Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery
Crown fire damage is a mixture of three principal fire-related components: charred material, scorched foliage, and unaltered green canopy. This study estimated the abundance of these physical alterations in two immediate post-fire Mediterranean forest contexts (Portugal and Italy) by applying linear...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2023-12-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/17538947.2023.2243900 |
| _version_ | 1797678462353276928 |
|---|---|
| author | Giandomenico De Luca Giuseppe Modica João M. N. Silva Salvatore Praticò José M.C. Pereira |
| author_facet | Giandomenico De Luca Giuseppe Modica João M. N. Silva Salvatore Praticò José M.C. Pereira |
| author_sort | Giandomenico De Luca |
| collection | DOAJ |
| description | Crown fire damage is a mixture of three principal fire-related components: charred material, scorched foliage, and unaltered green canopy. This study estimated the abundance of these physical alterations in two immediate post-fire Mediterranean forest contexts (Portugal and Italy) by applying linear spectral mixture analysis (LSMA) on Sentinel-2 imagery. The tree crowns fire damage was subsequently mapped, integrating fractional abundance information in a random forest (RF) algorithm, comparing the accuracy resulting from the adoption of generic or image spectral libraries as the primary investigative goal. Although image-derived endmembers resulted in more effectiveness in terms of fire-related components abundance quantification (LMSA-derived RMSE < 0.1), the F-scores always were ≥ 90% whether generic endmembers or image endmembers derived information was employed. The environmental heterogeneity of the two study areas affected the fire severity gradients, with a prevalence of the charred (PT) (45–46%) and green class (IT) (44–53%). Post-fire temporal monitoring was initialized by applying the proposed strategies, and the preliminary results showed a positive recovery trend in forest vegetation from the first year following the fire event, with a reduced charcoal predominance and an increasing proportion of green components. |
| first_indexed | 2024-03-11T23:00:07Z |
| format | Article |
| id | doaj.art-fbf89e30fc9d4761a512471dcb5025ca |
| institution | Directory Open Access Journal |
| issn | 1753-8947 1753-8955 |
| language | English |
| last_indexed | 2024-03-11T23:00:07Z |
| publishDate | 2023-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj.art-fbf89e30fc9d4761a512471dcb5025ca2023-09-21T15:09:03ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552023-12-011613162319810.1080/17538947.2023.22439002243900Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imageryGiandomenico De Luca0Giuseppe Modica1João M. N. Silva2Salvatore Praticò3José M.C. Pereira4Università degli Studi Mediterranea di Reggio CalabriaUniversità degli Studi di MessinaUniversity of LisbonUniversità degli Studi Mediterranea di Reggio CalabriaUniversity of LisbonCrown fire damage is a mixture of three principal fire-related components: charred material, scorched foliage, and unaltered green canopy. This study estimated the abundance of these physical alterations in two immediate post-fire Mediterranean forest contexts (Portugal and Italy) by applying linear spectral mixture analysis (LSMA) on Sentinel-2 imagery. The tree crowns fire damage was subsequently mapped, integrating fractional abundance information in a random forest (RF) algorithm, comparing the accuracy resulting from the adoption of generic or image spectral libraries as the primary investigative goal. Although image-derived endmembers resulted in more effectiveness in terms of fire-related components abundance quantification (LMSA-derived RMSE < 0.1), the F-scores always were ≥ 90% whether generic endmembers or image endmembers derived information was employed. The environmental heterogeneity of the two study areas affected the fire severity gradients, with a prevalence of the charred (PT) (45–46%) and green class (IT) (44–53%). Post-fire temporal monitoring was initialized by applying the proposed strategies, and the preliminary results showed a positive recovery trend in forest vegetation from the first year following the fire event, with a reduced charcoal predominance and an increasing proportion of green components.http://dx.doi.org/10.1080/17538947.2023.2243900post-fire assessmentfire severitypost-fire vegetation recoveryrandom forest (rf)scikit-learnfraction image extractionspectral unmixingendmemberscrown fire damage mappingfully constrained least squarespixel purity index |
| spellingShingle | Giandomenico De Luca Giuseppe Modica João M. N. Silva Salvatore Praticò José M.C. Pereira Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery International Journal of Digital Earth post-fire assessment fire severity post-fire vegetation recovery random forest (rf) scikit-learn fraction image extraction spectral unmixing endmembers crown fire damage mapping fully constrained least squares pixel purity index |
| title | Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery |
| title_full | Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery |
| title_fullStr | Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery |
| title_full_unstemmed | Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery |
| title_short | Assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on Sentinel-2 imagery |
| title_sort | assessing tree crown fire damage integrating linear spectral mixture analysis and supervised machine learning on sentinel 2 imagery |
| topic | post-fire assessment fire severity post-fire vegetation recovery random forest (rf) scikit-learn fraction image extraction spectral unmixing endmembers crown fire damage mapping fully constrained least squares pixel purity index |
| url | http://dx.doi.org/10.1080/17538947.2023.2243900 |
| work_keys_str_mv | AT giandomenicodeluca assessingtreecrownfiredamageintegratinglinearspectralmixtureanalysisandsupervisedmachinelearningonsentinel2imagery AT giuseppemodica assessingtreecrownfiredamageintegratinglinearspectralmixtureanalysisandsupervisedmachinelearningonsentinel2imagery AT joaomnsilva assessingtreecrownfiredamageintegratinglinearspectralmixtureanalysisandsupervisedmachinelearningonsentinel2imagery AT salvatorepratico assessingtreecrownfiredamageintegratinglinearspectralmixtureanalysisandsupervisedmachinelearningonsentinel2imagery AT josemcpereira assessingtreecrownfiredamageintegratinglinearspectralmixtureanalysisandsupervisedmachinelearningonsentinel2imagery |