Gridded precipitation products are naturally uncertain due to different factors like measurement errors, precipitation undercatch, and errors introduced by the different interpolation algorithms. Hydrological modelling is significantly influenced by the uncertainty in meteorological data. The effectiveness of advanced data assimilation systems and other tools in land surface and hydrological modeling is limited due to the lack of quantitative estimates of uncertainty for hydro-meteorological data. We have created a high-resolution ensemble precipitation dataset, Indian Precipitation Ensemble Dataset (IPED), at 0.1° resolution from 1991 to 2023 over the Indian region. This dataset is derived from observation-based precipitation data using a locally weighted linear regression algorithm. However, its potential in evaluating the uncertainties in hydrological modeling is yet to be explored. This study investigates the impact of uncertainties in precipitation data on hydrological modeling across 18 basins in India by utilizing IPED dataset into Indian Land Data Assimilation System (ILDAS), a hydrologic hydrodynamic model. The ensemble simulation employs IPED's probabilistic estimates to perform uncertainty analysis. The results highlight the extent, spatial distribution, and magnitude of uncertainties in precipitation and streamflow variables from 1991 to 2023. A key finding is that uncertainties in streamflow are significantly affected by uncertainties in precipitation. Both spatial and temporal averaging have distinctive effects on the uncertainty of different variables across the study area. In conclusion, this investigation provides a thorough understanding of how IPED dataset improves and quantifies the uncertainties in streamflow over Indian river basins that arise from precipitation data uncertainties. 

How to cite: Peringiyil, A., Saharia, M., and Deka, P.: The Impact of Precipitation Uncertainty on Hydrological Modeling: An Analysis over Indian basins, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-780, https://doi.org/10.5194/egusphere-egu25-780, 2025.

Over time, many hydrologic models have been developed, ranging from physical-based to system-theoretic approaches. A simplified system-level understanding of the hydrologic process can be represented using conceptual models. Budyko framework, a lumped first-order representation of precipitation partitioning, has been widely applied to evaluate water balance. The Budyko equations are characterized by specific parameters representing the climatic and catchment characteristics. Therefore, the reliability of the framework is highly influenced by the accurate estimation of the parameters. The different input data sources can lead to varied estimates of the model parameter. The study examines the parametric uncertainties arising from various meteorological data sources. With the uncertainty attributed to precipitation, temperature, and potential evapotranspiration data sources, the study highlights the need to select and validate data sources carefully. In addition, the study highlights the challenges in parameter estimation and in capturing the underlying hydrologic processes within the Budyko framework. 

How to cite: Raghava Panikkar, R. U. and Srivastav, R.: Impact of Meteorological Data Variability on Budyko Parameter Estimations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8367, https://doi.org/10.5194/egusphere-egu25-8367, 2025.

Groundwater dependent ecosystems (GDEs) exhibit complex spatiotemporal dependency on multiple variables including precipitation, surface and soil water, and groundwater availability. While spatial relationships have been explored in many previous studies, the temporal lag between GDE vegetation health and hydrology is not well understood across a scale of anthropogenic influence. Data from three groundwater dependent ecosystems (Arkansas River in Colorado, USA; South Platte River in Colorado, USA; and the San Pedro River in Arizona, USA), each differing in magnitude of anthropogenic encroachment, are collected and analyzed to discern which hydrologic variables have the strongest correlation in time with phreatophyte health (e.g., Populus fremontii and Populus deltoides). Phreatophyte health serves as a surrogate for overall GDE health. The response variable used to estimate GDE health is a daily time series (2016-2023) of plant health that is quantified by computing the normalized difference vegetation index (NDVI) using Planet Imagery (3-m spatial resolution). The covariates are groundwater depth (meters below land surface), discharge (cubic meter per second), precipitation (mm) and temperature (deg. C). We examine the temporal dependence between the response variable and each covariate by first pre-whitening each data series using a Bayesian hierarchical autoregressive model and then applying a cross-correlation analysis to the residuals. Initial results indicate the correlation in time between NDVI and groundwater depth are highest at time t and t-1, regardless of the magnitude of anthropogenic influence, on a monthly time scale. We anticipate that at higher temporal frequency (i.e., daily) the correlation between the response and the covariates will show distinct patterns owing to alterations in the natural flow regime from agricultural practices and reservoir management. Our research highlights the complex temporal relationship between phreatophyte health and hydrology in groundwater dependent ecosystems encroached by differing magnitudes of anthropogenic influence. This research can aid conservationists in understanding the lagged impact that environmental flows, drought, land use change, and pumping-induced water table decline can have on phreatophyte health. Additionally, this research informs the choice of temporal scale (i.e., daily or monthly) at which to model groundwater dependent ecosystems in spatially distributed parameter modeling schemes.

How to cite: Lurtz, M., Ronayne, M., and Huete, A.: Comparison of temporal dependence between phreatophyte health and hydrometeorological variables for three groundwater dependent ecosystems, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11726, https://doi.org/10.5194/egusphere-egu25-11726, 2025.

The main goal of this study was to determine the reliability of the double-clustering approach using hierarchical cluster analyses (HCA) to delineate the abundance of major and trace elements in groundwater of an arid watershed in northcentral Mexico, where the As geogenic contamination leads an important deterioration of the groundwater quality. For that, fifty-five groundwater samples were collected from wells within the watershed and physicochemical parameters such as pH, conductivity, temperature, and ORP were measured in situ. In the laboratory, major ions, metalloids, and trace elements were measured by ion chromatography and ICP-MS. For the development of the double-clustering approach, all the data were log-transformed and standardized to approximate normality. A first HCA was performed for clustering variables. This HCA produced six groups of variables. Then, an HCA of cases was applied to each group of variables, which delineated maps describing the magnitude of each group of variables in the aquifer. In general, the double-clustering approach was effective for identifying processes (lithogenic/anthropogenic) controlling the abundance of major and trace elements in groundwater of the above-mentioned watershed. This method identified hotspot of As, Sb, Ge, V, and W in the alluvial aquifer, suggesting a concomitant release to these elements to groundwater. In addition, the applied approach identified mountainous areas with high concentrations of HCO₃^-, Ca, Mg, K, Sr, Rb, Ga, Ba, Cs, Pb, Ni, Y, and U, indicating that the weathering of carbonate/silicate rocks plays an important role in the abundance of these ions/elements in groundwater. The double-clustering approach was also successful in delineating disperse areas where the salinity and the levels of Na, Cl^-, SO₄^2-, NO₃^-, B, Li, and the chalcophilic elements Cu, Re, and Se in groundwater were elevated, mainly related to processes such as evaporite dissolution and increasing concentrations due to the irrigation return flow. Overall, the double-clustering was also compared with spatial statistical techniques such as the Moran Index and the Local Indicator for Spatial Association (LISA), which demonstrated that the double-clustering is a powerful tool capable of visualizing zones where specific natural/anthropogenic processes may threaten the groundwater quality.

How to cite: Mora, A., Torres-Martínez, J. A., and Mahlknecht, J.: Mapping trace element abundance in groundwater using a double-clustering approach, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3875, https://doi.org/10.5194/egusphere-egu25-3875, 2025.

Understanding compound events involves analyzing the interactions between different climate variables, assessing their probability of co-occurrence, and evaluating their cumulative impacts. 

This field of study has gained attention in recent years due to the increasing frequency and severity of extreme weather events, which are often linked to climate change.

Regionalization, in the study of compound events, refers to the process of tailoring analyses and models to specific geographic regions. This approach is vital because the characteristics and impacts of compound events can vary significantly across different areas due to variations in climate, geography, socio-economic conditions, and infrastructure resilience. Regionalization methods seek to identify sub-regions that display similar patterns in the variables of interest.

The objective of this talk is to offer a regionalization of intricate spatial climatological datasets, particularly when considering compound extremes.

To this end, a clustering algorithm is introduced to group time series of maxima for paired random variables observed at different stations. The approach requires different types of dissimilarity measures. In particular, it relies on the copula approach and on the use of the related Kendall distributions that are compared with the Wasserstein distance.

As an illustration, using data on daily maximum temperature (in Celsius) and daily maximum evapotranspiration (mm/day) from the ERA5 dataset, collected across various municipalities throughout Italy, enhanced estimation of climate-related metrics at specific locations are obtained by leveraging regions with statistically similar characteristics.

How to cite: Castrovilli, R., Durante, F., Gallo, D., and Salvadori, G.: Regionalization methods for compound events based on Wasserstein distance, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19012, https://doi.org/10.5194/egusphere-egu25-19012, 2025.

It is frequently the case that direct and indirect measurements have to be combined to deliver meaningful estimates of the variable of interest. Linear co-regionalization, which assumes that all variables share common spatial structures, has widely been used in geostatistics to model the correlated spatial random fields. The underlying linearity assumption, however, is restrictive with respect to the choice of the direct and cross variograms, as it assumes very similar spatial structure for the direct and indirect variables. In this contribution, a new method of non-linear co-regionalization based on Fourier transformation is presented. First, the coherence of the corresponding fields based on their power spectra is introduced. The coherence gives a variogram-dependent upper and lower limit for the correlation of the random fields. The direct variograms of the two fields depend on their phase spectrum. The phase differences of these phase spectra determine the cross-variogram. A simulation method for generating correlated random fields with given direct and cross variograms is presented. The method allows the use of different models for the direct variograms as well as for the cross variogram. Further, the method enables the consideration of non-Gaussian copula-based spatial features, such as different types of spatial asymmetries. This enables the simulation of correlated fields with value-dependent correlations. A real world and various theoretical examples with different Gaussian and non-Gaussian copula-based dependence structures will be used to illustrate the methodology and its flexibility.

How to cite: Hoerning, S., Lauzon, D., and Bárdossy, A.: Spectral methods for non-linear co-regionalization, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5250, https://doi.org/10.5194/egusphere-egu25-5250, 2025.

The analysis of spatial processes in environmental and hydrological sciences is often informed by remote sensing observations, provided by satellite or airborne sensors. However, the raw data obtained by such means can present gaps, for instance due to the orbital characteristics of a satellite or to specificities of the aircraft flight path. Many applications and modeling workflows require complete, gapless data. Geostatistical approaches are often used to fill these data gaps, however the sheer size of modern remote sensing datasets make the application of traditional geostatistical approaches challenging due to computational constraints (high-resolution, broad spatial coverage) and to data characteristics (complexity of features, non-stationarity).

In this work, we develop a new approach based on multiple-point geostatistics to fill gaps in very large and non-stationary data sets. It is based on a strategy of partition of the domain in overlapping tiles. This makes the problem computationally more affordable, while additionally enabling parallelization. It also alleviates issues related to non-stationarity, since the assumption of stationarity is more likely to be valid on a small tile than on a large domain.

The approach is illustrated on a dataset that is based on acquisitions by the AVIRIS-NG hyperspectral airborne sensor in Switzerland. The data present significant gaps, and at the same time the domain is extremely large, comprising over 300 million pixels. The simulation results are visually realistic and corroborate independent validation data.

How to cite: Mariethoz, G., Gerber, L., Lambiel, A., and Külling, N.: A method for gap-filling very large spatial datasets: application to AVIRIS-based airborne data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4442, https://doi.org/10.5194/egusphere-egu25-4442, 2025.

Abstract

Typhoons, accompanied by strong winds and heavy rainfall, are among the most devastating natural disasters, causing significant loss of life and property damage. To prepare for disaster situations caused by typhoons, predicting Typhoon-induced Accumulated Rainfall (TAR) is crucial. Many previous studies have attempted to predict TAR by evaluating the similarity between typhoon tracks and rainfall patterns using various methods. In this study, we utilized time series data evaluating methodologies (e.g. Dynamic Time Warping, Cosine Similarity, etc.), for assessing similarity of typhoon tracks. Using the best typhoon track data from the Regional Specialized Meteorology Center, Tokyo from 1979 to 2024 (1,157 in the Western North Pacific and the South China Sea), and National Hurricane Center (727 in Atlantic and 816 in Northeast and North Central Pacific), and precipitation data from the National Oceanic and Atmospheric Administrations Climate Prediction Center. The similarity of typhoon tracks was evaluated based on the latitude and longitude of the typhoon center and various meteorological properties such as pressure, translation speed. Typhoons with highly similar tracks were clustered, and the average TAR of the clustered typhoons was used to predict the TAR. To optimize the number of typhoons included within a single cluster, we determined the Optimal Ensemble Number (OEN) based on the root mean square error between observed TAR and predicted TAR. In this process, each typhoon’s trajectory and region are considered. Using OEN, we predicted TAR and validated the performance of our method by selecting typhoons which have different tracks and rainfall characteristics. The results demonstrated that the proposed methodology achieved performance comparable to that of previous studies. These findings suggest that methodologies for evaluating the similarity of time series data can comprehensively account for not only typhoon tracks but also unique meteorological attributes, contributing to improved TAR prediction.

Acknowledgement

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (RS-2024-00334564).

How to cite: Ku, S., Baik, J., Byun, J., Kim, J.-S., and Jun, C.: Optimizing typhoon-induced accumulated rainfall prediction through track similarity and meteorological properties, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5147, https://doi.org/10.5194/egusphere-egu25-5147, 2025.

Understanding hydrological extremes and the factors that condition them is crucial to 
promoting the adaptability of urban watersheds. Furthermore, few studies investigate 
the spatial variability of these factors and their explanatory power. This study analyzed 
the Tamanduateí River Basin, located in São Paulo, Brazil, using data from multiple 
sources to explore the spatial relationships between inundation points occurrence and 
geo-environmental factors. Spatial autocorrelation models, as Global and Local 
Moran’s Index were applied in these points to identify patterns and areas most 
susceptible to these events. To assess the explanatory power and interaction among 
11 geo-environmental factors - including Topographic Position Index (TPI), Terrain 
Roughness Index (TRI), Sediment Transport Index (STI), Stream Power Index (SPI), 
Topographic Wetness Index (TWI), Drainage Density (DD), Height Above Nearest 
Drain (HAND), Slope, Hillshade, Distance to River (DR) and Cumulative Expanded 
Area (AEXPAND) - the Geodetector geostatistical tool was used. Subsequently, the 
Multiscale Geographically Weighted Regression (MGWR) algorithm was used to 
examine the most relevant factors, allowing a detailed analysis of the spatial interaction 
among them. The results indicated a strong spatial dependence of inundation points 
occurrence and showed significant simultaneous effects of the factors analyzed. Flat 
areas with consolidated anthropogenic use had a higher incidence of these events, 
with variables such as HAND and AEXPAND standing out. These findings reinforce the 
importance of topography and land use in the dynamics of hydrological extremes. This 
study offers an integrated approach to understanding the spatial heterogeneity of 
hydrological extremes in urban areas, contributing to the mapping of these events. In 
addition, the proposed methodology can be replicated in other regions, especially 
those with scarce spatial data, expanding the possibilities for preventing and adapting 
to extreme events in different urban contexts.

How to cite: Mira, Í., Santos, L., Monteiro, A. M., and Rennó, C.: Multivariate spatial analysis of hydrological extremes in urban watersheds, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5186, https://doi.org/10.5194/egusphere-egu25-5186, 2025.