Various forms of staking, guying, and root ball anchoring are used to prevent post-transplant tree destabilization in the landscape, but little scientific evidence exists to support this practice. This experiment tested the efficacy of three generic tree stabilization systems (TSS) and their effect on tree growth and establishment. In spring 2006, 48 balled and burlapped, 6.4 cm (2.5 inch) diameter, white ash (Fraxinus americana L. Autumn Purpleâ ) were transplanted to a field site in Blacksburg, VA. At planting, one of four TSS treatments (staking, guying, root ball anchoring, or non-stabilized) was installed on each tree. After five weeks, tree pulling tests were conducted on 24 trees to simulate a strong wind load using a cable winch mounted to a skid-steer loader. After one growing season, change in tree height, trunk diameter, and trunk taper were compared among the 24 remaining trees. Soil cores were taken and the length, diameter, and dry weight of roots within the cores were analyzed. TSS were then removed and tree pulling tests were conducted using the same method. The five week tests showed that destabilization was significantly greater for non-stabilized trees (mean of 16 degrees from vertical) than for trees with TSS (all means less than 3 degrees from vertical). Yet after one growing season, there were no significant differences among any treatments in tree stability. We conclude that in locations with high wind speeds, TSS may be necessary for trees similar to those in our study, but only for a very short period of time. Results also indicated that staking, guying, and root ball anchoring were equally effective, very robust, very durable, caused no tree injuries, and did not impact tree growth or establishment after one growing season. Practical considerations may therefore play a more important role when choosing which TSS to use. Although the time required for TSS installation was similar for each system, staking was more than twice as expensive as guying or root ball anchoring. / Master of Science

The WikiLeaks Afghanistan war logs contain nearly 77,000 reports of incidents in the US-led Afghanistan war, covering the period from January 2004 to December 2009. The recent growth of data on complex social systems and the potential to derive stories from them has shifted the focus of journalistic and scientific attention increasingly toward data-driven journalism and computational social science. In this paper we advocate the usage of modern statistical methods for problems of data journalism and beyond, which may help journalistic and scientific work and lead to additional insight. Using the WikiLeaks Afghanistan war logs for illustration, we present an approach that builds intelligible statistical models for interpretable segments in the data, in this case to explore the fatality rates associated with different circumstances in the Afghanistan war. Our approach combines preprocessing by Latent Dirichlet Allocation (LDA) with model trees. LDA is used to process the natural language information contained in each report summary by estimating latent topics and assigning each report to one of them. Together with other variables these topic assignments serve as splitting variables for finding segments in the data to which local statistical models for the reported number of fatalities are fitted. Segmentation and fitting is carried out with recursive partitioning of negative binomial distributions. We identify segments with different fatality rates that correspond to a small number of topics and other variables as well as their interactions. Furthermore, we carve out the similarities between segments and connect them to stories that have been covered in the media. This gives an unprecedented description of the war in Afghanistan and serves as an example of how data journalism, computational social science and other areas with interest in database data can benefit from modern statistical techniques. (authors' abstract)

Les tempêtes hivernales causent des pertes en bois qui s’élèvent à 50% du volume des dégâts dans les forêts européennes. Les phénomènes de déracinement des arbres (chablis) sont les plus fréquents or ils sont encore mal compris. Cette thèse vise à mieux comprendre le processus de déracinement de l’arbre et à identifier les traits structuraux et matériels (racines, sol) ayant un effet du premier ordre sur l’ancrage racinaire dans le cas du Pinus pinaster. Un modèle d’éléments finis a été développé et permis de simuler et suivre la chronologie des ruptures successives au cours du déracinement. Un seuil de rupture globale de l’ancrage est ainsi défini comme une résultante de l’architecture et de la résistance des matériaux en jeu (racines, sol). Cela devrait permettre à terme d’améliorer les modèles de risque au vent qui actuellement n’incluent pas de relation mécaniste pour le chablis. Dans la même logique, nous nous sommes appuyés sur les données expérimentales pour construire une architecture simplifiée du système racinaire du P. pinaster. L’importance des différentes composantes sur le mécanisme d’ancrage a été étudiée et le rôle essentiel joué par le pivot et les racines traçantes montré. Ce résultat confirme de nombreuses études expérimentales et théoriques et pour la première fois permet de quantifier ces effets. Le nombre de paramètres pertinents pourra ainsi être réduit pour exprimer l’ancrage. Cela ouvre des perspectives intéressantes pour simplifier l’utilisation du modèle pour l’appliquer à d’autres espèces, d’autres conditions de sol et différentes pratiques sylvicoles. / Winter storms cause 50% of wood damage by volume to European forests. Tree uprooting isthe most frequent phenomenon during storms ; however the mechanism is not well understood.This thesis aims to better understand the tree uprooting process and to identify both rootstructural features and material properties which have first-order effects on tree anchoragestrength for the case ofPinus pinaster. A Finite Element Model has been developed and allowedsimulating and tracking the sequential root breakage during the course of tree overturning. Anoverall tree anchorage strength is thus defined as the resultant of contribution of root systemarchitecture and material strength (roots, soil). This would allow improving the risk modelswhich currently don’t include any mechanistic relationships to describe tree uprooting. In thesame spirit, we have relied on root architectural data to build a simplified root system patternwith features ofP. pinaster. Importance of different root components has been studied andthe essential role of the taproot and shallow roots demonstrated. This result has confirmednumerous experimental and theoretical studies and for the first time quantified these impacts.Therefore the number of relevant parameters can be reduced to express overall root anchorage.This opens new prospects to simplify the model in order to apply to other species under othersoil conditions and considering different silvicultural practices