Application of Deep Learning for Classification of Intertidal Eelgrass from Drone-Acquired Imagery
, Nam Nguyen 2, Jonathan Ventura 2,†, Andrew Fricker 3,†, Sadie Calhoun 3, Jennifer O’Leary 4, Mauriça Fitzgibbons 5, Ian Robbins 6 and Ryan K. Walter 6,†
Round 1
Reviewer 1 Report
This research introduced the application of DeepLab for the classification of intertidal eelgrass from drone-acquired imagery. However, the author should add more background and details of the model. I suggest rewriting the result and discussion to support the application(Such as showing the final whole map of the study area and testing the model parameter sensitivity, comparing this result to other related maps in a proper way).
Author Response
Thank you so much, Reviewer 1! We are attaching our responses here.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors have demonstrated application of drone surveys on mapping eelgrass (Zostera marina) meadows in the Morro Bay environment. The manuscript could be improved once authors try to address the following issues:
a) Introduction section:
Authors have sufficiently described usefulness of drone-based remote sensing, but they should also critically review limitations or problems associated with drone image classification that motivated them to undertake this study. Is there any issue associated with tide level or water clarity (turbidity) or sun glint affect on drone image quality and processing?
Drone is less useful for large-scale seagrass mapping. Authors have missed some recent papers that have recently shown applications of low-cost Landsat remote sensing for large scale underwater substrate mapping. For example:
Gallagher, A.J., Brownscombe, J.W., Alsudairy, N.A., Casagrande, A.B., Fu, C., Harding, L., Harris, S.D., Hammerschlag, N., Howe, W., Huertas, A.D., Kattan, S., Kough, A.S., Musgrove, A., Payne, N.L., Phillips, A., Shea, B.D., Shipley, O.N., Sumaila, U.R., Hossain, M.S., Duarte, C.M., 2022. Tiger sharks support the characterization of the world’s largest seagrass ecosystem. Nat. Commun. 13, 6328. https://doi.org/10.1038/s41467-022-33926-1
Wabnitz, C.C., Andréfouët, S., Torres-Pulliza, D., Müller-Karger, F.E., Kramer, P.A., 2008. Regional-scale seagrass habitat mapping in the Wider Caribbean region using Landsat sensors: Applications to conservation and ecology. Remote Sens. Environ. 112, 3455–3467. https://doi.org/http://dx.doi.org/10.1016/j.rse.2008.01.020
O’Neill, J.D., Costa, M., 2013. Mapping eelgrass (Zostera marina) in the Gulf Islands National Park Reserve of Canada using high spatial resolution satellite and airborne imagery. Remote Sens. Environ. 133, 152–167. https://doi.org/https://doi.org/10.1016/j.rse.2013.02.010
b) Discussion section: authors should discuss what are the limitations of drone-based remote sensing for regional scale mapping; how drone data can aid country to regional scale mapping.
If drone images acquired in two different times with different tide heights, what are the remote sensing considerations to remove light attenuation effects as a result of tidal variations?
Author Response
Thank you so much, Reviewer 2! We are attaching our responses here.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The revised manuscript is very interesting and I have no further questions.
