The illegal timber trade has significant impact on the survival of endangered tropical hardwood species like
- Award ID(s):
- 1920908
- NSF-PAR ID:
- 10211603
- Date Published:
- Journal Name:
- Remote Sensing
- Volume:
- 12
- Issue:
- 11
- ISSN:
- 2072-4292
- Page Range / eLocation ID:
- 1739
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract Dalbergia spp. (rosewood), a world‐wide protected genus from the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Due to increased threat toDalbergia spp., and lack of action to reduce threats, port of entry analysis methods are required to identifyDalbergia spp. Handheld laser‐induced breakdown spectroscopy (LIBS) has been shown to be capable of identifying species and establishing provenance ofDalbergia spp. and other tropical hardwoods, but analysis methods for this work have yet to be investigated in detail. The present work investigates five well‐known algorithms—partial least squares discriminant analysis (PLS‐DA), classification and regression trees (CART),k ‐nearest neighbor (k ‐NN), random forest (RF), and support vector machine (SVM)—two training/test set sampling regimes, and data collection at two signal‐to‐noise (S/N) ratios to assess the potential for handheld LIBS analyses. Additionally, imbalanced classes are addressed. For this application, SVM and RF yield near identical results (though RF takes nearly 100 longer to compute), while the S/N ratio has a significant effect on model success assuming all else is equal. It was found that forming a training set with replicate low S/N analyses can perform as well as higher precision training sets for true prediction, even if the predicted samples have low signal to noise! This work confirms handheld LIBS analyzers can provide a viable method for classification of hardwood species, even within the same genus. -
null (Ed.)Identifying dust aerosols from passive satellite images is of great interest for many applications. In this study, we developed five different machine-learning (ML) based algorithms, including Logistic Regression, K Nearest Neighbor, Random Forest (RF), Feed Forward Neural Network (FFNN), and Convolutional Neural Network (CNN), to identify dust aerosols in the daytime satellite images from the Visible Infrared Imaging Radiometer Suite (VIIRS) under cloud-free conditions on a global scale. In order to train the ML algorithms, we collocated the state-of-the-art dust detection product from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) with the VIIRS observations along the CALIOP track. The 16 VIIRS M-band observations with the center wavelength ranging from deep blue to thermal infrared, together with solar-viewing geometries and pixel time and locations, are used as the predictor variables. Four different sets of training input data are constructed based on different combinations of VIIRS pixel and predictor variables. The validation and comparison results based on the collocated CALIOP data indicate that the FFNN method based on all available predictor variables is the best performing one among all methods. It has an averaged dust detection accuracy of about 81%, 89%, and 85% over land, ocean and whole globe, respectively, compared with collocated CALIOP. When applied to off-track VIIRS pixels, the FFNN method retrieves geographical distributions of dust that are in good agreement with on-track results as well as CALIOP statistics. For further evaluation, we compared our results based on the ML algorithms to NOAA’s Aerosol Detection Product (ADP), which is a product that classifies dust, smoke, and ash using physical-based methods. The comparison reveals both similarity and differences. Overall, this study demonstrates the great potential of ML methods for dust detection and proves that these methods can be trained on the CALIOP track and then applied to the whole granule of VIIRS granule.more » « less
-
more » « less
Abstract Questions What are the primary biotic and abiotic factors driving composition and abundance of naturally regenerated tree seedlings across forest landscapes of Maine? Do seedling species richness (
SR ) and density (SD ) decrease with improved growing conditions (climate and soil), but increase with increased diversity of overstorey composition and structure? Does partial harvesting disproportionately favour relative dominance of shade‐intolerant hardwoods (PIHD ) over shade‐tolerant softwoods (PTSD )?Location Forest landscapes across the diverse eco‐regions and forest types of Maine,
USA .Methods This study used
USDA Forest Service Forest Inventory Analysis permanent plots (n = 10 842), measured every 5 yr since 1999. The best models for each response variable (SR ,SD ,PIHD andPTSD ) were developed based onAIC and biological interpretability, while considering 35 potential explanatory variables incorporating climate, soil, site productivity, overstorey structure and composition, and past harvesting.Results Mean annual temperature was the most important abiotic factor, whereas overstorey tree size diversity was the most important biotic factor for
SR andSD . Both mean annual temperature and overstorey tree size diversity had a curvilinear relationship withSR andSD . Average overstorey shade tolerance and percentage tolerant softwood basal area in the overstorey were the top predictor variables ofPIHD andPTSD , respectively. Partial harvesting favouredPIHD but notPTSD .Conclusions This is one of the first studies to comprehensively evaluate a number of factors influencing naturally established tree seedlings at a broad landscape scale in the Northern Forest region of the eastern
USA and Canada. Despite limitations associated with relatively small plot size, large seedling size class and lack of direct measurements of light, water and nutrients, this study documents the influence of these factors amid high variability associated with patterns of natural regeneration. The curvilinear relationship between mean annual temperature withSR andSD supports the argument that species richness and abundance usually have unimodal relationships with productivity indicators, whereas the curvilinear relationship between overstorey tree size diversity andSR andSD suggest that moderate overstorey diversity incorporates multiple species as well as higher seedling individuals. -
more » « less
Abstract Interspecific interactions can provoke temporal and spatial avoidance, ultimately affecting population densities and spatial distribution patterns. The ability (or inability) of species to coexist has consequences for diversity and ultimately ecosystem stability. Urbanization is predicted to change species interactions but its relative impact is not well known. Urbanization gradients offer the opportunity to evaluate the effect of humans on species interactions by comparing community dynamics across levels of disturbance.
We used camera traps deployed by citizen scientists to survey mammals along urbanization gradients of two cities (Washington, DC and Raleigh, NC, USA). We used a multispecies occupancy model with four competing predator species to test whether forest fragmentation, interspecific interactions, humans or prey had the greatest influence on carnivore distribution.
Our study produced 6,413 carnivore detections from 1,260 sites in two cities, sampling both private and public lands. All species used all levels of the urbanization gradient to a similar extent, but co‐occurrence of urban‐adapted foxes with less urban‐adapted bobcats and coyotes was dependent on the availability of green space, especially as urbanization increased. This suggests green space allows less urban‐adapted species to occupy suburban areas, but focuses their movements through remaining forest patches, leading to more species interactions.
Synthesis and applications . Species interactions, forest fragmentation and human‐related covariates were important determinants of carnivore occupancy across a gradient of urbanization with the relative importance of forest fragmentation being highest. We found evidence of both positive and negative interactions across the gradient with some dependent on available green space, suggesting that fragmentation leads to higher levels of spatial interaction. Where green space is adequate, there appears to be sufficient opportunity for coexistence between carnivore species in an urban landscape. -
more » « less
Summary Species dominance and biodiversity in plant communities have received considerable attention and characterisation. However, species codominance, while often alleged, is seldom defined or quantified. Codominance is a common phenomenon and is likely to be an important driver of community structure, ecosystem function and the stability of both. Here we review the use of the term ‘codominance’ and find inconsistencies in its use, suggesting that the scientific community currently lacks a universal understanding of codominance. We address this issue by: (1) qualitatively defining codominance as mostly shared abundance that is distinctively isolated within a subset of a community, and (2) presenting a novel metric for quantifying the degree to which relative abundances are shared among a codominant subset of plant species, while also accounting for the remaining species within a plant community. Using both simulated and real‐world data, we then demonstrate the process of applying the codominance metric to compare communities and to generate a quantitatively defensible subset of species to consider codominant within a community. We show that our metric effectively distinguishes the degree of codominance between four types of grassland ecosystems as well as simulated ecosystems with varying degrees of abundance sharing among community members. Overall, we make the case that increased research focusses on the conditions under which codominance occurs and the consequences for species coexistence, community structure and ecosystem function that would considerably advance the fields of community and ecosystem ecology.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/rs12111739
