2021
DOI: 10.1111/ddi.13244
|View full text |Cite
|
Sign up to set email alerts
|

No place to hide: Rare plant detection through remote sensing

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
14
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
6
2
1

Relationship

1
8

Authors

Journals

citations
Cited by 24 publications
(15 citation statements)
references
References 102 publications
1
14
0
Order By: Relevance
“…On the other hand, the wide variety of potentially relevant predictors for rare plants that can be derived from RS (related to vegetation, humidity, forest structure, topography, etc.) [ 90 ], can allow a more realistic approach to the environment-species relationship, which can be particularly useful for species with complex ecological niches. Thus, our methodology can play an important role in filling existing knowledge gaps on bryophyte distribution ranges, as well as their ecological preferences, in largely unexplored regions such as boreal forests [ 36 ].…”
Section: Discussionmentioning
confidence: 99%
“…On the other hand, the wide variety of potentially relevant predictors for rare plants that can be derived from RS (related to vegetation, humidity, forest structure, topography, etc.) [ 90 ], can allow a more realistic approach to the environment-species relationship, which can be particularly useful for species with complex ecological niches. Thus, our methodology can play an important role in filling existing knowledge gaps on bryophyte distribution ranges, as well as their ecological preferences, in largely unexplored regions such as boreal forests [ 36 ].…”
Section: Discussionmentioning
confidence: 99%
“…Confinement of the plants to specific life forms and consolidating these data within the dataset is a good tool for studies of species richness and its determinants (Liu et al, 2019;Sandanov et al, 2020), for analysis of growth-form plasticity in rare and endemic species (Mills & Schwartz, 2005), and for detection of rare plants using remote sensing (Cerrejón et al, 2021). Our results revealed the prevalence of short rhizome plants among rare and endangered species.…”
Section: Discussionmentioning
confidence: 66%
“…Evaluation metrics that are weighted by the number of samples per class, such as Weighted F1 which was one metric used in this competition, favor models that are most accurate for abundant classes. However, for many ecological questions and applications, having strong predictions across all species, especially the rare species is important (Leitão et al, 2016;Dee et al, 2019;Cerrejón et al, 2021), and therefore an evaluation score such as Macro F1 is most appropriate. Understanding patterns of taxonomic and functional diversity or evaluating the impact of climate changes and extreme disturbance events on species are examples where poor accuracy of rare species will impact the ability to use the predictions because of the uncertainty in the predictions.…”
Section: Classificationmentioning
confidence: 99%