Understory species map in Connecticut US

Published: 27 April 2023| Version 3 | DOI: 10.17632/rschxhwgvw.3
Xiucheng Yang,


We created maps of four understory classes (i.e., native shrubs of greenbrier and mountain laurel, and invasive shrubs of barberry and the assemblage of mixed invasive) at 10 m resolution in Connecticut’s deciduous forest in 2020. A harmonic time series model and three years of Sentinel-2 time series from 2019 to 2021 were used to classify understory species based on their unique, intra-annual phenology characteristics. The time series model coefficients captured the subtle phenology differences and created synthetic cloud-free images within a short temporal window (e.g., on the 100th and 120th day of year) in the spring prior to canopy leaf-on (hereafter called ‘observe window’), in which Sentinel-2 data penetrated the deciduous overstory canopy and observed the unique trajectories of different understory species due to their phenology differences. As the different conditions of leaf growth in the observation window presented distinct spatial patterns within deciduous forests, we also calculated multiple texture features (i.e., mean, second moment, and contrast from gray level co-occurrence matrix) based on the synthetic images created within the observation window. By using the spectral, temporal, and spatial features as input variables from dense Sentinel-2 data, auxiliary data (i.e., LiDAR and soil drainage layer), a random forest classifier, and a new strategy to iteratively select representative samples (namely ISRS), understory species maps were created with an overall accuracy of approximately 93%, and the user’s and producer’s accuracies varied from 39% to 99% for the three mapped understory species and one assemblage of species. The proposed method created an accurate binary map of understory presence with an overall accuracy of 95%, a producer’s accuracy of 84%, and user’s accuracy of 68%. Additionally, we separated the invasive (i.e. barberry and mixed invasive of multi-flora rose, oriental bittersweet, honeysuckle, winged euonymus, and autumn olive) and native (greenbrier and mountain laurel) species with an overall accuracy of 94%. We estimated that the invasive species cover an area of 649.33±140.59 km2, which occupies a large proportion (~53%) of the shrub understory in Connecticut’s deciduous forests. With enough accurate training data collected, this classification strategy has the potential to be applied at a much larger spatial extent than Connecticut within the Sentinel-2 era.



University of Connecticut


Remote Sensing, Plant Phenology, Time Series, Vegetation Mapping, Shrub, Forest