Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis
Remote sensing approaches that could identify species of submerged aquatic vegetation (SAV) and measure their extent in lake littoral zones would greatly enhance SAV study and management, especially if these approaches can provide faster or more accurate results than traditional field methods. Remot...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/10/2336 |
| _version_ | 1827666847537823744 |
|---|---|
| author | Colin Brooks Amanda Grimm Amy M. Marcarelli Nicholas P. Marion Robert Shuchman Michael Sayers |
| author_facet | Colin Brooks Amanda Grimm Amy M. Marcarelli Nicholas P. Marion Robert Shuchman Michael Sayers |
| author_sort | Colin Brooks |
| collection | DOAJ |
| description | Remote sensing approaches that could identify species of submerged aquatic vegetation (SAV) and measure their extent in lake littoral zones would greatly enhance SAV study and management, especially if these approaches can provide faster or more accurate results than traditional field methods. Remote sensing with multispectral sensors can provide this capability, but SAV identification with this technology must address the challenges of light extinction in aquatic environments where chlorophyll, dissolved organic carbon, and suspended minerals can affect water clarity and the strength of the sensed light signal. Here, we present an uncrewed aerial system (UAS)-enabled methodology to identify the extent of the invasive SAV species <i>Myriophyllum spicatum</i> (Eurasian watermilfoil, or EWM), primarily using a six-band Tetracam multispectral camera, flown over sites in the Les Cheneaux Islands area of northwestern Lake Huron, Michigan, USA. We analyzed water chemistry and light data and found our sites clustered into sites with higher and lower water clarity, although all sites had relatively high water clarity. The overall average accuracy achieved was 76.7%, with 78.7% producer’s and 77.6% user’s accuracy for the EWM. These accuracies were higher than previously reported from other studies that used remote sensing to map SAV. Our study found that two tested scale parameters did not lead to significantly different classification accuracies between sites with higher and lower water clarity. The EWM classification methodology described here should be applicable to other SAV species, especially if they have growth patterns that lead to high amounts of biomass relative to other species in the upper water column, which can be detected with the type of red-edge and infrared sensors deployed for this study. |
| first_indexed | 2024-03-10T01:57:26Z |
| format | Article |
| id | doaj.art-f7bb01f04781458aa3d6ac88cf60d812 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T01:57:26Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-f7bb01f04781458aa3d6ac88cf60d8122023-11-23T12:54:29ZengMDPI AGRemote Sensing2072-42922022-05-011410233610.3390/rs14102336Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery AnalysisColin Brooks0Amanda Grimm1Amy M. Marcarelli2Nicholas P. Marion3Robert Shuchman4Michael Sayers5Michigan Tech Research Institute, Michigan Technological University, 3600 Green Court, Suite 100, Ann Arbor, MI 48105, USAGreat Lakes Commission, 1300 Victors Way, Suite 1350, Ann Arbor, MI 48108, USADepartment of Biological Sciences, Michigan Technological University, 1400 Townsend Dr., Houghton, MI 49931, USAMichigan Tech Research Institute, Michigan Technological University, 3600 Green Court, Suite 100, Ann Arbor, MI 48105, USAMichigan Tech Research Institute, Michigan Technological University, 3600 Green Court, Suite 100, Ann Arbor, MI 48105, USAMichigan Tech Research Institute, Michigan Technological University, 3600 Green Court, Suite 100, Ann Arbor, MI 48105, USARemote sensing approaches that could identify species of submerged aquatic vegetation (SAV) and measure their extent in lake littoral zones would greatly enhance SAV study and management, especially if these approaches can provide faster or more accurate results than traditional field methods. Remote sensing with multispectral sensors can provide this capability, but SAV identification with this technology must address the challenges of light extinction in aquatic environments where chlorophyll, dissolved organic carbon, and suspended minerals can affect water clarity and the strength of the sensed light signal. Here, we present an uncrewed aerial system (UAS)-enabled methodology to identify the extent of the invasive SAV species <i>Myriophyllum spicatum</i> (Eurasian watermilfoil, or EWM), primarily using a six-band Tetracam multispectral camera, flown over sites in the Les Cheneaux Islands area of northwestern Lake Huron, Michigan, USA. We analyzed water chemistry and light data and found our sites clustered into sites with higher and lower water clarity, although all sites had relatively high water clarity. The overall average accuracy achieved was 76.7%, with 78.7% producer’s and 77.6% user’s accuracy for the EWM. These accuracies were higher than previously reported from other studies that used remote sensing to map SAV. Our study found that two tested scale parameters did not lead to significantly different classification accuracies between sites with higher and lower water clarity. The EWM classification methodology described here should be applicable to other SAV species, especially if they have growth patterns that lead to high amounts of biomass relative to other species in the upper water column, which can be detected with the type of red-edge and infrared sensors deployed for this study.https://www.mdpi.com/2072-4292/14/10/2336multispectralaquaticinvasiveremote sensingUASmacrophyte |
| spellingShingle | Colin Brooks Amanda Grimm Amy M. Marcarelli Nicholas P. Marion Robert Shuchman Michael Sayers Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis Remote Sensing multispectral aquatic invasive remote sensing UAS macrophyte |
| title | Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis |
| title_full | Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis |
| title_fullStr | Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis |
| title_full_unstemmed | Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis |
| title_short | Classification of Eurasian Watermilfoil (<i>Myriophyllum spicatum</i>) Using Drone-Enabled Multispectral Imagery Analysis |
| title_sort | classification of eurasian watermilfoil i myriophyllum spicatum i using drone enabled multispectral imagery analysis |
| topic | multispectral aquatic invasive remote sensing UAS macrophyte |
| url | https://www.mdpi.com/2072-4292/14/10/2336 |
| work_keys_str_mv | AT colinbrooks classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis AT amandagrimm classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis AT amymmarcarelli classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis AT nicholaspmarion classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis AT robertshuchman classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis AT michaelsayers classificationofeurasianwatermilfoilimyriophyllumspicatumiusingdroneenabledmultispectralimageryanalysis |