About

Our faculty are engaged and dedicated educators, advisors, and mentors and have been honored with numerous university-wide and national teaching awards. Our classes emphasize the latest research coupled with cutting-edge technology and practices making our graduates among the most competitive candidates in the country for natural resource professions. Our curricula include everything from protected lands management and urban forestry, to industrial forestry operations and ecology. Small class sizes and faculty dedicated to teaching afford FREC students the chance to get to know their professors personally. Wide varieties of academic and professional opportunities are available through research, student organizations, and public outreach programs organized by the faculty.

Research topics

• Computer Science
• Environmental science
• Geography
• Remote sensing
• Forestry
• Geology
• Machine Learning
• Agronomy
• Biology
• Engineering
• Business
• Botany
• Economics
• Agricultural economics
• Agroforestry
• Statistics
• Cartography
• Mathematics
• Natural resource economics

Selected publications

• Factors Influencing the Consistency in Crowdsourced Interpretations of Aerial Photographs to Measure Tree Canopy Cover

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• Characterizing the influence of varying functional traits from remotely sensed data on forest productivity acquired from selected NEON sites

  Maryland Shared Open Access Repository (USMAI Consortium) · 2025-12-01

  articleOpen access
  PublisherDOI

• FOLU-Net: A novel framework using long short-term memory networks to predict future forestry and other land use

  Ecological Informatics · 2025-07-05

  articleOpen accessSenior author

  The objective of this study is to predict future tropical forest cover presence and types using multitemporal imaging spectroscopy data. Accurately predicting land cover changes with image time series is vital for assessing the effects of climate change and land management on forest resources. Data were obtained from the DLR Earth Sensing Imaging Spectrometer (DESIS) covering a region in the West Godavari district of Andhra Pradesh, India. DESIS, mounted on the International Space Station, records Earth observation data in 235 channels over a 400–1000 nm spectral range. Five overlapping cloud-free images were selected, capturing the seasonal variability among land covers to form the multitemporal image stack. 1070 randomly generated training points spanning the five dates were visually classified into four land cover classes: forest plantation, palm plantation, natural forest, and non-forest. Future land cover was predicted using the following steps: (1) A recurrent neural network with long short-term memory (LSTM) was used to predict future reflectance values of the 235 bands for each point. This model had an R 2 coefficient of 83.0 %. (2) A multi-layer neural network using Keras was trained on the classified points from each image with 5-fold cross-validation, achieving an accuracy of 73.0 %. (3) The classification model from step 2 was then applied to the reflectance data generated from the LSTM (step 1) to predict the future land type at each point. The combined land cover prediction framework, titled Forestry and Other Land Use Neural Network (FOLU-Net), enables predictions of land use change without the need for potentially error-prone land use classifications at each prior time step necessitated by approaches such as Markov chain analysis. Our findings demonstrate a robust framework for characterizing the evolution of land cover using multitemporal imaging spectroscopy. • Multitemporal imaging spectroscopy data from DESIS used to predict land cover change. • Long short-term memory networks successfully predict future hyperspectral data. • Deep learning provides robust framework for characterizing future land use. • FOLU-Net eliminates error-prone classifications per timestep unlike Markov analysis.

  PublisherDOI

• Characterizing the influence of varying functional traits from remotely sensed data on forest productivity acquired from selected NEON sites

  Science of Remote Sensing · 2025-07-24 · 2 citations

  articleOpen access

  Gross primary productivity (GPP) describes total photosynthesis (carbon fixation) in an ecosystem and is key to the global land carbon budget. To reduce uncertainties in carbon accounting for different forest ecosystems, it is crucial to analyze the health and productivity of forested ecosystems. Plant functional traits, which are a combination of morphological, physiological, and environmental characteristics, have been shown to be predictive of forest ecosystem carbon dynamics. This study aimed to assess how well GPP can be predicted by remotely quantified functional traits across varying forested ecosystems. Airborne remote sensing observations and in situ flux tower measurements used in this analysis were acquired from selected forested sites from the National Ecological Observatory Network (NEON) data portal. We investigated hyperspectral indices and lidar derived products as proxies of remotely sensed plant functional traits. Average midday GPP around the date of flight was calculated by developing a relationship between night respiration and temperature and removing that component from the net surface-atmosphere CO 2 exchange (NSAE). We applied multiple linear regression with a best subset approach for three trait classes: morphological and environmental traits from lidar, physiological traits from hyperspectral data, and a combined functional trait model. The best-performing model, using lidar and hyperspectral traits, included CHM mean, DSM standard deviation, PRI standard deviation, and WBI mean producing a R 2 of 0.87, an adjusted R 2 of 0.84, a PRESS R 2 of 0.75 and RMSE of 3.48 μmol CO 2 /m 2 /s. Results show that a combination of plant functional traits are important predictors of forest productivity. • Integrating multiple plant functional traits is important to predict forest productivity accurately. • All three types of remotely sensed plant functional traits were significant predictors of forest productivity — morphological (canopy structure), environmental (topography), and physiological (leaf water content, light use efficiency). • GPP generally increases with canopy height, with the exception of an old-growth site. • Dominant vegetation types describe the range of forest functional traits contributing to productivity.

  PublisherDOI

• Factors influencing the consistency in crowdsourced interpretations of aerial photographs to measure tree canopy cover

  Ecological Informatics · 2025-07-15

  articleOpen access

  Machine learning models are typically data-hungry algorithms that require large data inputs for training. When they produce wall-to-wall remote sensing products, model validation also requires large sets of temporally harmonized field observations. Crowdsourcing may offer a potential solution for the collection of photointerpretations for the training and validation of spatial models of tree canopy cover (TCC), as it harnesses the power of a large anonymous crowd in the completion of repetitive discrete analyses or human intelligence tasks (HITs). This study explores the factors that determine the consistency of TCC interpretations collected by an anonymous crowd to those collected by a control group. The crowd interpretations were obtained through an anonymous platform with a task-reward framework, while those collected by the control group were collected by known interpreters in a more traditional setting. Both groups carried out this task using an interface developed for Amazon’s Mechanical Turk platform. We collected multiple interpretations at sample plot locations from both crowd and control interpreters, and sampled these data in a Monte Carlo framework to estimate a classification model predicting the consistency of each crowd interpretation with control interpretations. Using this model, we identified the most important variables in estimating the relationship between a location’s characteristics and interpretation behaviors which affect consistency in interpretations between crowd workers our control group. Overall, we show low agreement between crowdsourced and control interpretations, as well as interpretations from individual control group members. This warrants caution in considering the crowdsourced photointerpretation of TCC as a data source for model training and validation without adequate interpreter training as well as significant quality control measures and consistency standards. We show that the number of plots interpreted was the strongest indicator of the reliability of an individual’s interpretations, further evidenced by apparent fatigue effects in crowd interpretations. The second most important variable related to the use of the false color display during interpretation followed by a variable related to the use of the natural color display during interpretation, reflecting the differences in interpretation methodologies used by crowd workers and control group interpreters and the impact display has on the interpretation of tree canopy cover. Finally, we discuss recommendations for further study and future implementations of crowdsourced photointerpretation. These include the enhanced use of existing mechanisms within Mechanical Turk such as worker qualifications to identify and reward more attentive workers, as well as enhanced attention to quality control measures throughout the data collection process and measures to increase intrinsic motivation. For our study we also recommend a minimum time on task or other measures to reduce the punishment of access to HITs for workers who took their time providing detailed interpretations. We also recommend using optimized default interface settings instead of providing a variety of options to the interpreter. • We evaluated the photointerpretation of tree canopy cover (a continuous variable) based on application utilization and various worker behaviors. • Photointerpretation of tree canopy cover can result in variability among control interpreters as well as crowd workers, and so should be used with caution. • Statistical tests show that reliability of crowd photointerpretation of tree canopy cover is related most significantly to interpreter fatigue, followed by interface utilization and time-money motivations. • For studies with similar applications and methods, we recommend robust quality control measures, evaluation of fair wages, mechanisms to identify and reward attentive workers, and interface design for optimal interpreter performance.

  PublisherDOI

• Attributing vegetation disturbance change agent from Landsat time series in the arid and semi-arid ecosystem of Qilian Mountains, China

  Remote Sensing Applications Society and Environment · 2025-08-01

  article
  PublisherDOI

• Comparing Canopy Height Models from Regional-Scale Aerial Photogrammetry with Global Spaceborne Lidar-Derived Data for Estimating Forest Volume and Biomass

  Forest Science · 2025-05-27

  article
  PublisherDOI

• Effects of establishment fertilization on Landsat-assessed leaf area development of loblolly pine stands

  Forest Ecology and Management · 2024-02-01 · 1 citations

  articleOpen access
  PublisherOA PDFDOI

• Robust Identification of Vegetation Change Using Shapelet-Based Temporal Segmentation of Landsat Time-Series Stacks: A Case Study in the Qilian Mountains

  IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · 2024-12-23 · 2 citations

  articleOpen accessSenior author

  Although many algorithms have been developed for monitoring annual vegetation change, most require complex controlling parameters or can only detect abrupt forest disturbances. We developed a new shapelet-based temporal segmentation vegetation change detection algorithm (SVCD) to identify the pre-change baseline and directly measure changes using data-driven change detection rules. A temporal sliding window-based anomaly checking procedure is applied to the annual maximum composite of spectral vegetation indices in the growing season to remove the unmasked clouds, cloud shadows, and other temporary changes. Then, an iterative shapelet searching algorithm is performed on each given time series trajectory to filter out all atypical subseries. Finally, a multiple-spectral-index-based thresholding method is developed to measure the difference in spectral indices' statistical values between atypical and typical (stable) subseries. When those values exceed the predefined threshold values, the shapelet windows are flagged as having changed. SVCD was tested on a Landsat scene in the Qilian Mountains (WRS-2 Path 133 Row 34), a region with extensive natural and human-driven land cover changes over three decades. A stratified random sampling of 1012 pixels in persistent vegetation areas and 80 in changed areas showed SVCD achieved 99.0%<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>0.6% accuracy (95% confidence intervals). Applied across the entire Qilian Mountains, it outperformed LandTrendr with 99.5%<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>0.4% accuracy. These results highlight SVCD's strong potential for precise, noise-resistant vegetation change detection.

  PublisherDOI

• Estimation of individual stem volume and diameter from segmented UAV laser scanning datasets in<i>Pinus taeda</i>L. plantations

  International Journal of Remote Sensing · 2023-01-02 · 14 citations

  article

  The competitive neighbourhood surrounding an individual tree can have a significant influence on its diameter at breast height (DBH) and individual tree stem volume (SV). Distance dependent competition index metrics are rarely recorded in traditional field campaigns because they are laborious to collect and are spatially limited. Remote sensing data could overcome these limitations while providing estimation of forest attributes over a large area. We used unoccupied aerial vehicle laser scanning data to delineate individual tree crowns (ITCs) and calculated crown size and distance-dependent competition indices to estimate DBH and SV. We contrasted two methods: (i) Random Forest (RF) and (ii) backwards-stepwise, linear multiple regression (LMR). We utilized an existing experiment in Pinus taeda L. plantations including multiple planting densities, genotypes and silvicultural levels. While the tree planting density did affect the correct delineation of ITCs, between 61% and 99% (mean 86%) were correctly linked to the planting location. The most accurate RF and LMR models all included metrics related to ITC size and competitive neighbourhood. The DBH estimates from RF and LMR were similar: RMSE 3.05 and 3.13 cm (R2 0.64 and 0.62), respectively. Estimates of SV from RF were slightly better than for LMR: RMSE 0.06 and 0.07 m3 (R2 0.77 and 0.70), respectively. Our results provide evidence that ITC size and competition index metrics may improve DBH and SV estimation accuracy when analysing laser-scanning data. The ability to provide accurate, and near-complete, forest inventories holds a great deal of potential for forest management planning.

  PublisherDOI

Frequent coauthors

• Valerie A. Thomas

  Virginia Tech

  74 shared

• Christine E. Blinn

  Virginia Tech

  34 shared

• Layne T. Watson

  23 shared

• Rhonda D. Phillips

  Massachusetts Institute of Technology

  21 shared

• Thomas R. Fox

  Rayonix (United States)

  20 shared

• John W. Coulston

  20 shared

• Evan B. Brooks

  18 shared

• Timothy J. Albaugh

  Virginia Tech

  15 shared

Education

• Postdoc, NTL-LTER

  University of Wisconsin Madison

  1996

• Ph.D., Environmental Monitoring

  University of Wisconsin Madison

  1995

• M.S., Environmental Monitoring

  University of Wisconsin Madison

  1993

• B.S., Env. Science & Engineering

  University of North Carolina at Chapel Hill

  1986