http://mirror1.mcfns.com/index.php/Journal/issue/feedMathematical and Computational Forestry & Natural-Resource Sciences (MCFNS)2021-04-19T16:33:04+00:00MCFNS System Manageradmin@j.mcfns.comOpen Journal Systems<p>The mission of MCFNS is to publish peer-reviewed basic and applied research in Mathematical and Computational Forestry and Natural-Resource Sciences. This research can include analytical solutions, proofs, derivations, software developments, and simulations, in forest management, growth and yield modeling, and other natural resource related studies. Journal items are published collectively as part of an issue with its Table of Contents biannually, currently in <strong>March</strong> and <strong>October</strong>. </p>http://mirror1.mcfns.com/index.php/Journal/article/view/13.1ESTIMATING INDIVIDUAL TREE HEIGHTS AND DBHS FROM VERTICALLY DISPLACED SPHERICAL IMAGE PAIRS2021-04-14T19:03:43+00:00Haozhou Wanghaozhou-wang@outlook.comTing-Ru Yangb00605080@gmail.comJoni Waldyjoni.waldy@unb.caJohn A Kershaw Jr.kershaw@unb.caIndividual tree parameters, such as diameter at breast height (DBH) and tree height, are fundamental measurements in forest inventory, and often labour intensive and require significant financial expenditures. Applying digital imaging in forest inventory is an efficient way to decrease the workload. In this study, spherical images taken using a novel commercial 360° camera (Ricoh Theta S) and stereographic geometry were applied to obtain these parameters directly without stitching multiple images from common cameras. This technology was validated in both a sparse urban forest (pairwise comparison) and denser real forest (distributional comparison) in Atlantic Canada. The DBH (r2 > 0.76) and height (Dmax < 0.25, K-S test) showed high correspondence with field measures. The spherical camera represents a low-cost option to terrestrial laser scanning and has potential to produce more accurate forest-level estimates with quicker field and office processing time than current under-canopy remote sensing technologies.<div id="gtx-trans" style="position: absolute; left: 263px; top: 69.5985px;"> </div>2021-03-30T00:00:00+00:00Copyright (c) 2021 Mathematical and Computational Forestry & Natural-Resource Sciences (MCFNS)http://mirror1.mcfns.com/index.php/Journal/article/view/13.2Spatial analysis of airborne laser scanning point clouds for predicting forest structure2021-04-19T16:33:04+00:00Henrike Häbelhenrike.habel@ki.fiAndras Balazsandras.balazs@luke.fiMari Myllymäkimari.myllymaki@luke.fi<p>The spatial structure of forest, which can be understood as the arrangement of trees with respect to each other, plays a role in various forestry decisions. In this study the spatial structure is summarized by three different indices which were compared on the example of a study site with circular field plots with 9 m radius in Central Finland. The aim was to predict the indices by airborne laser scanning (ALS) and study usefulness of spatial or horizontal summaries of the ALS point cloud. Thus, in addition to commonly used vertical summaries of the point clouds, we explored summaries of the horizontal distribution of the pulse returns through canopy height models thresholded at different height levels. We used these summaries the well-known K-nn estimation method to predict the indices. In this study, we show that quantifying the spatial structure from small sample plots is challenging. Still, we present evidence that the use of spatial metrics improved the prediction of spatial structure of forests, and has potential for improvements possibly for also other variables related to gap structures.</p>2021-03-30T00:00:00+00:00Copyright (c) 2021 Mathematical and Computational Forestry & Natural-Resource Sciences (MCFNS)http://mirror1.mcfns.com/index.php/Journal/article/view/13.3CORRECTING TREE COUNT BIAS FOR OBJECTS SEGMENTED FROM LIDAR POINT CLOUDS2021-04-15T20:02:10+00:00Mike Robert Strubmike.r.strub@gmail.comNathaniel Osbornenathaniel.osborne@rayonier.com<p>We introduce a new statistical distribution for modeling tree count in segmented LiDAR point clouds. The new distribution is based on the Poisson distribution as a logical basis since the Poisson is based on the premise of rare events from a large population. The probability a particular tree falls in a given point cloud segment is small and the number of trees is large. The purpose of segmentation is to provide segments that contain a single tree. This implies that a Poisson with deflated probability of zero occurrences and inflated probability of one occurrences is appropriate. LiDAR point cloud data on twenty ground truth plots are used to show the utility of this approach.</p>2021-03-30T00:00:00+00:00Copyright (c) 2021 Mathematical and Computational Forestry & Natural-Resource Sciences (MCFNS)http://mirror1.mcfns.com/index.php/Journal/article/view/13.4UTADA: Unified Theory of the Algebraic Differences Approaches---Derivation of Dynamic Site Equations from Yield-Site Relationships2021-04-14T23:00:02+00:00Chris J. Cieszewskibiomat@uga.edu<p>Dynamic-equation-based self-referencing models of the form: <em>Y=f(y<sub>0</sub>,t<sub>0</sub>,t)</em> describe changes in <em>Y</em> as a function of two variables: one longitudinal variable <em>t</em>, and one unobservable cross-sectional variable<em> X</em>. Traditionally, <em>X</em> is represented implicitly by its substitution of a snapshot value of <em>Y, (y<sub>0</sub>),</em> at an arbitrary value of <em>t, (t<sub>0</sub>)</em>. The unobservable variable <em>X</em> represents the environment potential, which cannot be directly measured or precisely defined due to its extreme complexity and variability. While the most elusive and difficult in handling, <em>X</em> is the most critical variable of the dynamic site equations due to its disproportionate impact on the modeled dynamics, yet, all traditional approaches to such modeling are predominantly based on a detailed analysis of primarily longitudinal relationships <em>Y=u(t)</em>, which subsequently, to be helpful in practice, are modified into the self-referencing forms. All the former approaches devote little to no effort to explicitly model the cross-sectional relationships governed by the unobservable variable <em>X. </em></p><p>The presented approach unifies the modeling efforts of defining yield and site relationships equally by focusing primarily on direct mathematical formulations describing the theory of their interaction. <em>This approach</em> considers the variable <em>t</em> only in the secondary analysis, adding it to the model through modifying the final model parameters. Despite the somewhat elusive nature of exploring the unobservable properties of <em>the site,</em> the new approach appears to be highly empowering by analyzing more direct yet more robust relationships between <em>Y</em> and <em>X</em> as opposed to those between <em>Y</em> and <em>t</em>.</p>2021-03-30T00:00:00+00:00Copyright (c) 2021 Mathematical and Computational Forestry & Natural-Resource Sciences (MCFNS)