Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks
We propose an approach to develop a solar radiation model with spatial portability based on deep neural networks (DNNs). Weather station networks in South Korea between 33.5–37.9° N latitude were used to collect data for development and internal testing of the DNNs, respectively. Multiple sets of we...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | Environmental Research Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/1748-9326/acf6d4 |
| _version_ | 1797671636910997504 |
|---|---|
| author | Dae Gyoon Kang Yean-Uk Kim Shinwoo Hyun Kwang Soo Kim Junhwan Kim Chung-Kuen Lee Atsushi Maruyama Robert M Beresford David H Fleisher |
| author_facet | Dae Gyoon Kang Yean-Uk Kim Shinwoo Hyun Kwang Soo Kim Junhwan Kim Chung-Kuen Lee Atsushi Maruyama Robert M Beresford David H Fleisher |
| author_sort | Dae Gyoon Kang |
| collection | DOAJ |
| description | We propose an approach to develop a solar radiation model with spatial portability based on deep neural networks (DNNs). Weather station networks in South Korea between 33.5–37.9° N latitude were used to collect data for development and internal testing of the DNNs, respectively. Multiple sets of weather station data were selected for cross-validation of the DNNs by standard distance deviation (SDD) among training sites. The DNNs tended to have greater spatial portability when a threshold of spatial dispersion among training sites, e.g. 190 km of SDD, was met. The final formulation of the deep solar radiation (DSR) model was obtained from training sites associated with the threshold of SDD. The DSR model had RMSE values <4 MJ m ^−2 d ^−1 at external test sites in Japan that were within ±6° of the latitude boundary of the training sites. The relative difference between the outputs of crop yield simulations using observed versus estimated solar radiation inputs from the DSR model was about 4% at the test sites within the given boundary. These results indicate that the identification of the spatial dispersion threshold among training sites would aid the development of DNN models with reasonable spatial portability for estimation of solar radiation. |
| first_indexed | 2024-03-11T21:18:26Z |
| format | Article |
| id | doaj.art-f9b7009092f94d0d836b9c66826fac77 |
| institution | Directory Open Access Journal |
| issn | 1748-9326 |
| language | English |
| last_indexed | 2024-03-11T21:18:26Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Environmental Research Letters |
| spelling | doaj.art-f9b7009092f94d0d836b9c66826fac772023-09-28T15:25:02ZengIOP PublishingEnvironmental Research Letters1748-93262023-01-01181010402010.1088/1748-9326/acf6d4Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networksDae Gyoon Kang0https://orcid.org/0000-0001-9056-5272Yean-Uk Kim1https://orcid.org/0000-0002-9431-8575Shinwoo Hyun2https://orcid.org/0000-0002-8321-7648Kwang Soo Kim3https://orcid.org/0000-0003-2284-4389Junhwan Kim4https://orcid.org/0000-0003-2155-5294Chung-Kuen Lee5https://orcid.org/0000-0001-8699-4576Atsushi Maruyama6https://orcid.org/0000-0002-5901-9529Robert M Beresford7https://orcid.org/0000-0003-1854-4236David H Fleisher8https://orcid.org/0000-0002-0631-3986National Center for Agro-Meteorology , Seoul, Republic of KoreaLeibniz Centre for Agricultural Landscape Research (ZALF) , Müncheberg, Germany; Research Institute of Agriculture and Life Sciences, Seoul National University , Seoul, Republic of KoreaDepartment of Agriculture, Forestry and Bioresources, Seoul National University , Seoul, Republic of KoreaResearch Institute of Agriculture and Life Sciences, Seoul National University , Seoul, Republic of Korea; Department of Agriculture, Forestry and Bioresources, Seoul National University , Seoul, Republic of KoreaKorea National University of Agriculture and Fisheries , Jeonju-si, Jeollabuk-do, Republic of KoreaDivision of Crop Post-Harvest Technology Research, National Institute of Crop Science , Suwon-si, Kyeonggi-do, Republic of KoreaInstitute for Agro-Environmental Sciences, National Agriculture and Food Research Organization , Tsukuba, Ibaraki, JapanThe New Zealand Institute for Plant and Food Research Limited , Private Bag 92169, Auckland 1142, New ZealandAdaptive Cropping Systems Laboratory, United States Department of Agriculture—Agricultural Research Service (USDA-ARS) , Beltsville, MD, United States of AmericaWe propose an approach to develop a solar radiation model with spatial portability based on deep neural networks (DNNs). Weather station networks in South Korea between 33.5–37.9° N latitude were used to collect data for development and internal testing of the DNNs, respectively. Multiple sets of weather station data were selected for cross-validation of the DNNs by standard distance deviation (SDD) among training sites. The DNNs tended to have greater spatial portability when a threshold of spatial dispersion among training sites, e.g. 190 km of SDD, was met. The final formulation of the deep solar radiation (DSR) model was obtained from training sites associated with the threshold of SDD. The DSR model had RMSE values <4 MJ m ^−2 d ^−1 at external test sites in Japan that were within ±6° of the latitude boundary of the training sites. The relative difference between the outputs of crop yield simulations using observed versus estimated solar radiation inputs from the DSR model was about 4% at the test sites within the given boundary. These results indicate that the identification of the spatial dispersion threshold among training sites would aid the development of DNN models with reasonable spatial portability for estimation of solar radiation.https://doi.org/10.1088/1748-9326/acf6d4solar radiationspatial portabilityartificial intelligenceempirical modelcrop model |
| spellingShingle | Dae Gyoon Kang Yean-Uk Kim Shinwoo Hyun Kwang Soo Kim Junhwan Kim Chung-Kuen Lee Atsushi Maruyama Robert M Beresford David H Fleisher Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks Environmental Research Letters solar radiation spatial portability artificial intelligence empirical model crop model |
| title | Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| title_full | Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| title_fullStr | Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| title_full_unstemmed | Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| title_short | Identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| title_sort | identification of a spatial distribution threshold for the development of a solar radiation model using deep neural networks |
| topic | solar radiation spatial portability artificial intelligence empirical model crop model |
| url | https://doi.org/10.1088/1748-9326/acf6d4 |
| work_keys_str_mv | AT daegyoonkang identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT yeanukkim identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT shinwoohyun identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT kwangsookim identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT junhwankim identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT chungkuenlee identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT atsushimaruyama identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT robertmberesford identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks AT davidhfleisher identificationofaspatialdistributionthresholdforthedevelopmentofasolarradiationmodelusingdeepneuralnetworks |