Ross Searle
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ross Searle has a total of 194 co-authors. Here are the top co-authors by number of shared publications:
- Brendan Philip Malone
CSIRO Agriculture and Food, Black Mountain, ACT, Australia
5 shared publications - David Clifford
CSIRO Mathematical and Information Sciences, Locked Bag 17, North Ryde, NSW 1670, Australia
5 shared publications Mike Grundy
3 shared publications- Mike Grundy
ACSIRO Agriculture 306 Carmody Road, St Lucia, Qld 4067, Australia
3 shared publications - Raphael A. Viscarra Rossel
Soil and Landscape Science, School of Molecular and Life Sciences, Curtin University, Bentley, WA 6102, Australia
3 shared publications - Chengrong Chen
Australian Rivers Institute and School of Environment and Science Griffith University Nathan Queensland Australia
2 shared publications
Latest Publications
- Open Access
- 7 March 2026
- Published by Elsevier in SSRN Electronic Journal
The Soil Spectral Selection System: A Practical Platform to operationalise Soil Spectroscopy
Soil spectroscopy is a recognised and mature method for characterising a wide range of soil physical, chemical, and biological properties. Owing to its ability to deliver rapid, scalable, and cost-effective soil analysis, it is increasingly used in applications such as soil health monitoring, precision agriculture, carbon sequestration verification and natural capital accounting. However, practical implementation remains constrained by the diversity of spectrometer types, measurement conditions, and limited access to harmonised spectral libraries and predictive models.This paper introduces the CSIRO Soil Spectral Selection System (C4S)—a platform developed under open-science principles to load, access, manage, query, and analyse soil spectral data. C4S provides built-in, instrument- and condition-agnostic spectral inference using pre-calibrated Vis–NIR, NIR, and MIR machine-learning models derived from Australia’s national soil spectral library. Users can also combine local spectra with selected subsets of national data to build localised regression models, improving model transferability and prediction accuracy.The system integrates a multi-API architecture combining modern web technologies with R-based scientific processing for spectral harmonisation, uncertainty estimation, and sample optimisation. Spectral inference workflows employ random forest models with empirical uncertainty estimates derived from bootstrap variance across ensemble predictions.C4S enables consistent and scalable soil property estimation from both laboratory and in-field spectra, providing an extensible framework for co-development across research and industry. By bridging spectral science and digital infrastructure, C4S integrates modern artificial intelligence techniques in the management of soil information, combining data sampling and instance-selection techniques for tailored model calibration.
- 7 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Soil Spectral Selection System: A Practical Platform to operationalise Soil Spectroscopy
Soil spectroscopy is a recognised and mature method for characterising a wide range of soil physical, chemical, and biological properties. Owing to its ability to deliver rapid, scalable, and cost-effective soil analysis, it is increasingly used in applications such as soil health monitoring, precision agriculture, carbon sequestration verification and natural capital accounting. However, practical implementation remains constrained by the diversity of spectrometer types, measurement conditions, and limited access to harmonised spectral libraries and predictive models.This paper introduces the CSIRO Soil Spectral Selection System (C4S)—a platform developed under open-science principles to load, access, manage, query, and analyse soil spectral data. C4S provides built-in, instrument- and condition-agnostic spectral inference using pre-calibrated Vis–NIR, NIR, and MIR machine-learning models derived from Australia’s national soil spectral library. Users can also combine local spectra with selected subsets of national data to build localised regression models, improving model transferability and prediction accuracy.The system integrates a multi-API architecture combining modern web technologies with R-based scientific processing for spectral harmonisation, uncertainty estimation, and sample optimisation. Spectral inference workflows employ random forest models with empirical uncertainty estimates derived from bootstrap variance across ensemble predictions.C4S enables consistent and scalable soil property estimation from both laboratory and in-field spectra, providing an extensible framework for co-development across research and industry. By bridging spectral science and digital infrastructure, C4S integrates modern artificial intelligence techniques in the management of soil information, combining data sampling and instance-selection techniques for tailored model calibration.
- Open Access
- 30 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-19
Spatial Soil Moisture Prediction From In Situ Data Upscaled to Landsat Footprint: Assessing Area of Applicability of Machine Learning Models
The inherent spatial mismatch between satellite-derived and ground-observed near-surface soil moisture (SM) data necessitates cautious interpretation of point-to-pixel comparisons. Though data-driven upscaling of point-scale SM may enable statistically sound comparisons, the uncertainty across a spatial domain was less explored in previous studies. This gap underscores the need of addressing the spatial prediction uncertainties when extrapolating SM information to a broader spatial scale. Accordingly, this study presents a spatial prediction approach integrating machine learning (ML) and spatiotemporal fusion, which enables the characterisation of SM variability at the Landsat satellite footprint. Spatially clustered SM from 28 in-situ stations was extrapolated to a 100 km × 100 km area at 100 m resolution over a cross-validation period (2016-2019) and an independent test period (2020-2021). The area of applicability (AOA), which represents the spatial extent within which a prediction model is considered reliable, was determined for two ML models; Random Forests (RF) and eXtreme Gradient Boosting (XGB). The AOA of RF and XGB models encompassed 43.1% and 41.5% of the study area, respectively. The spatial SM predictions were further evaluated against multiple independent datasets, including field campaign data, in-situ SM from different networks, and satellite retrievals. Specifically, RF-predicted SM achieved a spatial R of 0.62-0.64 against field campaign data, temporal R of 0.84-0.91 against network-recorded data, and spatiotemporal R of 0.87 against SMAP L2 data during the cross-validation period. SM predictions within the AOA showed markedly lower uncertainties, which were further validated across an extended area (300 km × 300 km) with diverse physiographic conditions. Overall, this study demonstrated the use of AOA in delineating the statistically reliable spatial extent for ML-based SM predictions.
- 30 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-19
Open AccessSpatial Soil Moisture Prediction From In Situ Data Upscaled to Landsat Footprint: Assessing Area of Applicability of Machine Learning Models
The inherent spatial mismatch between satellite-derived and ground-observed near-surface soil moisture (SM) data necessitates cautious interpretation of point-to-pixel comparisons. Though data-driven upscaling of point-scale SM may enable statistically sound comparisons, the uncertainty across a spatial domain was less explored in previous studies. This gap underscores the need of addressing the spatial prediction uncertainties when extrapolating SM information to a broader spatial scale. Accordingly, this study presents a spatial prediction approach integrating machine learning (ML) and spatiotemporal fusion, which enables the characterisation of SM variability at the Landsat satellite footprint. Spatially clustered SM from 28 in-situ stations was extrapolated to a 100 km × 100 km area at 100 m resolution over a cross-validation period (2016-2019) and an independent test period (2020-2021). The area of applicability (AOA), which represents the spatial extent within which a prediction model is considered reliable, was determined for two ML models; Random Forests (RF) and eXtreme Gradient Boosting (XGB). The AOA of RF and XGB models encompassed 43.1% and 41.5% of the study area, respectively. The spatial SM predictions were further evaluated against multiple independent datasets, including field campaign data, in-situ SM from different networks, and satellite retrievals. Specifically, RF-predicted SM achieved a spatial R of 0.62-0.64 against field campaign data, temporal R of 0.84-0.91 against network-recorded data, and spatiotemporal R of 0.87 against SMAP L2 data during the cross-validation period. SM predictions within the AOA showed markedly lower uncertainties, which were further validated across an extended area (300 km × 300 km) with diverse physiographic conditions. Overall, this study demonstrated the use of AOA in delineating the statistically reliable spatial extent for ML-based SM predictions.
- Open Access
- 30 July 2024
- Published by CSIRO Publishing in Soil Research
- Vol. 62 (5)
An experiential account with recommendations for the design, installation, operation and maintenance of a farm-scale soil moisture sensing and mapping system
Context: The research explores the benefits of real time tracking of soil moisture for various land management contexts and the importance of spatio-temporal modelling and mapping to gain clear and visual understanding of soil moisture fluxes across a farm. Aims: This research aims to outline the key processes required for building an operational on-farm soil moisture monitoring system where the product is highly granular daily soil moisture maps depicting variations temporally, spatially and vertically. Methods: We describe processes of capacitance soil moisture probe installation, data collection infrastructure, sensor calibration, spatio-temporal modelling, and mapping. Key results: An out-of-bag soil moisture evaluation modelling system was tested for nearly 2 years. We found a model accuracy (RMSE) estimate of 0.002 cm−3 cm−3 and concordance of 0.96 were found. This result is averaged over this period but fluctuated daily, and related to rainfall patterns across the target farm, which were not directly incorporated into the modelling framework. As expected, incorporating prior estimates of soil moisture into the modelling framework contributed to very accurate estimates of real time available soil moisture. Conclusions: This research promotes the importance of iterative improvements to the soil moisture monitoring system, particularly in areas of sensor recalibration and spatio-temporal modelling. We stress the need for a longer-term view and plan for ongoing maintenance and improvement of such systems in the emerging digital farming ecosystem. Implications: The results of this research will be useful for researchers and practitioners involved in the design and implementation of on-farm soil monitoring systems.
- 30 July 2024
- Published by CSIRO Publishing in Soil Research
- Vol. 62 (5)
Open AccessAn experiential account with recommendations for the design, installation, operation and maintenance of a farm-scale soil moisture sensing and mapping system
Context: The research explores the benefits of real time tracking of soil moisture for various land management contexts and the importance of spatio-temporal modelling and mapping to gain clear and visual understanding of soil moisture fluxes across a farm. Aims: This research aims to outline the key processes required for building an operational on-farm soil moisture monitoring system where the product is highly granular daily soil moisture maps depicting variations temporally, spatially and vertically. Methods: We describe processes of capacitance soil moisture probe installation, data collection infrastructure, sensor calibration, spatio-temporal modelling, and mapping. Key results: An out-of-bag soil moisture evaluation modelling system was tested for nearly 2 years. We found a model accuracy (RMSE) estimate of 0.002 cm−3 cm−3 and concordance of 0.96 were found. This result is averaged over this period but fluctuated daily, and related to rainfall patterns across the target farm, which were not directly incorporated into the modelling framework. As expected, incorporating prior estimates of soil moisture into the modelling framework contributed to very accurate estimates of real time available soil moisture. Conclusions: This research promotes the importance of iterative improvements to the soil moisture monitoring system, particularly in areas of sensor recalibration and spatio-temporal modelling. We stress the need for a longer-term view and plan for ongoing maintenance and improvement of such systems in the emerging digital farming ecosystem. Implications: The results of this research will be useful for researchers and practitioners involved in the design and implementation of on-farm soil monitoring systems.
- Top Cited PapersOpen Access
- 3 September 2019
- Published by Taylor & Francis in Carbon Management
- Vol. 10 (6) , 567-587
Quantifying carbon for agricultural soil management: from the current status toward a global soil information system
The importance of building/maintaining soil carbon, for soil health and CO2 mitigation, is of increasing interest to a wide audience, including policymakers, NGOs and land managers. Integral to any approaches to promote carbon sequestering practices in managed soils are reliable, accurate and cost-effective means to quantify soil C stock changes and forecast soil C responses to different management, climate and edaphic conditions. While technology to accurately measure soil C concentrations and stocks has been in use for decades, many challenges to routine, cost-effective soil C quantification remain, including large spatial variability, low signal-to-noise and often high cost and standardization issues for direct measurement with destructive sampling. Models, empirical and process-based, may provide a cost-effective and practical means for soil C quantification to support C sequestration policies. Examples are described of how soil science and soil C quantification methods are being used to support domestic climate change policies to promote soil C sequestration on agricultural lands (cropland and grazing land) at national and provincial levels in Australia and Canada. Finally, a quantification system is outlined – consisting of well-integrated data-model frameworks, supported by expanded measurement and monitoring networks, remote sensing and crowd-sourcing of management activity data – that could comprise the core of a new global soil information system.
- 3 September 2019
- Published by Taylor & Francis in Carbon Management
- Vol. 10 (6) , 567-587
Top Cited PapersOpen AccessQuantifying carbon for agricultural soil management: from the current status toward a global soil information system
The importance of building/maintaining soil carbon, for soil health and CO2 mitigation, is of increasing interest to a wide audience, including policymakers, NGOs and land managers. Integral to any approaches to promote carbon sequestering practices in managed soils are reliable, accurate and cost-effective means to quantify soil C stock changes and forecast soil C responses to different management, climate and edaphic conditions. While technology to accurately measure soil C concentrations and stocks has been in use for decades, many challenges to routine, cost-effective soil C quantification remain, including large spatial variability, low signal-to-noise and often high cost and standardization issues for direct measurement with destructive sampling. Models, empirical and process-based, may provide a cost-effective and practical means for soil C quantification to support C sequestration policies. Examples are described of how soil science and soil C quantification methods are being used to support domestic climate change policies to promote soil C sequestration on agricultural lands (cropland and grazing land) at national and provincial levels in Australia and Canada. Finally, a quantification system is outlined – consisting of well-integrated data-model frameworks, supported by expanded measurement and monitoring networks, remote sensing and crowd-sourcing of management activity data – that could comprise the core of a new global soil information system.
- Open Access
Soil legacy data rescue via GlobalSoilMap and other international and national initiatives
- Open Access
- 1 December 2017
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 21 (12) , 6049-6067
Multiscale soil moisture estimates using static and roving cosmic-ray soil moisture sensors
Soil moisture plays a critical role in land surface processes and as such there has been a recent increase in the number and resolution of satellite soil moisture observations and the development of land surface process models with ever increasing resolution. Despite these developments, validation and calibration of these products has been limited because of a lack of observations on corresponding scales. A recently developed mobile soil moisture monitoring platform, known as the rover, offers opportunities to overcome this scale issue. This paper describes methods, results and testing of soil moisture estimates produced using rover surveys on a range of scales that are commensurate with model and satellite retrievals. Our investigation involved static cosmic-ray neutron sensors and rover surveys across both broad (36 × 36 km at 9 km resolution) and intensive (10 × 10 km at 1 km resolution) scales in a cropping district in the Mallee region of Victoria, Australia. We describe approaches for converting rover survey neutron counts to soil moisture and discuss the factors controlling soil moisture variability. We use independent gravimetric and modelled soil moisture estimates collected across both space and time to validate rover soil moisture products. Measurements revealed that temporal patterns in soil moisture were preserved through time and regression modelling approaches were utilised to produce time series of property-scale soil moisture which may also have applications in calibration and validation studies or local farm management. Intensive-scale rover surveys produced reliable soil moisture estimates at 1 km resolution while broad-scale surveys produced soil moisture estimates at 9 km resolution. We conclude that the multiscale soil moisture products produced in this study are well suited to future analysis of satellite soil moisture retrievals and finer-scale soil moisture models.
- 1 December 2017
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 21 (12) , 6049-6067
Open AccessMultiscale soil moisture estimates using static and roving cosmic-ray soil moisture sensors
Soil moisture plays a critical role in land surface processes and as such there has been a recent increase in the number and resolution of satellite soil moisture observations and the development of land surface process models with ever increasing resolution. Despite these developments, validation and calibration of these products has been limited because of a lack of observations on corresponding scales. A recently developed mobile soil moisture monitoring platform, known as the rover, offers opportunities to overcome this scale issue. This paper describes methods, results and testing of soil moisture estimates produced using rover surveys on a range of scales that are commensurate with model and satellite retrievals. Our investigation involved static cosmic-ray neutron sensors and rover surveys across both broad (36 × 36 km at 9 km resolution) and intensive (10 × 10 km at 1 km resolution) scales in a cropping district in the Mallee region of Victoria, Australia. We describe approaches for converting rover survey neutron counts to soil moisture and discuss the factors controlling soil moisture variability. We use independent gravimetric and modelled soil moisture estimates collected across both space and time to validate rover soil moisture products. Measurements revealed that temporal patterns in soil moisture were preserved through time and regression modelling approaches were utilised to produce time series of property-scale soil moisture which may also have applications in calibration and validation studies or local farm management. Intensive-scale rover surveys produced reliable soil moisture estimates at 1 km resolution while broad-scale surveys produced soil moisture estimates at 9 km resolution. We conclude that the multiscale soil moisture products produced in this study are well suited to future analysis of satellite soil moisture retrievals and finer-scale soil moisture models.