HS3.1

CrowdWater is a citizen science project that uses a mobile phone application (app) to collect hydrological data. The project aims to develop and test methods for hydrological measurements that do not require sensors. So far, the app is used to collect data about stream water levels, the state of intermittent streams, soil moisture, and plastic pollution. The app can also be used to record general characteristics of streams, such as the color, the presence of fish and other living beings, or water pollution.

Perhaps without realizing it, many people make visual water quality assessments for their daily decisions. Based on visual aspects of the water, people decide whether the water is suitable for swimming or drinking. This presents an opportunity to explore the potential of a visual approach through citizen science-based water quality observations. Water color and clarity are one of the most frequently used indicators for the visual assessment of water quality. Remote sensing studies have shown that water color is changing in many areas, which suggests that it is useful to characterize visual water quality aspects throughout time and to relate these to perceptions of water quality and how this affects water use. However, most of the time this perception is not just based only on the current characteristics of the water but also on local environmental knowledge, such as the presence of outlets that discharge waste water or sewage overflows, the water quality in the past, etc.

This presentation will describe the visual approach for water quality that fits within the CrowdWater philosophy of not using any sensors so that observations can be made by anyone at any place. We will also present a first evaluation of the method. This includes discussing how consistent people are in their assessment of water color and whether they can assess differences in water color over time or between sites.

How to cite: Blanco Ramírez, S., van Meerveld, I., Seibert, J., Scheller, M., and Schwarzenbach, F.: A visual approach for water quality monitoring within the CrowdWater project, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-523, https://doi.org/10.5194/egusphere-egu22-523, 2022.

In recent decades, the uncertainty associated with characteristics (i.e., frequency, intensity, severity, and duration) of extreme events (e.g., droughts, floods) has increased considerably due to the changing global climatic condition and intensification of anthropogenic activities. Effective in-situ monitoring of the hydrometeorological drivers (e.g., precipitation, temperature) is crucial for precise prediction/forecasting and early warnings to initiate measures for mitigating the adverse effects of these extreme events.  However, due to the increased availability of satellite-based data products and economic constraints, the density of in-situ gauges has reduced drastically over the past few decades. Against this backdrop, this study proposes a multivariate hydrometeorological gauge network design methodology to facilitate integrated monitoring of dry and wet conditions. It harnesses the advantages of multi-objective optimization and fuzzy concepts and involves multi-level clustering and the use of multiple ground- and satellite-based hydrometeorological products.  The multi-level clustering is based on (i) a newly proposed multi-objective Non-dominated Sorting Genetic Algorithm III (NSGA-III) based fuzzy optimization clustering and (ii) fuzzy ensemble clustering. The key stations in the designed network were selected based on the Drought/Wetness Gauge Demand Index (DWGDI), which accounts for the region's drought/wetness characteristics and crop yield.  It also offers scope to consider additional attributes based on the specific purpose of the network design. The potential of the proposed methodology is illustrated through Monte Carlo simulations on a hypothetical region and a case study on Karnataka state (~191,791 km²) in India to arrive at gauge network monitoring three hydrometeorological variables (precipitation, maximum and minimum temperature, and soil moisture). A random forest-based merging procedure is considered to obtain hydrometeorological time-series at ungauged locations using ground-based measurements and multiple gridded/satellite-based products (CRU, CPC, IMD, CHIRPS and IMERG). Overall, the proposed network design methodology appears promising for application to small as well as large data-sparse areas. To the best of our knowledge, this is the first study of its kind, which proposes a multivariate gauge network design procedure for integrated monitoring of dry and wet conditions. The proposed methodology yielded wet/dry condition-specific monitoring networks for the Karnataka state. Additionally, the key stations crucial for monitoring both wet and dry conditions are identified. The counts of precipitation, temperature and soil moisture stations in the network designed for monitoring (i) dry conditions are 1059, 1059, and 552, respectively, (ii) wet conditions are 1144, 1144, and 664, respectively. Stations in the two networks were prioritized by assigning ranks based on DWGDI. The information could be helpful to decision-makers in identifying potential locations for the installation of new gauges accounting for budgetary constraints. The real-time observations from the designed gauge network could be helpful for various purposes, such as better water management to meet irrigation demands, monitoring droughts and floods, and forecasting natural hazards like wildfires, soil erosion, and landslides. 

How to cite: Vijay, S. and Venkata Vemavarapu, S.: An Approach to Integrate Ground- and Satellite-based Products for Multivariate Hydrometeorological Network Design to Monitor Dry and Wet conditions, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-610, https://doi.org/10.5194/egusphere-egu22-610, 2022.

Hydrological data are often sparse and incomplete for large northern watershed with difficult access. Landscape hydrology approaches are useful for the indirect assessment of their hydrological characteristics by analysing the landscape properties of the watersheds. In this study, we use unsupervised Geographic Object-Based Image Analysis (GeOBIA) paired with the Fuzzy C-Means (FCM) clustering algorithm to produce a total of seven high-resolution territorial classifications for the 1985-2019 time-period. Each classification spans 5-year period and is based on key hydro-geomorphic metrics. Our application site is the George River watershed (GRW), draining a 42 000 km² area and is located in Nunavik, northern Québec (Canada). The retrieved subwatersheds within the GRW are used as the objects of the GeOBIA and are classified in function of their hydrological similarities.

First, classification results for the time-period 2015-2019 show that the GRW is composed of two main types of subwatersheds distributed along a latitudinal gradient. This indicates differences in water balance, and hydrological regime and response. Second, six other classifications are then computed for the period 1985-2014 to investigate past changes in hydrological behavior. The seven-classification time series present an expansion of the southern-type subwatersheds northwards, principally along the George River’s main channel. This expansion is due to increases of (i) vegetation production and (ii) moisture content in soil and canopy. These are the major changes occurring in the land cover metrics of the GRW. We speculate that a rise in vegetation production contributes to evapotranspiration increase and therefore induces changes in water balance, which could explain the measured decrease of about 1% in the George River’s discharge since the mid-1970s.

How to cite: Sicaud, E., Franssen, J., Dedieu, J.-P., and Fortier, D.: Remote Sensing and Clustering Applications in Landscape Hydrology: Characterizing a Subarctic Watershed in Nunavik (Canada), EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-716, https://doi.org/10.5194/egusphere-egu22-716, 2022.

Detailed investigations across time scales and variable types can progress our understanding of hydroclimate. In this work, we analyse temperature, precipitation and streamflow time series at nine time scales (i.e., the 1-day, 2-day, 3-day, 7-day, 0.5-month, 1-month, 2-month, 3-month and 6-month ones). The analyses are performed over the continental United States, and in terms of temporal dependence, temporal variation, “forecastability”, lumpiness, stability, nonlinearity (and linearity), trends, spikiness, curvature and seasonality, among others. Thus, they facilitate extensive characterizations of the cross-scale properties of the temperature, precipitation and streamflow variables. Based on these characterizations, various similarities and differences are identified between the examined variables regarding the evolution patterns of their features with increasing (or decreasing) time scale. Moreover, the computed features are used as inputs to unsupervised random forests to detect any meaningful clusters between the time series. The clustering is performed separately for each set {time scale, variable type}, and allows the investigation of the spatial variability of the temperature, precipitation and streamflow features across the examined continental-scale region and across time scales, with the spatial patterns emerging from it being largely analogous across time scales. Lastly, explainable machine learning is applied to compare the features with respect to their importance-usefulness in the clustering. For most of the features, this usefulness can vary to a notable degree across time scales and variable types, thereby implying the need for conducting multifaceted time series characterizations for the study of hydroclimatic similarity.

How to cite: Papacharalampous, G., Tyralis, H., Markonis, Y., and Hanel, M.: Hydroclimatic time series analysis and clustering at multiple time scales, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-936, https://doi.org/10.5194/egusphere-egu22-936, 2022.

Both the grouping of hydroclimatic time series (often required, e.g., for technical and operational purposes) and the identification of spatial hydroclimatic patterns can be formalized and automated through algorithmic clustering methodologies. In this presentation, we focus on a new family of such methodologies that can be applied to various types of hydroclimatic variables (e.g., temperature, precipitation and streamflow) and at various temporal scales (e.g., the daily, monthly, seasonal, annual and climatic ones) with minimal adaptations. Aiming to exploit the largest part possible of the total information encompassed in the hydroclimatic time series, this family of clustering methodologies primarily relies on massive feature extraction, a concept sourced from the data science field. Once a compilation of numerous and diverse time series features (comprising autocorrelation, long-range dependence, entropy, temporal variation, seasonality, trend, lumpiness, stability, nonlinearity, linearity, spikiness, curvature and more features) has been computed, the clustering upon them is performed using a selected machine and statistical learning algorithm, with unsupervised random forests being an appealing choice for the task. Explainable machine learning can also be applied, as part of wider methodological frameworks, for ranking the features from the most to the least informative ones in obtaining the clusters, thereby facilitating the interpretability of the clustering outcomes in a comprehensive manner. We extensively discuss the above-outlined approach to hydroclimatic time series clustering emphasizing its main similarities and differences with the current well-established approaches in hydrology (e.g., from the catchment hydrology field), as well as its strengths and current limitations. Our discussions are well-supported by global-scale and other large-scale investigations, which have been conducted for temperature, precipitation and streamflow variables at several temporal scales.

How to cite: Papacharalampous, G. and Tyralis, H.: Feature-based clustering of hydroclimatic time series, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-937, https://doi.org/10.5194/egusphere-egu22-937, 2022.

Streamflow sensitivity to different hydrologic processes varies in both space and time. In numerical modeling of streamflow, this sensitivity manifests as parameter sensitivity, which is typically model-specific.

In this study, we apply a novel analysis over more than 3000 basins across North America enabling the estimation of the process sensitivities on streamflow based on basin characteristics that can be derived from physiographic and climatologic data without needing to perform the expensive sensitivity analysis itself. This continental-scale analysis allows for high-level conclusions as to the importance of water cycle components on streamflow predictions, as the analysis considers a flexible model structure rather than an individual model. This work derives the sensitivity of streamflow simulation to entire hydrologic processes rather than only specific parameters. Process sensitivities are computed and provided for each day of the year over a wide range of physiographic and climatologic regimes, enabling future hydrologic model improvement at the continental scale.

A few highlight results are: 1) Baseflow and other sub-surface processes are of low importance across North America- especially when time points of high flows are of interest. 2) Percolation, evaporation, and infiltration show very similar patterns with increased importance in South-eastern US and west of the Rocky Mountains. 3) Up to 30% of the overall model variability can be attributed to snow melt in regions that are snow dominated (Northern Canada and Rocky mountains). Potential melt shows a similar gradient as snow melt with sensitivities of above 60% in the Province of Quebec and the Rocky Mountains. 4) Direct runoff (quickflow) is the most sensitive of all hydrologic processes- especially in South-Eastern US it is responsible for more than 80% of the model variability. 5) The derived functional relationship to estimate the process sensitivities based on basin characteristics has predictive power of at least 0.8 in Pearson correlation coefficients based on more than 1000 basins used for validation.

How to cite: Mai, J., Craig, J. R., Tolson, B. A., and Arsenault, R.: The Sensitivity of Simulated Streamflow to Individual Hydrologic Processes Across North America, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-1827, https://doi.org/10.5194/egusphere-egu22-1827, 2022.

In recent years, artificial intelligence (AI) tools have gained popularity as forecasting and predictive tools able to approximate with high accuracy trends and outcomes in many fields such as robotics, climatology, and hydrology, including water resources. AI models have shown remarkable performances handling big data and dealing with their nonlinearity and nonstationary features in monitoring and forecasting water quality, complementing the traditional numerical water quality models that provide precise parametrizations of near-shore and off-shore processes and their complex interactions.

In this research, we examine the accuracy of different machine learning techniques in estimating and predicting dissolved oxygen concentration (DO) in water bodies. DO is a crucial water quality variable that influences the living conditions of all aquatic organisms requiring oxygen. Low DO concentration, when persistent, can cause eutrophic conditions, thus altering the normal nutrient cycle, favoring the formation of algal blooms and furtherly reducing water quality and affecting the entire ecosystems, also causing fish mortality.

The Random Forest (RF) and the generalized regression neural network (GRNN) are explored and compared. The two models are developed using high frequency in situ data collected from Andromeda Group, a leading company in the aquaculture sector in Greece, at four different stations in the Greek Mediterranean Sea. The input variables used for the two models are temperature and currents. The performances of the models are evaluated using root mean square errors (RMSE) and mean absolute error (MAE). The RF and GRNN showed similar performances, with the best fit obtained using the GRNN model. Results are also compared with a traditional numerical model developed with the DELFT3D-WAQ modeling suite. The AI models show better performances in estimating daily changes of the DO concentration and by being less computationally expensive than the numerical model, enhance the water quality monitoring and provide aquaculture and farmers managers with a forecasting tool.

Acknowledgments:

The work has been conducted within the framework of the HiSea project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 821934 and of the Water Harmony project, an ERA-NET WATERJPI Co-fund Action. Funding received via the Dutch Research Council - NWO Project number ENWWW.2018.1

How to cite: Spinosa, A. and El Serafy, G.: Enhancing the modeling of dissolved oxygen concentration using machine learning, a case study in the Greek Mediterranean Sea., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-1912, https://doi.org/10.5194/egusphere-egu22-1912, 2022.

Uncertainties in efficiently managing contaminated groundwater still pertains. Several in-situ remediation technologies can be applied to treat nitrate contamination; however, those cannot be applied at aquifer which has high hydraulic conductivity. In this study, we apply Simulation-Optimization modeling (S-O modeling) to suggest optimal in-situ flushing integrated to pump-and-treat design that can reduce and manage nitrate contamination of groundwater. MODFLOW-2005 and MT3D-USGS are used to simulate groundwater flow and nitrate transport. Genetic Algorithm(GA) is used to suggest proper well location and pumping rate for reducing nitrate contamination. Our optimization modeling based on the field tests conducted on a multi-layered aquifer. There is an uncontaminated lower aquifer and an upper aquifer contaminated with nitrate. At the first stage, we generated hydraulic gradient for collecting pre-existed contamination with installing pumping well at the lower aquifer and injection well at the upper aquifer contaminated with nitrate. In the second stage, 1 moving well of variable location and pumping rate was used for pump-and treat well. Cost function satisfies minimizing the total expenses of drilling, pumping, injection, water treatment, and penalty for violating nitrate concentration. The early stage result of S-O modeling shows nitrate moves along the groundwater flow and is captured at the moving well that is located in the center of nitrate plume.

Keyword: Simulation-Optimization modeling (S-O modeling) ∙ Remediation ∙ On-site flushing ∙ Groundwater modeling

Acknowledgement: This work was supported by Korea Environment Industry & Technology Institute(KEITI) through "Activation of remediation technologies by application of multiple tracing techniques for remediation of groundwater in fractured rocks" funded by Korea Ministry of Environment (MOE) (Grant number:20210024800002/1485017890), Korea Environment Industry & Technology Institute(KEITI) through the Demand Responsive Water Supply Service Program (RE20191097) funded by the Korea Ministry of Environment (MOE) and Korea Ministry of Environment as "The SEM projects; RE2020002470001/1485017133".

How to cite: Lee, S.-H., Park, I.-W., Lee, S.-S., and Lee, K.-K.: Application of Simulation-Optimization modeling to suggest optimal hybrid in-situ flushing for nitrate attenuation, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-3334, https://doi.org/10.5194/egusphere-egu22-3334, 2022.

Water plays a vital role in Earth’s ecosystems, hence, the management and monitoring of water resources are significantly important. Ireland has more than 12200 lakes, 3,192 river water bodies including rivers, streams, and tributaries that exceed in total 70,000 km. However, few of them are currently monitored (2% of lake and 1.8% of river). Earth observation (EO) has shown promise in understanding and monitoring water resources. However, the sizes of Ireland’s water bodies have remained a challenge for EO monitoring.

In this research, three different platforms (Landsat 8-OLI, Sentinel-2A and PlanetScope Ortho Scene 3B product) are used to calculate their individual spatial coverage of water bodies across Ireland and quantity their usefulness for water quality monitoring. To explore if water extraction methods impact the results, four different water extract methods (NDWI, NDVI, MNDWI, and AWE have been used to create water masks for Ireland. These water masks created for each platform were then compared with existing map of river network, lakes, and water monitoring stations. The results indicated that AWE’s water mask is the best performing extraction method compared to the existing maps, while the high-resolution platforms (Planetscope and Sentinel-2) clearly outperforms Landsat, Landsat is still able to detect at least 51.90% and 1.63% of rivers. This shows that even the coarser resolution Landsat imagery is useful in monitoring water quality across Ireland.

How to cite: Zhao, M. and O'Loughlin, F.: Mapping Ireland’s Surface Water from Space: Comparison of three remote sensing platforms , EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-3863, https://doi.org/10.5194/egusphere-egu22-3863, 2022.

Partitioning a dataset in multivariate analysis is one of the key points to better understanding the hydrological process. Different regions of a catchment may bring floods variously due to distinct types of floods or their simultaneous occurrence. Therefore, it is needed to determine the spatial extent floods brought together. In a multidimensional space, it is demanding to investigate floods. It is not clear which kind of clustering methods or dimension reduction techniques are appropriate for visualizing initial similarities among measured peaks. Two methods are applied to reduce dimensions and compare their differences in this research. Multidimensional scaling (MDS) and t-distributed Stochastic Neighbor Embedding (tSNE) are the employed models for 55 years of extreme floods in the Neckar catchment. MDS is based on Principal Component Analysis (PCA), which is a linear technique. While tSNE is a non-linear dimensionality reduction method. In theory, tSNE can handle outliers and perplexity and preserve the local structure of data. As a result, compared to the MDS, both methods react similarly in soliciting an additional algorithm to cluster data in 2D space. It is another challenge that has to be investigated in future research. Due to the fatter and heavier tails, the t-student distributions have a greater chance for extreme values than normal distributions. Therefore, tSNE can better visualize data in a high-dimensional space for assessing extreme events. However, these algorithms must be run in different climates and deal with distinct hydrometeorological variables.

How to cite: Modiri, E. and Bárdossy, A.: Multidimensional flood analysis challenges and similarities utilizing linear and non-linear approaches, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-3921, https://doi.org/10.5194/egusphere-egu22-3921, 2022.

In this study, the concept of nonlinear dynamics is used to predict runoff in the Savitri River basin, Maharashtra, India. Hourly runoff from four stations in the basin, namely Kangule, Bhave, Birwadi, and Kokkare, are studied. The nonlinear prediction method with a local approximation approach is employed, and one-hour-ahead runoff predictions are made. The method uses (1) reconstruction of the single-dimensional runoff series in a multi-dimensional phase space; (2) determination of the Euclidean distances between the reconstructed vectors; and (3) prediction using a nearest-neighbor local approximation approach, with consideration of different number of neighbours. For each of the four streamflow stations, data observed during the period 2000–2009 are used for phase space reconstruction, and predictions are made for the year 2010. Three statistical evaluation measures, correlation coefficient (CC), Nash-Sutcliffe efficiency (NSE), and normalized root mean square error (NRMSE), are used to determine the performance of the method. The prediction results for the four stations indicate very good accuracy, with the CC values ranging between 0.980 and 998, the NSE values between 0.961 and 0.995, and the NRMSE values between 0.010 to 0.014. The optimal embedding dimensions (i.e. the dimensions yielding the best predictions) for the Kangule, Bhave, Birwadi, and Kokkare are 9, 13, 6, and 9, respectively. These dimensions suggest that the complexity of the dynamics of hourly runoff in the Savitri River basin is medium-to-high dimensional. The outcomes from the present study are certainly encouraging to further enhance the application of nonlinear dynamic concepts for studying the runoff dynamics in the Savitri River basin.

Keywords: Runoff prediction, Nonlinear dynamics, Chaos, Phase space reconstruction, Local approximation prediction, Savitri River basin

How to cite: Saji, N., Jothiprakash, V., and Sivakumar, B.: Prediction of hourly runoff in the Savitri River basin in India: Use of a local approximation approach, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4930, https://doi.org/10.5194/egusphere-egu22-4930, 2022.

Direct measurement of evapotranspiration (ET) is costly and difficult to implement on a large scale. It is therefore a necessity to count on reliable approaches to estimate it. Among such approaches, Machine learning models (MLMs) are easily applicable and computationally inexpensive, especially for broadscale analyses. In this study, three different types of MLMs, namely Random Forest, Light Gradient Boosting Machine and Neural Networks are assessed for their estimation accuracy on unseen locations (i.e. generalization power). Estimates of ET from these MLMs are compared against direct observation from 143 eddy-covariance flux towers spanning across a broad range of climate and vegetation types. We initially hypothesized that the MLMs, provided that they are trained using data from a wide variety of climate and vegetation types, are able to accurately estimate ET on unseen locations (default experiment).  The MLMs are benchmarked against Penman, Priestley-Taylor, and Oudin ET formulas/models. The results show that the MLMs indeed perform satisfactorily on the majority of the test locations, but not in all of them, yielding on average a 15% lower normalized mean-absolute-error (NMAE) than the Priestley-Taylor formula. Moreover, we compared the performance of the MLMs trained and tested using different data splitting strategies. When training and testing data are not spatially separated, the results show that the Random Forest model has a 7% lower NMAE compared to when the spatial separation is done (the default experiment). This suggests that the MLMs are prone to overfit to site-specific patterns that might not be relevant for other locations. In conclusion, the results of this large scale study points toward reliability of the MLMs as far as their generalization power is concerned. At the same time, they also show that different data splitting strategies can lead to significantly different results. 

How to cite: Amani, A., Boucher, M.-A., Cabral, A. R., and Nadeau, D. F.: Assessing the generalization power of three machine learning models and three evapotranspiration formulas using 143 FLUXNET towers data, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5580, https://doi.org/10.5194/egusphere-egu22-5580, 2022.

Estuaries are transition zones between rivers and oceans that provide essential ecosystem services with high economic, environmental and social importance. These areas are exploited by fishing, maritime transportation and tourism industries, having banks usually highly urbanized and, thus, their waters are exposed to anthropogenic activities. They also serve as nursery areas for many fish and marine birds species, providing shelter and food in their early stages of life. Therefore, the preservation of the quality of estuarine natural resources is essential for all stakeholders, being of utmost importance to promote its sustainable use. However, the anthropogenic pressure threatens the availability of these natural resources, demanding a continuous monitoring effort.

Morphological conditions are determinants for most of the services provided by estuaries. Though its forecasting is a very challenging task due to the involved physical-process complexity. At the same time, its characteristic timescales demand long-term simulations with high computational costs to achieve relevant and accurate results. Anticipating the morphology evolution, allows, for example, to optimize dredge operations to maintain navigation channels, being necessary for efficient flood management.

Available state-of-the-art physical-based numerical models can be applied to predict estuarine morphology. However, the sediment transport component usually requires additional computational resources, increasing the CPU time and limiting their application for short to medium-term forecasts. An artificial intelligence (AI) emulator based solely on the hydrodynamic component results could be a solution to minimize the total morphodynamic forecast CPU time.

This work implemented a convolutional neural network (CNN) to emulate the morphodynamic evolution of the Minho river estuary, located at the northern Portuguese coast. AI-based methods demand considerable time to be implemented, mainly during dataset preprocessing and training tasks, but their simulation performance could be superior when compared to numerical models if sufficient data are available for training the algorithm. In this proof of concept work, the CNN used the estuarine currents average velocity and direction and the bottom stress extracted from the Delft3D numerical model runs as input to forecast the accumulated sedimentation/erosion. The hydrodynamic numerical model was automatically calibrated using the OpenDA tool, determining the best value combination for the numerical parameters. Simulations were performed considering a 20-year return period flood event, with hourly generated outputs. The network was implemented using the TensorFlow open-source platform and was composed of an input layer, for reading the results of the hydrodynamic model, a filter layer, for simplifying the inputs, a hidden layer, for learning and processing the input information and, lastly, an output layer, for generating the accumulated erosion/accretion patterns within the estuary.

The results demonstrated the emulator’s capacity to reproduce the sandbar patterns inside the estuary, revealing to be a promising approach to forecast estuarine morphodynamics in a shorter computational time. The mean absolute percentage error of the CNN model was 0.80 during training and 0.77 during testing. While the numerical model requires 38 minutes to simulate a one-month simulation period, the emulator needed only a couple of seconds. Future works will analyze the networks hyperparameters, aiming to increase the emulator performance and accuracy.

How to cite: Weber de Melo, W., Pinho, J. L., Iglesias, I., Gomes, A., Vieira, J., Bio, A., Bastos, L., Veloso-Gomes, F., and Avilez-Valente, P.: Estuarine morphodynamics forecast using a numerical model emulator based on deep learning methods. A first approach., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6277, https://doi.org/10.5194/egusphere-egu22-6277, 2022.

CrowdWater is a citizen science project in which we investigate how the public can be involved in the collection of hydrological data, such as stream water levels, soil moisture conditions and the presence of water in temporary streams. Another important part is to study the value of the collected data for hydrological forecasts. Therefore, we have evaluated the potential value of citizen science observations, which might be uncertain and spotty in time, in several studies. The project's long-term goal is to collect a large number of observations and thus improve the prediction of hydrological events, such as drought or flooding, by using data collected by the public in hydrological model calibration. In this presentation, we discuss our experiences from the CrowdWater project with regard to app-based data collection and evaluation of these data. We also highlight methods to ensure data quality, including a gamified approach and machine learning for the analyses of the photos that are submitted through the app. Additionally, we will give an update on new activities in the CrowdWater project.

How to cite: Seibert, J., Blanco, S., Scheller, M., Schwarzenbach, F., Ze, W., and van Meerveld, I.: Engaging the public for water data collection – experiences from the CrowdWater project, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6319, https://doi.org/10.5194/egusphere-egu22-6319, 2022.

There is a current concern for the health of river ecosystems due to their vulnerability and increasing deterioration from human pressures, as well as the interest in achieving freshwater environmental sustainability in the well-known climate change challenges.

Analysis of international monitoring frameworks of river health have highlighted the need to increase data availability and frequency as well as reduce data uncertainty. With this, new aggregation, standardization, and classification methods are required as the development of technologies have grown and reached citizens at different social and cultural levels, their participation have increased in the recent years, showing important time-cost advantages. However, still there are no clear protocols to implement as assessment using mobile phone tools and platforms. 

This study aims to develop a dynamic framework for smart river health ecosystem monitoring, employing citizen science and remote sensing. This concept uses hydro-morphological and biological river indicators, combined with machine learning algorithms to analyze spatiotemporal data. 

The smart framework for assessment presented here aims to be provide to 1) Characterized  natural and non-natural changes of river ecosystem health; 2) Improve river monitoring methods linking local observation and remote sensing data; 3) Develop databases and data visualization of river condition components; 4) Enable citizens to become a large sensor network to contribute to river health monitoring; and 5) Determine and georeferenced the causes of  river health changes to support nature-based solutions for river ecosystem management.

How to cite: Enriquez Brambila, E. R., Corzo Perez, G., McClain, M., and Solomatine, D.: A framework for smart assessment of river health using machine learning, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6656, https://doi.org/10.5194/egusphere-egu22-6656, 2022.

The Integrated Surface and Subsurface Hydrologic Modeling (ISSHM) approach, based on the coupling of physical-based surface and subsurface routing processes, evolved significantly in the last decades, also thanks to the continuously increasing capabilities offered by high-performance computing (HPC). The Extended Cellular Automata (XCA) paradigm perfectly fits the needs of HPC infrastructures, due to its inherent aptitude for parallel computing and other specific features like asynchronism that allow not only parallelization but also the reduction of the computational cost.

We present HydroCAL, a new ISSHM based on the Extended Cellular Automata paradigm linking a two-dimensional weighted XCA surface routing model with a three-dimensional XCA subsurface model. The model was implemented in the parallel software library Open Computing Abstraction Layer (OpenCAL), which allows users to exploit several parallelization strategies, hardware architectures and XCA features.

Preliminarily, the subsurface model was tested in several thousand synthetical test cases to assess the effects produced by an asynchronous functionality based on a fixed threshold rule on the hydraulic head difference. The results show the high efficiency of the asynchronous XCA model in terms of elapsed time, preserving the accuracy of the results.

Then, the coupled surface and subsurface HydroCAL modules were tested with high-resolution (10¹ m) simulations in a small headwater Mediterranean catchment characterized by high hydrogeological heterogeneity. The model parameters were calibrated and validated using different events, characterized by several discharge peaks, during two years.

The results show that the model can accurately catch the hydrological response, reproducing multi-peak events with correct peak times and discharge values, simulating adequately also the recession phases. At the same time, the XCA model implementation permits highly detailed coupled simulations with computational times adequate to operational (even real-time) purposes.

Further study will regard the application of different asynchronism rules on both the surface and subsurface modules and the addition of other modules concerning subsurface-groundwater and land surface-atmosphere interaction.

How to cite: Furnari, L., Senatore, A., De Rango, A., and Mendicino, G.: HydroCAL: An Integrated Surface-Subsurface Cellular Automata Hydrological Model, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8548, https://doi.org/10.5194/egusphere-egu22-8548, 2022.

Lack of information and dependence on unscientific techniques for measuring soil moisture has resulted in water loss and reduced crop yield for smallholder farmers in developing countries. Existing sensors are expensive, eco-unfriendly, and require an external power source hence cannot be used in off-grid and rural areas. The objective of this work is to design a wireless low-cost, biodegradable, and environment-friendly paper-based soil moisture sensor powered by microbes present in the soil which will transmit this moisture data to smartphones. The expected outcome of this work is the real-time soil moisture monitoring system accessible in off-grid areas based on microbial fuel cells (MFC). The fundamental assumption here is that the current generated by MFC and signal sent by the flexible near field communication (NFC) tag will be a function of soil moisture. This document describes the empirical procedure followed to execute this study. A lab-scale proof of concept is presented where the current is generated by a paper battery fabricated using cellulose paper and conductive ink along with microbes present in the soil and nitrates. Future plans of embedding the paper battery with NFC tags for designing soil moisture sensor using MFC technology is also presented.

How to cite: Meshram, S., Pande, Dr. S., and Jourdin, Dr. L.: Design of wireless soil moisture sensor powered by bacteria charged paper battery., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10140, https://doi.org/10.5194/egusphere-egu22-10140, 2022.

Distributed hydrology models are suitable tools for understanding the hydrological processes, which take place on heterogeneous media under ever-changing internal (e.g. land use change) and boundary conditions (e.g. climate change). The generally accepted practice for applying such models is to calibrate their parameters using observed data. Still in many locations, even in developed countries, observed data is lacking or unreliable. Regionalization is a way around this problem. In this research, we built a Random Forest (RF) model to regionalize the parameters of a distributed hydrology models (Hydrotel), which is the operational model at Quebec government. Using the RF model, the following three hypotheses were tested regarding the efficiency and spatio-temporal variability of the proposed regionalization technique: (1) A finer time-step adds more information to the calibrated parameters and therefore improves the efficiency of the regionalization method; (2) The parameters approximated by RF are spatially consistent and therefore transferrable across spatial scales (i.e. from lumped to sub-catchment to hydrological response units); (3) Using more spatially representative predictors (i.e. by refining the spatial resolution of CDs) to reflect heterogeneity of the catchment will improve the performance of regionalization at internal ungauged locations. All these hypotheses were tested on three groups of nested catchments at 3- and 24-hour time-steps. The results show that for simulations at sub-daily time-steps, the calculated loss of regionalization efficiency (with respect to calibration) is less than that of the 24-hour time-step (12% improvement). Approximating the parameters at different levels of spatial discretization demonstrates that the parameters are spatially consistent as the distribution of parameters and catchment descriptors are spatially correlated. Finally, we found a consistent improvement of simulations when we replace lumped with fully distributed parameters, for simulations with a 24-hour time step. This improvement in the efficiency is higher for catchments with a higher degree of spatial heterogeneity (up to 12%). However, no significant improvement in the efficiency of simulation from lumped to distributed parameters has been observed when the time-step of the simulation was reduced to 3-hour.

How to cite: Pouryousefi-Markhali, S., Poulin, A., and Boucher, M.-A.: Regionalization of a Distributed Hydrology Model Using Random Forest, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10239, https://doi.org/10.5194/egusphere-egu22-10239, 2022.

Representing the physical heterogeneity of the land surface in Earth System Models (ESM) remains a persistent challenge due to its relevance to represent weather and climate dynamics and the hydrological cycle accurately. To address this challenge, the HydroBlocks Land Surface Model (LSM) [1] uses a hierarchical tiling scheme that defines its Hydrologic Response Units (HRUs) by clustering high-resolution global environmental data (e.g., 30-m land cover, topography, soil properties). The recently implemented reach-based routing scheme in HydroBlocks enables a two-way coupling between the high-resolution river network and the land HRUs. However, preliminary results show that the implementation is only computationally manageable for a limited number of river reaches (~5,000) per macroscale grid cell (1x1 arc degree). This hinders the scheme generalization and scalability over continental scales. As such, further simplification of the river network structure in routing schemes is required to ensure the feasibility of the approach in ESMs.

This presentation will explore simplification alternatives for river network topologies using clustering analysis. Initially, given that a significant fraction of the total river reaches on any domain are first-order, we propose an approach that clusters these streams based on average basins’ physical and environmental features (e.g., slope, upslope contributing area, aspect, average precipitation, and land cover), and channels’ geometry. The river network topology is simplified by depicting all the clusters’ members as single equivalent channels. The clustering is performed using K-means, and the number of clusters depends on a maximum target number of reaches required to provide computational tractability. Although useful, this approach will not be enough to sufficiently reduce the computational burden, for which solving the second- and even third-order reaches remain a substantial load. Therefore, the second approach relies on clustering the river channel structure over sets of interconnected reaches (i.e., topologies including first-, second-, and third-order streams). The performance of the proposed approach will be compared to the original HydroBlocks implementation for the temporal evolution of the streamflow, inundation height and the resulting computation times.

[1]      N. W. Chaney, L. Torres-Rojas, N. Vergopolan, and C. K. Fisher, “HydroBlocks v0.2: enabling a field-scale two-way coupling between the land surface and river networks in Earth system models,” Geosci. Model Dev., vol. 14, no. 11, pp. 6813–6832, Nov. 2021, doi: 10.5194/gmd-14-6813-2021.

How to cite: Torres-Rojas, L., Vergopolan, N., Guyumus, D., and Chaney, N. W.: Clustering fine-scale river network topologies for use in Earth system models, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10782, https://doi.org/10.5194/egusphere-egu22-10782, 2022.

Evapotranspiration is a key process in the water cycle. Evapotranspiration is influenced by several hydro-meteorological variables in complex and nonlinear ways and, therefore, its estimation is often very challenging. This study employs a chaotic time series approach to predict evapotranspiration. Measured monthly evapotranspiration data over a period of 40 years (1976–2015) from the Rietholzbach monitoring station in Switzerland are analysed. The nonlinear local approximation prediction method, which uses nearest neighbours, is employed. The method involves the following steps: (1) Phase-space reconstruction of a single-variable time series in a multi-dimensional phase space using delay embedding; (2) Identification of the nearest reconstructed vectors using Euclidean distance; and (3) Prediction of the future value based on the evolution of the nearest neighbours in the phase-space. The phase-space reconstruction is done with embedding dimension (m) from 1 to 10, and nearest neighbours (k) varying from 1 to 300 are used for prediction. Out of the 480 monthly evapotranspiration values available, the first 320 values are used for phase-space reconstruction and prediction, and the remaining 160 values are used for checking the prediction accuracy. The performance of the prediction method is evaluated using correlation coefficient and root mean square error. The results generally indicate very good predictions. The prediction accuracy generally increases with an increase in the embedding dimension up to a certain point and then somewhat saturates beyond that point. The best predictions are achieved when the embedding dimension is five and the number of neighbours is 10, with a correlation coefficient value of 0.86 and root mean square error value of 14.64 mm. The low embedding dimension and small number of neighbours yielding the best predictions suggest that the dynamics of monthly evapotranspiration in the Rietholzbach station exhibit chaotic behaviour dominated by five governing variables. The optimal embedding dimension value obtained from the prediction method also matches with the optimal embedding dimension estimated using the False Nearest Neighbour (FNN) algorithm, which is a dimensionality-based approach. The results from this study have important implications for modelling and prediction of evapotranspiration.

Keywords:

Evapotranspiration, Chaos, Local approximation prediction, Phase space reconstruction, False nearest neighbour algorithm

How to cite: Vamsi Krishna, G., Jothiprakash, V., and Sivakumar, B.: Prediction of evapotranspiration using a nonlinear local approximation approach, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11467, https://doi.org/10.5194/egusphere-egu22-11467, 2022.

Point cloud is made up of a multitude of three-dimensional (3D) points with one or more attributes attached. Point cloud is the third data paradigm in addition to the well-established object (vector) and gridded (raster) representations, since point cloud data can be directly collected, computed, stored, and analyzed without converting to other types. Modern ways of data acquisition, including laser scanning from airborne, mobile, or static platforms, multi-beam echo-sounding, and dense image matching from photos, generate millions to trillions of 3D points with attached attributes. If the collection is carried out in different periods, one of the essential attributes is precisely time, allowing spatiotemporal analysis to be performed. Its use is widespread in some fields such as metrology and quality inspection, virtual reality, indoor/outdoor navigation, object detection, vegetation monitoring, building modeling, cultural heritage, and diverse visualization applications. There are some examples in fields related to hydroinformatics, mainly related to terrain modeling. Due to its nature of big data, over the past decades, a series of developments have been carried out in the different processing chains for the optimal use of point cloud. This research seeks to introduce the various point cloud developments from which the hydroinformatics community and research could benefit. A review of recent advances is made, mainly including the analysis and visualization of point cloud for dealing with water-related problems. Potential areas of application and development in hydroinformatics are identified. These include, for example, the topics of coastal monitoring, coastal erosion, shallow water assessment, ice sheet change analysis, sea-level rise assessment, monitoring of levels in water bodies, crop and vegetation monitoring, analysis of the effects of groundwater depletion, detail tracing of basins and channels, analysis of floods with detailed terrain models, and drought monitoring in crops and forests. The challenges to overcome and ongoing developments regarding point cloud application in hydroinformatics are also discussed.

How to cite: Diaz, V., Liu, H., van Oosterom, P., Meijers, M., Verbree, E., Baart, F., Pronk, M., and van Lankveld, T.: Point clouds and Hydroinformatics, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-12880, https://doi.org/10.5194/egusphere-egu22-12880, 2022.