Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study
Soil analysis using near-infrared spectroscopy has shown great potential to be an alternative to traditional laboratory analysis, and there is continuously increasing interest in building large-scale soil spectral libraries (SSLs). However, due to issues such as high non-linearity in soil spectral d...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/7/1256 |
| _version_ | 1797212036000645120 |
|---|---|
| author | Tianyu Miao Wenjun Ji Baoguo Li Xicun Zhu Jianxin Yin Jiajie Yang Yuanfang Huang Yan Cao Dongheng Yao Xiangbin Kong |
| author_facet | Tianyu Miao Wenjun Ji Baoguo Li Xicun Zhu Jianxin Yin Jiajie Yang Yuanfang Huang Yan Cao Dongheng Yao Xiangbin Kong |
| author_sort | Tianyu Miao |
| collection | DOAJ |
| description | Soil analysis using near-infrared spectroscopy has shown great potential to be an alternative to traditional laboratory analysis, and there is continuously increasing interest in building large-scale soil spectral libraries (SSLs). However, due to issues such as high non-linearity in soil spectral data and complexity in soil spatial variation, the establishment of robust prediction models for soil spectral libraries remains a challenge. This study aimed to investigate the performance of deep learning algorithms, including long short-term memory (LSTM) and LSTM–convolutional neural networks (LSTM–CNN) integrated models, to predict the soil organic matter (SOM) of a provincial-scale SSL, and compare it to the normally used local weighted regression (LWR) model. The Hebei soil spectral library (HSSL) contains 425 topsoil samples (0–20 cm), of which every 3 soil samples were collected from dry land, irrigated land, and paddy fields, respectively, in different counties of Hebei Province, China. The results show that the accuracy of the validation dataset rank as follows: LSTM–CNN (R<sup>2</sup><sub>p</sub> = 0.96, RMSE<sub>p</sub> = 1.66 g/kg) > LSTM (R<sup>2</sup><sub>p</sub> = 0.83, RMSE<sub>p</sub> = 3.42 g/kg) > LWR (R<sup>2</sup><sub>p</sub> = 0.82, RMSE<sub>p</sub> = 3.79 g/kg). The LSTM–CNN model performed the best, mainly due to its comprehensive ability to effectively extract spatial and temporal features. Meanwhile, the LSTM model achieved higher accuracy than the LWR model, owing to its built-in memory unit and its advantage of faster feature band extraction. Thus, it was suggested to use deep learning algorithms for SOM predictions in SSLs. However, their performance on larger-scale SSLs such as continental/global SSLs still needs to be further investigated. |
| first_indexed | 2024-04-24T10:35:59Z |
| format | Article |
| id | doaj.art-b406ba0a6c8140f38f28abab11bdcccf |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-04-24T10:35:59Z |
| publishDate | 2024-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-b406ba0a6c8140f38f28abab11bdcccf2024-04-12T13:25:46ZengMDPI AGRemote Sensing2072-42922024-04-01167125610.3390/rs16071256Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case StudyTianyu Miao0Wenjun Ji1Baoguo Li2Xicun Zhu3Jianxin Yin4Jiajie Yang5Yuanfang Huang6Yan Cao7Dongheng Yao8Xiangbin Kong9College of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege Resources and Environment, Shandong Agricultural University, Taian 271001, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaInstitute of Botany, Chinese Academy of Sciences, Beijing 100093, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100193, ChinaSoil analysis using near-infrared spectroscopy has shown great potential to be an alternative to traditional laboratory analysis, and there is continuously increasing interest in building large-scale soil spectral libraries (SSLs). However, due to issues such as high non-linearity in soil spectral data and complexity in soil spatial variation, the establishment of robust prediction models for soil spectral libraries remains a challenge. This study aimed to investigate the performance of deep learning algorithms, including long short-term memory (LSTM) and LSTM–convolutional neural networks (LSTM–CNN) integrated models, to predict the soil organic matter (SOM) of a provincial-scale SSL, and compare it to the normally used local weighted regression (LWR) model. The Hebei soil spectral library (HSSL) contains 425 topsoil samples (0–20 cm), of which every 3 soil samples were collected from dry land, irrigated land, and paddy fields, respectively, in different counties of Hebei Province, China. The results show that the accuracy of the validation dataset rank as follows: LSTM–CNN (R<sup>2</sup><sub>p</sub> = 0.96, RMSE<sub>p</sub> = 1.66 g/kg) > LSTM (R<sup>2</sup><sub>p</sub> = 0.83, RMSE<sub>p</sub> = 3.42 g/kg) > LWR (R<sup>2</sup><sub>p</sub> = 0.82, RMSE<sub>p</sub> = 3.79 g/kg). The LSTM–CNN model performed the best, mainly due to its comprehensive ability to effectively extract spatial and temporal features. Meanwhile, the LSTM model achieved higher accuracy than the LWR model, owing to its built-in memory unit and its advantage of faster feature band extraction. Thus, it was suggested to use deep learning algorithms for SOM predictions in SSLs. However, their performance on larger-scale SSLs such as continental/global SSLs still needs to be further investigated.https://www.mdpi.com/2072-4292/16/7/1256deep learninglong short-term memory (LSTM)long short-term memory–convolutional neural networks (LSTM–CNN)near-infrared (NIR)soil spectral library |
| spellingShingle | Tianyu Miao Wenjun Ji Baoguo Li Xicun Zhu Jianxin Yin Jiajie Yang Yuanfang Huang Yan Cao Dongheng Yao Xiangbin Kong Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study Remote Sensing deep learning long short-term memory (LSTM) long short-term memory–convolutional neural networks (LSTM–CNN) near-infrared (NIR) soil spectral library |
| title | Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study |
| title_full | Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study |
| title_fullStr | Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study |
| title_full_unstemmed | Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study |
| title_short | Advanced Soil Organic Matter Prediction with a Regional Soil NIR Spectral Library Using Long Short-Term Memory–Convolutional Neural Networks: A Case Study |
| title_sort | advanced soil organic matter prediction with a regional soil nir spectral library using long short term memory convolutional neural networks a case study |
| topic | deep learning long short-term memory (LSTM) long short-term memory–convolutional neural networks (LSTM–CNN) near-infrared (NIR) soil spectral library |
| url | https://www.mdpi.com/2072-4292/16/7/1256 |
| work_keys_str_mv | AT tianyumiao advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT wenjunji advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT baoguoli advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT xicunzhu advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT jianxinyin advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT jiajieyang advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT yuanfanghuang advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT yancao advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT donghengyao advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy AT xiangbinkong advancedsoilorganicmatterpredictionwitharegionalsoilnirspectrallibraryusinglongshorttermmemoryconvolutionalneuralnetworksacasestudy |