Joseph Janssen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Joseph Janssen has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Ali A. Ameli
Department of Earth, Ocean, and Atmospheric Science The University of British Columbia Vancouver BC Canada
3 shared publications Aella Gong
Independent Researcher, Vancouver, Canada
2 shared publicationsAli Ameli
Department of Earth, Ocean, and Atmospheric Sciences University of British Columbia Vancouver BC Canada
2 shared publicationsArdalan Tootchi
2 shared publications- Jiguo Cao
Department of Statistics and Actuarial Science, Simon Fraser University
2 shared publications Rosemary Knight
Department of Geophysics Stanford University Stanford California USA
2 shared publications
Latest Publications
- Open Access
- 10 October 2026
- Published by American Geophysical Union (AGU) in Water Resources Research
- Vol. 62 (10)
On the Adversarial Robustness of Hydrological Models
The evaluation of hydrological models is essential for both model selection and reliability assessment. However, comparing predictions with observations alone is insufficient to fully characterize model behavior. This limitation is particularly relevant to deep learning (DL) models because of their complex internal structures. In risk‐averse operational settings, water managers require models whose reliability and failure modes have been carefully evaluated because unstable predictions can put critical infrastructure at risk. Motivated by the need to assess model reliability before operational deployment, we introduce and explore adversarial robustness analysis for hydrological modeling and evaluate whether small, targeted perturbations to meteorological forcings induce substantial changes in simulated discharge. We compare a physical‐conceptual model (HBV) with a deep learning‐based model (LSTM) across 1,347 German catchments using perturbations of varying magnitudes generated by the fast gradient sign method (FGSM). We find that FGSM perturbations generally reduce Kling–Gupta efficiency and increase mean squared error. However, widespread catastrophic failures under the specific FGSM perturbations are not observed, and LSTM generally exhibits less performance degradation than HBV. In the examined catchments, changes in the predicted hydrographs and time‐averaged internal model states often vary approximately linearly with perturbation magnitude, providing a compact summary of how errors grow under such perturbations. Similar patterns are observed along random perturbation directions in selected catchments, suggesting that, in those catchments, small input changes produce approximately proportional changes in model output. Overall, these findings characterize local predictive stability under controlled stress tests and motivate future studies of model robustness in real‐world operational settings.
- 10 October 2026
- Published by American Geophysical Union (AGU) in Water Resources Research
- Vol. 62 (10)
Open AccessOn the Adversarial Robustness of Hydrological Models
The evaluation of hydrological models is essential for both model selection and reliability assessment. However, comparing predictions with observations alone is insufficient to fully characterize model behavior. This limitation is particularly relevant to deep learning (DL) models because of their complex internal structures. In risk‐averse operational settings, water managers require models whose reliability and failure modes have been carefully evaluated because unstable predictions can put critical infrastructure at risk. Motivated by the need to assess model reliability before operational deployment, we introduce and explore adversarial robustness analysis for hydrological modeling and evaluate whether small, targeted perturbations to meteorological forcings induce substantial changes in simulated discharge. We compare a physical‐conceptual model (HBV) with a deep learning‐based model (LSTM) across 1,347 German catchments using perturbations of varying magnitudes generated by the fast gradient sign method (FGSM). We find that FGSM perturbations generally reduce Kling–Gupta efficiency and increase mean squared error. However, widespread catastrophic failures under the specific FGSM perturbations are not observed, and LSTM generally exhibits less performance degradation than HBV. In the examined catchments, changes in the predicted hydrographs and time‐averaged internal model states often vary approximately linearly with perturbation magnitude, providing a compact summary of how errors grow under such perturbations. Similar patterns are observed along random perturbation directions in selected catchments, suggesting that, in those catchments, small input changes produce approximately proportional changes in model output. Overall, these findings characterize local predictive stability under controlled stress tests and motivate future studies of model robustness in real‐world operational settings.
- Open Access
- 10 December 2025
- Published by IWA Publishing in Hydrology Research
- Vol. 57 (1) , 61-77
How to out-perform default random forest regression: choosing hyperparameters for applications in large-sample hydrology
Predictions are a central part of water resources research. Despite their strong theoretical basis, the effective application of physically based models to catchment-scale processes remains an ongoing challenge and there are some important prediction problems that are not easily amenable to a first-principles representation. As such, machine learning (ML) models have been seen as a valid alternative in recent years. In spite of their availability, well-optimized state-of-the-art ML strategies are not widely used in water resources research. Further, some analyses require many model trainings, so sometimes computational time prevents properly optimized hyperparameters. To leverage data and use it effectively to drive scientific advances in the field, it is essential to make ML models accessible to users that may lack a deep understanding of ML by improving automated machine learning resources. ML models such as XGBoost have been recently shown to outperform random forest (RF) models which are traditionally used in water resources research. In this study, based on over 150 water-related datasets, we extensively compare XGBoost and RF. This study provides water scientists with access to quick user-friendly RF and XGBoost model optimization.
- 10 December 2025
- Published by IWA Publishing in Hydrology Research
- Vol. 57 (1) , 61-77
Open AccessHow to out-perform default random forest regression: choosing hyperparameters for applications in large-sample hydrology
Predictions are a central part of water resources research. Despite their strong theoretical basis, the effective application of physically based models to catchment-scale processes remains an ongoing challenge and there are some important prediction problems that are not easily amenable to a first-principles representation. As such, machine learning (ML) models have been seen as a valid alternative in recent years. In spite of their availability, well-optimized state-of-the-art ML strategies are not widely used in water resources research. Further, some analyses require many model trainings, so sometimes computational time prevents properly optimized hyperparameters. To leverage data and use it effectively to drive scientific advances in the field, it is essential to make ML models accessible to users that may lack a deep understanding of ML by improving automated machine learning resources. ML models such as XGBoost have been recently shown to outperform random forest (RF) models which are traditionally used in water resources research. In this study, based on over 150 water-related datasets, we extensively compare XGBoost and RF. This study provides water scientists with access to quick user-friendly RF and XGBoost model optimization.
- 14 November 2025
- Published by Wiley in Hydrological Processes
- Vol. 39 (11)
Identifying Baseflow Source Areas Using Remotely Sensed and Ground‐Based Hydrologic Data and Models
Understanding how rainfall and snowmelt influence baseflow, the groundwater‐fed component of streamflow, is essential for sound water resources management. Current approaches that aim to uncover the spatial couplings between these processes and baseflow are limited. The most commonly used methods include geochemical tracers and hydrologic models. A key limitation of the first is cost, while the second is limited by the need for simplifying assumptions. This study developed a data‐driven approach which leverages satellite Earth observation data and ground‐based data to assess the degree to which baseflow is influenced by upstream rainfall and snowmelt in California's Sierra Nevada. The procedure involved: (1) Separation of baseflow from streamflow time series using a low‐pass filtering technique, (2) application of time series and information theory methods to identify the areas which have the greatest impacts on baseflow through both rainfall and snowmelt, and (3) characterisation of the elevation zones which have a prevailing influence on baseflow. Results suggest that areas which have the strongest impact on baseflow through rainfall and snowmelt are not necessarily the areas which experience the highest annual rates of snowmelt or rainfall; snowmelt occurring in the 3000–3700 m elevation range was found to be the most important driver of baseflow.
- 14 November 2025
- Published by Wiley in Hydrological Processes
- Vol. 39 (11)
Identifying Baseflow Source Areas Using Remotely Sensed and Ground‐Based Hydrologic Data and Models
Understanding how rainfall and snowmelt influence baseflow, the groundwater‐fed component of streamflow, is essential for sound water resources management. Current approaches that aim to uncover the spatial couplings between these processes and baseflow are limited. The most commonly used methods include geochemical tracers and hydrologic models. A key limitation of the first is cost, while the second is limited by the need for simplifying assumptions. This study developed a data‐driven approach which leverages satellite Earth observation data and ground‐based data to assess the degree to which baseflow is influenced by upstream rainfall and snowmelt in California's Sierra Nevada. The procedure involved: (1) Separation of baseflow from streamflow time series using a low‐pass filtering technique, (2) application of time series and information theory methods to identify the areas which have the greatest impacts on baseflow through both rainfall and snowmelt, and (3) characterisation of the elevation zones which have a prevailing influence on baseflow. Results suggest that areas which have the strongest impact on baseflow through rainfall and snowmelt are not necessarily the areas which experience the highest annual rates of snowmelt or rainfall; snowmelt occurring in the 3000–3700 m elevation range was found to be the most important driver of baseflow.
- Open Access
- 23 May 2025
- Published by Wiley
Identifying Baseflow Source Areas Using Remotely Sensed and Ground-Based Hydrologic Data
Understanding how rainfall and snowmelt influence baseflow, the groundwater-fed component of streamflow, is essential for sound water resources management. Current approaches that aim to uncover the spatial couplings between these processes and baseflow are limited. The most commonly used methods include geochemical tracers and hydrologic models. A key limitation of the first is cost, while the second is limited by the need for simplifying assumptions. This study developed a data-driven approach which leverages satellite Earth observation data and ground-based data to assess the degree to which baseflow is influenced by upstream rainfall and snowmelt in California’s Sierra Nevada. The procedure involved: (1) separation of baseflow from streamflow time series using a low-pass filtering technique, (2) application of time series and information theory methods to identify the areas which have the greatest impacts on baseflow through both rainfall and snowmelt, and (3) characterization of the elevation zones which have a prevailing influence on baseflow. Results suggest that areas which have the strongest impact on baseflow through rainfall and snowmelt are not necessarily the areas which experience the highest annual rates of snowmelt or rainfall; snowmelt occurring in the 3000 – 3700 m elevation range was found to be the most important driver of baseflow.
- 23 May 2025
- Published by Wiley
Open AccessIdentifying Baseflow Source Areas Using Remotely Sensed and Ground-Based Hydrologic Data
Understanding how rainfall and snowmelt influence baseflow, the groundwater-fed component of streamflow, is essential for sound water resources management. Current approaches that aim to uncover the spatial couplings between these processes and baseflow are limited. The most commonly used methods include geochemical tracers and hydrologic models. A key limitation of the first is cost, while the second is limited by the need for simplifying assumptions. This study developed a data-driven approach which leverages satellite Earth observation data and ground-based data to assess the degree to which baseflow is influenced by upstream rainfall and snowmelt in California’s Sierra Nevada. The procedure involved: (1) separation of baseflow from streamflow time series using a low-pass filtering technique, (2) application of time series and information theory methods to identify the areas which have the greatest impacts on baseflow through both rainfall and snowmelt, and (3) characterization of the elevation zones which have a prevailing influence on baseflow. Results suggest that areas which have the strongest impact on baseflow through rainfall and snowmelt are not necessarily the areas which experience the highest annual rates of snowmelt or rainfall; snowmelt occurring in the 3000 – 3700 m elevation range was found to be the most important driver of baseflow.
- Open Access
- 15 May 2025
- Published by Wiley in Environmetrics
- Vol. 36 (4)
Learning From Limited Temporal Data: Dynamically Sparse Historical Functional Linear Models With Applications to Earth Science
Scientists and statisticians often seek to understand the complex relationships that connect two time‐varying variables. Recent work on sparse functional historical linear models confirms that they are promising as a tool for obtaining complex and interpretable inferences, but several notable limitations exist. Most importantly, previous works have imposed sparsity on the historical coefficient function, but have not allowed the sparsity, hence lag, to vary with time. We simplify the framework of sparse functional historical linear models by using a rectangular coefficient structure along with Whittaker smoothing, then reduce the assumptions of the previous frameworks by estimating the dynamic time lag from a hierarchical coefficient structure. We motivate our study by aiming to extract the physical rainfall–runoff processes hidden within hydrological data. We show the promise and accuracy of our method using eight simulation studies, further justified by two real sets of hydrological data.
- 15 May 2025
- Published by Wiley in Environmetrics
- Vol. 36 (4)
Open AccessLearning From Limited Temporal Data: Dynamically Sparse Historical Functional Linear Models With Applications to Earth Science
Scientists and statisticians often seek to understand the complex relationships that connect two time‐varying variables. Recent work on sparse functional historical linear models confirms that they are promising as a tool for obtaining complex and interpretable inferences, but several notable limitations exist. Most importantly, previous works have imposed sparsity on the historical coefficient function, but have not allowed the sparsity, hence lag, to vary with time. We simplify the framework of sparse functional historical linear models by using a rectangular coefficient structure along with Whittaker smoothing, then reduce the assumptions of the previous frameworks by estimating the dynamic time lag from a hierarchical coefficient structure. We motivate our study by aiming to extract the physical rainfall–runoff processes hidden within hydrological data. We show the promise and accuracy of our method using eight simulation studies, further justified by two real sets of hydrological data.
- Open Access
- 11 April 2025
- Published by Elsevier in Advances in Water Resources
- Vol. 201
Tackling water table depth modeling via machine learning: From proxy observations to verifiability
- 11 April 2025
- Published by Elsevier in Advances in Water Resources
- Vol. 201
Open AccessTackling water table depth modeling via machine learning: From proxy observations to verifiability
- Open Access
- 16 March 2025
- Published by California Digital Library (CDL) in EarthArXiv
Tackling water table depth modeling via machine learning: From proxy observations to verifiability
Spatial patterns of water table depth (WTD) play a crucial role in shaping ecological resilience, hydrological connectivity, and human-centric systems. Generally, a large-scale (e.g., continental or global) continuous map of static WTD can be simulated using either physically-based (PB) or machine learning-based (ML) models. We construct three fine-resolution (500 m) ML simulations of WTD, using the XGBoost algorithm and more than 20 million real and proxy observations of WTD, across the United States and Canada. The three ML models were constrained using known physical relations between WTD's drivers and WTD and were trained by sequentially adding real and proxy observations of WTD. Through an extensive (pixel-by-pixel) evaluation across the study region and within ten major ecoregions of North America, we demonstrate that our models (corr=0.6-0.75) can more accurately predict unseen real and proxy observations of WTD compared to two available PB simulations of WTD (corr=0.21-0.40). However, we still argue that currently-available large-scale simulations of static WTD could be uncertain within data-scarce regions such as steep mountainous regions. We reason that biased observational data mainly collected from low-elevation floodplains and the over-flexibility of available models can negatively affect the verifiability of large-scale simulations of WTD. Ultimately, we thoroughly discuss future directions that may help hydrogeologists decide how to improve machine learning-based WTD estimations. In particular, we advocate for the use of proxy satellite data, the incorporation of physical laws, the implementation of better model verification standards, the development of novel globally-available emergent indices, and the collection of more reliable observations.
- 16 March 2025
- Published by California Digital Library (CDL) in EarthArXiv
Open AccessTackling water table depth modeling via machine learning: From proxy observations to verifiability
Spatial patterns of water table depth (WTD) play a crucial role in shaping ecological resilience, hydrological connectivity, and human-centric systems. Generally, a large-scale (e.g., continental or global) continuous map of static WTD can be simulated using either physically-based (PB) or machine learning-based (ML) models. We construct three fine-resolution (500 m) ML simulations of WTD, using the XGBoost algorithm and more than 20 million real and proxy observations of WTD, across the United States and Canada. The three ML models were constrained using known physical relations between WTD's drivers and WTD and were trained by sequentially adding real and proxy observations of WTD. Through an extensive (pixel-by-pixel) evaluation across the study region and within ten major ecoregions of North America, we demonstrate that our models (corr=0.6-0.75) can more accurately predict unseen real and proxy observations of WTD compared to two available PB simulations of WTD (corr=0.21-0.40). However, we still argue that currently-available large-scale simulations of static WTD could be uncertain within data-scarce regions such as steep mountainous regions. We reason that biased observational data mainly collected from low-elevation floodplains and the over-flexibility of available models can negatively affect the verifiability of large-scale simulations of WTD. Ultimately, we thoroughly discuss future directions that may help hydrogeologists decide how to improve machine learning-based WTD estimations. In particular, we advocate for the use of proxy satellite data, the incorporation of physical laws, the implementation of better model verification standards, the development of novel globally-available emergent indices, and the collection of more reliable observations.
- Open Access
- 14 February 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 74 (4) , 946-968
A Gaussian sliding windows regression model for hydrological inference
Statistical models are an essential tool to model, forecast, and understand the hydrological processes in watersheds. In particular, the understanding of time lags associated with the delay between rainfall occurrence and subsequent changes in streamflow is of high practical importance. Since water can take a variety of flow paths to generate streamflow, a series of distinct runoff pulses may combine to create the observed streamflow time series. Current state-of-the-art models are not able to sufficiently confront the problem complexity with interpretable parametrization, thus preventing novel insights about the dynamics of distinct flow paths from being formed. The proposed Gaussian Sliding Windows Regression Model targets this problem by combining the concept of multiple windows sliding along the time axis with multiple linear regression. The window kernels, which indicate the weights applied to different time lags, are implemented via Gaussian-shaped kernels. As a result, straightforward process inference can be achieved since each window can represent one flow path. Experiments on simulated and real-world scenarios underline that the proposed model achieves accurate parameter estimates and competitive predictive performance, while fostering explainable and interpretable hydrological modelling.
- 14 February 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 74 (4) , 946-968
Open AccessA Gaussian sliding windows regression model for hydrological inference
Statistical models are an essential tool to model, forecast, and understand the hydrological processes in watersheds. In particular, the understanding of time lags associated with the delay between rainfall occurrence and subsequent changes in streamflow is of high practical importance. Since water can take a variety of flow paths to generate streamflow, a series of distinct runoff pulses may combine to create the observed streamflow time series. Current state-of-the-art models are not able to sufficiently confront the problem complexity with interpretable parametrization, thus preventing novel insights about the dynamics of distinct flow paths from being formed. The proposed Gaussian Sliding Windows Regression Model targets this problem by combining the concept of multiple windows sliding along the time axis with multiple linear regression. The window kernels, which indicate the weights applied to different time lags, are implemented via Gaussian-shaped kernels. As a result, straightforward process inference can be achieved since each window can represent one flow path. Experiments on simulated and real-world scenarios underline that the proposed model achieves accurate parameter estimates and competitive predictive performance, while fostering explainable and interpretable hydrological modelling.
- Open Access
How to out-perform default random forest regression: choosing hyperparameters for applications in large-sample hydrology
Predictions are a central part of water resources research. Historically, physically-based models have been preferred; however, they have largely failed at modeling hydrological processes at a catchment scale and there are some important prediction problems that cannot be modeled physically. As such, machine learning (ML) models have been seen as a valid alternative in recent years. In spite of their availability, well-optimized state-of-the-art ML strategies are not being widely used in water resources research. This is because using state-of-the-art ML models and optimizing hyperparameters requires expert mathematical and statistical knowledge. Further, some analyses require many model trainings, so sometimes even expert statisticians cannot properly optimize hyperparameters. To leverage data and use it effectively to drive scientific advances in the field, it is essential to make ML models accessible to subject matter experts by improving automated machine learning resources. ML models such as XGBoost have been recently shown to outperform random forest (RF) models which are traditionally used in water resources research. In this study, based on over 150 water-related datasets, we extensively compare XGBoost and RF. This study provides water scientists with access to quick user-friendly RF and XGBoost model optimization.
Open AccessHow to out-perform default random forest regression: choosing hyperparameters for applications in large-sample hydrology
Predictions are a central part of water resources research. Historically, physically-based models have been preferred; however, they have largely failed at modeling hydrological processes at a catchment scale and there are some important prediction problems that cannot be modeled physically. As such, machine learning (ML) models have been seen as a valid alternative in recent years. In spite of their availability, well-optimized state-of-the-art ML strategies are not being widely used in water resources research. This is because using state-of-the-art ML models and optimizing hyperparameters requires expert mathematical and statistical knowledge. Further, some analyses require many model trainings, so sometimes even expert statisticians cannot properly optimize hyperparameters. To leverage data and use it effectively to drive scientific advances in the field, it is essential to make ML models accessible to subject matter experts by improving automated machine learning resources. ML models such as XGBoost have been recently shown to outperform random forest (RF) models which are traditionally used in water resources research. In this study, based on over 150 water-related datasets, we extensively compare XGBoost and RF. This study provides water scientists with access to quick user-friendly RF and XGBoost model optimization.
- Open Access
- 15 September 2021
- Published by Frontiers Media SA in Frontiers in Earth Science
- Vol. 9
Assessment of Future Risks of Seasonal Municipal Water Shortages Across North America
While anthropogenic climate change poses a risk to freshwater resources across the globe through increases in evapotranspiration and temperature, it is essential to quantify the risks at local scales in response to projected trends in both freshwater supply and demand. In this study, we use empirical modeling to estimate the risks of municipal water shortages across North America by assessing the effects of climate change on streamflow and urban water demand. In addition, we aim to quantify uncertainties in both supply and demand predictions. Using streamflow data from both the US and Canada, we first cluster 4,290 streamflow gauges based on hydrograph similarity and geographical location. We develop a set of multiple linear regression (MLR) models, as a simplified analog to a distributed hydrological model, with minimum input data requirements. These MLR models are calibrated to simulate streamflow for the 1993–2012 period using the ERA5 climate reanalysis data. The models are then used to predict streamflow for the 2080–2099 period in response to two climate scenarios (RCP4.5 and RCP8.5) from five global climate models. Another set of MLR models are constructed to project seasonal changes in municipal water consumption for the clustered domains. The models are calibrated against collected data on urban water use from 47 cities across the study region. For both streamflow and water use, we quantified uncertainties in our predictions using stochastic weather generators and Monte Carlo methods. Our study shows the strong predictive power of the MLR models for simulating both streamflow regimes (Kling-Gupta efficiency >0.5) and urban water use (correlation coefficient ≈0.7) in most regions. Under the RCP4.5 (RCP8.5) emissions scenario, the West Coast, the Southwest, and the Deep South (South-Central US and the Deep South) have the highest risk of municipal water shortages. Across the whole domain, the risk is the highest in the summer season when demand is high. We find that the uncertainty in projected changes to the water demand is substantially lower than the uncertainty in the projected changes to the supply. Regions with the highest risk of water shortages should begin to investigate mitigation and adaptation strategies.
- 15 September 2021
- Published by Frontiers Media SA in Frontiers in Earth Science
- Vol. 9
Open AccessAssessment of Future Risks of Seasonal Municipal Water Shortages Across North America
While anthropogenic climate change poses a risk to freshwater resources across the globe through increases in evapotranspiration and temperature, it is essential to quantify the risks at local scales in response to projected trends in both freshwater supply and demand. In this study, we use empirical modeling to estimate the risks of municipal water shortages across North America by assessing the effects of climate change on streamflow and urban water demand. In addition, we aim to quantify uncertainties in both supply and demand predictions. Using streamflow data from both the US and Canada, we first cluster 4,290 streamflow gauges based on hydrograph similarity and geographical location. We develop a set of multiple linear regression (MLR) models, as a simplified analog to a distributed hydrological model, with minimum input data requirements. These MLR models are calibrated to simulate streamflow for the 1993–2012 period using the ERA5 climate reanalysis data. The models are then used to predict streamflow for the 2080–2099 period in response to two climate scenarios (RCP4.5 and RCP8.5) from five global climate models. Another set of MLR models are constructed to project seasonal changes in municipal water consumption for the clustered domains. The models are calibrated against collected data on urban water use from 47 cities across the study region. For both streamflow and water use, we quantified uncertainties in our predictions using stochastic weather generators and Monte Carlo methods. Our study shows the strong predictive power of the MLR models for simulating both streamflow regimes (Kling-Gupta efficiency >0.5) and urban water use (correlation coefficient ≈0.7) in most regions. Under the RCP4.5 (RCP8.5) emissions scenario, the West Coast, the Southwest, and the Deep South (South-Central US and the Deep South) have the highest risk of municipal water shortages. Across the whole domain, the risk is the highest in the summer season when demand is high. We find that the uncertainty in projected changes to the water demand is substantially lower than the uncertainty in the projected changes to the supply. Regions with the highest risk of water shortages should begin to investigate mitigation and adaptation strategies.