Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes
Soil moisture is a critical variable in the hydrological cycle and the climate system, significantly impacting water resources, ecosystem functioning, and the occurrence of extreme events. However, soil moisture data are often scarce, and soil water dynamics are not fully understood in mountainous r...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/16/6/832 |
| _version_ | 1797239034059161600 |
|---|---|
| author | Diego Escobar-González Marcos Villacís Sebastián Páez-Bimos Gabriel Jácome Juan González-Vergara Claudia Encalada Veerle Vanacker |
| author_facet | Diego Escobar-González Marcos Villacís Sebastián Páez-Bimos Gabriel Jácome Juan González-Vergara Claudia Encalada Veerle Vanacker |
| author_sort | Diego Escobar-González |
| collection | DOAJ |
| description | Soil moisture is a critical variable in the hydrological cycle and the climate system, significantly impacting water resources, ecosystem functioning, and the occurrence of extreme events. However, soil moisture data are often scarce, and soil water dynamics are not fully understood in mountainous regions such as the tropical Andes of Ecuador. This study aims to model and predict soil moisture dynamics using in situ-collected hydrometeorological data for training and data-driven machine-learning techniques. Our results highlight the fundamental role of vegetation in controlling soil moisture dynamics and significant differences in soil water balance related to vegetation types and topography. A baseline model was developed to predict soil moisture dynamics using neural network techniques. Subsequently, by employing transfer-learning techniques, this model was effectively applied to different soil horizons and profiles, demonstrating its generalization capacity and adaptability. The use of neural network schemes and knowledge transfer techniques allowed us to develop predictive models for soil moisture trained on in situ-collected hydrometeorological data. The transfer-learning technique, which leveraged the knowledge from a pre-trained model to a model with a similar domain, yielded results with errors on the order of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup><mo><</mo><mi>ϵ</mi><mo><</mo><mn>1</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>3</mn></mrow></msup></mrow></semantics></math></inline-formula>. For the training data, the forecast of the base network demonstrated excellent results, with the lowest magnitude error metric RMSE equal to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>4.77</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup></mrow></semantics></math></inline-formula>, and NSE and KGE both equal to 0.97. These models show promising potential to accurately predict short-term soil moisture dynamics with potential applications for natural hazard monitoring in mountainous regions. |
| first_indexed | 2024-04-24T17:45:07Z |
| format | Article |
| id | doaj.art-7a199d5d28134687867db5b4e2888f26 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-04-24T17:45:07Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-7a199d5d28134687867db5b4e2888f262024-03-27T14:08:17ZengMDPI AGWater2073-44412024-03-0116683210.3390/w16060832Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical AndesDiego Escobar-González0Marcos Villacís1Sebastián Páez-Bimos2Gabriel Jácome3Juan González-Vergara4Claudia Encalada5Veerle Vanacker6Departamento de Gestión de Recursos Hídricos, Empresa Pública Metropolitana de Agua Potable y Saneamiento de Quito, EPMAPS Agua de Quito, Quito 170509, EcuadorDepartamento de Ingeniería Civil y Ambiental & Centro de Investigación y Estudios en Ingeniería de los Recursos Hídricos, Escuela Politécnica Nacional, Quito 170525, EcuadorDepartamento de Ingeniería Civil y Ambiental & Centro de Investigación y Estudios en Ingeniería de los Recursos Hídricos, Escuela Politécnica Nacional, Quito 170525, EcuadorLaboratorio de Geociencias y Medio Ambiente (GEOMA), Carrera de Recursos Naturales Renovables, Facultad de Ingeniería en Ciencias Agropecuarias y Ambientales, Universidad Técnica del Norte (UTN), Av. 17 de Julio 5-21 y Gral. José María Córdova, Ibarra 100150, EcuadorFondo Para la Protección del Agua (FONAG), Mariana de Jesús N32 y Martín de Utreras, Quito 170509, EcuadorDepartamento de Gestión de Recursos Hídricos, Empresa Pública Metropolitana de Agua Potable y Saneamiento de Quito, EPMAPS Agua de Quito, Quito 170509, EcuadorEarth and Climate Research, Earth and Life Institute, UCLouvain, 1348 Louvain-la-Neuve, BelgiumSoil moisture is a critical variable in the hydrological cycle and the climate system, significantly impacting water resources, ecosystem functioning, and the occurrence of extreme events. However, soil moisture data are often scarce, and soil water dynamics are not fully understood in mountainous regions such as the tropical Andes of Ecuador. This study aims to model and predict soil moisture dynamics using in situ-collected hydrometeorological data for training and data-driven machine-learning techniques. Our results highlight the fundamental role of vegetation in controlling soil moisture dynamics and significant differences in soil water balance related to vegetation types and topography. A baseline model was developed to predict soil moisture dynamics using neural network techniques. Subsequently, by employing transfer-learning techniques, this model was effectively applied to different soil horizons and profiles, demonstrating its generalization capacity and adaptability. The use of neural network schemes and knowledge transfer techniques allowed us to develop predictive models for soil moisture trained on in situ-collected hydrometeorological data. The transfer-learning technique, which leveraged the knowledge from a pre-trained model to a model with a similar domain, yielded results with errors on the order of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup><mo><</mo><mi>ϵ</mi><mo><</mo><mn>1</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>3</mn></mrow></msup></mrow></semantics></math></inline-formula>. For the training data, the forecast of the base network demonstrated excellent results, with the lowest magnitude error metric RMSE equal to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>4.77</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>6</mn></mrow></msup></mrow></semantics></math></inline-formula>, and NSE and KGE both equal to 0.97. These models show promising potential to accurately predict short-term soil moisture dynamics with potential applications for natural hazard monitoring in mountainous regions.https://www.mdpi.com/2073-4441/16/6/832soil moistureneural networkstransfer learningpáramosoil water |
| spellingShingle | Diego Escobar-González Marcos Villacís Sebastián Páez-Bimos Gabriel Jácome Juan González-Vergara Claudia Encalada Veerle Vanacker Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes Water soil moisture neural networks transfer learning páramo soil water |
| title | Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes |
| title_full | Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes |
| title_fullStr | Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes |
| title_full_unstemmed | Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes |
| title_short | Soil Moisture Forecast Using Transfer Learning: An Application in the High Tropical Andes |
| title_sort | soil moisture forecast using transfer learning an application in the high tropical andes |
| topic | soil moisture neural networks transfer learning páramo soil water |
| url | https://www.mdpi.com/2073-4441/16/6/832 |
| work_keys_str_mv | AT diegoescobargonzalez soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT marcosvillacis soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT sebastianpaezbimos soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT gabrieljacome soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT juangonzalezvergara soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT claudiaencalada soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes AT veerlevanacker soilmoistureforecastusingtransferlearninganapplicationinthehightropicalandes |