Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database
The development and provision of soil spectral library (SSL) could facilitate the application of near infrared (NIR) spectroscopy for economical, accurate, and efficient determination of soil organic matter (SOM). In this work, the performances of partial least squares regression (PLSR) and convolut...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Czech Academy of Agricultural Sciences
2023-08-01
|
| Series: | Soil and Water Research |
| Subjects: | |
| Online Access: | https://swr.agriculturejournals.cz/artkey/swr-202303-0002_strategies-and-methods-for-predicting-soil-organic-matter-at-the-field-scale-based-on-the-provincial-near-infra.php |
| _version_ | 1797730324836253696 |
|---|---|
| author | Shengyao Jia Chunbo Hong Hongyang Li Yuchan Li Siyuan Hu |
| author_facet | Shengyao Jia Chunbo Hong Hongyang Li Yuchan Li Siyuan Hu |
| author_sort | Shengyao Jia |
| collection | DOAJ |
| description | The development and provision of soil spectral library (SSL) could facilitate the application of near infrared (NIR) spectroscopy for economical, accurate, and efficient determination of soil organic matter (SOM). In this work, the performances of partial least squares regression (PLSR) and convolutional neural network (CNN) combined with the datasets of Zhejiang provincial SSL (ZSSL) and the feature subset (FS) were compared for the prediction of SOM at the target field. The FS dataset was chosen from ZSSL based on similarity to the spectral characteristics of the target samples. The results showed that compared with modelling using ZSSL, modelling using FS can greatly improve the prediction accuracy of the PLSR model, but the impact on the performance of the CNN model was limited. The method of mean squared Euclidean distance (MSD) was an effective way for determining the optimal spiking sample size for the PLSR model only using the spectral data of the spiking subset and the prediction set. The PLSR model combined with the FS dataset and the spiking subset determined by MSD achieved the optimal prediction results among all developed models, which is an accurate and easy-to-implement solution for the SOM determination based on ZSSL. |
| first_indexed | 2024-03-12T11:43:37Z |
| format | Article |
| id | doaj.art-8671a9e504bf412daa2ae0dd812961be |
| institution | Directory Open Access Journal |
| issn | 1801-5395 1805-9384 |
| language | English |
| last_indexed | 2024-03-12T11:43:37Z |
| publishDate | 2023-08-01 |
| publisher | Czech Academy of Agricultural Sciences |
| record_format | Article |
| series | Soil and Water Research |
| spelling | doaj.art-8671a9e504bf412daa2ae0dd812961be2023-08-31T12:44:29ZengCzech Academy of Agricultural SciencesSoil and Water Research1801-53951805-93842023-08-0118315816810.17221/133/2022-SWRswr-202303-0002Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral databaseShengyao Jia0Chunbo Hong1Hongyang Li2Yuchan Li3Siyuan Hu4College of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, P.R. ChinaCollege of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, P.R. ChinaCollege of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, P.R. ChinaCollege of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, P.R. ChinaZhejiang Provincial Emergency Management Science Research Institute, Hangzhou, P.R. ChinaThe development and provision of soil spectral library (SSL) could facilitate the application of near infrared (NIR) spectroscopy for economical, accurate, and efficient determination of soil organic matter (SOM). In this work, the performances of partial least squares regression (PLSR) and convolutional neural network (CNN) combined with the datasets of Zhejiang provincial SSL (ZSSL) and the feature subset (FS) were compared for the prediction of SOM at the target field. The FS dataset was chosen from ZSSL based on similarity to the spectral characteristics of the target samples. The results showed that compared with modelling using ZSSL, modelling using FS can greatly improve the prediction accuracy of the PLSR model, but the impact on the performance of the CNN model was limited. The method of mean squared Euclidean distance (MSD) was an effective way for determining the optimal spiking sample size for the PLSR model only using the spectral data of the spiking subset and the prediction set. The PLSR model combined with the FS dataset and the spiking subset determined by MSD achieved the optimal prediction results among all developed models, which is an accurate and easy-to-implement solution for the SOM determination based on ZSSL.https://swr.agriculturejournals.cz/artkey/swr-202303-0002_strategies-and-methods-for-predicting-soil-organic-matter-at-the-field-scale-based-on-the-provincial-near-infra.phpconvolutional neural networksoil organic contentsoil spectral libraryspiking sample sizestrategy |
| spellingShingle | Shengyao Jia Chunbo Hong Hongyang Li Yuchan Li Siyuan Hu Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database Soil and Water Research convolutional neural network soil organic content soil spectral library spiking sample size strategy |
| title | Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| title_full | Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| title_fullStr | Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| title_full_unstemmed | Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| title_short | Strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| title_sort | strategies and methods for predicting soil organic matter at the field scale based on the provincial near infrared spectral database |
| topic | convolutional neural network soil organic content soil spectral library spiking sample size strategy |
| url | https://swr.agriculturejournals.cz/artkey/swr-202303-0002_strategies-and-methods-for-predicting-soil-organic-matter-at-the-field-scale-based-on-the-provincial-near-infra.php |
| work_keys_str_mv | AT shengyaojia strategiesandmethodsforpredictingsoilorganicmatteratthefieldscalebasedontheprovincialnearinfraredspectraldatabase AT chunbohong strategiesandmethodsforpredictingsoilorganicmatteratthefieldscalebasedontheprovincialnearinfraredspectraldatabase AT hongyangli strategiesandmethodsforpredictingsoilorganicmatteratthefieldscalebasedontheprovincialnearinfraredspectraldatabase AT yuchanli strategiesandmethodsforpredictingsoilorganicmatteratthefieldscalebasedontheprovincialnearinfraredspectraldatabase AT siyuanhu strategiesandmethodsforpredictingsoilorganicmatteratthefieldscalebasedontheprovincialnearinfraredspectraldatabase |