期刊
SCIENCE OF THE TOTAL ENVIRONMENT
卷 861, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.scitotenv.2022.160622
关键词
Deep learning; Geomorphon; LiDAR; Model transferability; Remote sensing; U-Net
Accurate and up-to-date wetland maps are crucial for landscape scale wetland conservation. This study successfully developed a deep learning model to automatically map wetlands at high resolution using free data and efficient deep learning models that do not require manual feature engineering. This approach can be a game changer in informing restoration and development decisions for wetlands due to the dynamic nature and important ecosystem services they provide.
Landscape scale wetland conservation requires accurate, up-to-date wetland maps. The most useful approaches to cre-ating such maps are automated, spatially generalizable, temporally repeatable, and can be applied at large spatial scales. However, mapping wetlands with predictive models is challenging due to the highly variable characteristics of wetlands in both space and time. Currently, most approaches are limited by coarse resolution, commercial data, and geographic specificity. Here, we trained a deep learning model and evaluated its ability to automatically map wet-lands at landscape scale in a variety of geographies. We trained a U-Net architecture to map wetlands at 1-meter spatial resolution with the following remotely sensed covariates: multispectral data from the National Agriculture Imagery Program and the Sentinel-2 satellite system, and two LiDAR-derived datasets, intensity and geomorphons. The full model mapped wetlands accurately (94 % accuracy, 96.5 % precision, 95.2 % AUC) at 1-meter resolution. Post hoc model evaluation showed that the model correctly predicted wetlands even in areas that had incorrect label/train-ing data, which penalized the recall rate (90.2 %). Applying the model in a new geography resulted in poor perfor-mance (precision = similar to 80 %, recall = 48 %). However, limited retraining in this geography improved model performance substantially, demonstrating an effective means to create a spatially generalizable model. We demon-strate wetlands can be mapped at high-resolution (1 m) using free data and efficient deep-learning models that do not require manual feature engineering. Including LiDAR and geomorphons as input data improved model accuracy by 2 %, and where these data are unavailable a simpler model can efficiently map wetlands. Given the dynamic nature of wetlands and the important ecosystem services they provide, high-resolution mapping can be a game changer in terms of informing restoration and development decisions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据