Runoff is a crucial water cycle component that contributes to the water resources to sustain human life. Historical trends in runoff, when examining climate change scenarios, provide vital information about past variability and support the design of adaptation measures. However, hydrological models based on climate data, such as the Budyko model, can be biased in estimating annual runoff due to input data uncertainty. Therefore, it is vital to utilize advanced machine learning-based computing models to reduce uncertainty and reconstruct climate variables over a long period of time and sufficiently large spatial coverage, preferably at a continental scale. We propose and test a novel machine learning-based framework called Hybrid Ensemble Multi-Model Framework (HEMMF) to reconstruct the gridded runoff of Europe over a 500-year historical period (1500 to 1999). The HEMMF combines non-parametric extended data pattern recognition and data-driven methods. The extended data patterns are computed using Moran's spatial autocorrelation (SPA) index of the climate variable fields and the Budyko models output, whereas the data-driven methods contain nine different machine learning (ML) algorithms and four ensembles of ML. The extended data patterns are jointly ingested with climate-reconstructed data (precipitation, temperature, Palmer's drought severity index) as predictor variables, which serve as input for the data-driven methods. To assess the impact and contribution of SPA, the runoff is simulated based on three different input training datasets in the HEMMF: (1) a dataset containing only precipitation, temperature, Palmer's drought severity index, and four different estimates of runoff from the Budyko model, (2) a dataset containing only SPA of the first input datasets, and (3) a dataset created by merging the first and second datasets. The HEMMF offers the best reconstruction performance when using the third input dataset. This reconstructed runoff helps to explain the runoff trend, drought propagation, and runoff's link with the climate variables. The proposed methodology has the potential to be applied to past hydroclimatic data and related analyses across different temporal periods, climate scenarios, and geographical scales.

<p><span dir="ltr" role="presentation">Runoff is the key hydrological process, which is vital to the sustaining of human life on earth in examining</span><br role="presentation" /><span dir="ltr" role="presentation">the climate change scenario. There are a lot of hydrological models available to simulate the runoff, but</span><br role="presentation" /><span dir="ltr" role="presentation">these models’ outputs have biases due to uncertainty. Most machine learning algorithms cannot capture</span><br role="presentation" /><span dir="ltr" role="presentation">the runoff generated by the real-world complex hydrological system accurately. The hybrid model combines</span><br role="presentation" /><span dir="ltr" role="presentation">the efficiency of hydrological, machine learning, and ensemble modeling to minimize the bias of output [1],</span><br role="presentation" /><span dir="ltr" role="presentation">[2].</span> <span dir="ltr" role="presentation">The recent development of evolutionary computation in hybrid modelling frameworks combines the</span><br role="presentation" /><span dir="ltr" role="presentation">efficiency of different components such as hydrological models, spatial autocorrelation, machine learning,</span><br role="presentation" /><span dir="ltr" role="presentation">and machine learning ensemble to estimate robust and less biased runoff [1]. However, these components</span><br role="presentation" /><span dir="ltr" role="presentation">need to significantly capture the heterogeneity and similarity of the catchment properties, which are highly</span><br role="presentation" /><span dir="ltr" role="presentation">linked with the spatial variation of various hydrological patterns. Clustering is a technique that can group</span><br role="presentation" /><span dir="ltr" role="presentation">similar types of hydrological patterns, which can be integrated within a hybrid modeling framework.</span><br role="presentation" /><span dir="ltr" role="presentation">However, there is rarely found literature on the hybrid framework, which consists of different clustering</span><br role="presentation" /><span dir="ltr" role="presentation">techniques and their ensemble. These clustering algorithms are based on different categories. We proposed</span><br role="presentation" /><span dir="ltr" role="presentation">the hybrid ensemble framework based on extended input data, hydrological models, different clustering</span><br role="presentation" /><span dir="ltr" role="presentation">algorithms, deep learning, and an ensemble of deep learning to reconstruct the minimum biased surface</span><br role="presentation" /><span dir="ltr" role="presentation">runoff.</span> <span dir="ltr" role="presentation">We tested our proposed hybrid framework, which is robust compared to previously developed</span><br role="presentation" /><span dir="ltr" role="presentation">frameworks. This proposed hybrid framework methodology will help to develop a new hybrid algorithm</span><br role="presentation" /><span dir="ltr" role="presentation">to estimate the less biased surface runoff using various available climate data to understand the dynamics</span><br role="presentation" /><span dir="ltr" role="presentation">of surface runoff for different spatial-temporal scales and climates.</span></p> <p> </p> <div class="textLayer"><span dir="ltr" role="presentation">[1]</span> <span dir="ltr" role="presentation">U. Singh, P. Maca, M. Hanel,</span> <span dir="ltr" role="presentation">et al.</span><span dir="ltr" role="presentation">, “Hybrid multi-model ensemble learning for reconstructing gridded</span><br role="presentation" /><span dir="ltr" role="presentation">runoff of europe for 500 years,” vol. Available at SSRN:</span> <span dir="ltr" role="presentation">doi</span><span dir="ltr" role="presentation">:</span> <span dir="ltr" role="presentation">10 . 2139 / ssrn . 4188518</span><span dir="ltr" role="presentation">. [Online].</span><br role="presentation" /><span dir="ltr" role="presentation">Available:</span> <span dir="ltr" role="presentation">http://dx.doi.org/10.2139/ssrn.4188518</span><span dir="ltr" role="presentation">.</span><br role="presentation" /><span dir="ltr" role="presentation">[2]</span> <span dir="ltr" role="presentation">S. M. Hauswirth, M. F. Bierkens, V. Beijk, and N. Wanders, “The suitability of a hybrid framework</span><br role="presentation" /><span dir="ltr" role="presentation">including data driven approaches for hydrological forecasting,”</span> <span dir="ltr" role="presentation">Hydrology and Earth System Sciences</span><br role="presentation" /><span dir="ltr" role="presentation">Discussions</span><span dir="ltr" role="presentation">, pp. 1–20, 2022.</span><br role="presentation" /> <div class="endOfContent"> </div> </div> <div class="annotationLayer"> </div>

Abstract Achieving accurate precipitation estimation at high spatial and temporal resolution is critical in hydrology and meteorology, particularly in regions experiencing water resource degradation. Integrating original products of generally coarse-resolution hydrological parameters from multiple satellites is a promising technology for producing massive repositories of space-time varying datasets such as precipitation. A methodological modelling framework commonly known as downscaling or disaggregation is evolving as a viable approach for generating hydrological datasets with spatio-temporal scales suitable for operational usage. In this study, we propose a non-parametric method for generating a high-resolution dataset from a coarse-resolution integrated multiple satellite-based precipitation dataset. The disaggregation involves using a hybrid Extreme Gradient Boosting (XGBoost) approach combined with multivariate spatial-temporal Fuzzy clustering. This clustering relies on Integrated Multi-satellite Retrievals for GPM (IMERG) precipitation and Shuttle Radar Topography Mission (SRTM) Digital Elevation Data to establish eight distinct clusters.The proposed method is experimentally demonstrated implementing to downscale 255 months (June 2000 to September 2021) of IMERG satellite data from 11km to 1km spatial resolution over the Czech Republic. We utilized eight stations, one per cluster, for training and validation purposes, with the remaining 19 stations used solely for validation. Our findings demonstrate a strong agreement between the disaggregated monthly precipitation over the 20 years and ground-observed precipitation, suggesting that our proposed methodology substantially enhances the accuracy of IMERG precipitation data. This method holds promise for applications in other regions with remotely sensed data, especially where ground-measured station data is sparse, facilitating the generation of high-resolution, accurate precipitation data.

Decadal-scale, high-resolution geodetic measurements of glacier thinning have transformed our understanding of glacier response to climate change. Annual glacier mass balance can be estimated using remote-sensing proxies like snow-line altitude. These methods require field data for calibration, which are not available for most glaciers. Here we propose a method that combines multiple remotely-sensed proxies to obtain robust estimates of the annual glacier-wide balance using only remotely-sensed decadal-scale geodetic mass balance for calibration. The method is tested on Chhota Shigri, Argentiere and Saint-Sorlin glaciers in the Himalaya and the Alps between 2001 and 2020, using four remotely-sensed proxies - the snow-line altitude, the minimum summer albedo over the glacier and two statistics of normalised difference snow index over the off-glacier area around the ablation zone. The reconstructed mass balance compares favourably with the corresponding glaciological field data (correlation coefficient 0.81 - 0.90, p < 0.001; root mean squared error 0.38 - 0.43 m w.e. a(-1)). The method presented may be useful to study interannual variability in mass balance on glaciers where no field data are available.

Since the beginning of this century, Europe has been experiencing severe drought events (2003, 2007, 2010, 2018 and 2019) which have had adverse impacts on various sectors, such as agriculture, forestry, water management, health and ecosystems. During the last few decades, projections of the impact of climate change on hydroclimatic extremes have often been used for quantification of changes in the characteristics of these extremes. Recently, the research interest has been extended to include reconstructions of hydroclimatic conditions to provide historical context for present and future extremes. While there are available reconstructions of temperature, precipitation, drought indicators, or the 20th century runoff for Europe, multi-century annual runoff reconstructions are still lacking. In this study, we have used reconstructed precipitation and temperature data, Palmer Drought Severity Index and available observed runoff across 14 European catchments in order to develop annual runoff reconstructions for the period 1500–2000 using two data-driven and one conceptual lumped hydrological model. The comparison to observed runoff data has shown a good match between the reconstructed and observed runoff and their characteristics, particularly deficit volumes. On the other hand, the validation of input precipitation fields revealed an underestimation of the variance across most of Europe, which is propagated into the reconstructed runoff series. The reconstructed runoff is available via Figshare, an open-source scientific data repository, under the DOI https://doi.org/10.6084/m9.figshare.15178107, (Sadaf et al., 2021).

The aridity index, also known as the Budyko index, describes spatiotemporal changes in the hydroclimatic system in the long-term perspective. Defined as the ratio between potential evapotranspiration and precipitation, it can be used to determine wet (humid) and dry (arid) regions. In this study, we evaluated the aridity index estimated in different temporal scales, investigated its spatial patterns, and highlighted the long-term changes in Europe using three gridded data sets (CRU, E–OBS, and ERA5). A significant dry region expansion is evident in all data sets since the late 1980s. The extent of the dry regions has increased in Western, Central, and Eastern Europe, especially at low and medium altitudes. The results show the long-term development of the European hydroclimatic system and which areas have changed from wet to dry.

Abstract. Since the beginning of this century, Europe has been experiencing severe drought events (2003, 2007, 2010, 2018, and 2019) which have had adverse impacts on various sectors, such as agriculture, forestry, water management, health, and ecosystems. During the last few decades, projections of the impact of climate change on hydroclimatic extremes were often capable of reproducing changes in the characteristics of these extremes. Recently, the research interest has been extended to include reconstructions of hydro-climatic conditions to provide historical context for present and future extremes. While there are available reconstructions of temperature, precipitation, drought indicators, or the 20th century runoff for Europe, long-term runoff reconstructions are still lacking (e.g, monthly or daily runoff series for short periods are commonly available). Therefore, we considered reconstructed precipitation and temperature fields for the period between 1500 and 2000 together with reconstructed scPDSI, natural proxy data, and observed runoff over 14~European catchments to calibrate and validate the semi-empirical hydrological model GR1A and two data-driven models (Bayesian recurrent and long short-term memory neural network). The validation of input precipitation fields revealed an underestimation of the variance across most of Europe. On the other hand, the data-driven models have been proven to correct this bias in many cases, unlike the semi-empirical hydrological model GR1A. The comparison to observed historical runoff data has shown a good match between the reconstructed and observed runoff and between the runoff characteristics, particularly deficit volumes. The reconstructed runoff is available via figshare, an open source scientific data repository under the DOI https://doi.org/10.6084/m9.figshare.15178107, (Sadaf et al., 2021).

Water resource management is of paramount importance for sustainable agricultural and socioeconomic development. Agriculture is also one of the prominent factors responsible for the deterioration in the water quality mostly due to poor water management practices and lack of proper knowledge about soil-plant-atmosphere relationship. As such, optimally designed techniques and careful selection of irrigation system can ensure high efficiency and uniform distribution of applied water. Advanced planning and proper management of water could lead us towards sustainable agricultural development with optimal crop production even under physical, environmental, financial and technological restrictions. Therefore, to discuss some of the irrigation-through-computer approaches as a tool for better agricultural water management in this report, we present a detailed description of some of these advanced techniques including decision support systems such as Hydra, Hydrus, DSSAT, CropSyst and MOPECO and irrigation practices such as drip, sprinkler and mulching systems.

The study deals with probabilities of transitions from arid to humid environment and vice versa in Europe. Aridity index, defined as a ratio of potential evapotranspiration and precipitation and representing the ratio between energy availability and water availability, is used to characterize humid (wet) and arid (dry) regions and allows us to study transitions between individual periods (wet-wet, wet-dry, dry-dry, dry-wet). Three gridded datasets – CRU (UEA, 2020), E-OBS (ECAD, 2020) and ERA5 (ECMWF, 2020) – are used for this purpose. The aim of the study is to compare the three datasets as to transitions between wet and dry conditions, which are determined according to the aridity index, and evaluate the variability in Europe over 1950–2019. The changes in the aridity index since 1950 are found to be most pronounced in Northern and Central Europe. references: ECAD, 2020: E-OBS gridded dataset, available from . UEA, 2020: University of East Anglia – Climatic Research Unit, available from . ECMWF, 2020: European Centre for Medium-Range Weather Forecasts – ERA5, available from .

In present paper we compare the reconstructed gridded seasonal precipitation (P) and temperature (T) for Europe [1,2] to the available station data from the GHCN [3,4] network going back to 1800. The basic statistical properties at various time-scales ranging from 1/4 to 30 years are examined. It is shown, that there are significant biases in the reconstructed P and T and the bias in mean and variability considerably vary over the time-scales. The same applies for considered drought indices. We further investigate how the simulation of hydrological model driven by reconstructed data compares to that based on station data and runoff from GRDC database. In addition, a set of data-driven methods is used to link the reconstructed and observed P and T data to observed runoff, the results are validated and a reconstruction back to 1500 is provided. Finally, we check to what extent the raw proxy data can be used for drought reconstruction.[1] https://doi.org/10.1007/s00382-005-0090-8[2] https://doi.org/10.1126/science.1093877[3] https://doi.org/10.1175/JCLI-D-18-0094.1[4] doi:10.7289/V5X34VDR

Currently, there are changes in the hydroclimatic system, with most of Europe affected by droughts. Recent reconstructions on historical precipitation and temperature fields can be used for determination of impacts of meteorological, hydrological and agricultural droughts. Those reconstructions are available for European continent in gridded form (Casty et al.,2007). Aridity index, defined as a fraction of potential evapotranspiration and precipitation, can be used for characterization of humid – wet -- and arid – dry -- regions. It represents the ratio between energy availability and water availability. This study deals with conditional probabilities of transitions from arid to humid environment and vice versa. The aridity index was used to determine the transitions annual basis for the European continent for the period 1766 - 2015. The probabilities were calculated for each year, and for 10-year, 20-year and 30-year periods. It is shown that the recent droughts followed the drying of substantial part of Europe starting in 2014 (Hanel et al., 2018). The changes are most pronounced in Northern and Central Europe. references: Casty C., Raible Ch. C., Stocker T. F., Wanner H., Luterbacher J., 2007: A European pattern climatology 1766-2000. Climate Dynamics 29. 791-805. Hanel, M., Rakovec, O., Markonis, Y., Máca, P., Samaniego, L., Kyselý, J., Kumar, R., 2018: Revisiting the recent European droughts in a long-term perspective. Scientific Report 8, 9499.

The SCATSAT-1 satellite data can be used for various applications in the field of agriculture. The main aim of the study is to investigate the water cloud model (WCM) for backscattering simulation by using the field-measured soil moisture in order to validate the SCATSAT-1 measured backscattering. WCM requires various input datasets for simulation of backscattering such as vegetation parameters A and B and soil parameters C and D, which can be estimated by Non-linear least square fitting method by using with experimental dataset. The results showed that the simulated WCM values are well correlated with the backscattering of SCATSAT-1 satellite data. However, it can be further improved when each parameter of WCM is generated by using the ground-based measurements. In this study, some progress has been made toward backscattering simulations using the SCATSAT-1; however, it can be further refined with the advancement in the retrieval algorithms and sensor sensitivity.

Hyperspectral acquisition provides the spectral response in narrow and continuous spectral channel. The high number of contiguous bands in hyperspectral remote sensing provides significant improvements in assessing subtle changes as compared to the multispectral satellite datasets in context of spectral resolution. Therefore, the main goal of the present research is to evaluate the sensitivity of the artificial neural networks (ANNs) for chlorophyll prediction in the winter wheat crop using different hyperspectral spectral indices. For evaluating relative variable significance in the study, the Olden's function has been applied. The Lek's profile method is used for sensitivity analysis of ANNs for chlorophyll prediction using the vegetation indices such as Red Edge Inflection Point (REIP), Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), and Structure-Insensitive Pigment Index (SIPI) derived from hyperspectral radiometer. The analysis indicates a high sensitivity of SAVI followed by NDVI, REIP and SIPI for chlorophyll retrieval using ANNs. The statistical performance indices obtained from calibration (RMSE = 0.27; index of agreement = 0.96) and validation (RMSE = 0.66; index of agreement = 0.83) suggested that the ANN model is appropriate for chlorophyll prediction with good efficiency. The outcome of this work can be used by upcoming hyperspectral missions such as Airborne Visible Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) and Hyperspectral Infrared Imager (HyspIRI) for large-scale estimation of chlorophyll and could help in the real-time monitoring of crop health status.

Annual runoff is important information on water balance in the catchment and large river basin scale. It forms the boundary conditions for mathematical modelling of hydrological balance on a finer temporal and spatial scale. It is important for the assessment of climate change on water resources. Currently, there are several datasets on global gridded runoff fields available. GRUN and E-RUN provide monthly estimates of runoff rate with the spatial resolution of 0.5 degree. The GRUN is global dataset and E-RUN is covering Europe 1,2.In this study, we evaluate the capability of paleoclimate reconstructions on precipitation, PDSI, and temperature, which are available in the form of gridded fields, to estimate annual surface runoff using selected machine learning techniques. For this purpose, we use as a benchmark runoff information GRUN and E-RUN data sets. Both data are aggregated on the annual time scale for the period 1902 – 2014 (GRUN) and 1952-2015 (E-RUN). Following machine learning algorithms were tested: Random forests, SVM, MLP, LDA and Extra Trees. Reconstructed precipitation, temperature, PDSI3 and runoff estimated using selected Budyko models with different spatial aggregation served as inputs4–7 . Different combinations of inputs were analysed.Our results show that the estimated surface runoff is in good agreement with E-RUN and GRUN datasets for analysed periods. The result and newly tested approach based on derived machine learning models can be further applied to the estimation of paleoclimatic reconstructions of runoff fields. References: 1. Ghiggi, G., Humphrey, V., Seneviratne, S. I. & Gudmundsson, L. GRUN: an observation-based global gridded runoff dataset from 1902 to 2014. Earth Syst. Sci. Data 11, 1655–1674 (2019). 2. Gudmundsson, L. & Seneviratne, S. I. Observation-based gridded runoff estimates for Europe (E-RUN version 1.1). Earth Syst. Sci. Data 8, 279–295 (2016). 3. Cook, E. R. et al. Old World megadroughts and pluvials during the Common Era, Sci. Adv., 1, e1500561. (2015). 4. Schreiber, P. Über die Beziehungen zwischen dem Niederschlag und der Wasserführung der Flüsse in Mitteleuropa. Z Meteorol 21, 441–452 (1904). 5. Ol’Dekop, E. M. On evaporation from the surface of river basins. Trans. Meteorol. Obs. 4, 200 (1911). 6. Turc, L. Le bilan d’eau des sols: relations entre les précipitations, l’évaporation et l’écoulement. (1953). 7. Pike, J. G. The estimation of annual run-off from meteorological data in a tropical climate. J. Hydrol. 2, 116–123 (1964).

Several studies have revealed that rainfall and temperature are highly correlated with malaria spread. There are several studies relating the combined effect of hydrological and meteorological information for the malaria diseases 1–4 . In this study, attempts are being made for assessing the combined effect of hydro-meteorological variables on malaria disease at the regional scale. It reveals that evaporation is one of the essential climatic variables in this context, which is jointly derived by hydrological and meteorological variables. To our best knowledge, there are very few studies which have been performed to analyse the relations between malaria and the ratio of precipitation (P) and actual evaporation (AET). This study analyses the impact of the ratio of P and actual AET on malaria diseases. The work has performed at regional scale using annual data of malaria disease over the Tirap district of Arunachal Pradesh in India. Annual P data from Indian Meteorological (IMD) and GRUN 5 global surface runoff during the period of 1995 to 2012 are used for this analysis. The AET was estimated as difference e between P and runoff time series. The AET and P relationship with Plasmodium vivax (PV), Plasmodium falciparum (PF) is analysed. The sum of PV and PF is BSB indicator, it shows the total number of people affected by malaria. The study has revealed that fraction P/AET is negatively correlated with PV, PB and BSB. In comparison to hydrological and meteorological variables like P, surface runoff, AET and AET/P which are mostly positively correlated with BSB, PV and PF. This preliminary result will be further explored in order to find a connection on improving the forecast of malaria diseases using hydrometeorological inputs for better health management In the studied district.

Soil moisture represents a vital component of the ecosystem, sustaining life-supporting activities at micro and mega scales. It is a highly required parameter that may vary significantly both spatially and temporally. Due to this fact, its estimation is challenging and often hard to obtain especially over large, heterogeneous surfaces. This study aimed at comparing the performance of four widely used interpolation methods in estimating soil moisture using GPS-aided information and remote sensing. The Distance Weighting (IDW), Spline, Ordinary Kriging models and Kriging with External Drift (KED) interpolation techniques were employed to estimate soil moisture using 82 soil moisture field-measured values. Of those measurements, data from 54 soil moisture locations were used for calibration and the remaining data for validation purposes. The study area selected was Varanasi City, India covering an area of 1535 km(2). The soil moisture distribution results demonstrate the lowest RMSE (root mean square error, 8.69%) for KED, in comparison to the other approaches. For KED, the soil organic carbon information was incorporated as a secondary variable. The study results contribute towards efforts to overcome the issue of scarcity of soil moisture information at local and regional scales. It also provides an understandable method to generate and produce reliable spatial continuous datasets of this parameter, demonstrating the added value of geospatial analysis techniques for this purpose.

The leaf area index (LAI) is a crucial parameter that governs the physical and biophysical processes of plant canopies and acts as an input variable in land surface and soil moisture modeling. The ScatSat-1 is the latest microwave Ku-band scatterometer mission of Indian Space Research Organization (ISRO), provides data at a higher temporal and spatial resolution for various applications. Due to its all-weather operational capability, it could be used as an alternative to the optical/IR sensors for the LAI estimation. In the technical literature domain, no testing has been done to estimate the LAI using ScatSat-1 scatterometer data. Therefore, the objective of this study is to retrieve the LAI using the ScatSat-1 backscattering by modifications of two different models viz. water cloud model (WCM) and the recently developed Oveisgharan et al. model and compared against the PROBA-V, MODIS, and ground-based LAI products. To assess the performance of these models, coefficient of determination ( $R<^>{2}$ ), root-mean-squared error (RMSE) and bias are computed. For Oveisgharan et al., the values of $R<^>{2}$ , RMSE and bias were obtained as 0.87, 0.57 m(2)m(-2), and 0.05 m(2)m(-2) respectively, whereas for WCM model, the values were found as 0.82, 0.67 m(2)m(-2), and 0.32 m(2)m(-2) respectively. This investigation showed that the modifications in Oveisgharan et al. model provide marginally better results in the retrieval of LAI using ScatSat-1 data than the WCM model. The models' limitation may be less serious for crop management studies because the majority of crops attains its maturity at LAI values less than 6 m(2)/m(2).