RRRachel Riemann

Rachel Riemann

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Rachel Riemann has a total of 124 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Maryland Forests 2018: Summary Report

    by Tonya Lister, Thomas A. Albright, Brett J. Butler, Susan Crocker, Jesse Caputo, Thomas Goff, Cassandra Kurtz, Andrew J. Lister, Randall Morin, Mark D. Nelson, et al.

    This summary report provides an overview of the current conditions, changing trends, issues, and threats to Statewide forest resources based on findings from the 2018 inventory of Maryland’s forests based on data collected from 2012 through 2018. There are an estimated 2.4 million acres of forest land in Maryland primarily in private ownership (72 percent). Forest land is dominated by the oak/hickory forest-type group which occupies 59 percent of the total forest land area. Since 2013, growing-stock volume increased by 3 percent, totaling 5.7 billion cubic feet in 2018. Annual net growth of growing-stock trees on timberland averages 110 million cubic feet, and the ratio of net growth-to-removals (2.9:1.0) increased slightly from what was reported in 2008 (2.7:1.0). This report includes additional information on forest attributes, land-use change, carbon, timber products, and forest health. A detailed interactive report and supplemental information, including (1) tables summarizing quality assurance, (2) a core set of tabular estimates for a variety of forest resources, and (3) user and database guides can be accessed at https://doi.org/10.2737/NRS-RB-148 . Access interactive report: Maryland Forests 2018: Interactive Report Supplemental materials: Download Maryland Forests 2018: Quality Assurance Tables (Excel file, 86.6 KB) Download Maryland Forests 2018: Summary Tables (Excel File, 579.6 KB) Download Maryland Forests 2018: Database CSV Files (Zip File 19.4 MB) Download Maryland Forests 2018: User and Database Guides (Zip File, 12 MB)

  • New Jersey forests 2018: summary report

    by Susan J. Crocker, Sjana L. Schanning, Christopher W. Woodall, Brett J. Butler, Jesse Caputo, Thomas C. Goff, Cassandra M. Kurtz, Tonya W. Lister, Randall S. Morin, Mark D. Nelson, et al.

  • Open Access

    Comparing Machine Learning and Statistical Models for Remote Sensing-Based Forest Aboveground Biomass Estimations

    by Shashika Himandi Gardeye Lamahewage, Chandi Witharana, Rachel Riemann, Robert Fahey, Thomas Worthley

    Understanding the distribution of forest aboveground biomass (AGB) is pivotal for carbon monitoring. Field-based inventorying is time-consuming and costly for large-area AGB estimations. The integration of multimodal remote sensing (RS) observations with single-year, field-based forest inventory analysis (FIA) data has the potential to improve the efficiency of large-scale AGB modeling and carbon monitoring initiatives. Our main objective was to systematically compare the AGB prediction accuracies of machine learning algorithms (e.g., random forest (RF) and support vector machine (SVM)) with those of conventional statistical methods (e.g., multiple linear regression (MLR)) using multimodal RS variables as predictors. We implemented a method combining AGB estimates of actual FIA subplot locations with airborne LiDAR, National Agriculture Imagery Program (NAIP) aerial imagery, and Sentinel-2 satellite images for model training, validation, and testing. The hyperparameter-tuned RF model produced a root mean square error (RMSE) of 27.19 Mgha−1 and an R2 of 0.41, which outperformed the evaluation metrics of SVM and MLR models. Among the 28 most important explanatory variables used to build the best RF model, 68% of variables were derived from the LiDAR height data. The hyperparameter-tuned linear SVM model exhibited an R2 of 0.10 and an RMSE of 32.17 Mgha−1. Additionally, we developed an MLR using eight explanatory variables, which yielded an RMSE of 22.59 Mgha−1 and an R2 of 0.22. The linear ensemble model, which was developed using the predictions of all three models, yielded an R2 of 0.79. Our results suggested that more field data are required to better generalize the ensemble model. Overall, our findings highlight the importance of variable selection methods, the hyperparameter tuning of ML algorithms, and the integration of multimodal RS data in improving large-area AGB prediction models.


  • Open Access

    Aboveground biomass estimation using multimodal remote sensing observations and machine learning in mixed temperate forest

    by Shashika Himandi Gardeye Lamahewage, Chandi Witharana, Rachel Riemann, Robert Fahey, Thomas Worthley

    Plants sequester carbon in their aboveground components, making aboveground tree biomass a key metric for assessing forest carbon storage. Traditional methods of aboveground biomass (AGB) estimation via Forest Inventory and Analysis (FIA) plots lack sufficient sampling intensity to directly produce accurate estimates at fine granularities. Increasing the sampling intensity with additional FIA plots would be labor and time intensive, particularly for large-scale carbon studies. Utilizing remote sensing (RS) data, such as Airborne Light Detection and Ranging (LiDAR), aerial imagery, and satellite images can significantly enhance the efficiency of forest carbon monitoring efforts. The principal objective of this study is to utilize the random forest (RF) algorithm to build predictive AGB models. We utilized 67 explanatory variables, which were extracted from three RS sources resulting in nine RF models. Each RF model was subjected to variable selection, hyperparameter tuning, and model evaluation. The optimum model considered 28 explanatory variables, with root mean square error (RMSE) of 27.19 Mgha− 1 and R2 of 0.41. Combining LiDAR with image metrics increased the accuracy of prediction models, serving as a pivotal tool for large area biomass mapping and carbon related decision making.

  • Indiana forests 2018: summary report

    by Dale D. Gormanson, Bryan J. Hemmer, Daniel J. Gallion, Thomas A. Albright, Brett J. Butler, Susan J. Crocker, James D. Garner, Thomas C. Goff, Cassandra M. Kurtz, Tonya W. Lister, et al.


  • West Virginia forests 2018: summary report

    by Randall S. Morin, Thomas A. Albright, Brett J. Butler, Jesse Caputo, Susan J. Crocker, Thomas C. Goff, Cassandra M. Kurtz, Tonya W. Lister, Mark D. Nelson, Rachel Riemann, et al.


  • Pennsylvania forests 2019: summary report

    by Thomas A. Albright, Brett J. Butler, Jesse Caputo, Susan J. Crocker, Thomas C. Goff, Cassandra M. Kurtz, Shawn Lehman, Tonya W. Lister, William G. Luppold, Randall S. Morin, et al.


  • Open Access

    Wisconsin Forests 2019

    by Cassandra M. Kurtz, Thomas A. Albright, Brian D. Anderson, Brett J. Butler, Collin Buntrock, Daniel Buckler, Susan J. Crocker, Andrea L. Diss-Torrance, Thomas C. Goff, Bryan Hemmer, et al.


  • Michigan forests 2019: summary report

    by Scott A. Pugh, Jagdish Poudel, Thomas A. Albright, Brett J. Butler, Jesse Caputo, Susan J. Crocker, Cassandra M. Kurtz, Tonya W. Lister, Randall S. Morin, Mark D. Nelson, et al.

  • Open Access

    Expanding wildland-urban interface alters forest structure and landscape context in the northern United States

    by Nancy Falxa Sonti, Rachel Riemann, Miranda H Mockrin, Grant Michael Domke

    The wildland-urban interface (WUI), where housing intermingles with wildland vegetation, is the fastest-growing land use type in the United States. Given the ecological and social benefits of forest ecosystems, there is a growing need to more fully understand how such development alters the landscape context and structure of these WUI forests. In a space-for-time analysis we utilized land cover data, forest inventory plots, and housing density data over time to examine differences in forest characteristics of the northern US across three WUI change classes: (a) forest that has been in WUI housing density levels since at least 1990 (old-WUI), (b) forest where development crossed the WUI housing density threshold after 1990 (new-WUI), and (c) forest with little to no housing development (non-WUI). Of the 184 million acres of forest in the study area, 34 million acres (19%) were in old-WUI, 12 million acres (7%) were new-WUI, and 136 million acres (74%) were non-WUI. In general, as areas transitioned from non-WUI to newer WUI to older more established WUI, the forest was associated with decreased spatial integrity, increased forest-developed edges, and lower proportions of forest in the surrounding landscape. Forest in the WUI had greater carbon storage, with greater aboveground biomass, relative stand density, and more live trees per hectare than non-WUI forest, suggesting greater capacity to sequester carbon compared to non-WUI forest. At the same time, WUI forest also had significantly reduced structural diversity compared to non-WUI forest, with fewer saplings, seedlings, and dead trees per hectare. Forest that more recently crossed the WUI housing density threshold appeared to be on a trajectory towards that of old-WUI forest. These differences in forest structure across the northern US suggest reduced capacity for forest regeneration in the WUI and the potential for changes in other ecological functions.