Overlooked and extensive ghost forest formation across the US Atlantic coast - data
收藏资源简介：
This repository contains the derived data products for the paper 'Overlooked and extensive ghost forest formation across the US Atlantic coast'. All data products were derived from National Agriculture Imagery Program (NAIP) images in 2020 (+/- 2 years), spanning 11 states along the U.S. Atlantic coast. The study extent is limited to coastal forests located below 50 m above sea-level and within 10 km from the coast. The image acquisition year is state-dependent: CT (2018), DE (2018), MA (2018), MD (2021), NC (2022), NH (2018), NJ (2019), NY (2019), RI (2018), SC (2019), and VA (2021). All usage of the data must be attributed and should be cited with the paper citation. For methodological details, please refer to the paper (Link: https://lnkd.in/efWCXaAx; Online version: https://rdcu.be/eSWxu). Further information on each data layer: Main products: Dead tree density (100m_deadTreeDenisity_50m10km_forest.tif): dead tree density (per hectare), aggregated from individual dead trees detected with a supervised deep learning model. Hotspot (100m_hotspot_50m10km_forest.tif): Hotspots of clustered dead trees (1 = mortality hotspot, 0 = non-hotspot), produced with local hotspot analysis. Note that in our study, only hotspots located below 5 m elevation were considered as ghost forests QA products: Acquisition Timing (align_100m_acquisitionTiming.tif): Phenological timing of NAIP imagery used for analysis, with QA labels ranging from 1 to 5 (1: Peak growing, 2: Senescence, 3: Maturity, 4: Greenup/down, 5: Out-of-season). We recommend careful use of data in category 4 and 5. Eccentricity (align_100m_eccentricity500m.tif): Eccentricity (i.e., elongation) of all dead trees calculated over a 500 m grid cells. We recommend filtering highly pixels with high eccentricity (e.g. state mean + 1 s.d.). Note: This is Version 1 of the dataset. We hope to make improvements to the QA layers (e.g., acquisition timing classification). For updates, you may contact hchyeung@virginia.edu.
