An artificial neural network approach for soil moisture retrieval using passive microwave data

Soil moisture is a key variable that defines land surface-atmosphere (boundary layer) interactions, by contributing directly to the surface energy and water balance. Soil moisture values derived from remote sensing platforms only accounts for the near surface soil layers, generally the top 5cm. Pass...

Full description

Bibliographic Details
Main Author: Chai, Soo See
Format: Thesis
Language:English
Published: Curtin University 2010
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/1416
_version_ 1848743661878116352
author Chai, Soo See
author_facet Chai, Soo See
author_sort Chai, Soo See
building Curtin Institutional Repository
collection Online Access
description Soil moisture is a key variable that defines land surface-atmosphere (boundary layer) interactions, by contributing directly to the surface energy and water balance. Soil moisture values derived from remote sensing platforms only accounts for the near surface soil layers, generally the top 5cm. Passive microwave data at L-band (1.4 GHz, 21cm wavelength) measurements are shown to be a very effective observation for surface soil moisture retrieval. The first space-borne L-band mission dedicated to observing soil moisture, the European Space Agency's (ESA) Soil Moisture and Ocean Salinity (SMOS) mission, was launched on 2nd November 2009.Artificial Neural Network (ANN) methods have been used to empirically ascertain the complex statistical relationship between soil moisture and brightness temperature in the presence of vegetation cover. The current problem faced by this method is its inability to predict soil moisture values that are 'out-of-range' of the training data.In this research, an optimization model is developed for the Backpropagation Neural Network model. This optimization model utilizes the combination of the mean and standard deviation of the soil moisture values, together with the prediction process at different pre-determined, equal size regions to cope with the spatial and temporal variation of soil moisture values. This optimized model coupled with an ANN of optimum architecture, in terms of inputs and the number of neurons in the hidden layers, is developed to predict scale-to-scale and downscaling of soil moisture values. The dependency on the accuracy of the mean and standard deviation values of soil moisture data is also studied in this research by simulating the soil moisture values using a multiple regression model. This model obtains very encouraging results for these research problems.The data used to develop and evaluate the model in this research has been obtained from the National Airborne Field Experiments in 2005.
first_indexed 2025-11-14T05:49:07Z
format Thesis
id curtin-20.500.11937-1416
institution Curtin University Malaysia
institution_category Local University
language English
last_indexed 2025-11-14T05:49:07Z
publishDate 2010
publisher Curtin University
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-14162021-01-28T05:25:19Z An artificial neural network approach for soil moisture retrieval using passive microwave data Chai, Soo See surface soil moisture Backpropagation Neural Network space-borne L-band mission European Space Agency‟s (ESA) Soil Moisture and Ocean Salinity (SMOS) surface energy soil moisture water balance Artificial Neural Network (ANN) Soil moisture is a key variable that defines land surface-atmosphere (boundary layer) interactions, by contributing directly to the surface energy and water balance. Soil moisture values derived from remote sensing platforms only accounts for the near surface soil layers, generally the top 5cm. Passive microwave data at L-band (1.4 GHz, 21cm wavelength) measurements are shown to be a very effective observation for surface soil moisture retrieval. The first space-borne L-band mission dedicated to observing soil moisture, the European Space Agency's (ESA) Soil Moisture and Ocean Salinity (SMOS) mission, was launched on 2nd November 2009.Artificial Neural Network (ANN) methods have been used to empirically ascertain the complex statistical relationship between soil moisture and brightness temperature in the presence of vegetation cover. The current problem faced by this method is its inability to predict soil moisture values that are 'out-of-range' of the training data.In this research, an optimization model is developed for the Backpropagation Neural Network model. This optimization model utilizes the combination of the mean and standard deviation of the soil moisture values, together with the prediction process at different pre-determined, equal size regions to cope with the spatial and temporal variation of soil moisture values. This optimized model coupled with an ANN of optimum architecture, in terms of inputs and the number of neurons in the hidden layers, is developed to predict scale-to-scale and downscaling of soil moisture values. The dependency on the accuracy of the mean and standard deviation values of soil moisture data is also studied in this research by simulating the soil moisture values using a multiple regression model. This model obtains very encouraging results for these research problems.The data used to develop and evaluate the model in this research has been obtained from the National Airborne Field Experiments in 2005. 2010 Thesis http://hdl.handle.net/20.500.11937/1416 en Curtin University fulltext
spellingShingle surface soil moisture
Backpropagation Neural Network
space-borne L-band mission
European Space Agency‟s (ESA) Soil Moisture and Ocean Salinity (SMOS)
surface energy
soil moisture
water balance
Artificial Neural Network (ANN)
Chai, Soo See
An artificial neural network approach for soil moisture retrieval using passive microwave data
title An artificial neural network approach for soil moisture retrieval using passive microwave data
title_full An artificial neural network approach for soil moisture retrieval using passive microwave data
title_fullStr An artificial neural network approach for soil moisture retrieval using passive microwave data
title_full_unstemmed An artificial neural network approach for soil moisture retrieval using passive microwave data
title_short An artificial neural network approach for soil moisture retrieval using passive microwave data
title_sort artificial neural network approach for soil moisture retrieval using passive microwave data
topic surface soil moisture
Backpropagation Neural Network
space-borne L-band mission
European Space Agency‟s (ESA) Soil Moisture and Ocean Salinity (SMOS)
surface energy
soil moisture
water balance
Artificial Neural Network (ANN)
url http://hdl.handle.net/20.500.11937/1416