This session aims to bring together freshwater temperature researchers from both the surface (i.e. springs, rivers, streams, lakes and reservoirs) and subsurface waters (i.e. soil and groundwater) community. It aims to shed light on the natural and anthropogenic processes governing fresh water temperatures, and the resulting implications for ecosystem health, water resources management and sectoral uses. We are interested in spatio-temporal patterns and historic or future trends of water temperature and its extremes, their impact on aquatic biochemistry and ecosystems, as well as different types of water temperature modelling approaches (e.g. process-based, machine learning).
We welcome submissions that investigate water temperature dynamics across various temporal and spatial scales (local to global) in any inland water bodies using field and lab experiments, large-scale and -sample analyses, and modelling approaches.
Orals: Tue, 5 May, 08:30–10:15 | Room 2.15
Riverine heat waves receive far less attention compared to air heat waves. Our understanding of riverine heat waves has been limited by poorly consolidated water temperature records, despite the substantial advances in river water temperature monitoring. To address this gap, we developed a long short-term memory (LSTM) model for reconstructing historical daily mean water temperature in 1471 sites across the contiguous United States from 1980 to 2022. Using these temporally complete records, we analyzed the characteristics of riverine heat waves and their long-term trends, and evaluated how they compare with air heat waves. Our analysis revealed that riverine heat waves occur less frequently (2.3 versus 4.6 events/year), and with lower intensity (2.6 versus 7.7 °C/event) than air heat waves, but persist for longer durations (7.2 versus 4.0 days/event). More importantly, the frequency, duration and intensity of riverine heat waves have been increasing at rates 2–4 times higher than air heat waves. These increases in riverine heat waves are primarily driven by climate factors including rising air temperatures and declining snowpacks, while anthropogenic activities such as river regulation by dams and agriculture further modulate response of rivers to climate change. The pronounced rise in riverine heat waves highlights the urgent need for global monitoring of river water temperatures and inclusion of riverine thermal extremes in climate risk assessments and mitigation strategies.
How to cite: Sadayappan, K. and Li, L.: Riverine heat waves on the rise, outpacing air heat waves, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8460, https://doi.org/10.5194/egusphere-egu26-8460, 2026.
How to cite: Poulain--Auzéau, L., Kraft, B., Gudmundsson, L., and Seneviratne, S.: Spatially contiguous reconstruction of water temperature and discharge in Switzerland using deep learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7754, https://doi.org/10.5194/egusphere-egu26-7754, 2026.
Accurate simulations of lake thermal structure are essential for understanding the impacts of climate change and predicting ecosystem responses. We provide operational simulations of the thermal dynamics of all major and selected smaller Swiss lakes using the one-dimensional hydrodynamic model Simstrat. These simulations are updated daily with five-day forecasts and made openly accessible via the Alplakes platform, which also offers downloadable input and output files.
Model performance is generally excellent for the larger lakes, with RMSE values around 1.0 °C at the surface and 0.5 °C in the deep water, compared to extensive monitoring data. However, notable challenges persist for: (i) lakes with short residence times strongly influenced by inflow dynamics, (ii) managed reservoirs with operational procedures outside the model’s scope, and (iii) small lakes where local interactions (e.g., groundwater exchange, snowmelt) substantially affect thermal properties.
Comparisons with high-resolution data from new temperature monitoring stations and with 3D simulations available for selected lakes on Alplakes provide multiple opportunities for model evaluation, enhancing process understanding and guiding model improvement. Furthermore, coupling Simstrat with biogeochemical models leverages its robust physical simulations for water quality projections. A beta version of an oxygen model is already running operationally with promising results, paving the way for integrated ecological and geochemical forecasting.
How to cite: Schmid, M., Bärenbold, F., Runnalls, J., and Bouffard, D.: Operational modelling of the thermal structure of Swiss Lakes: Achievements, challenges, and new horizons, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12579, https://doi.org/10.5194/egusphere-egu26-12579, 2026.
Lake and river water temperatures provide a framework for understanding the ecological, biogeochemical, and physical functioning of these aquatic ecosystems. In particular, knowledge of the current global distribution of lake and river thermal characteristics can serve as a critical baseline to assess impacts of projected future changes. However, site-specific observations of lake and river temperatures are not readily available for most locations in the world and are especially scarce for small lakes and streams. This presentation provides an overview of several assessments at global scale and in high spatial resolution to estimate the surface water temperatures of all lakes and rivers worldwide as contained in the HydroATLAS database.
First, the seasonality of lake surface temperatures was derived from Landsat 8 thermal radiance observations between 2013 and 2021 for ~1.4 million lakes in the world that are ≥10 ha in surface area; resulting in a dataset termed LakeTEMP. Furthermore, mixing regime types were estimated for all lakes with a deterministic, physically-based model using the satellite-derived, lake-specific surface temperatures of LakeTEMP combined with other lake properties (ice cover, transparency, wind, solar radiation, and mean lake depth); resulting in a dataset termed LakeMIX. LakeTEMP and LakeMIX fill a crucial spatial data gap in large-scale limnological research, especially for the incorporation of small lakes and understudied geographies of remote regions. The data are in an analysis-ready format and freely available at https://www.hydrosheds.org/products/laketemp.
Second, we developed a global high-resolution model to estimate the long-term monthly average water temperatures of every river reach within the global digital river network of HydroATLAS, representing all rivers and streams exceeding either 10 km2 in upstream catchment area or 100 L/sec in average flow. The hybrid model uses a geostatistical approach to estimate the water temperature of headwater streams based on air temperature as well as a physical model component that includes streamflow routing and snow and groundwater contributions.
Our global, high-resolution water temperature datasets are intended to serve as a baseline that can aid in large-scale assessments of lake and river ecosystem conditions by categorizing different thermal regime types. Results are globally consistent and expected to be particularly valuable as first-order proxies in remote and data sparse regions, but less adequate for smaller scale studies due to local inaccuracies.
How to cite: Lehner, B., Korver, M., and Han, Z.: Lake and river temperature regimes at global scale derived from remote sensing imagery and geospatial modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15362, https://doi.org/10.5194/egusphere-egu26-15362, 2026.
Climate change constraints impose to identify streams and rivers that are, either resilient to atmospheric warming and can serve as ecological refuges, or highly sensitive and require restoration measures, e.g. tree planting, restoration of groundwater connections. In order to characterise the streams’ resilience or sensitivity, predict their evolution under climate change, and analyse scenarios of riparian restorations, we used data-driven analysis and physical-based modelling using thermal sensitivity (TS), as a key indicator linking climate forcing, hydrological processes and ecological relevance.
We first aggregated and harmonised a large database of French streams temperatures, i.e. 5515 (summer records) and 3143 (annual records) monitoring sites recorded between 2008 and 2024. We developed linear regressions for each site to predict weekly water temperature from weekly air temperature. The slope of the regression was qualified as the sensitivity of stream temperature (TS) of a given site to change in air temperature. Based on this large dataset, we identified several predictors of the TS, including surface area upstream sampling stations, baseflow index, riparian vegetation, summer air temperature. We found a significant positive correlation between TS and summer water temperatures, i.e. cold-streams are less sensitive to air temperature, and could therefore be important refuge areas for native species conservation.
We then used TS indicator to evaluate how a physically based thermal model (T-NET) coupled with a hydrological model (J2000) on a meso-scale contrasted lithology catchment could simulate daily water temperatures and discharges. TS enables the attribution of water temperature model’s biases to specific hydrological processes, rather than merely quantifying overall model performance. In particular, low-TS streams, typically groundwater-dominated headwaters, highlight structural limitations in the representation of groundwater-driven heat fluxes.
Finally, we used the coupled model in sub-catchments with high-TS to study the influence of several riparian vegetation restoration scenarios on the stream temperatures. We could rank streams and rivers according to the temperature attenuation gain for specific riparian scenarios.
Our results suggest that TS indicator could be a relevant metric for biological applications, such as stream species distribution modelling.
How to cite: Moatar, F., Hevin, G., Moussay, M., Duffet, J.-C., Valette, L., Marçais, J., de Lavenne, A., and Branger, F.: Thermal sensitivity of rivers, an indicator for ecological refuges and hydrological model diagnostic, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17531, https://doi.org/10.5194/egusphere-egu26-17531, 2026.
Stream temperature is an important variable for the aquatic system as it plays a key role for habitats of aquatic species and interacts with multiple other variables such as dissolved oxygen and nutrients. Therefore, it directly and indirectly affects fish, invertebrate communities, primary production and other biological processes. Due to global climate variability, increased stream temperatures are becoming more relevant in ecohydrological research. However, spatial and temporal stream temperature dynamics are impacted by a complex interplay between climate, hydrological processes, and catchment characteristics. In many ecohydrological model applications, this interplay is simplified or neglected. To address these challenges, a more detailed representation of stream temperature and its spatio-temporal variability is required.
Our study addresses the limitations of a simplified stream temperature model by using the ecohydrological model SWAT+ to demonstrate how process representation can improve simulations. SWAT+ currently predicts stream temperature based on a relatively simple linear relationship with air temperature. Important factors are not considered, since the influence of runoff components, heat exchange, and riparian shading are neglected in stream temperature predictions. To improve the representation of temperature related processes, we modified the SWAT+ model source code (version 60.5.4) and included mixing of hydrological processes (Ficklin et al., 2012), heat transfer processes (Du et al., 2018), and shading (Noa-Yarasca et al., 2023). The enhanced SWAT+ model was tested at 23 stations in the medium-sized mountainous Kinzig catchment (Central Germany) with high-resolution observed stream temperature data.
The enhanced model performed significantly better than the default model, achieving a mean KGE of 0.8 across multiple calibration sites (improved from 0.72 with the default linear model). We investigated, improved, and tested previous advances in stream temperature modelling within this work, highlighting the importance of accurate process representation. Furthermore, our results emphasize the necessity of a good hydrological calibration for a satisfactory stream temperature model performance. The resulting model serves as a valuable tool for ecological research and catchment management. By replacing empirical simplifications with process-based source code modifications, we provide a methodology for improving stream temperature representation that is transferable to other models.
How to cite: Peters, K., Kiesel, J., Oswald, I., Guse, B., Noa-Yarasca, E., Arnold, J. G., Osorio Leyton, J. M., Bieger, K., and Fohrer, N.: The integration of hydrological and heat exchange processes improves stream temperature simulations , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19585, https://doi.org/10.5194/egusphere-egu26-19585, 2026.
The strength of subsurface urban heat islands (SubSUHII) hasn't been studied much, even though this could be a key indicator of how urban infrastructure and human activities affect groundwater temperatures. Moreover, understanding SubSUHII is crucial for assessing the impact of cities on underground temperature patterns and what this means for groundwater resources.
This study tackles two main issues. First, we analyse how to quantify SubSUHII reliably. Similar to the intensity of the atmospheric urban heat island, SubSUHII is described as the difference in temperature between the average annual groundwater in urban wells and in a rural setting. Particular attention is given to how urban boundaries should be defined, and how measurement depths and rural background groundwater temperatures (GWT) should be conceptually addressed in the subsurface, highlighting both the similarities and fundamental differences with atmospheric urban heat islands.
Secondly, we explore how SubSUHII spatial patterns can be compared across multiple cities using an updated global groundwater temperature dataset by Benz et al. (2024) in conjunction with several observed groundwater temperature datasets. While our data covers more than 40 cities in total, this pilot study particularly focuses on Karlsruhe (Germany), Turin (Italy) and Cardiff (UK) where data availability is highest. These cities cover a wide range of climatic, geological and urban contexts, allowing for a comparative analysis of SubSUHII under different conditions. At the urban level, we analyse variations in SubSUHII as a function of local climate zones (LCZ) and groundwater measurement depth. Finally, we investigate the role of urban morphology and infrastructure, including building types and population density, to identify recurrent patterns of SubSUHII across cities and countries.
By providing a consistent framework for calculating SubSUHII and identification of local and global patterns, this study helps to provide a more comprehensive understanding of urban thermal impacts in the subsurface.
How to cite: Dohmwirth, V., Egidio, E., Patton, A. M.-E., Cocca, D., De Luca, D. A., Lasagna, M., and Benz, S. A.: Patterns of Subsurface Urban Heat Island Intensity (SubSUHII), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11067, https://doi.org/10.5194/egusphere-egu26-11067, 2026.
Urban streams are exposed to strong anthropogenic pressures and frequently exhibit extensive morphological modifications, including embankments, channel straightening, and elevated longitudinal slopes. Especially in urban areas hydraulic capacity and requirements on flood control are driving factors for channel geometry and morphology, leaving strongly limited lateral space availability. Despite these pressures, restoration has gained increasing attention also for urban streams. Hydraulic conditions change, roughness is getting higher, slope might be reduced. Such morphological alterations have a direct influence on hydraulic behavior. Frequently, a trade-off exists between riparian vegetation and hydraulic capacity, since flow capacity is generally highest where woody vegetation is absent from banks and adjacent zones. In urban stream systems thermal energy input during low flow situations is mainly connected to short wave radiation and to air temperature. Restoration works slow down the system, increasing hydraulic residence time. In consequence slower systems exhibit a higher energy input, which makes them more sensitive to heat stress. Riparian vegetation can buffer energy input via shading effects. When addressing the trade-off between the services provided by riparian vegetation and hydraulic capacity, linking hydraulic and thermal aspects could provide valuable insights. The research done in the INTERLAYER project dedicates the impact of riparian vegetation on water temperature during low flow conditions at the small to medium size urban river, Liesingbach. Based on a water temperature model the vegetation effects on water temperature through shade during a low flow period with high air temperature values are quantified and compared to a scenario in absence of riparian vegetation. A HEC-RAS water quality model was set up for 8 km along the urban flow path of the river Liesingbach using the full energy budget approach. The model was calibrated and validated by measured water temperature data for two river heat periods in two consecutive years. Vegetation data was collected by hemispherical photographs. The prevailing vegetation was tested against a total absence of vegetation. The results show that effects of shade through riparian vegetation were able to buffer of 0.7 to 1.3°C daily water temperature peaks. An impact on daily minimum values was not detected. Based on our results vegetation effects are not limited to dense riparian forests. Even single tree lines, or shrubby vegetation implemented through soil and water bioengineering techniques have remarkable influence on water temperature. Vegetation can clearly support ecological aspects during low flows. With the prevailing water temperature model, we created a tool to quantify the impacts and predict potential changes. Building on this integration of hydraulic and vegetation modelling will enable us to improve ecological outcomes during low-flow periods.
How to cite: Mueller, H., Ludwiczek, A., von der Thannen, M., and Rauch, H. P.: Vegetation effects on low flow water temperature small to the medium size urban stream Liesingbach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7506, https://doi.org/10.5194/egusphere-egu26-7506, 2026.
Posters on site: Tue, 5 May, 10:45–12:30 | Hall A
Austria hosts a relatively dense network of stream water temperature stations (n = 680), most operating since 2006. Measurement frequency is highly irregular both across stations and within individual stations, ranging from 5-minute records to sporadic single observations with imprecise timestamps. These characteristics complicate analysis and modeling, and systematic assessments of trends and modeling approaches for stream water temperature remain scarce in Austria.
We present a workflow that addresses these obstacles to produce extended daily and monthly stream water temperature datasets for all available stations. Missing values are imputed on an hourly basis for days with at least one observation per station. We fit a station-specific diurnal spline weighted by daily meteorological predictors to reconstruct the diurnal cycle. Results are compared to an approach using LSTM autoencoder. From these reconstructions, we derive daily minimum, maximum, and mean temperatures together with day-specific uncertainty. Monthly statistics (e.g., quantiles, maximum, mean) are obtained via Monte Carlo simulation. To extend and regionalize the datasets, we use LSTMs for daily resolution and a combination of Model-based boosting and Topkriging for monthly estimates.
The resulting products enable robust trend analyses and the evaluation of stream water temperature models across Austria.
How to cite: Laimighofer, J., Hohenstein, L., and Laaha, G.: Stream water temperature in Austria – From irregular observations to regionalized monthly and daily datasets, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18598, https://doi.org/10.5194/egusphere-egu26-18598, 2026.
Quantifying the joint impacts of climate change and intensive cascade regulation on river thermal regimes is critical for managing ecological risks and optimizing hydropower production. However, most existing attribution studies primarily document broad, seasonally asymmetric warming and cooling patterns, offering limited mechanistic understanding of how specific reservoir operation strategies—particularly the widely implemented clear-water impoundment in China—regulate cross-seasonal heat storage and downstream winter warming. Here we developed an LSTM-based attribution framework to reconstruct counterfactual “no-dam” river temperatures and to quantify the relative contributions of anthropogenic regulation (ΔANT), long-term climatic warming (ΔTrend), and intra-annual climatic variability (ΔNCV) to downstream temperature changes in the lower Jinsha River, China. In addition, a suite of thermal-timing metrics is proposed to characterize seasonal heat states and to diagnose the cross-seasonal heat-storage processes responsible for the pronounced winter warming.
Results indicate that anthropogenic regulation (ΔANT) is the dominant driver of observed downstream thermal changes, inducing substantial autumn–winter warming of up to ~2°C while dampening summer temperature extremes. Long-term climatic warming (ΔTrend) provides a persistent background increase, whereas intra-annual climatic variability (ΔNCV) imposes strong seasonal and interannual modulation. Notably, the magnitude of winter warming varies markedly among years and is strongly controlled by antecedent thermal-storage conditions, with thermal-timing metrics linking earlier autumn impoundment and greater cumulative heat storage to enhanced downstream winter temperatures (Pearson's r≈0.62). Overall, these findings elucidate the coupled roles of climate change and cascade reservoir regulation in shaping river thermal regimes and provide a mechanistic basis for optimizing multi-reservoir operations to balance hydropower generation with downstream thermal and ecological requirements.
How to cite: Qin, Y., Chen, Y., Ren, H., Luo, J., Xiao, Z., Zhang, H., and Liu, Z.: Cross-Seasonal Heat Storage Driven by Autumn Impoundment: Thermal-Timing Evidence in Large Chinese Cascade Reservoirs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2206, https://doi.org/10.5194/egusphere-egu26-2206, 2026.
Riverine heatwaves, that is, periods of anomalously high water temperature, have become more frequent and intense over the past decades across Europe. To improve our understanding of their spatial distribution and to analyze potential shifts in the causes of such events, we require long time-series of river temperature over a large spatial domain. However, records of measurement stations are often of limited length or, in some regions of Europe, not available at all.
To fill these temporal and spatial data gaps, we train a deep learning Long Short-Term Memory (LSTM) model to reconstruct historic time-series (1985-2020) of daily river temperature from meteorological records and catchment characteristics. To train and evaluate the model for the simulation of riverine heatwaves, we use the new TempER (Temperature of European Rivers) dataset that contains over 4000 temperature measurement stations in Europe. TempER covers 26 European countries and contains daily water temperature records of 1 to 72 years length. 48% of the stations additionally provide streamflow observations that we integrate into the model to enhance the fidelity of the reconstructed data. In regions not covered by TempER we use streamflow records from EStreams[1] to guide the temperature simulation.
We analyze the reconstructed river temperatures with respect to trends in riverine heatwave characteristics (frequency, duration, intensity) and their spatial distribution across Europe. With our findings we intend to improve the understanding of the hydro-meteorological processes that drive riverine heatwaves and provide a continuous dataset allowing further analysis of river temperatures in Europe.
[1] do Nascimento, T.V.M., Rudlang, J., Höge, M. et al. EStreams: An integrated dataset and catalogue of streamflow, hydro-climatic and landscape variables for Europe. Sci Data 11, 879 (2024).
How to cite: Frank, C., Gauch, M., Chartier-Rescan, C., Grundmann, M., and Brunner, M.: Mapping riverine heatwave trends across Europe using reconstructed river temperature time-series, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14281, https://doi.org/10.5194/egusphere-egu26-14281, 2026.
River and groundwater temperatures (RT and GWT) play a critical role in aquatic ecosystem functioning, drinking-water resources, and geothermal potential. Their dynamics reflect delayed responses to atmospheric temperature as well as interactions between surface water and groundwater, while for GWT, geological and geomorphological controls may be particularly important. Under ongoing climate change, understanding the dominant controls on RT and GWT at regional scales is essential for anticipating and mitigating the impacts of warming water resources.
In this study, we develop two deep learning models to predict and analyse daily RT and GWT across Germany. For river temperature, we use low-frequency (biweekly to monthly) observations from the QUADICA v2 dataset (Ebeling et al., 2024) at more than 300 locations, combined with hydroclimatic variables from CAMELS-DE and catchment descriptors as predictors. For groundwater temperature, we use daily sensor measurements from the federal state of Thuringia. After data quality control, 77 monitoring wells are retained, and model inputs include groundwater levels, atmospheric temperature, precipitation, and static well characteristics such as depth and surface elevation. We employ an LSTM architecture to account for delayed responses to atmospheric forcing, which are known to be characteristic of water temperature dynamics. All variables are transformed using a Box-Cox transformation to approximate normal distributions. Model hyperparameters are tuned using a train-validation-test split (60%, 15%, and 25%, respectively) by minimizing the root mean squared error on the validation set, and overfitting is mitigated through early stopping.
Preliminary results show that the LSTM model for river temperature achieves a median Kling–Gupta efficiency (KGE) of 0.88 for unseen periods, indicating a high predictive skill. In contrast, the groundwater temperature model yields a median KGE below zero, highlighting substantially higher complexity of the system response compared to river temperature. This reduced performance is likely attributable to a combination of data limitations and missing site-specific controls, including river proximity, land use, human abstractions, and recharge processes. These findings highlight the need for larger groundwater datasets and richer explanatory variables to better understand and predict regional-scale GWT variability.
How to cite: Shah, H., Saavedra, F., Wang, Z., Merz, R., and Siebert, C.: Building regional LSTM models to predict river and groundwater temperature in Germany, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18771, https://doi.org/10.5194/egusphere-egu26-18771, 2026.
Surface water temperature (SWT) is a key factor of aquatic ecosystem balance but also of domestic,industrial and agricultural water uses, particularly in densely populated regions with intense humanactivities. SWT is affected by multiple drivers, including meteorological conditions (e.g. airtemperature, precipitation, solar radiation, wind speed) and anthropogenic activities (e.g. wastewatertreatment plant effluents, nuclear reactor cooling, datacenter cooling, and urban cooling systems).These drivers exhibit strong spatial and temporal variability. Consequently, SWT exhibits pronouncedvariability at short-term (e.g. day-to-day) timescales, influencing the structure of aquatic communities(Bonacina et al., 2023) and human uses, such as energy production (Du et al., 2026), as well as atlong-term timescales, particularly since the Industrial Revolution, in response to ongoing globalchange. Characterizing these dynamics at the scale of large river basins is essential for understandingclimate-change impacts, assessing risks of critical thermal thresholds, and informing adaptive waterand energy resources management strategies. Conducting such characterization is particularlychallenging in the Seine River Basin (France), a 76,238 km² basin that includes the Paris metropolitanarea, hosts 17 million inhabitants, and faces competing water and energy uses. This study aimed tocharacterize long-term daily river water temperature dynamics across the Seine River Basin from1958 to 2025. Continuous daily SWT time series from 1958 to 2025 were reconstructed at nearly 80monitoring stations using an LSTM model. The time series were subsequently analysed with respectto critical thermal thresholds identified for different water uses (drinking water supply, industrialcooling, irrigation, and ecosystem preservation), in order to assess the risks of reaching or exceedingthese thresholds during the study period. This approach enables the analysis of long-term thermaltrends and paves the way for identifying adaptation levers supporting sustainable, multi-use water-resource management.
How to cite: Rivière, A., Metayer, G., Corral, D., Roy, V., and Thomas, W.: Assessing long-term variability in daily river water temperature in the Seine Basin, including the Paris metropolitan area, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20927, https://doi.org/10.5194/egusphere-egu26-20927, 2026.
River water temperature governs aquatic metabolism, oxygen availability, and habitat suitability, yet monitoring remains sparse and uneven especially for small tributaries leaving much of the river network unobserved. Across Central Europe, long-term in situ observations are spatially heterogeneous, constraining assessments for many ecologically important rivers and tributaries that are only a few tens of meters wide. We address these limitations by developing a machine-learning framework that combines satellite thermal infrared land surface temperature (LST) with hydroclimatic, topographic, and land-cover predictors to generate high-resolution river water temperature (Tw) estimates across Central Europe, with a focus on Germany. We collocate quality-controlled in situ water temperature records with ECOSTRESS overpasses using a river-corridor sampling strategy that minimizes mixed-pixel contamination and propagates cloud and retrieval uncertainty. To ensure geographic transferability, we apply geographically structured training validation that limits spatial leakage. We then fit and compare machine-learning models to predict Tw from daily to monthly timescales. The resulting product provides spatially explicit Tw fields at ECOSTRESS native resolution (order 70 m) along the river network, resolving fine-scale thermal gradients and identifying localized hotspots associated with urbanization, flow regulation, and riparian alteration. By quantifying the combined influence of climate variability and anthropogenic modification on river heating and cooling capacity, this work supports ecological risk assessment and climate-adaptation planning, and offers a transferable template for other data-limited river systems.
How to cite: Goyal, S. S., Schmidt, C., Nguyen, T., and Kumar, R.: High-resolution contemporary river temperature atlas for Central Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9900, https://doi.org/10.5194/egusphere-egu26-9900, 2026.
River temperature (Tw) is of fundamental importance to freshwater ecosystem health and the services this provides to society. Yet anthropogenically-induced Tw transformations from pressures like flow regulation, deforestation and climate change induce various thermal impacts globally. As such, evidence-led management approaches are needed to mitigate Tw modifications, but these are often hindered by a paucity of reliable data across river networks. Modelling Tw regimes across unmonitored rivers and forecasting future change is critical for helping safeguard freshwater ecosystems. Hybrid Tw models offer a promising scientific avenue to embed process-based insights within spatially transferrable statistical frameworks, but few studies have applied this across national-scales. However, current hybrid architectures often rely on simplified equations as rigid structural priors which may constrain their flexibility in capturing complex thermal dynamics. In light of this, we have developed a novel hybrid Tw method based on Differentiable Physics-Informed Neural Networks (DPINNs) and applied this to multi-decadal data (1980-2020) from 78 sites across the conterminous United States. This approach integrates a zero-dimensional (0D) heat advection-dispersion equation within a neural network (NN) framework and utilizes the neural network to estimate river heat exchange processes. This capability allows the method to be applied to data from a single site, where establishing a physical process-based model is typically difficult. We observed a Mean Absolute Error (MAE) of 0.68 °C when comparing DPINN model predictions versus observed Tw values. Our results indicated this approach outperformed the established air2stream Tw model and traditional neural networks approaches like MLP (MAE = 0.79 °C) and LSTM (MAE = 0.93 °C). Our results thus highlight that by embedding physical priors to incorporate explicit heat transfer mechanisms, it enhances Tw modelling performance while also reducing the need for large environmental datasets. The strong performance of this innovative DPINN Tw model on a national-scale highlights its potential transferability across a broad range of river environments, and thus could be a vital tool to help predict large-scale Tw dynamics to help underpin effective, ‘climate proofed’ management interventions.
How to cite: Li, G., Shi, P., Kong, L., White, J., Zhu, S., Sun, Y., Qu, S., and Yang, Q.: A Differentiable Physics-Informed Neural Network (DPINN) for National-Scale River Temperature Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13760, https://doi.org/10.5194/egusphere-egu26-13760, 2026.
Abrupt river water temperature (Tw) transitions are a key driver of freshwater ecosystem health as this governs how quickly and readily biota can adapt to shifting thermal conditions. However, despite rapid Tw increases (‘surges’) or decreases (‘plummets’) occurring prevalently across various river typologies globally, they remain poorly explored. Most research to date has isolated hydropower, snowmelt and summer storm controls on surges and plummets in individual catchments, but multiple environmental drivers governing such rapid Tw changes across broad geographic domains have been seldom explored. To address this gap, we leveraged high resolution Tw data from 77 locations spanning the conterminous United States, whereby 5159 surges and 4020 plummets (∆±1 °C in a 15-minute period) were identified between 2008-2022. Subsequently, we quantified the effects of natural and anthropogenic environmental controls on surge and plummet tallies. For this, hydropower activity was identified a key driver nationally, as evidenced by its influence at the two monitoring stations exhibiting by far the highest number of surges and plummets (n = 950-1608). Catchment-wide urban cover and snowmelt influences were more consistently associated with surges and plummets, respectively. The same environmental drivers were also tested against different event-based Tw metrics derived for each surge / plummet. For this, climate exerted significant effects on Tw magnitudes and averages (i.e., minimum, mean and maximum Tw), while catchment influences like hydropower and geology yielded stronger influences on Tw variations occurring within each event (e.g., maximum rate of change, the number of Tw increases and decreases). This paper provides a better understanding of hydroclimatic and river catchment conditions governing surges and plummets. Such evidence could help inform management interventions by targeting river environments most sensitive to rapidly fluctuating Tw regimes, which could become more volatile with a changing climate.
How to cite: White, J., Hannah, D., Ficklin, D., and Adelsperger, S.: Unravelling environmental drivers governing rapid warming and cooling in rivers spanning the conterminous United States, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18436, https://doi.org/10.5194/egusphere-egu26-18436, 2026.
Climate change adaptation in urban areas often involves the expansion and further development of blue-green infrastructure. In cities where stream systems are routed in the underground, there is potential to adapt public spaces to the growing requirements caused by climate change by daylighting these streams. Usually, urban streams are used as climate change adaptation in terms of PET (physically equivalent temperature) reductions. Beside their usage in cooling public spaces, streams, especially spring-fed streams, also provide potential for thermal utilization as a climate change adaptation measure in the built environment. In case streams originating from thermal springs temperature patterns are inverted. Showing the highest temperatures at the spring and cooling down on their way downstream. Climate change adaptation initiatives include efforts to utilize their thermal potential for geothermal energy. One example is MaDoKli Project, focusing on Mannersdorfergraben in Eastern Austria. The region is characterized through a focus on agriculture and open fields next to the city of Mannersdorf. The initial conditions of the project area show an interesting situation with a constant discharge in a range of 14 – 27 l/s measured with 23° C year-round at the spring. Therefore, the warm stream is intended to be integrated into the municipality’s energy system through heat exchange for thermal energy extraction. The Mannersdorfergraben is monitored over its 6 km long flow path. Water temperature is measured at 300 m intervals using HOBO Bluetooth Onset 1-800-loggers, covering the springs, the flow path in the underground and the open channel. The inverse temperature regime was investigated over a six-month period in 2025 at 15-min interval. The measurements are intended to support a sensitivity analysis assessing how fluctuations of +/- 5° C impact the total water temperature regime of the stream. The data allows the identification of the longitudinal extent affected by thermal stress, manifested as river temperature reduction in heating period and thermal pollution during the summer period. Early results indicate that approximately halfway along the open flow section, the effect of thermal energy extraction diminishes and the water temperature returns to its initial baseline.
Additionally, the results are intended to help determine the impact of the thermal intervention in the flow system on the stream and its ecosystem.
How to cite: Kleiner, J., Müller, H., and Rauch, J. P.: Water temperature aspects of a (geo)thermal stream in context of climate change adaptation , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11177, https://doi.org/10.5194/egusphere-egu26-11177, 2026.
The present study aims to quantify the thermal regime of a highly regulated Alpine river at the catchment scale, some of its ecological implications, to support decision making in water management, particularly in relation to revising existing ecological flows protocols. The chosen case study is the Sarca river (NorthEastern Italian Alps), where a complex hydropower diversion scheme, consisting of more than 60 km of tunnels and penstocks, two large storage reservoirs and two major hydropower plants, were constructed in the 1950s.
The Sarca River originates from the Adamello glacier, flowing for approximately 80 km to Lake Garda. The study spans from Spiazzo (1100m asl) to Sarche (150m asl), including some major lateral tributaries such as the glacier-fed Bedù stream. River water temperature was monitored since May 2025 using 14 continuous HOBO sensors conveniently distributed along the study area, on the basis of criteria that account for existing hydrometric stations, the ability to capture relevant spatial and temporal variability, and physical accessibility. In addition to such year-round catchment-scale distributed monitoring, short-term temperature monitoring at five sites along the middle course of the Sarca River was designed to assess local thermal variability associated with riparian vegetation shading, and valley slopes morphology and exposure comparing sensors placed in locations with different sunlight exposure with a spatially explicit assessment of hourly radiation.
The measured thermal regime reveals that river water temperatures already achieve values of concern for the local fish community, particularly during summer heatwaves. An analysis of continuous event duration under threshold allows to estimate some possible effects on the autochthonous marble trout. Reaches with a North-South alignment and a wider valley floor show higher daily thermal oscillations despite their higher elevation. The effect of local hydrology, valley morphology, riparian vegetation, in creating local thermal refugia for target species such as brown and marble trout is quantified. By integrating watershed-scale analyses, thermal monitoring, fish thermal requirements, spatially explicit year-round shading models, this study shows the relevance of combining ecological, hydrological, and water management perspectives to understand the thermal effects of water diversions and climate change on Alpine riverine ecosystems, providing the basis to design possible mitigation measures to increased river warming.
How to cite: Bica, A., Crivellaro, M., Schiavi Cappello, N., and Zolezzi, G.: Thermal regime of a heavily regulated Alpine river at multiple spatial scales, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11260, https://doi.org/10.5194/egusphere-egu26-11260, 2026.
River water temperature integrates atmospheric forcing with hydrological and operational controls, yet the extent to which large-scale climate modes shape river thermal variability in regulated, cold region basins is still not well constrained. We examine the imprint of two dominant Pacific modes, El Niño-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO), on multi-decadal river temperature variability across the 47,200 km2 Nechako River Basin of western Canada, where a major reservoir and flow regulation may alter the transmission of climate signals to the river network. Monthly water temperature series were assembled for 10 stations over the period 1950-2024 using a combination of Air2Stream model hindcast simulations and available observations. We then used multiscale time-frequency and coherence diagnostics to characterize variability from interannual to decadal bands and to isolate the relative influence of ENSO and PDO while accounting for shared variability and the confounding effects of regulation. The analyses indicate a clear scale separation in climate-temperature linkages: ENSO is associated with intermittent interannual modulation of river temperatures, whereas PDO relates more consistently to lower-frequency variability, with spatially heterogeneous expression across the basin. Regulated reaches show reduced persistence of low-frequency thermal variability compared with less regulated sites, consistent with reservoir storage and operations that damp longer timescale climate imprints and reshape seasonal sensitivity. In particular, the combined effects of large-scale climate variability and regulation emerge most strongly during warm-season conditions, when thermal habitat constraints are most relevant. Overall, the results show that teleconnection controls on river thermal regimes are strongly scale-dependent and can be substantially modified by regulation in cold-region systems. Resolving these interacting controls provides a basis for interpreting past thermal changes and for improving climate-informed river management and warming risk assessments under future variability.
How to cite: Sanikhani, H., Khorsandi, M., Déry, S. J., and St-Hilaire, A.: Multiscale teleconnection controls on river water temperature variability in a cold region regulated basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15389, https://doi.org/10.5194/egusphere-egu26-15389, 2026.
