Evaluating individual tree species classification performance across diverse environments
Vegetation species mapping using airborne imaging spectroscopy yields accurate results and is important for advancing conservation objectives and biogeographic studies. As these data become more readily available owing to the upcoming launch of spaceborne imaging spectrometers, it is necessary to un...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2024-01-01
|
| Series: | Environmental Research: Ecology |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2752-664X/ad1f49 |
| _version_ | 1797313065517056000 |
|---|---|
| author | M M Seeley N R Vaughn G A Asner |
| author_facet | M M Seeley N R Vaughn G A Asner |
| author_sort | M M Seeley |
| collection | DOAJ |
| description | Vegetation species mapping using airborne imaging spectroscopy yields accurate results and is important for advancing conservation objectives and biogeographic studies. As these data become more readily available owing to the upcoming launch of spaceborne imaging spectrometers, it is necessary to understand how these data can be used to consistently classify species across large geographic scales. However, few studies have attempted to map species across multiple ecosystems; therefore, little is known regarding the effect of intra-specific variation on the mapping of a single species across a wide range of environments and among varying backgrounds of other non-target species. To explore this effect, we developed and tested species classification models for Metrosideros polymorpha , a highly polymorphic canopy species endemic to Hawai’i, which is found in a diverse array of ecosystems. We compared the accuracies of support vector machine (SVM) and random forest models trained on canopy reflectance data from each of eight distinct ecosystems (ecosystem-specific) and a universal model trained on data from all ecosystems. When applied to ecosystem-specific test datasets, the ecosystem-specific models outperformed the universal model; however, the universal model retained high (>81%) accuracies across all ecosystems. Additionally, we found that models from ecosystems with broad variation in M. polymorpha canopy traits, as estimated using chemometric equations applied to canopy spectra, accurately predicted M. polymorpha in other ecosystems. While species classifications across ecosystems can yield accurate results, these results will require sampling procedures that capture the intra-specific variation of the target species. |
| first_indexed | 2024-03-08T02:23:13Z |
| format | Article |
| id | doaj.art-856fae3af99c47908733ddd95ac854f0 |
| institution | Directory Open Access Journal |
| issn | 2752-664X |
| language | English |
| last_indexed | 2024-03-08T02:23:13Z |
| publishDate | 2024-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Environmental Research: Ecology |
| spelling | doaj.art-856fae3af99c47908733ddd95ac854f02024-02-13T14:22:35ZengIOP PublishingEnvironmental Research: Ecology2752-664X2024-01-013101100110.1088/2752-664X/ad1f49Evaluating individual tree species classification performance across diverse environmentsM M Seeley0https://orcid.org/0000-0003-1945-1162N R Vaughn1https://orcid.org/0000-0003-0428-2909G A Asner2https://orcid.org/0000-0001-7893-6421Center for Global Discovery and Conservation Science, Arizona State University , Tempe, AZ 85281, United States of America; School of Geographic Sciences and Urban Planning, Arizona State University , Tempe, AZ 85281, United States of AmericaCenter for Global Discovery and Conservation Science, Arizona State University , Tempe, AZ 85281, United States of AmericaCenter for Global Discovery and Conservation Science, Arizona State University , Tempe, AZ 85281, United States of America; School of Geographic Sciences and Urban Planning, Arizona State University , Tempe, AZ 85281, United States of AmericaVegetation species mapping using airborne imaging spectroscopy yields accurate results and is important for advancing conservation objectives and biogeographic studies. As these data become more readily available owing to the upcoming launch of spaceborne imaging spectrometers, it is necessary to understand how these data can be used to consistently classify species across large geographic scales. However, few studies have attempted to map species across multiple ecosystems; therefore, little is known regarding the effect of intra-specific variation on the mapping of a single species across a wide range of environments and among varying backgrounds of other non-target species. To explore this effect, we developed and tested species classification models for Metrosideros polymorpha , a highly polymorphic canopy species endemic to Hawai’i, which is found in a diverse array of ecosystems. We compared the accuracies of support vector machine (SVM) and random forest models trained on canopy reflectance data from each of eight distinct ecosystems (ecosystem-specific) and a universal model trained on data from all ecosystems. When applied to ecosystem-specific test datasets, the ecosystem-specific models outperformed the universal model; however, the universal model retained high (>81%) accuracies across all ecosystems. Additionally, we found that models from ecosystems with broad variation in M. polymorpha canopy traits, as estimated using chemometric equations applied to canopy spectra, accurately predicted M. polymorpha in other ecosystems. While species classifications across ecosystems can yield accurate results, these results will require sampling procedures that capture the intra-specific variation of the target species.https://doi.org/10.1088/2752-664X/ad1f49Metrosideros polymorphaspecies classificationssupport vector machinerandom forestecosystemsvegetation mapping |
| spellingShingle | M M Seeley N R Vaughn G A Asner Evaluating individual tree species classification performance across diverse environments Environmental Research: Ecology Metrosideros polymorpha species classifications support vector machine random forest ecosystems vegetation mapping |
| title | Evaluating individual tree species classification performance across diverse environments |
| title_full | Evaluating individual tree species classification performance across diverse environments |
| title_fullStr | Evaluating individual tree species classification performance across diverse environments |
| title_full_unstemmed | Evaluating individual tree species classification performance across diverse environments |
| title_short | Evaluating individual tree species classification performance across diverse environments |
| title_sort | evaluating individual tree species classification performance across diverse environments |
| topic | Metrosideros polymorpha species classifications support vector machine random forest ecosystems vegetation mapping |
| url | https://doi.org/10.1088/2752-664X/ad1f49 |
| work_keys_str_mv | AT mmseeley evaluatingindividualtreespeciesclassificationperformanceacrossdiverseenvironments AT nrvaughn evaluatingindividualtreespeciesclassificationperformanceacrossdiverseenvironments AT gaasner evaluatingindividualtreespeciesclassificationperformanceacrossdiverseenvironments |