遇见数据集

Locations of floating macroalgae on the West Greenland Shelf and selected NOAA Global Drifter Program Surface Drifter Trajectories

收藏
Zenodo2025-07-29 更新2026-05-26 收录
官方服务:

资源简介:

This dataset contains two files: floating_mat_locations.txt and NOAA_GDP_select.zip floating_mat_locations.txt This text file contains all the locations of floating mats of macroalgae that were detected by analyzing Sentinel-2 satellite imagery of the West Greenland Shelf. The longitude and latitude of the centroids of each floating mat are reported. Sentinel-2 images of the SW Greenland shelf were obtained from the Copernicus Data Space Ecosystem using the CDSETool package for Python (https://github.com/CDSETool/CDSETool). The Sentinel-2 collection was queried by tile id, month, processing level, and cloud cover. Thumbnails of Level-1C (L1C) images with less than 50% cloud cover that were obtained during the months of March to October were returned. Before downloading L1C images, thumbnails were visually checked for clouds, as the cloud cover algorithm does not appear to function well over water. Inspection of the thumbnails also allowed small partial tiles to be excluded. 1380 L1C Sentinel-2 images from 26 tiles that cover the eastern part of the Davis Strait and the SW Greenland shelf were downloaded. The L1C top-of-atmosphere images were processed using Acolite (version 20231023; Vanhellemont, 2019) to compute Rayleigh-corrected reflectance (Rrc). Land, clouds, fog, ice, and sunglint were masked. Adjacency effects were accounted for by dilating mask edges by eight pixels. The Rayleigh-corrected reflectance was then used to identify mats of floating macroalgae using the floating algae index (FAI; Hu, 2009). The input bands were denoised using a 7x7 median filter. Floating mats of macroalgae were identified by constructing a binary image where FAI pixels values greater than or equal to 0.002 were set to 1 and values below this threshold were set to 0. Connected components in the binary image with areas greater than 10x10 pixels (1000 m2) were identified and the coordinates of their centroids and their areas were recorded. For tiles that contained land, only centroid positions seaward of the 100 m isobath were considered. Centroid positions within fjords and in the nearshore region, therefore, were removed. The coordinates of the remaining centroids of floating mats were then stored in a log file. Overlapping Sentinel-2 tiles produced some duplicate entries, which were identified and removed. NOAA_GDP_select.zip Surface drifter trajectories were obtained from the National Oceanic and Atmospheric Administration (NOAA) Global Drifter Program (GDP; Elipot et al., 2022). Drifter trajectories were obtained from the 1 hour quality controlled database using ERDDAP (https://erddap.aoml.noaa.gov/gdp/erddap/tabledap/drifter_hourly_qc.html). The GDP ERDDAP server was queried to return drifter trajectories between 54°N to 68°N and 68°W to 30°W with drogues at 15 m depth and that recorded their positions using Global Positioning System (GPS) satellites, producing over 300 trajectories between 2016 and 2022. This initial dataset was further processed to remove drifter trajectories that were in the study area for less than 2 weeks, leaving 279 drifter trajectories available for analysis.

提供机构:
Zenodo
创建时间:
2025-07-29
二维码
社区交流群
二维码
科研交流群
商业服务