Improving plant-available water estimation using model averaging of national soil water models

IntroductionMultiple operational soil water balance (SWB) models provide real-time estimates of soil moisture across Australia, yet differences in model structure and outputs introduce uncertainty for end users. Model averaging offers a potential pathway to improve predictions, but previous studies...

Full description

Saved in:
Bibliographic Details
Main Authors: Brendan P. Malone, Ross D. Searle, Siyuan Tian, Thomas F. Bishop, Yi Yu
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-08-01
Series:Frontiers in Soil Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fsoil.2025.1629686/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849222446041792512
author Brendan P. Malone
Ross D. Searle
Siyuan Tian
Thomas F. Bishop
Yi Yu
Yi Yu
Yi Yu
author_facet Brendan P. Malone
Ross D. Searle
Siyuan Tian
Thomas F. Bishop
Yi Yu
Yi Yu
Yi Yu
author_sort Brendan P. Malone
collection DOAJ
description IntroductionMultiple operational soil water balance (SWB) models provide real-time estimates of soil moisture across Australia, yet differences in model structure and outputs introduce uncertainty for end users. Model averaging offers a potential pathway to improve predictions, but previous studies have largely applied static weighting schemes. This study investigates a temporally dynamic implementation of the Granger–Ramanathan (GRA) model averaging approach to improve in situ and spatial estimates of plant-available water (PAW) in southeastern and southern Australia.MethodsTwo hypotheses were tested: (1) that GRA model averaging improves point-scale PAW predictions compared to individual models, and (2) that spatially scaling GRA coefficients produces more accurate PAW maps than equal-weight averaging. Soil moisture sensor networks from three study regions were used to evaluate GRA performance at the probe scale. Spatial implementations of GRA were developed using temporally varying coefficients, with and without environmental covariates, and compared against static models and simple averaging.ResultsAt the point scale, GRA consistently outperformed individual SWB models and equal weighting, achieving higher concordance with sensor observations (e.g., mean concordance of 0.87 at Boorowa, 0.73 at Muttama, and 0.90 at Eyre Peninsula, compared to 0.29–0.53 for individual models and 0.05–0.60 for equal weighting). Spatial GRA with dynamic coefficients improved mapping performance relative to static approaches, but incorporating environmental covariates did not consistently enhance accuracy and in some cases reduced model generalizability.DiscussionDynamic GRA model averaging provides a practical framework for integrating multiple national-scale SWB models to improve real-time PAW prediction, particularly at well-instrumented locations. However, scaling these benefits to landscape mapping remains challenging when sensor networks are sparse or unevenly distributed. The approach has potential applications in agricultural decision-making and environmental monitoring, but further refinement is needed to optimise spatial implementations.
format Article
id doaj-art-9f5b78aa0a1943f1a26edf32fc8e45ef
institution Kabale University
issn 2673-8619
language English
publishDate 2025-08-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Soil Science
spelling doaj-art-9f5b78aa0a1943f1a26edf32fc8e45ef2025-08-26T05:27:50ZengFrontiers Media S.A.Frontiers in Soil Science2673-86192025-08-01510.3389/fsoil.2025.16296861629686Improving plant-available water estimation using model averaging of national soil water modelsBrendan P. Malone0Ross D. Searle1Siyuan Tian2Thomas F. Bishop3Yi Yu4Yi Yu5Yi Yu6CSIRO Agriculture and Food, Black Mountain, ACT, AustraliaCSIRO Agriculture and Food, St Lucia, QLD, AustraliaFenner School of Environment and Society, Australian National University, Canberra, ACT, AustraliaSydney Institute of Agriculture, School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW, AustraliaCSIRO Agriculture and Food, Black Mountain, ACT, AustraliaFenner School of Environment and Society, Australian National University, Canberra, ACT, AustraliaSydney Institute of Agriculture, School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW, AustraliaIntroductionMultiple operational soil water balance (SWB) models provide real-time estimates of soil moisture across Australia, yet differences in model structure and outputs introduce uncertainty for end users. Model averaging offers a potential pathway to improve predictions, but previous studies have largely applied static weighting schemes. This study investigates a temporally dynamic implementation of the Granger–Ramanathan (GRA) model averaging approach to improve in situ and spatial estimates of plant-available water (PAW) in southeastern and southern Australia.MethodsTwo hypotheses were tested: (1) that GRA model averaging improves point-scale PAW predictions compared to individual models, and (2) that spatially scaling GRA coefficients produces more accurate PAW maps than equal-weight averaging. Soil moisture sensor networks from three study regions were used to evaluate GRA performance at the probe scale. Spatial implementations of GRA were developed using temporally varying coefficients, with and without environmental covariates, and compared against static models and simple averaging.ResultsAt the point scale, GRA consistently outperformed individual SWB models and equal weighting, achieving higher concordance with sensor observations (e.g., mean concordance of 0.87 at Boorowa, 0.73 at Muttama, and 0.90 at Eyre Peninsula, compared to 0.29–0.53 for individual models and 0.05–0.60 for equal weighting). Spatial GRA with dynamic coefficients improved mapping performance relative to static approaches, but incorporating environmental covariates did not consistently enhance accuracy and in some cases reduced model generalizability.DiscussionDynamic GRA model averaging provides a practical framework for integrating multiple national-scale SWB models to improve real-time PAW prediction, particularly at well-instrumented locations. However, scaling these benefits to landscape mapping remains challenging when sensor networks are sparse or unevenly distributed. The approach has potential applications in agricultural decision-making and environmental monitoring, but further refinement is needed to optimise spatial implementations.https://www.frontiersin.org/articles/10.3389/fsoil.2025.1629686/fullmodel averaging methodsoil moisturesoil moisture sensingGranger-Ramanathan averagingdigital soil mappingspatio-temporal modelling
spellingShingle Brendan P. Malone
Ross D. Searle
Siyuan Tian
Thomas F. Bishop
Yi Yu
Yi Yu
Yi Yu
Improving plant-available water estimation using model averaging of national soil water models
Frontiers in Soil Science
model averaging method
soil moisture
soil moisture sensing
Granger-Ramanathan averaging
digital soil mapping
spatio-temporal modelling
title Improving plant-available water estimation using model averaging of national soil water models
title_full Improving plant-available water estimation using model averaging of national soil water models
title_fullStr Improving plant-available water estimation using model averaging of national soil water models
title_full_unstemmed Improving plant-available water estimation using model averaging of national soil water models
title_short Improving plant-available water estimation using model averaging of national soil water models
title_sort improving plant available water estimation using model averaging of national soil water models
topic model averaging method
soil moisture
soil moisture sensing
Granger-Ramanathan averaging
digital soil mapping
spatio-temporal modelling
url https://www.frontiersin.org/articles/10.3389/fsoil.2025.1629686/full
work_keys_str_mv AT brendanpmalone improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT rossdsearle improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT siyuantian improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT thomasfbishop improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT yiyu improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT yiyu improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels
AT yiyu improvingplantavailablewaterestimationusingmodelaveragingofnationalsoilwatermodels