scholarly journals Towards hybrid modeling of the global hydrological cycle

2021 ◽  
Author(s):  
Basil Kraft ◽  
Martin Jung ◽  
Marco Körner ◽  
Sujan Koirala ◽  
Markus Reichstein
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Soil Moisture ◽  
Data Streams ◽  
Hydrological Modeling ◽  
Water Cycle ◽  
Hybrid Modeling ◽  
Hydrological Models ◽  
Global Water Cycle ◽  
Global Water

Abstract. Progress in machine learning in conjunction with the increasing availability of relevant Earth observation data streams may help to overcome uncertainties of global hydrological models due to the complexity of the processes, diversity, and heterogeneity of the land surface and subsurface, as well as scale-dependency of processes and parameters. In this study, we exemplify a hybrid approach to global hydrological modeling that exploits the data-adaptiveness of machine learning for representing uncertain processes within a model structure based on physical principles like mass conservation. Our H2M model simulates the dynamics of snow, soil moisture, and groundwater pools globally at 1º spatial resolution and daily time step where simulated water fluxes depend on an embedded recurrent neural network. We trained the model simultaneously against observational products of terrestrial water storage variations (TWS), runoff, evapotranspiration, and snow water equivalent with a multi-task learning approach. We find that H2M is capable of reproducing the key patterns of global water cycle components with model performances being at least on par with four state-of-the-art global hydrological models. The neural network learned hydrological responses of evapotranspiration and runoff generation to antecedent soil moisture state that are qualitatively consistent with our understanding and theory. Simulated contributions of groundwater, soil moisture, and snowpack variability to TWS variations are plausible and within the large range of traditional GHMs. H2M indicates a somewhat stronger role of soil moisture for TWS variations in transitional and tropical regions compared to GHMs. Overall, we present a proof of concept for global hybrid hydrological modeling in providing a new, complementary, and data-driven perspective on global water cycle variations. With further increasing Earth observations, hybrid modeling has a large potential to advance our capability to monitor and understand the Earth system by facilitating a data-adaptive yet physically consistent, joint interpretation of heterogeneous data streams.

scholarly journals A global water cycle reanalysis (2003–2012) merging satellite gravimetry and altimetry observations with a hydrological multi-model ensemble

2014 ◽  
Vol 18 (8) ◽  
pp. 2955-2973 ◽  
Author(s):  
A. I. J. M. van Dijk ◽  
L. J. Renzullo ◽  
Y. Wada ◽  
P. Tregoning
Keyword(s):  
Data Assimilation ◽  
Mass Loss ◽  
Water Level ◽  
Water Cycle ◽  
Water Storage ◽  
Subsurface Water ◽  
Groundwater Depletion ◽  
Hydrological Models ◽  
Global Water Cycle ◽  
Global Water

Abstract. We present a global water cycle reanalysis that merges water balance estimates derived from the Gravity Recovery And Climate Experiment (GRACE) satellite mission, satellite water level altimetry and off-line estimates from several hydrological models. Error estimates for the sequential data assimilation scheme were derived from available uncertainty information and the triple collocation technique. Errors in four GRACE storage products were estimated to be 11–12 mm over land areas, while errors in monthly storage changes derived from five global hydrological models were estimated to be 17–28 mm. Prior and posterior water storage estimates were evaluated against independent observations of river water level and discharge, snow water storage and glacier mass loss. Data assimilation improved or maintained agreement overall, although results varied regionally. Uncertainties were greatest in regions where glacier mass loss and subsurface storage decline are both plausible but poorly constrained. We calculated a global water budget for 2003–2012. The main changes were a net loss of polar ice caps (−342 Gt yr−1) and mountain glaciers (−230 Gt yr−1), with an additional decrease in seasonal snowpack (−18 Gt yr−1). Storage increased due to new impoundments (+16 Gt yr−1), but this was compensated by decreases in other surface water bodies (−10 Gt yr−1). If the effect of groundwater depletion (−92 Gt yr−1) is considered separately, subsurface water storage increased by +202 Gt yr−1 due particularly to increased wetness in northern temperate regions and in the seasonally wet tropics of South America and southern Africa. The reanalysis results are publicly available via www.wenfo.org/wald/.

scholarly journals A global water cycle reanalysis (2003–2012) reconciling satellite gravimetry and altimetry observations with a hydrological model ensemble

2013 ◽  
Vol 10 (12) ◽  
pp. 15475-15523 ◽  
Author(s):  
A. I. J. M. van Dijk ◽  
L. J. Renzullo ◽  
Y. Wada ◽  
P. Tregoning
Keyword(s):  
Data Assimilation ◽  
Mass Loss ◽  
Water Level ◽  
Water Cycle ◽  
Water Storage ◽  
Groundwater Depletion ◽  
Hydrological Models ◽  
Satellite Mission ◽  
Global Water Cycle ◽  
Global Water

Abstract. We present a global water cycle reanalysis that reconciles water balance estimates derived from the GRACE satellite mission, satellite water level altimetry and off-line estimates from several hydrological models. Error estimates for the sequential data assimilation scheme were derived from available uncertainty information and the triple collocation technique. Errors in four GRACE storage products were estimated to be 11–12 mm over land areas, while errors in monthly storage changes derived from five global hydrological models were estimated to be 17–28 mm. Prior and posterior estimates were evaluated against independent observations of river water level and discharge, snow water storage and glacier mass loss. Data assimilation improved or maintained agreement overall, although results varied regionally. Uncertainties were greatest in regions where glacier mass loss and sub-surface storage decline are both plausible but poorly constrained. We calculated a global water budget for 2003–2012. The main changes were a net loss of polar ice (−341 Gt yr−1) and mountain glaciers (−185 Gt yr−1), with an additional decrease in seasonal snow pack (−19 Gt yr−1). Storage in lakes increased by +77 Gt yr−1, due to new reservoir impoundments (+87 Gt yr−1), water level change in the Caspian Sea (−27 Gt yr−1) and net increases in the remaining lakes combined (+17 Gt yr−1). There was no change in subsurface storage, because groundwater depletion (−90 Gt yr−1) was offset by increased water storage in the seasonally wet tropics of South America and southern Africa (+87 Gt yr−1), which agrees with observed and predicted changes in the tropical monsoon.

Historical Development of the Global Water Cycle as a Science Framework

2017 ◽  
Author(s):  
Richard G. Lawford ◽  
Sushel Unninayar
Keyword(s):  
Soil Moisture ◽  
Prediction Models ◽  
Hydrological Cycle ◽  
Water Cycle ◽  
Energy System ◽  
The United States ◽  
Global Perspective ◽  
Hydrological Processes ◽  
Global Water Cycle ◽  
Global Water

The global water cycle concept has its roots in the ancient understanding of nature. Indeed, the Greeks and Hebrews documented some of the most some important hydrological processes. Furthermore, Africa, Sri Lanka, and China all have archaeological evidence to show the sophisticated nature of water management that took place thousands of years ago. During the 20th century, a broader perspective was taken and the hydrological cycle was used to describe the terrestrial and freshwater component of the global water cycle. Data analysis systems and modeling protocols were developed to provide the information needed to efficiently manage water resources. These advances were helpful in defining the water in the soil and the movement of water between stores of water over land surfaces. Atmospheric inputs to these balances were also monitored, but the measurements were much more reliable over countries with dense networks of precipitation gauges and radiosonde observations. By the 1960s, early satellites began to provide images that gave a new perception of Earth processes, including a more complete realization that water cycle components and processes were continuous in space and could not be fully understood through analyses partitioned by geopolitical or topographical boundaries. In the 1970s, satellites delivered quantitative radiometric measurements that allowed for the estimation of a number of variables such as precipitation and soil moisture. In the United States, by the late 1970s, plans were made to launch the Earth System Science program, led by the National Aeronautics and Space Agency (NASA). The water component of this program integrated terrestrial and atmospheric components and provided linkages with the oceanic component so that a truly global perspective of the water cycle could be developed. At the same time, the role of regional and local hydrological processes within the integrated “global water cycle” began to be understood. Benefits of this approach were immediate. The connections between the water and energy cycles gave rise to the Global Energy and Water Cycle Experiment (GEWEX)1 as part of the World Climate Research Programme (WCRP). This integrated approach has improved our understanding of the coupled global water/energy system, leading to improved prediction models and more accurate assessments of climate variability and change. The global water cycle has also provided incentives and a framework for further improvements in the measurement of variables such as soil moisture, evapotranspiration, and precipitation. In the past two decades, groundwater has been added to the suite of water cycle variables that can be measured from space. New studies are testing innovative space-based technologies for high-resolution surface water level measurements. While many benefits have followed from the application of the global water cycle concept, its potential is still being developed. Increasingly, the global water cycle is assisting in understanding broad linkages with other global biogeochemical cycles, such as the nitrogen and carbon cycles. Applications of this concept to emerging program priorities, including the Sustainable Development Goals (SDGs) and the Water-Energy-Food (W-E-F) Nexus, are also yielding societal benefits.

An Integrative Information Aqueduct to Close the Gaps between Global Satellite Observation of Water Cycle and Local Sustainable Management of Water Resources (iAqueduct)

2020 ◽  
Author(s):  
yijian zeng ◽  
Keyword(s):  
Soil Moisture ◽  
Water Resources ◽  
Sustainable Management ◽  
Water Cycle ◽  
Satellite Observation ◽  
Global Scale ◽  
Multiple Sources ◽  
Management Level ◽  
Global Water Cycle ◽  
Global Water

<p>In the past decades, space-based Earth Observations (EO) have been rapidly advancing in monitoring the global water cycle, in particular for the variables related to precipitation, evapotranspiration and soil moisture, often at (tens of) kilometre scales. Whilst these data are highly effective to characterise water cycle variation at regional to global scale, they are less suitable for sustainable management of water resource, which needs more detailed information at local and field scale due to inhomogeneous characteristics of the soil and vegetation. To effectively exploit existing knowledge at different scales we thus need to answer the following questions: How to downscale the global water cycle products to local scale using multiple sources/scales of EO data? How to explore and apply the downscaled information at the management level for understanding soil-water-vegetation-energy processes? And how to use such fine-scale information to improve the management of soil and water resources? An integrative information aqueduct (iAqueduct) is proposed to close the gaps between global satellite observation of water cycle and local needs of information for sustainable management of water resources. iAqueduct aims to accomplish its goals by combining Copernicus satellite data (with intermediate resolutions) with high resolution Unmanned Aerial System (UAS) and in-situ observations to develop scaling functions for soil properties and soil moisture and evapotranspiration at high spatial resolution scales.</p>

The global water cycle and continental erosion during Phanerozoic time (570 my)

1989 ◽  
Vol 289 (4) ◽  
pp. 455-483 ◽  
Author(s):  
Y. Tardy ◽  
R. N'Kounkou ◽  
J.-L. Probst
Keyword(s):  
Water Cycle ◽  
Global Water Cycle ◽  
Continental Erosion ◽  
Global Water

scholarly journals Transferability Intercomparison: An Opportunity for New Insight on the Global Water Cycle and Energy Budget

2007 ◽  
Vol 88 (3) ◽  
pp. 375-384 ◽  
Author(s):  
E. S. Takle ◽  
J. Roads ◽  
B. Rockel ◽  
W. J. Gutowski ◽  
R. W. Arritt ◽  
...  
Keyword(s):  
Energy Budget ◽  
Climate Models ◽  
Climate Model ◽  
Water Cycle ◽  
Regional Climate ◽  
Climate Conditions ◽  
Continental Scale ◽  
Global Water Cycle ◽  
Wide Range ◽  
Global Water

A new approach, called transferability intercomparisons, is described for advancing both understanding and modeling of the global water cycle and energy budget. Under this approach, individual regional climate models perform simulations with all modeling parameters and parameterizations held constant over a specific period on several prescribed domains representing different climatic regions. The transferability framework goes beyond previous regional climate model intercomparisons to provide a global method for testing and improving model parameterizations by constraining the simulations within analyzed boundaries for several domains. Transferability intercomparisons expose the limits of our current regional modeling capacity by examining model accuracy on a wide range of climate conditions and realizations. Intercomparison of these individual model experiments provides a means for evaluating strengths and weaknesses of models outside their “home domains” (domain of development and testing). Reference sites that are conducting coordinated measurements under the continental-scale experiments under the Global Energy and Water Cycle Experiment (GEWEX) Hydrometeorology Panel provide data for evaluation of model abilities to simulate specific features of the water and energy cycles. A systematic intercomparison across models and domains more clearly exposes collective biases in the modeling process. By isolating particular regions and processes, regional model transferability intercomparisons can more effectively explore the spatial and temporal heterogeneity of predictability. A general improvement of model ability to simulate diverse climates will provide more confidence that models used for future climate scenarios might be able to simulate conditions on a particular domain that are beyond the range of previously observed climates.

Self-validating deep learning of continental hydrology through satellite gravimetry and altimetry

2021 ◽  
Author(s):  
Christopher Irrgang ◽  
Jan Saynisch-Wagner ◽  
Robert Dill ◽  
Eva Boergens ◽  
Maik Thomas
Keyword(s):  
Neural Network ◽  
Satellite Altimetry ◽  
Water Cycle ◽  
Global Water Cycle ◽  
Network Training ◽  
The Neural Network ◽  
Continental Hydrology ◽  
Terrestrial Water ◽  
Combine Machine ◽  
Groundwater Basins

<p>Space-borne observations of terrestrial water storage (TWS) are an essential ingredient for understanding the Earth's global water cycle, its susceptibility to climate change, and for risk assessments of ecosystems, agriculture, and water management. However, the complex distribution of water masses in rivers, lakes, or groundwater basins remains elusive in coarse-resolution gravimetry observations. We combine machine learning, numerical modeling, and satellite altimetry to build and train a downscaling neural network that recovers simulated TWS from synthetic space-borne gravity observations. The neural network is designed to adapt and validate its training progress by considering independent satellite altimetry records. We show that the neural network can accurately derive TWS anomalies in 2019 after being trained over the years 2003 to 2018. Specifically for validated regions in the Amazonas, we highlight that the neural network can outperform the numerical hydrology model used in the network training.</p><p>https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2020GL089258</p>

scholarly journals Multi-source data assimilation for physically-based hydrological modeling of an experimental hillslope

2018 ◽  
Author(s):  
Anna Botto ◽  
Enrica Belluco ◽  
Matteo Camporese
Keyword(s):  
Soil Moisture ◽  
Data Assimilation ◽  
Hydrological Modeling ◽  
Nonlinear Models ◽  
Pressure Head ◽  
Richards Equation ◽  
Hydrological Models ◽  
Filter Performance ◽  
Trade Offs ◽  
Source Data

Abstract. Data assimilation has been recently the focus of much attention for integrated surface-subsurface hydrological models, whereby joint assimilation of water table, soil moisture, and river discharge measurements with the ensemble Kalman filter (EnKF) have been extensively applied. Although the EnKF has been specifically developed to deal with nonlinear models, integrated hydrological models based on the Richards equation still represent a challenge, due to strong nonlinearities that may significantly affect the filter performance. Thus, more studies are needed to investigate the capabilities of the EnKF to correct the system state and identify parameters in cases where the unsaturated zone dynamics are dominant, as well as to quantify possible tradeoffs associated with assimilation of multi-source data. Here, the model CATHY (CATchment HYdrology) is applied to reproduce the hydrological dynamics observed in an experimental two-layered hillslope, equipped with tensiometers, water content reflectometer probes, and tipping bucket flow gages to monitor the hillslope response to a series of artificial rainfall events. Pressure head, soil moisture, and subsurface outflow are assimilated with the EnKF in a number of scenarios and the challenges and issues arising from the assimilation of multi-source data in this real-world test case are discussed. Our results demonstrate that the EnKF is able to effectively correct states and parameters even in a real application characterized by strong nonlinearities. However, multi-source data assimilation may lead to significant trade-offs: the assimilation of additional variables can lead to degradation of model predictions for other variables that were otherwise well reproduced. Furthermore, we show that integrated observations such as outflow discharge cannot compensate for the lack of well-distributed data in heterogeneous hillslopes.

scholarly journals Ocean Salinities Reveal Strong Global Water Cycle Intensification During 1950 to 2000

Science ◽  
2012 ◽  
Vol 336 (6080) ◽  
pp. 455-458 ◽  
Author(s):  
P. J. Durack ◽  
S. E. Wijffels ◽  
R. J. Matear
Keyword(s):  
Water Cycle ◽  
Global Water Cycle ◽  
Global Water
Sign in / Sign up

Export Citation Format

Share Document