Brazilian Beaches and Dunes Status - Three Decades of Detection Using Machine Learning.

IGARSS(2021)

引用 0|浏览1
暂无评分
摘要
Brazilian beaches play an essential role in maintaining human populations and conserving biodiversity, being primarily impacted by anthropogenic and natural activities, such as tourism, pollution, and sea-level variations. The present work uses the Random Forest (RF) algorithm to automatically and systematically classify sandy coastal bodies as “beaches or dunes” from 1985 to 2019 throughout the Brazilian Coastal Zone (BCZ). The areas covered by beaches and dunes reduced ∼ 714 km 2 (around 17% of the national cover) in Brazil. The South and Southeast regions experimented with more significant area reductions compared to the North and Northeast regions. Regarding concordance levels with pre-existent data, the general agreement of 1985 and 2018 data reached 0.84 and 0.97 of global agreement levels.
更多
查看译文
关键词
Beaches,Dunes,Machine Learning,Brazil
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要