Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning
<p>Soil moisture, an essential parameter for hydroclimatic studies, exhibits considerable spatial and temporal variability, which complicates its mapping at high spatiotemporal resolutions. Although current remote sensing products offer global estimates of soil moisture at fine temporal resolu...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2025-04-01
|
| Series: | SOIL |
| Online Access: | https://soil.copernicus.org/articles/11/287/2025/soil-11-287-2025.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850186532389912576 |
|---|---|
| author | M. T. Widyastuti J. Padarian B. Minasny M. Webb M. Taufik D. Kidd |
| author_facet | M. T. Widyastuti J. Padarian B. Minasny M. Webb M. Taufik D. Kidd |
| author_sort | M. T. Widyastuti |
| collection | DOAJ |
| description | <p>Soil moisture, an essential parameter for hydroclimatic studies, exhibits considerable spatial and temporal variability, which complicates its mapping at high spatiotemporal resolutions. Although current remote sensing products offer global estimates of soil moisture at fine temporal resolutions, they do so at a coarse spatial resolution. Deep learning (DL) techniques have recently been employed to produce high-resolution maps of various soil properties; however, these methods require substantial training data. This study sought to map daily soil moisture across Tasmania, Australia, at an 80 m resolution using a limited set of training data. We assessed three modeling strategies: DL models calibrated using an Australian dataset (51 411 observation points), models calibrated using the Tasmanian dataset (9825 observation points), and a transfer learning technique that transferred information from the Australian models to Tasmania using region-specific data. We also evaluated two DL approaches, i.e., multilayer perceptron (MLP) and long short-term memory (LSTM). The models included the Soil Moisture Active Passive (SMAP) dataset, weather data, an elevation map, land cover, and multilevel soil property maps as inputs to generate soil moisture at the surface (0–30 cm) and subsurface (30–60 cm) layers. Results showed that (1) models calibrated from the Australian dataset performed worse than Tasmanian models regardless of the type of DL approaches; (2) Tasmanian models, calibrated solely using local data, resulted in shortcomings in predicting soil moisture; and (3) transfer learning exhibited remarkable performance improvements (error reductions of up to 45 % and a 50 % increase in correlation) and resolved the drawbacks of the two previous models. The LSTM models with transfer learning had the highest overall performance with an average mean absolute error (MAE) of 0.07 <span class="inline-formula">m<sup>3</sup> m<sup>−3</sup></span> and a correlation coefficient (<span class="inline-formula"><i>r</i></span>) of 0.77 across stations for the surface layer as well as <span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M3" display="inline" overflow="scroll" dspmath="mathml"><mrow><mtext>MAE</mtext><mo>=</mo><mn mathvariant="normal">0.07</mn><mspace width="0.125em" linebreak="nobreak"/><mrow class="unit"><msup><mi mathvariant="normal">m</mi><mn mathvariant="normal">3</mn></msup><mspace width="0.125em" linebreak="nobreak"/><msup><mi mathvariant="normal">m</mi><mrow><mo>-</mo><mn mathvariant="normal">3</mn></mrow></msup></mrow></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="96pt" height="13pt" class="svg-formula" dspmath="mathimg" md5hash="b05688daef17e3a769d11816504850c5"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="soil-11-287-2025-ie00001.svg" width="96pt" height="13pt" src="soil-11-287-2025-ie00001.png"/></svg:svg></span></span> and <span class="inline-formula"><i>r</i>=0.69</span> for the subsurface layer. The fine-resolution soil moisture maps captured the detailed landscape variation as well as temporal variation according to four distinct seasons in Tasmania. The models were then applied to generate daily soil moisture maps of Tasmania, integrated into a near-real-time monitoring system to assist agricultural decision-making.</p> |
| format | Article |
| id | doaj-art-e700701a6e2a494caffd9a070cf7f1f6 |
| institution | OA Journals |
| issn | 2199-3971 2199-398X |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | SOIL |
| spelling | doaj-art-e700701a6e2a494caffd9a070cf7f1f62025-08-20T02:16:19ZengCopernicus PublicationsSOIL2199-39712199-398X2025-04-011128730710.5194/soil-11-287-2025Mapping near-real-time soil moisture dynamics over Tasmania with transfer learningM. T. Widyastuti0J. Padarian1B. Minasny2M. Webb3M. Taufik4D. Kidd5School of Life and Environmental Sciences & Sydney Institute of Agriculture, The University of Sydney, Sydney, New South Wales, AustraliaSchool of Life and Environmental Sciences & Sydney Institute of Agriculture, The University of Sydney, Sydney, New South Wales, AustraliaSchool of Life and Environmental Sciences & Sydney Institute of Agriculture, The University of Sydney, Sydney, New South Wales, AustraliaDepartment of Natural Resources and Environment Tasmania, Prospect, Tasmania, AustraliaDepartment of Geophysics and Meteorology, IPB University, Jalan Meranti Wing 19 Level 4 Darmaga Campus, Bogor, 16680, IndonesiaDepartment of Natural Resources and Environment Tasmania, Prospect, Tasmania, Australia<p>Soil moisture, an essential parameter for hydroclimatic studies, exhibits considerable spatial and temporal variability, which complicates its mapping at high spatiotemporal resolutions. Although current remote sensing products offer global estimates of soil moisture at fine temporal resolutions, they do so at a coarse spatial resolution. Deep learning (DL) techniques have recently been employed to produce high-resolution maps of various soil properties; however, these methods require substantial training data. This study sought to map daily soil moisture across Tasmania, Australia, at an 80 m resolution using a limited set of training data. We assessed three modeling strategies: DL models calibrated using an Australian dataset (51 411 observation points), models calibrated using the Tasmanian dataset (9825 observation points), and a transfer learning technique that transferred information from the Australian models to Tasmania using region-specific data. We also evaluated two DL approaches, i.e., multilayer perceptron (MLP) and long short-term memory (LSTM). The models included the Soil Moisture Active Passive (SMAP) dataset, weather data, an elevation map, land cover, and multilevel soil property maps as inputs to generate soil moisture at the surface (0–30 cm) and subsurface (30–60 cm) layers. Results showed that (1) models calibrated from the Australian dataset performed worse than Tasmanian models regardless of the type of DL approaches; (2) Tasmanian models, calibrated solely using local data, resulted in shortcomings in predicting soil moisture; and (3) transfer learning exhibited remarkable performance improvements (error reductions of up to 45 % and a 50 % increase in correlation) and resolved the drawbacks of the two previous models. The LSTM models with transfer learning had the highest overall performance with an average mean absolute error (MAE) of 0.07 <span class="inline-formula">m<sup>3</sup> m<sup>−3</sup></span> and a correlation coefficient (<span class="inline-formula"><i>r</i></span>) of 0.77 across stations for the surface layer as well as <span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M3" display="inline" overflow="scroll" dspmath="mathml"><mrow><mtext>MAE</mtext><mo>=</mo><mn mathvariant="normal">0.07</mn><mspace width="0.125em" linebreak="nobreak"/><mrow class="unit"><msup><mi mathvariant="normal">m</mi><mn mathvariant="normal">3</mn></msup><mspace width="0.125em" linebreak="nobreak"/><msup><mi mathvariant="normal">m</mi><mrow><mo>-</mo><mn mathvariant="normal">3</mn></mrow></msup></mrow></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="96pt" height="13pt" class="svg-formula" dspmath="mathimg" md5hash="b05688daef17e3a769d11816504850c5"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="soil-11-287-2025-ie00001.svg" width="96pt" height="13pt" src="soil-11-287-2025-ie00001.png"/></svg:svg></span></span> and <span class="inline-formula"><i>r</i>=0.69</span> for the subsurface layer. The fine-resolution soil moisture maps captured the detailed landscape variation as well as temporal variation according to four distinct seasons in Tasmania. The models were then applied to generate daily soil moisture maps of Tasmania, integrated into a near-real-time monitoring system to assist agricultural decision-making.</p>https://soil.copernicus.org/articles/11/287/2025/soil-11-287-2025.pdf |
| spellingShingle | M. T. Widyastuti J. Padarian B. Minasny M. Webb M. Taufik D. Kidd Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning SOIL |
| title | Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning |
| title_full | Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning |
| title_fullStr | Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning |
| title_full_unstemmed | Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning |
| title_short | Mapping near-real-time soil moisture dynamics over Tasmania with transfer learning |
| title_sort | mapping near real time soil moisture dynamics over tasmania with transfer learning |
| url | https://soil.copernicus.org/articles/11/287/2025/soil-11-287-2025.pdf |
| work_keys_str_mv | AT mtwidyastuti mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning AT jpadarian mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning AT bminasny mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning AT mwebb mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning AT mtaufik mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning AT dkidd mappingnearrealtimesoilmoisturedynamicsovertasmaniawithtransferlearning |