Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer

Abstract Pedo‐transfer functions (PTFs) relate soil/landscape static properties to a wide range of model inputs (e.g., soil hydraulic parameters) that are essential to soil hydrological modeling. Combining PTFs and hydrological models is a powerful strategy allowing the use of soil/landscape static...

Full description

Saved in:
Bibliographic Details
Main Authors: Peijun Li, Yuanyuan Zha, Yonggen Zhang, Chak‐Hau Michael Tso, Sabine Attinger, Luis Samaniego, Jian Peng
Format: Article
Language:English
Published: Wiley 2024-03-01
Series:Water Resources Research
Subjects:
Online Access:https://doi.org/10.1029/2023WR035543
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850211958370861056
author Peijun Li
Yuanyuan Zha
Yonggen Zhang
Chak‐Hau Michael Tso
Sabine Attinger
Luis Samaniego
Jian Peng
author_facet Peijun Li
Yuanyuan Zha
Yonggen Zhang
Chak‐Hau Michael Tso
Sabine Attinger
Luis Samaniego
Jian Peng
author_sort Peijun Li
collection DOAJ
description Abstract Pedo‐transfer functions (PTFs) relate soil/landscape static properties to a wide range of model inputs (e.g., soil hydraulic parameters) that are essential to soil hydrological modeling. Combining PTFs and hydrological models is a powerful strategy allowing the use of soil/landscape static properties for the generalization of large‐scale modeling. However, since the spatial scales of soil hydraulic parameters required for model inputs and soil/landscape static properties are often not identical, cross‐scale transfer is required, which can be a significant source of errors. Here, we investigate uncertainties in cross‐scale transfer and develop an approach that avoids them. The proposed method uses the convolutional neural network (CNN) as a cross‐scale transfer approach to directly map soil/landscape static properties to soil hydraulic parameters across different spatial scales. The proposed CNN approach is applied under two different estimation strategies to invert the hydraulic parameters of a soil‐water balance model and subsequently the quality of the parameters is assessed. Both synthetical and real‐world results around the conterminous United States indicate that in general the employed end‐to‐end strategy is superior to the two‐step strategy. The CNN‐based integrated model successfully reduces potential errors in cross‐scale transfer and can be applied to other areas lacking information on hydraulic parameters or observations. The proposed method can be extended to improve parameter estimation in earth system models and enhance our understanding of key hydrological processes.
format Article
id doaj-art-55dda767ca93416b883029dee9bae33b
institution OA Journals
issn 0043-1397
1944-7973
language English
publishDate 2024-03-01
publisher Wiley
record_format Article
series Water Resources Research
spelling doaj-art-55dda767ca93416b883029dee9bae33b2025-08-20T02:09:26ZengWileyWater Resources Research0043-13971944-79732024-03-01603n/an/a10.1029/2023WR035543Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale TransferPeijun Li0Yuanyuan Zha1Yonggen Zhang2Chak‐Hau Michael Tso3Sabine Attinger4Luis Samaniego5Jian Peng6State Key Laboratory of Water Resources Engineering and Management Wuhan University Wuhan ChinaState Key Laboratory of Water Resources Engineering and Management Wuhan University Wuhan ChinaInstitute of Surface‐Earth System Science School of Earth System Science Tianjin University Tianjin ChinaUK Centre for Ecology and Hydrology Lancaster UKDepartment of Computational Hydrosystems Helmholtz Centre for Environmental Research ‐ UFZ Leipzig GermanyDepartment of Computational Hydrosystems Helmholtz Centre for Environmental Research ‐ UFZ Leipzig GermanyDepartment of Remote Sensing Helmholtz Centre for Environmental Research—UFZ Leipzig GermanyAbstract Pedo‐transfer functions (PTFs) relate soil/landscape static properties to a wide range of model inputs (e.g., soil hydraulic parameters) that are essential to soil hydrological modeling. Combining PTFs and hydrological models is a powerful strategy allowing the use of soil/landscape static properties for the generalization of large‐scale modeling. However, since the spatial scales of soil hydraulic parameters required for model inputs and soil/landscape static properties are often not identical, cross‐scale transfer is required, which can be a significant source of errors. Here, we investigate uncertainties in cross‐scale transfer and develop an approach that avoids them. The proposed method uses the convolutional neural network (CNN) as a cross‐scale transfer approach to directly map soil/landscape static properties to soil hydraulic parameters across different spatial scales. The proposed CNN approach is applied under two different estimation strategies to invert the hydraulic parameters of a soil‐water balance model and subsequently the quality of the parameters is assessed. Both synthetical and real‐world results around the conterminous United States indicate that in general the employed end‐to‐end strategy is superior to the two‐step strategy. The CNN‐based integrated model successfully reduces potential errors in cross‐scale transfer and can be applied to other areas lacking information on hydraulic parameters or observations. The proposed method can be extended to improve parameter estimation in earth system models and enhance our understanding of key hydrological processes.https://doi.org/10.1029/2023WR035543soil moisturepedo‐transfer functionsoil hydraulic propertiesconvolutional neural networkscale conversiondata assimilation
spellingShingle Peijun Li
Yuanyuan Zha
Yonggen Zhang
Chak‐Hau Michael Tso
Sabine Attinger
Luis Samaniego
Jian Peng
Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
Water Resources Research
soil moisture
pedo‐transfer function
soil hydraulic properties
convolutional neural network
scale conversion
data assimilation
title Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
title_full Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
title_fullStr Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
title_full_unstemmed Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
title_short Deep Learning Integrating Scale Conversion and Pedo‐Transfer Function to Avoid Potential Errors in Cross‐Scale Transfer
title_sort deep learning integrating scale conversion and pedo transfer function to avoid potential errors in cross scale transfer
topic soil moisture
pedo‐transfer function
soil hydraulic properties
convolutional neural network
scale conversion
data assimilation
url https://doi.org/10.1029/2023WR035543
work_keys_str_mv AT peijunli deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT yuanyuanzha deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT yonggenzhang deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT chakhaumichaeltso deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT sabineattinger deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT luissamaniego deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer
AT jianpeng deeplearningintegratingscaleconversionandpedotransferfunctiontoavoidpotentialerrorsincrossscaletransfer