%0 Dataset
%T 30 meter resolution forest age map of China (2020)
%J National Cryosphere Desert Data Center
%I National Cryosphere Desert Data Center(www.ncdc.ac.cn)
%U http://www.ncdc.ac.cn/portal/metadata/793380a4-7a1b-4c98-aa7e-6a526ed27969
%W NCDC
%R 10.5281/zenodo.8354262
%A Guo Qinghua
%K Forest age;high-resolution;machine learning algorithms
%X High resolution and spatially clear forest age maps are crucial for quantifying forest carbon storage and carbon sequestration potential.. The nationwide forest age estimation work previously conducted in China has been limited by sparse resolution and incomplete forest ecosystem coverage. Due to the complex species composition, vast forest area, insufficient field measurements, and incomplete methods, the resolution is sparse and the forest ecosystem coverage is incomplete, which limits the estimation of forest age nationwide in China. To address these challenges, we have developed a framework that combines machine learning algorithms (MLAs) with remote sensing time series analysis to estimate the age of forests in China. Initially, we determined and developed the optimal MLA for estimating forest age for various vegetation classifications based on field measurements of forest height, climate, terrain, soil, and forest age, and used these MLA to determine forest age information. Subsequently, we applied LandTrendr time series analysis to detect forest disturbances from 1985 to 2020, and used the time since the last disturbance as a proxy for forest age. Finally, the LandTrendr forest age data was combined with the results of MLA to generate the 2020 Chinese forest age map.