Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia
Machine learning (ML) was used to assess and predict urban air temperature (T<sub>air</sub>) considering the complexity of the terrain features in Yerevan (Armenia). The estimation was performed based on the Partial Least-Squares Regression (PLSR) model with a high number (30) of input v...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-05-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/11/2795 |
_version_ | 1797596822456238080 |
---|---|
author | Garegin Tepanosyan Shushanik Asmaryan Vahagn Muradyan Rima Avetisyan Azatuhi Hovsepyan Anahit Khlghatyan Grigor Ayvazyan Fabio Dell’Acqua |
author_facet | Garegin Tepanosyan Shushanik Asmaryan Vahagn Muradyan Rima Avetisyan Azatuhi Hovsepyan Anahit Khlghatyan Grigor Ayvazyan Fabio Dell’Acqua |
author_sort | Garegin Tepanosyan |
collection | DOAJ |
description | Machine learning (ML) was used to assess and predict urban air temperature (T<sub>air</sub>) considering the complexity of the terrain features in Yerevan (Armenia). The estimation was performed based on the Partial Least-Squares Regression (PLSR) model with a high number (30) of input variables. The relevant parameters include a newly purposed modification of spectral index IBI-SAVI, which turned out to strongly impact T<sub>air</sub> prediction together with land surface temperature (LST). Cross-validation analysis on temperature predictions across a station-centered 1000 m circular area revealed quite a high correlation (R<sup>2</sup><sub>Val</sub> = 0.77, RMSE<sub>Val</sub> = 1.58) between the predicted and measured T<sub>air</sub> from the test set. It was concluded the remote sensing is an effective tool to estimate T<sub>air</sub> distribution where a dense network of weather stations is not available. However, further developments will include incorporation of additional weather parameters from the weather stations, such as precipitation and wind speed, as well as the use of non-parametric ML techniques. |
first_indexed | 2024-03-11T02:58:26Z |
format | Article |
id | doaj.art-3f93a5cfa6b34b4f8e965966dccff056 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-11T02:58:26Z |
publishDate | 2023-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-3f93a5cfa6b34b4f8e965966dccff0562023-11-18T08:28:47ZengMDPI AGRemote Sensing2072-42922023-05-011511279510.3390/rs15112795Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, ArmeniaGaregin Tepanosyan0Shushanik Asmaryan1Vahagn Muradyan2Rima Avetisyan3Azatuhi Hovsepyan4Anahit Khlghatyan5Grigor Ayvazyan6Fabio Dell’Acqua7Centre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaCentre for Ecological-Noosphere Studies, National Academy of Sciences of Armenia, Abovyan Street 68, Yerevan 0025, ArmeniaDepartment of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, ItalyMachine learning (ML) was used to assess and predict urban air temperature (T<sub>air</sub>) considering the complexity of the terrain features in Yerevan (Armenia). The estimation was performed based on the Partial Least-Squares Regression (PLSR) model with a high number (30) of input variables. The relevant parameters include a newly purposed modification of spectral index IBI-SAVI, which turned out to strongly impact T<sub>air</sub> prediction together with land surface temperature (LST). Cross-validation analysis on temperature predictions across a station-centered 1000 m circular area revealed quite a high correlation (R<sup>2</sup><sub>Val</sub> = 0.77, RMSE<sub>Val</sub> = 1.58) between the predicted and measured T<sub>air</sub> from the test set. It was concluded the remote sensing is an effective tool to estimate T<sub>air</sub> distribution where a dense network of weather stations is not available. However, further developments will include incorporation of additional weather parameters from the weather stations, such as precipitation and wind speed, as well as the use of non-parametric ML techniques.https://www.mdpi.com/2072-4292/15/11/2795urban air temperatureland surface temperaturemultiple independent variablesurban heatremote sensing datamachine learning (ML) |
spellingShingle | Garegin Tepanosyan Shushanik Asmaryan Vahagn Muradyan Rima Avetisyan Azatuhi Hovsepyan Anahit Khlghatyan Grigor Ayvazyan Fabio Dell’Acqua Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia Remote Sensing urban air temperature land surface temperature multiple independent variables urban heat remote sensing data machine learning (ML) |
title | Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia |
title_full | Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia |
title_fullStr | Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia |
title_full_unstemmed | Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia |
title_short | Machine Learning-Based Modeling of Air Temperature in the Complex Environment of Yerevan City, Armenia |
title_sort | machine learning based modeling of air temperature in the complex environment of yerevan city armenia |
topic | urban air temperature land surface temperature multiple independent variables urban heat remote sensing data machine learning (ML) |
url | https://www.mdpi.com/2072-4292/15/11/2795 |
work_keys_str_mv | AT garegintepanosyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT shushanikasmaryan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT vahagnmuradyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT rimaavetisyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT azatuhihovsepyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT anahitkhlghatyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT grigorayvazyan machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia AT fabiodellacqua machinelearningbasedmodelingofairtemperatureinthecomplexenvironmentofyerevancityarmenia |