More Like this

1. Tree++: Truncated Tree Based Graph Kernels

   https://doi.org/10.1109/TKDE.2019.2946149  

   Ye, Wei; Wang, Zhen; Redberg, Rachel; Singh, Ambuj (April 2021, IEEE Transactions on Knowledge and Data Engineering)
2. A Neuro-Symbolic Framework for Tree Crown Delineation and Tree Species Classification

   https://doi.org/10.3390/rs16234365  

   Harmon, Ira; Weinstein, Ben; Bohlman, Stephanie; White, Ethan; Wang, Daisy Zhe (December 2024, Remote Sensing)

   Neuro-symbolic models combine deep learning and symbolic reasoning to produce better-performing hybrids. Not only do neuro-symbolic models perform better, but they also deal better with data scarcity, enable the direct incorporation of high-level domain knowledge, and are more explainable. However, these benefits come at the cost of increased complexity, which may deter the uninitiated from using these models. In this work, we present a framework to simplify the creation of neuro-symbolic models for tree crown delineation and tree species classification via the use of object-oriented programming and hyperparameter tuning algorithms. We show that models created using our framework outperform their non-neuro-symbolic counterparts by as much as two F1 points for crown delineation and three F1 points for species classification. Furthermore, our use of hyperparameter tuning algorithms allows users to experiment with multiple formulations of domain knowledge without the burden of manual tuning. 

   more » « less
3. High-light demanding and rare tree species define tropical riparian tree communities

   Hudspeth, Virginia; Pastor, Alberto; Heartsill-Scalley, T. (August 2023, Acta Cientifica)

   Riparian zones are areas of transition between aquatic and terrestrial ecosystems, and their vegetation provides many important ecosystem services that are becoming of greater importance due to the changing climate. To understand riparian conditions along headwater tropical streams in terms of species composition and structure, we surveyed riparian tree communities and physical conditions in adjacent streams in the Luquillo Experimental Forest. We determined riparian substrate condition visually by estimating ground and vegetation attributes and measuring slope and elevation in a series of riparian plots at three stream sites. We determined riparian forest composition by identifying stem species and recording their height and diameter in plots. Our results demonstrated that elevation and slope were the main differences among sites, while all sites had similar ground and vegetation conditions. Species composition in all sites was characterized by hydrophytic species. However, although the most abundant species were generalists, we also found many species that were only observed once or twice per site. Overall, our results indicate that although recent and major hurricanes influence structure and composition at this site, riparian areas still maintain a large proportion of hydrophytic and rare tree species. 

   more » « less
4. Half-Tree: Halving the Cost of Tree Expansion in COT and DPF

   Guo, Xiaojie; Yang, Kang; Wang, Xiao; Zhang, Wenhao; Xie, Xiang; Zhang, Jiang; Liu, Zheli (April 2023, Springer)
5. Spider and caterpillar abundance, tree size, and canopy closure metrics in a tree diversity experiment

   https://doi.org/10.6084/m9.figshare.20997910.v1  

   Burghardt, Karin; Butz, Elizabeth; Schmitt, Lauren; Parker, John (January 2023, figshare)

   This data set includes spider abundances recorded on focal trees in a large-scale forest diversity manipulation at the Smithsonian Environmental Research Center in Edgewater, MD, USA. We repeatedly sampled spiders on 540 trees of 15 species planted in single or mixed species combinations (4 or 12) in June and August of 2019 and 2021. We took caterpillar abundance data, measured tree height, and took canopy closure measurements on each tree in 2021. Data associated with the paper: Positive tree diversity effects on arboreal spider abundance are tied to canopy cover in a forest experiment published in Ecology 

   more » « less