Hoshin Gupta

The University of Arizona | UA · Department of Hydrology and Water Resources

Under global climate change, urban flooding occurs frequently, leading to huge economic losses and human casualties. Extreme rainfall is one of the direct and key causes of urban flooding, and accurate rainfall estimates at high spatiotemporal resolution are of great significance for real‐time urban flood forecasting. Using existing rainfall intens...

Process-based modelling offers interpretability and physical consistency in many domains of geosciences but struggles to leverage large datasets efficiently. Machine-learning methods, especially deep networks, have strong predictive skills yet are unable to answer specific scientific questions. In this Perspective, we explore differentiable modelli...

Models play a pivotal role in advancing our understanding of Earth's physical nature and environmental systems, aiding in their efficient planning and management. The accuracy and reliability of these models heavily rely on data, which are generally partitioned into subsets for model development and evaluation. Surprisingly, how this partitioning i...

Building accurate rainfall–runoff models is an integral part of hydrological science and practice. The variety of modeling goals and applications have led to a large suite of evaluation metrics for these models. Yet, hydrologists still put considerable trust into visual judgment, although it is unclear whether such judgment agrees or disagrees with...

We study the sensitivity of aquifer‐scale estimates of transmissivity (T) and storativity (S) to the variance and correlation length scale of aquifer heterogeneity, when such estimates are obtained by the traditional approach of analyzing pumping test data. We consider both constant‐rate and variable‐rate pumping tests, and a variety of Theis‐based...

Regionalization is an issue that hydrologists have been working to solve for decades. It is used for example when we transfer parameters from one calibrated model to another, and we search for similarity between gauged to ungauged catchments. However, a unified method of regionalization that can successfully be applied to transfer parameters, and b...

It is typical to use a single portion of the available data to calibrate hydrological models, and the remainder for model evaluation. To minimize model‐bias, this partitioning must be performed so as to ensure distributional representativeness and mutual consistency. However, failure to account for data sampling variability (DSV) in the underlying...

It has been proposed that conservation laws might not be beneficial for accurate hydrological modeling due to errors in input (precipitation) and target (streamflow) data (particularly at the event time scale), and this might explain why deep learning models (which are not based on enforcing closure) can out‐perform catchment‐scale conceptual and p...

Process-Based Modeling (PBM) and Machine Learning (ML) are often perceived as distinct paradigms in the geosciences. Here we present differentiable geoscientific modeling as a powerful pathway toward dissolving the perceived barrier between them and ushering in a paradigm shift. For decades, PBM offered benefits in interpretability and physical con...

Key to model development is the selection of an appropriate representational system, including both the representation of what is observed (the data), and the formal mathematical structure used to construct the input‐state‐output mapping. These choices are critical, because they completely determine the questions we can ask, the nature of the analy...

Miscalculating the volumes of water withdrawn for irrigation, the largest consumer of freshwater in the world, jeopardizes sustainable water management. Hydrological models quantify water withdrawals, but their estimates are unduly precise. Model imperfections need to be appreciated to avoid policy misjudgements.

Building accurate rainfall-runoff models is an integral part of hydrological science and practice. The variety of modeling goals and applications have led to a large suite of evaluation metrics for these models. Yet, hydrologists still put considerable trust into visual judgment, although it is unclear whether such judgment agrees or disagrees with...

Data-driven hydrological models are widely used for many practical purposes. However, the reliability of such models depend heavily on the strategy used to partition available observations into model calibration and evaluation subsets. Unfortunately, available data splitting methods are poor at ensuring consistency of statistical properties between...

The most accurate rainfall–runoff predictions are currently based on deep learning. There is a concern among hydrologists that the predictive accuracy of data-driven models based on deep learning may not be reliable in extrapolation or for predicting extreme events. This study tests that hypothesis using long short-term memory (LSTM) networks and a...

Because physics‐based models of dynamical systems are constrained to obey conservation laws, they must typically be fed long sequences of temporally consecutive (TC) data during model calibration and evaluation. When memory time scales are long (as in many physical systems), this requirement makes it difficult to ensure distributional similarity wh...

Irrigation agriculture is the most important user of the global freshwater resources worldwide, which makes it one of the key actors conditioning sustainable development and water security. The anticipated future climate change, population growth, and rapidly rising global demand for food will likely lead to agricultural expansion by allowing the d...

Recent applications of the Long-Short Term Memory (LSTM) network-based machine-learning approach for streamflow prediction have demonstrated their ability to outperform traditional spatially-lumped process-based models. However, difficulties can arise when interpreting the internal processes and variables of the LSTM model, mainly due to the struct...

Wildfires elicit a diversity of hydrological changes, impacting processes that drive both water quantity and quality. As wildfires increase in frequency and severity, there is a need to assess the implications for the hydrological response. Wildfire‐related hydrological changes operate at three distinct timescales: the immediate fire aftermath, the...

Spatially distributed hydrologic models are useful for understanding the water balance dynamics of catchments under changing conditions, thereby providing important information for water resource management and decision making. However, in poorly gauged basins, the absence of reliable and overlapping in situ hydro-meteorological data makes the cali...

Accurate estimation of the spatio‐temporal distribution of snow water equivalent is essential given its global importance for understanding climate dynamics and climate change, and as a source of fresh water. Here, we explore the potential of using the Long Short‐Term Memory (LSTM) network for continental and regional scale modeling of daily snow a...

It has been proposed that conservation laws might not be beneficial for accurate hydrological modeling due to errors in input (precipitation) and target (streamflow) data, and this might explain why deep learning models (which are not based on enforcing closure) can out-perform catchment-scale conceptual and process-based models at predicting strea...

It has been proposed that conservation laws might not be beneficial for accurate hydrological modeling due to errors in input (precipitation) and target (streamflow) data, and this might explain why deep learning models (which are not based on enforcing closure) can out-perform catchment-scale conceptual and process-based models at predicting strea...

Accurate and timely estimation of precipitation is critical for issuing hazard warnings (e.g., for flash floods or landslides). Current remotely sensed precipitation products have a few hours of latency, associated with the acquisition and processing of satellite data. By applying a robust nowcasting system to these products, it is (in principle) p...

Satellite‐based precipitation products (SPPs) with short latencies provide a new opportunity for flood forecasting in basins with no precipitation stations. However, the larger uncertainties associated with these near‐real‐time SPPs can influence the accuracy of flood forecasts. Here, we propose a real‐time updating method, referred to as “Constrai...

The most accurate rainfall-runoff predictions are currently based on deep learning. There is a concern among hydrologists that data-driven models based on deep learning may not be reliable in extrapolation or for predicting extreme events. This study tests that hypothesis using Long Short-Term Memory networks (LSTMs) and an LSTM variant that is arc...

Accurate and timely estimation of precipitation is critical for issuing hazard warnings (e.g., for flash floods or landslides). Current remotely sensed precipitation products have a few hours of latency, associated with the acquisition and processing of satellite data. By applying a robust nowcasting system to these products, it is (in principle) p...

The NOAA National Water Model (NWM), maintained and executed by the NOAA National Weather Service (NWS) Office of Water Prediction, provides operational hydrological guidance throughout the Contiguous United States. Based on the WRF-Hydro model architecture developed by the National Center for Atmospheric Research (NCAR), the NWM was recently modif...

Process‐based hydrological models seek to represent the dominant hydrological processes in a catchment. However, due to unavoidable incompleteness of knowledge, the construction of “fidelius” process‐based models depends largely on expert judgment. We present a systematic approach that treats models as hierarchical assemblages of hypotheses (conser...

We confirm that energy dissipation weighting provides the most accurate approach to determining the effective hydraulic conductivity (Keff) of a binary K grid. A deep learning algorithm (UNET) can infer Keff with extremely high accuracy (R2 > 0.99). The UNET architecture could be trained to infer the energy dissipation weighting pattern from an ima...

We develop a simple Quantile Spacing (QS) method for accurate probabilistic estimation of one-dimensional entropy from equiprobable random samples, and compare it with the popular Bin-Counting (BC) and Kernel Density (KD) methods. In contrast to BC, which uses equal-width bins with varying probability mass, the QS method uses estimates of the quant...

Extreme hydrologic responses following wildfires can lead to floods and debris flows with costly economic and societal impacts. Process-based hydrologic and geomorphic models used to predict the downstream impacts of wildfire must account for temporal changes in hydrologic parameters related to the generation and subsequent routing of infiltration-...

We confirm that energy dissipation weighting provides the most accurate approach to determining the effective hydraulic conductivity (Keff) of a binary K grid. A deep learning algorithm (UNET) can infer Keff with extremely high accuracy (R2 > 0.99). The UNET architecture could be trained to infer the energy dissipation weighting pattern from an ima...

Runoff and sediment yield predictions using rainfall-runoff modeling systems play a significant role in developing sustainable rangeland and water resource management strategies. To characterize the behavior and predictive uncertainty of the KINEROS2 physically-based distributed hydrologic model, we assessed model parameters importance at the event...

This paper is derived from a keynote talk given at the Google's 2020 Flood Forecasting Meets Machine Learning Workshop. Recent experiments applying deep learning to rainfall‐runoff simulation indicate that there is significantly more information in large‐scale hydrological data sets than hydrologists have been able to translate into theory or model...

We develop a simple Quantile Spacing (QS) method for accurate probabilistic estimation of one-dimensional entropy from equiprobable random samples, and compare it with the popular Bin-Counting (BC) method. In contrast to BC, which uses equal-width bins with varying probability mass, the QS method uses estimates of the quantiles that divide the supp...

It is common for model-based simulations to be reported using prediction interval estimates that characterize the lack of precision associated with the simulated values. When based on Monte-Carlo sampling to approximate the relevant probability density function(s), such estimates can significantly underestimate the width of the prediction intervals...

Streamflow prediction is very important to the economic and human development of a country. For example, it is used in the quantification and distribution of the water resource, and in the design of new hydraulic infrastructure, risk quantification, rapid response to mitigate flooding, etc. For this reason, learning how to improve our estimation of...

Sensitivity analysis (SA) is en route to becoming an integral part of mathematical modeling. The tremendous potential benefits of SA are, however, yet to be fully realized, both for advancing mechanistic and data-driven modeling of human and natural systems, and in support of decision making. In this perspective paper, a multidisciplinary group of...

Drylands cover over 40% of the global land area and are home to more than 2 billion humans. Here, we use the terrestrial water storage (TWS) anomaly data derived from GRACE satellites to assess water storage changes globally and find that drylands lost ~15.9 ± 9.1 mm of water between April 2002 and January 2017. The TWS trends are more significant...

We present a detailed overview of the Multi-model Multi-product Streamflow Forecasting (MMSF) Platform, which has been developed recently at the University of Arizona under the NASA SERVIR Program, to ease its operational implementation. The platform is based on the use of multiple hydrologic models, satellite-based precipitation products, advanced...

Precipitation exhibits a large variability over a wide range of space and time scales: from seconds to years and decades in time and from the millimeter scale of microphysical processes to regional and global scales in space. It also exhibits a large variability in magnitude and frequency, from low extremes resulting in prolonged droughts to high e...

Hydrological modeling is typically characterized by deterioration in model performance when applied to the evaluation period datasets. This situation has only sporadically been studied and, in general, has become accepted as the “norm” when performing hydrological modeling. However, Machine Learning techniques are now available that can help us bet...

Pumping tests are widely used to estimate parameters such as transmissivity and storativity, using aquifer response equations that assume a time‐constant pumping rate. However, in actual practice the discharge rate will often vary erratically and follow a generally decreasing trend as the test proceeds. In such cases, if the discharge history is re...

To quantify the effects of rainfall and catchment characteristics on sediment yield is of essential importance for sustainable land management. Precipitation extremes are the external cause of the majority of serious soil erosion and suspended sediment yield (SSY) events in the Chinese Loess Plateau (CLP), which has undergone extensive land use/cov...

Conceptual rainfall‐runoff (CRR) models are widely used for runoff simulation and for prediction under a changing climate. The models are often calibrated with only a portion of all available data at a location and then evaluated independently with another part of the data for reliability assessment. Previous studies report a persistent decrease in...

Model evaluation and hypothesis testing are fundamental to any field of science. We propose here that by changing slightly the way we think and communicate about inference—from being fundamentally a problem of uncertainty quantification to being a problem of information quantification—allows us to avoid certain problems related to testing models as...

The basis for all knowledge is “information” that we compile about the world, expressed through models that support understanding, prediction, and decision making. This overview paper provides a contextual basis for the four papers that make up the “debate series” compiled under the above title. We briefly introduce Information Theory, discuss how...

Precipitation-extremes-driven floods, which compose an important proportion of streamflow but cause severe adverse impacts in the Loess Plateau of China, urged the progressive implementation of ecological restoration (ER) strategies in the Loess Plateau (LP) of China. Knowledge of the linkage between climate variables (especially precipitation extr...

We suggest that there is a potential danger to the hydrological sciences community in not recognizing how transformative machine learning could be for the future of hydrological modeling. Given the recent success of machine learning applied to modeling problems, it is unclear what the role of hydrological theory might be in the future. We suggest t...

Extreme precipitation (EP) is a major external agent driving various natural hazards in the Loess Plateau (LP), China. However, the characteristics of the spatiotemporal EP responsible for such hazardous situations remain poorly understood. We integrate universal multifractals with a segmentation algorithm to characterize a physically meaningful th...

To assess the predictive performance, robustness and generality of watershed-scale hydrological models, we conducted a detailed multi-objective evaluation of two conceptual rainfall-runoff models (the GRX model, based on the GR4J, and the MRX model, based on the MORDOR model), of differing complexity (with respectively, 5 and 11 free parameters in...

As the global demand for agricultural products (such as food and biofuel) grows at an unprecedented pace, the current area of arable land becomes a limiting factor, driving farmland expansion around the world. In Brazil, areas of undisturbed forest and savanna have been converted to farmland, and there is little evidence that agricultural expansion...

Suspended sediment yields (SSY) respond strongly to ecological restoration (ER) efforts, and significant improvements in SSY control have been achieved in the Loess Plateau of China. However, it remains challenging to quantify the net impacts of ER on SSY. Here, we formulate the notion of elasticity of sediment discharge, by associating SSY change...

Uncertainties in flood forecasts are inevitable, and the key issue is to develop probabilistic predictions so that the predictive uncertainty (PU) bounds can be estimated. We develop and test a general method for probabilistic forecasting and PU estimation that is based on a theoretical and practical analysis of the actual nature of the model resid...