drop the VPR signal-to-noise ratio below the VPR’s detection limit. This study used surface disdrometer observations and publically available T-Matrix scattering code to produce realistic VPR measurements at frequencies from 3 to 200 GHz.

These simulations found that in order to observe raindrops above a 3.5 km rain shaft, the constant rain rate needed to be less than 138, 67, 26, 14, and 4 mm/h for VPRs operating in the X-, Ku-, K-, Ka-, and W-bands, respectively (i.e., 9, 13.6, 24, 35.6, and 94 GHz). Additionally, due solely to atmospheric gas attenuation, the G-band (200 GHz) VPR return signal will be completely extinguished by 3.5 km. Preventing a G-band VPR from detecting raindrops above 3.5 km.

How to cite: Williams, C.: How far into a rain shaft can mm-wave vertically pointing radars detect raindrops?, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6386, https://doi.org/10.5194/egusphere-egu22-6386, 2022.

Observing, in a quantitative and robust way, the dynamic space-time pattern of precipitation in mountainous terrain presents a major challenge of great practical importance. The difficulties of this task are further exacerbated in mid to high latitudes where the typical melting layer for precipitation (i.e. the 0°C isotherm) is often close to the surface during winter months. One way to address this challenge is by improving observations made using networks of weather radars. Quantitative Precipitation Estimates (QPEs) derived from these instruments have many applications, for example as input to a hydrological model to simulate river flow for flood forecasting purposes. 

Here, a set of QPEs - obtained from an observation campaign using the National Centre for Atmospheric Science’s mobile X-band dual-polarisation Doppler weather radar (NXPol) in a mountainous area of Northern Scotland - are assessed with reference to observed river flows. Each form of QPE is used as an input to Grid-to-Grid (G2G), a distributed hydrological model used for flood forecasting across Great Britain, and the simulated river flows compared to observations. The location of the radar was specially chosen to infill an area of reduced coverage in the existing C-band radar network for the British Isles.

Assessments of radar QPE often only examine a final precipitation “best estimate” product and typically with reference to raingauges at specific locations. Here, we exploit the processing capabilities of NXPol and the hydrological modelling framework to investigate the benefits of ten separate processing methods that increase with complexity and make differing use of dual-polarisation variables. The role of the radar beam elevation and distance from the radar is investigated, and NXPol QPEs are compared to that provided by the radar network. Additionally, a preliminary investigation is carried out into the role of the drop-size distribution on the relationship between radar-reflectivity and rain-rate using disdrometer data.

The hydrological assessment reported on here has the benefit of integrating the precipitation over space and time which serves to complement and extend a previous meteorological assessment using raingauge data alone. The assessment proves to be insensitive to issues affecting both raingauges (e.g. representativity, wind-induced under-catch) and local artefacts in the space-time radar-rainfall field. It facilitates a direct assessment of whether potential benefits in the new QPEs are carried forward to an end-use such as flood forecasting, providing fresh insights for the development of new dual-polarisation radar QPE methods.

How to cite: Wallbank, J. R., Cole, S. J., Moore, R. J., Dufton, D., Neely III, R. R., and Bennett, L.: Dual-polarisation X-band radar estimates of precipitation assessed using a distributed hydrological model for mountainous catchments in Scotland, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2840, https://doi.org/10.5194/egusphere-egu22-2840, 2022.

The purpose of this study is to investigate the impact of assimilating dual-polarimetric parameters, i.e. differential reflectivity (ZDR) and specific differential phase (KDP), in addition to reflectivity (ZH) and radial wind (Vr) in a severe weather system. A squall line case forced by the synoptic southwesterly wind is selected to conduct the assimilation experiments. Besides, different microphysics parameterization schemes, including GCE, MOR, WSM6 and WDM6, are examined in the experiments. The results of the analysis field show that assimilating additional ZDR with single moment schemes (GCE and WSM6) can capture better mean raindrop size, yet it deteriorates the intensity of simulated ZH and KDP. Differ from GCE and WSM6, assimilating additional ZDR with double moment schemes (MOR and WDM6) would not lead to significant deterioration in the simulated ZH and KDP since the prognostic hydrometeor variables in double moment schemes include both mixing ratio and total number concentration. There will be more flexibility in adjusting microphysical states with two independent prognostic hydrometeor variables. The results of the short-term quantitative precipitation forecast (QPF) show that assimilating additional dual-polarimetric parameters with either single or double moment schemes increases the maximum of accumulated rainfall and the probability of heavy rainfall. In conclusion, double moment schemes can make better use of the extra information from dual-polarimetric parameters; furthermore, assimilating additional dual-polarimetric parameters, even with single moment schemes, can improve the performance of QPF, especially heavy rainfall events.

How to cite: Zhuang, B.-X., Chung, K.-S., and Tsai, C.-C.: Impact of Additional Assimilation of Dual-Polarimetric Parameters: Analysis and Forecasts in a Real Case, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4756, https://doi.org/10.5194/egusphere-egu22-4756, 2022.

Forecasting weather systems are capable to model atmospheric phenomena at various space-time scales. At very short space-time nowcasting techniques are still relying on measured data processing from ground-based microwave radars and satellite-based geostationary spectrometers. In this respect, precipitation field nowcasting from a few minutes up to a few hours is one of the most challenging goals to provide rapid and accurate updated features for civil prevention and protection decision-makers (e.g., from emergency services, marine services, sport, and cultural events, air traffic control, emergency management, agricultural sector and moreover flood early-warning system). Deep learning precipitation nowcasting models, based on weather radar network reflectivity measurements, have recently exceeded the overall performance of traditional extrapolation models, becoming one of the hottest topics in this field. This work proposes a novel network architecture to increase the performance of deep learning mesoscale precipitation prediction. Since precipitation nowcasting can be viewed as a video prediction problem, we present an architecture based on Graph Convolutional Neural Network (GCNN) for video frame prediction. Our solution exploits, as a cornerstone, the topology of Space-Time-Separable Graph-Convolutional- Network (STS-GCN), originally used for posing forecasting. We have applied our model on the TAASRAD19 radar data set with the aim of comparing our performance with other models, namely the Stacked Generalization (SG) Trajectory Gated Recurrent Unit (TrajGRU) and S-PROG Spectral Lagrangian extrapolation program (S-PROG).

The proposed model, named STSU-GCN (Space-Time-Separable Unet3d Graph Convolutional Network), has a structure composed of an encoder, decoder, and forecaster. The role of the encoder and decoder are accomplished by a Unet3d a structure borrowed with the specific purpose of modifying the spatial component, but not the temporal component. In the bottleneck of this Unet3D network, we use a graph-based forecaster. The performance of the STSU-GCN has been quantified using conventional metrics, such as the Critical Success Index (CSI), widely used in the meteorological community for the nowcasting task. Using TAASRAD19 radar data set and literature data, these CSI metrics have been applied to 4 different classes of rain rate, that is 5, 10, 20, 30 mm/h. Our STSU-GCN model has overperformed both TrajGRU and S-PROG in the classes 10 mm/h and 20 mm/h obtaining a CSI respectively of 0.148 and 0.097. On the other hand, STSU-GCN is underperforming in class 5mm per hour getting a CSI respectively of 0.099. Our STSU-GCN model is aligned with the results of the S-PROG benchmark, for the class 30 mm/h confirming a model skillful for classes with a high rain rate. In this work, we will also illustrate the results of the proposed STSU-GCN algorithm using case studies in the area of interest of the Italian Central Apennines during the summer of 2021. Statistical performances, potential developments, and critical issues of the STSU-GCN algorithm will be also discussed.

How to cite: Trappolini, D., Scofano, L., Sampieri, A., Messina, F., Galasso, F., Di Fabio, S., and Marzano, F. S.: Mesoscale precipitation nowcasting from weather radar data using space-time-separable graph convolutional networks, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5361, https://doi.org/10.5194/egusphere-egu22-5361, 2022.

How to cite: Drouen, G., Schertzer, D., Gires, A., and Tchiguirinskaia, I.: The Fresnel Platform for increasing the Greater Paris resilience to spatio-temporal variability of local rainfall, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9096, https://doi.org/10.5194/egusphere-egu22-9096, 2022.

The Tensift basin in Morocco is prominent for its ecological and hydrological diversity. This diversity is marked by rivers flowing into areas such as Ourika. In addition to agriculture, the basin is a hub of variable land use systems. It is important to have a better understanding of the relationship between simulated and observed precipitation measurements in this region to be able to better understand the role of precipitation in the variability of the climate and water resources in the basin. This study aims at evaluating the performance of multi-source satellite products against weather stations precipitation in the basin. In this work, the satellite product based data were first culled for seven satellite products namely PERSIANN, PERSIANN CDR, TRMM3B42, ARC2, RFE2, CHIRPS, and ERA5 (simulated precipitation) from, CHRS iRain, RainSphere, NASA, EUMETSAT, NOAA, FEWS NET, ECMWF respectively. Precipitation observations data from six weather stations, located at Tachedert (2343 m), Imskerbour (1404 m), Asni (1170 m), Grawa (550 m), Agdal (489 m), and Agafay (487 m) at different altitudes, latitudes and temporal scales (1D, 1M, 1Y), over the period 13/05/2007 and 31/09/2019, at Tensift basin were used. The data were compared and analyzed through inferential statistics such as Nash-Sutcliffe Efficiency Coefficient, Bias, Root Mean Square Error (RMSE), Root Mean Square Deviation (RMSD), the standard deviation, the Correlation Coefficient (R) and the Coefficient of Determination (R²) and visualized through taylor diagrams and scatter plots to have a visual idea of the closeness between the seven satellite products and the observed precipitation data. A second analysis was carried out on the monthly precipitation resulting from the six weather stations based on standardized precipitation index (SPI) in order to  determine the onset, duration, and magnitude of the meteorological drought. The results show that PERSIANN CDR performs best and is more reliable with regrad to its ability to estimate precipitation rates over a wide spatial and temporal scale over the basin. The precipitation of Persiann CDR  has significant rates for the different statistics (Bias: -0.05 (Daily asni), RMSE: 2.86 (Daily Agdal), R: 0.83, R²:0.687 (Monthly Agdal)). However, most of the time, this product records low or negative Nash values (-6.06 (Annual Grawa)), due to the insufficient weather station data in the study area (Tensift). It  was observed that TRMM overestimates precipitation during heavy precipitation and underestimates during low precipitation. This makes it important for the latter observations to be viewed with caution due to the quality of annual comparison results and underscores the need to develop more efficient precipitation comparison approaches. Also, the performance of the satellite products is better at low altitudes and during wet years. Finally, it was concluded from the SPI that Tensift Region has experienced 13 drought periods over the study period, with the longest event of 12 months was from Marsh 2015 to February 2016 and  the most intense event with the highest drought severity (19.6) and the lowest SPI value (-2.66) was in 2019.

How to cite: salih, W., chehbouni, A., and Epule, T. E.: Evaluation of the Performance of Multi-Source Satellite Products in Simulating Precipitation over the Tensift Basin in Morocco, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-961, https://doi.org/10.5194/egusphere-egu22-961, 2022.

Data availability and accuracy is predominantly an issue for building hydrological applications, particularly in data-scare regions, like Africa. This is further one of the challenges that hinders understanding the climate variability and its subsequent extreme flood and drought events. Forcing data from different sources, e.g., satellite sensors, in-situ observations, or reanalysis products, are required to derive hydrological models. Reanalysis products have recently become an alternative tool of meteorological data given their long record at various temporal and spatial scales. The overarching goal of this study is to evaluate the primary forcing data for hydrological models; precipitation, as produced by six different reanalysis data (JRA55, 20CRv3, ERA5, ERA-20C, MERRA, NCEP/NCAR). We here focused our evaluation on the major river basins in Africa during a 15-year period spanning from 2001 to 2015. The five major river basins include the Nile River, Congo River, Zambezi River, Orange River, and Niger River basins. Our evaluation method is summarized as follows: Firstly, precipitation data is compared with the gridded gauged data, e.g., CHIRPS for precipitation. Secondly, statistical indices, including categorical and continuous statistical metrics, will be used to assess the accuracy of reanalysis products over each of the major basins. Finally, we present the intercomparison of reanalysis products for extreme events including floods and droughts. The results from our evaluation will pinpoint the skill of reanalysis products and thus benefit the future development of hydrological modeling over the river basins in Africa.

How to cite: Abdelmoneim, H. and Eldardiry, H.: Intercomparison of reanalysis products during extreme flood and drought events: evaluation over the major river basins of Africa, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-970, https://doi.org/10.5194/egusphere-egu22-970, 2022.

Gathering accurate precipitation data is an important task for setting up hydrological models. In Norway, the gauge network density is higher in the southern parts and decreases in the north. Furthermore, the amount of high evaluated precipitation gauges is rather scarce. Radar data is available but lacks an accurate reflectivity-precipitation relation and errors in precipitation estimation are caused for example by beam blockage.

For modelling purposes, this study aims to evaluate whether the application of radar derived data gives any benefit, especially when modelling in a higher temporal resolution. The results of this study can give decision support for modellers having difficulties choosing the precipitation product. For that cause, spatial interpolated precipitation products were evaluated and compared in terms of performance in hydrological models. The Meteorological Institute Norway publishes gridded hourly datasets covering the Norwegian mainland: seNorge2, where gauge data is interpolated using an optimal interpolation, and the numerical weather prediction product (NWP), a combination of gauge data, radar data and a numerical weather model. Five different catchments were simulated in the numerical precipitation-runoff model HYPE with both datasets for comparison. The catchments vary in area, hydrological regime and availability of nearby gauges. The simulation was done in an hourly time step in order to compare precipitation variability on a small time scale.

In this study, a calibration method was developed that generates comparable and stable performance results in terms of the Kling–Gupta efficiency (KGE) for each catchment and dataset. The resulting discharges and water balances of the catchments were analysed and compared. Additionally, selected precipitation events, where the precipitation products were not able to describe atmospheric processes appropriately, were analysed. The datasets were further compared by spatially accumulating annual precipitation sums over the catchments, by using a private weather station to evaluate the fit of the data and by comparing the runoff and precipitation volume of the basins.

Preliminary results show the significant differences in water volume and spatial distribution of precipitation between these products. Furthermore, when comparing a private gauge with the precipitation products at an ungauged area, daily precipitation data tends to be more accurate than hourly data.

How to cite: Sienel, J., Schönfelder, L., and Seidel, J.: Using radar-derived precipitation data for hydrological modelling in selected study sides in Norway, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8119, https://doi.org/10.5194/egusphere-egu22-8119, 2022.

Understanding the role of forest on rainfall interception is fundamental for a correct analysis and modelling of runoff generation and catchment hydrological response. Despite many studies were carried out at the stand and hillslope scale, very little is known about the role of hillslope topography and the associated tree population characteristics on throughfall spatio-temporal variability. Therefore, this work aimed at better understanding the dominant factors on throughfall variability and on the temporal persistence of throughfall spatial patterns along a transect on a steep hillslope characterized by trees with different size and density.

The experimental activities were carried out in the upper part of the densely-forested Re della Pietra catchment, Tuscany Apennines, Central Italy. The hillslope is roughly 110 m long and 60 m wide, has a mean slope of 30°, and is dominantly covered by beech trees and by sparce individuals of oak trees. A grid of 126 throughfall collectors was installed in July 2020 and divided in three sub-plots: two plots of 144 m² with 2-m spaced 49 collectors at the bottom and the top of the hillslope, and a transect of 28 1-m spaced collectors from the bottom to the top of the hillslope. A survey was conducted to measure the diameter and basal area of the stand. Throughfall was manually measured from the collectors approximately monthly from June 2020 to November 2021, and compared with gross precipitation measured by a rain gauge placed outside the vegetation cover. Moreover, five automatic gauges connected to 1.5 m-long gutter to increase the collection area were installed in November 2021 along the hillslope to measure throughfall at high temporal resolution.

Preliminary results from 25 manual measurements over the experimental grid highlighted a large temporal variability of interception (mean: 17%, standard deviation: ±31%), reflecting the variable seasonal precipitation pattern of Mediterranean areas and the phenological stage of trees (leaves/no leaves). Overall, the spatial variability in throughfall increased with increasing gross precipitation. Particularly, the bottom plot, characterized by lower tree density and larger tree size compared to the top plot, showed a lower spatial variability with respect to the top plot, while the longitudinal transect exhibited an intermediate variability. Analogously, the temporal stability analysis revealed that the most temporally-stable and representative measurement points laid on the transect that, overall, captured the different tree characteristics along the hillslope.

Future work will make use of the high-resolution measurements of the five gauges to assess their representativeness compared to the manual grid and to test and validate an interception model at the hillslope scale to be possibly upscaled to the entire catchment.

How to cite: Verdone, M., Borga, M., Dani, A., Preti, F., Trucchi, P., Zuecco, G., van Meerveld, I., Massari, C., and Penna, D.: Throughfall variability at the hillslope scale: the role of topography and tree characteristics, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-526, https://doi.org/10.5194/egusphere-egu22-526, 2022.

The aim of the study is to reduce the uncertainty of the influence of Palmer-type drought indices in estimating seasonal discharge in the lower Danube basin. For this, four indices were considered: Palmer Drought Severity Index (PDSI), Palmer Hydrological Drought Index (PHDI), Weighted PDSI (WPLM) and Palmer Z-index (ZIND). These indices were quantified by PC1 of EOF decomposition, obtained from 15 stations located along the Danube basin.

The influences of these indices on the Danube discharge were tested, both simultaneously and with certain lags, by linear and nonlinear methods applying the elements of information theory. Nonstationarity was tested by wavelet analysis. The results differ depending on the season and the Palmer index.

The linear connections are generally obtained for synchronous links, and the nonlinear and nonstationary ones for the predictors considered with certain lags (in advance) compared to the discharge predictand. This result is useful for estimating the discharge, as Palmer indices can be estimated from the simulated data by the General Circulation Models or Regional Climate Models.

How to cite: Mares, I., Mares, C., Dobrica, V., and Demetrescu, C.: Testing nonlinearity and nonstationarity of the connection between Palmer drought indices and Danube discharge in the lower basin, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4319, https://doi.org/10.5194/egusphere-egu22-4319, 2022.