Large woody debris (LWD) is an important component of forest ecosystems and provides structural complexity to small streams. Riparian buffers are intended to provide long term supplies of LWD, but post harvest windthrow often occurs. To document the impacts of windthrow in riparian buffers and identify the components needed for small stream LWD recruitment modeling, I sampled 39 small streams at the Malcolm Knapp Research Forest (MKRF) and on Vancouver Island. I took two basic approaches. In the small stream experiment at MKRF a series of small clearcuts were harvested in 1998 in a 70 year old second growth stand. I measured LWD in 10m and 30m buffer treatments, and in the unharvested control. I added samples in mature and old-growth stands for comparison. In the second approach, I retrospectively sampled buffers that were exposed by harvesting from 0-20yrs ago on southwestern and northeastern Vancouver Island. In both studies, all logs greater than 7.5 cm diameter at mid-creek, in decay class 1 to 4 that spanned at least part of stream channel width were measured. There was no significant difference in the number of spanning and in-creek logs in 10m and 30m buffer given the short term monitoring of woody debris in the buffers. The majority of windthrown trees were still suspended above the stream channel years after a windthrow event. The height above stream was negatively correlated with log decay class and the buffer age class. The number of logs was higher in immature stands than mature stands. As the stems per hectare in riparian stands increases, so does the frequency of spanning LWD. The frequency of logs in decay classes 3 and 4 was higher in older buffers, and deciduous LWD decayed more quickly than conifers. Interestingly, the log length was found to be shorter in advance stage of decay. Key elements in a conceptual model of LWD recruitment via windthrow are the geometry of initial log position, log size, species and decay rate.

Large woody debris (LWD) is an important component of forest ecosystems and provides structural complexity to small streams. Riparian buffers are intended to provide long term supplies of LWD, but post harvest windthrow often occurs. To document the impacts of windthrow in riparian buffers and identify the components needed for small stream LWD recruitment modeling, I sampled 39 small streams at the Malcolm Knapp Research Forest (MKRF) and on Vancouver Island. I took two basic approaches. In the small stream experiment at MKRF a series of small clearcuts were harvested in 1998 in a 70 year old second growth stand. I measured LWD in 10m and 30m buffer treatments, and in the unharvested control. I added samples in mature and old-growth stands for comparison. In the second approach, I retrospectively sampled buffers that were exposed by harvesting from 0-20yrs ago on southwestern and northeastern Vancouver Island. In both studies, all logs greater than 7.5 cm diameter at mid-creek, in decay class 1 to 4 that spanned at least part of stream channel width were measured. There was no significant difference in the number of spanning and in-creek logs in 10m and 30m buffer given the short term monitoring of woody debris in the buffers. The majority of windthrown trees were still suspended above the stream channel years after a windthrow event. The height above stream was negatively correlated with log decay class and the buffer age class. The number of logs was higher in immature stands than mature stands. As the stems per hectare in riparian stands increases, so does the frequency of spanning LWD. The frequency of logs in decay classes 3 and 4 was higher in older buffers, and deciduous LWD decayed more quickly than conifers. Interestingly, the log length was found to be shorter in advance stage of decay. Key elements in a conceptual model of LWD recruitment via windthrow are the geometry of initial log position, log size, species and decay rate.

Large woody debris (LWD) is an important component of forest ecosystems and provides structural complexity to small streams. Riparian buffers are intended to provide long term supplies of LWD, but post harvest windthrow often occurs. To document the impacts of windthrow in riparian buffers and identify the components needed for small stream LWD recruitment modeling, I sampled 39 small streams at the Malcolm Knapp Research Forest (MKRF) and on Vancouver Island. I took two basic approaches. In the small stream experiment at MKRF a series of small clearcuts were harvested in 1998 in a 70 year old second growth stand. I measured LWD in 10m and 30m buffer treatments, and in the unharvested control. I added samples in mature and old-growth stands for comparison. In the second approach, I retrospectively sampled buffers that were exposed by harvesting from 0-20yrs ago on southwestern and northeastern Vancouver Island. In both studies, all logs greater than 7.5 cm diameter at mid-creek, in decay class 1 to 4 that spanned at least part of stream channel width were measured. There was no significant difference in the number of spanning and in-creek logs in 10m and 30m buffer given the short term monitoring of woody debris in the buffers. The majority of windthrown trees were still suspended above the stream channel years after a windthrow event. The height above stream was negatively correlated with log decay class and the buffer age class. The number of logs was higher in immature stands than mature stands. As the stems per hectare in riparian stands increases, so does the frequency of spanning LWD. The frequency of logs in decay classes 3 and 4 was higher in older buffers, and deciduous LWD decayed more quickly than conifers. Interestingly, the log length was found to be shorter in advance stage of decay. Key elements in a conceptual model of LWD recruitment via windthrow are the geometry of initial log position, log size, species and decay rate. / Forestry, Faculty of / Graduate

Thesis (Ph.D.)--Boston University / PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / This study is a quantitative analysis of the effect of fire and wind-throw on a spruce-fir-northern hardwoods forest on Peaks Island, Casco Bay, Maine. Six years after the burn, data for the effect of fire on approximately 170 acres of this forest type, were collected and categorized. The qualitative characteristics of the herb and shrub layers were noted and the soil analyzed. A random sample of the tree density was obtained by using 10 x 10 meter quadrats. In this sampling three size classes were recognized, seedling, sapling, and 1.0-4.0 inch d.b.h. For each size class the following distributions were determined: percentage frequency, percentage density, percentage basal area, density per acre, basal area per acre; and cumulative figures for the relative density, relative basal area, and relative frequency of each species over 1 inch d.b.h. A similar study was carried out on 8.8 acres of a wind-throw area, and on 144 acres of spruce-fir-northern hardwoods forest. In the latter, additional size classes up to 24.0 inches d.b.h. were added. Determining frequencies in randomly distributed quadrats resolved in a positive direction the question of whether fire and wind-throw significantly affect the plant successional trend. Since the tree species in each of the three areas were found to be nearly identical, the Chi-square criterion was employed to decide whether the density distribution was independent of the area. Two significant factors resulted: (1) Of the twenty-one species in the spruce-fir-northern hardwoods forest, thirteen were represented among the eighteen species of the wind-throw area. The distribution of these thirteen species differed significantly in the two areas. (2) Thirteen of the species of the spruce- fir- northern hardwoods forest, were represented in the burn which contained altogether fifteen species. Again, the density distribution differed significantly for the two areas. The hypothesis that the frequency distribution was independent of the area was rejected in both tests since the value of the Chi-square calculated, for the densities in each size class, was significantly different from what could be expected by chance. This substantiated the conclusion that the three unique ecological environments accounted for the quantitative distribution of their species. [TRUNCATED] / 2031-01-01

This study examined Hurricane Katrina damage in southeast Mississippi to identify stand and site characteristics that may contribute to wind-related damage. Aggregated forest plot-level biometrics were coupled with storm meteorology, topographical features, and soil attributes using GIS techniques to produce damage maps for specific tree species. Regression Tree Analysis was utilized to explore the relationship between damage type and distance variables (distance to coast/storm track). Results indicated that the total damage class had the greatest relationship with distance variables; individual damage classes (shear and blowdown) displayed a better relationship with stand-level variables (Quadratic Mean Diameter, Lorey’s Mean Height, Trees Per Hectare). Logistic regressions identified a negative relationship between damage and height variation, elevation, slope, and aspect and a positive relationship with TPH. For plots/stands nearest to the coast and storm track height variation, TPH, QMD, and LMH consistently predicted damage levels for most species examined.

The increasing frequency and severity of windthrow events affecting the forestry industry in New Zealand have raised important management issues surrounding the rate of colonisation of fallen trees by sapstain fungi and the time available for salvage harvesting before sapstain degradation limits potential economic returns. These fungi are known to be spread by a multitude of factors including wind, rain splash, harvesting processes and insect vectoring. Apart from the ecological interest in these interactions between fungi, plants and insects, sapstain fungi are also economically important because their hyphae discolour the sapwood and reduce the overall quality of the timber. The amount of time available to salvage harvest damaged trees is unknown, especially on seasonal and regional scales. Manipulative experiments were established in Pinus radiata forests to examine this seasonal and regional variation in sapstain attack following windthrow, and to investigate the importance of bark beetles as vectors of sapstain fungi. A range of methods were implemented to assess the role of bark beetles as vectors and to ascertain which sapstain fungi are associated with them. Experimental billet logs were caged to exclude beetles and subsequently analyse fungal attack in comparison with identical logs left exposed to beetles. In addition, individual beetles were sampled directly to determine whether they carried spores of particular fungal species and to assess the degree of association in vector-fungal dynamics. Finally, a novel application of DNA melt peak analysis was developed to investigate variation among the fungal communities associated with beetles potentially involved in vectoring sapstain spores. The moisture content of fallen trees was found to be the main factor regulating sapstain development, and when moisture content drops below 100% (on dry weight basis) sapstain fungi grew rapidly. The speed at which this level drops depends on the season, with much faster drying occurring in the warmer months of spring and summer. As a result, trees that fell in the previous winter or autumn did not develop significant sapstain levels until temperatures rose in the following summer, suggesting that storm-damaged trees that fall in winter can be left safely until just before the next summer before they are no longer suitable for salvage harvest. In New Zealand, the bark beetle species acting as vectors of sapstain fungi are not behaviourally adapted to colonisation of logs that are not in contact with the ground. Following windthrow events in pine forests, trees generally lie with their stems suspended above the level of the ground by their branches. As a result, under these circumstances, beetle colonisation of windthrown timber was low, and bark beetles were not a significant vector of stain. The caged and un-caged experimental log billets, however, were in contact with the ground, resulting in colonisation of the un-caged logs. In this case, bark beetles did play an important role in contributing to sapstain intensity, and the stain distribution within the logs mirrored that of the stain distribution. However, this effect may be due to the provision of access points for wind- or water-borne spores of the non-insect vectored stain fungus Diplodia pinea, or to the spread of hyphae through the tunnelling and feeding activities of beetles within the tree, rather than by bark beetles acting as vectors of spores. Bark beetles were confirmed as sapstain vectors with the isolation of seven different ophiostomatoid stain fungi from them, five of which were also found in wood. Finally, the development of a laboratory based, rapid species identification method was developed to identify fungal DNA. Melt peak analysis allowed the species-specific DNA melt temperatures to be compared with the melt temperatures of known species to be able to rapidly, and cheaply, identify an unknown species. Bark beetles are vectors of sapstain fungi in P. radiata forests, however the bark beetle species naturalised in New Zealand prefer to colonise wood when it is in contact with the ground. Following windthrow, trees are generally not attacked by beetles as they are held from the ground by their branches, leaving them to be stained predominantly be wind and rain dispersed stain fungus D. pinea. Stain did not occur until the moisture content of fallen trees dropped below 100%, which only happens in the warm months of summer and spring. In New Zealand, there are interactions between trees and bark beetles, and bark beetles with fungal pathogens from all around the globe resulting in a unique novel assemblage of species together for the first time. Understanding the dynamics of these species in their novel environment is crucial to effectively responding to potential pest threats.

The uprooting of trees due to high winds, or blowdown, is a common occurrence throughout the world. Rare, large scale blowdown events have the ability to drastically alter forest landscapes on a timescale of hours. Alone or when combined with other forest disturbances such as fires and insect outbreaks, blowdowns have the potential to modify existing watershed characteristics. Little information is known about how large-scale blowdowns affect the physical environment in general and watershed-wide sediment yield and deposition rates in particular. The few studies that have taken place have yielded inconclusive or somewhat contradictory results. Given the large scale of some events and the potential impact of such events it is important that the geomorphological effects of blowdowns be better understood, not only from a basic process but also from an applied perspective. Lakes receive and store a significant fraction of the disturbed sediment formerly found in the soil column and lake sediments have proven to be reliable and accurate records of sedimentation characteristics within their catchments. For the purpose of this study, it was hypothesized that areas subjected to blowdown will have increased erosion rates. It was also hypothesized, that lakes within blowdown areas will be subjected to accelerated sedimentation rates following events. In order to test these hypotheses and isolate the effects of a blowdown on sedimentation rate three small mountain lakes in the Mt. Zirkel Wilderness Area, Colorado were cored. The lakes selected consist of a control lake with no known recent history of blowdown in its watershed and two lakes that receive direct runoff from watersheds and slopes that were affected by a major blowdown event in 1997, the Routt-Divide Blowdown. One core was dated using the 210Pb method and multiple cores were analyzed for organic and inorganic matter concentrations to determine if changes in sedimentation rates and/or sediment compositions occurred following the event. In the case of the two lakes studied that had catchment slopes affected by blowdown, somewhat surprisingly, no significant change was observed with respect to sedimentation rates or sediment composition. In fact, sedimentation rates obtained from the dated core exhibited a decreasing trend that began decades prior to and continued through the blowdown event. This suggests that in the area affected by the Routt-Divide Blowdown it is likely that long-term changes in climate and the effect these changes have on the amount and type of vegetation present are the dominant influence on sedimentation rates, and that the Routt-Divide Blowdown of 1997 did not increase the rates. Based on further evaluation of the sites and the event it appears that slope characteristics and treefall orientation may be significant factors controlling the amount and type of disturbed sediment that moves downslope. In the case of future large-scale blowdowns, where their effect on water quality or sedimentation rates is in question, dominant treefall orientation with respect to slope should be an integral part of initial assessment, especially in remote area, and be considered when interpreting data from any similar study.

Magistro darbe tiriama 2010 m. rugpjūčio mėnesį Dubravos EMMU audros pažeisti medynai ir juose pagaminta apvaliosios medienos kokybė. Darbo objektas – 2010 m. rugpjūčio mėnesį škvalo Dubravos EMMU pažeisti medynai. Medienos kokybės bei medžių stiebų sortimentinės struktūros tyrimai atliekami Dubravos EMMU, Vaišvydavos girininkijoje esančiame škvalo pažeistame brandžiame eglyne. Darbo tikslas – Įvertinti vėjavartų apimtį ir medienos kokybės pokyčius Dubravos EMMU. Darbo metodai – Darbe naudoti dokumentų analizės, duomenų grupavimo, palyginimo, vidurkių skaičiavimo ir apibendrinimo metodai. Darbo rezultatai. Išanalizavus Dubravos EMMU audros likvidavimo metu 2010 m. 10-12 mėn. pagamintų pjautinųjų sortimentų kiekius nustatėme, kad pagaminta 76% daugiau negu visada, prastesnės C klasės medienos. Dubravos EMMU, Šilėnų ir Vaišvydavos girininkijose įvertinus vėjo pažeidimus pagal plotą skirtingose augavietėse 60 brandžių medynų, nustatėme, kad intensyviausiai pažeisti medynai, kurie sudaro 35% yra Lcp augavietėse. Įvertinus pažeistus medynus pagal vyraujančią medžių rūšį nustatyta, kad didžiausią pažeistų medynų dalį 46.9% sudaro eglynai. Sortimentuojant 26 vėjo pažeistus medžius nustatyta, kad dėl pakitusios sortimentinės struktūros dėl lūžių taškuose suskeldėjusios medienos, prarandama iš viso 19.8% likvidinės medienos, bei vidutiniškai vieno medžio kirtimo atliekų kiekis padidėja 2.1 karto. Iš vėjo pažeistų medžių, pjautinųjų sortimentų medienos kokybė lyginant su galimai... [toliau žr. visą tekstą] / Study object – 2010.08.08 storm damaged forest stand at Dubrava forest enterprise. Windblown timber quality assessment and storm damaged tree round wood structure assessment was done at Vaišvydava local forest district, at Dubrava forest enterprise. Study aim - to quantify most damaged forest types and to assess the round wood quality losses due to stem breakage and other defects which reduced merchantable timber volume during logging at damaged stands, after 2010.08.08 storm at Dubrava forest enterprise. Methodology – Data grouping, comparison, generalization, document’s analysis. Results - As the main changes in round wood quality during the period of salvage clear cuttings of storm damaged timber at Norway spruce stands at DFE were seen a significant increase of C quality class Norway spruce (Picea abies L) sawn logs – 76 per cent of all sawn logs. Evaluation of windthrow damage on DFE results showed that nearly half (46.9 percent) of all damages occurred on Norway spruce stands. As the most endangered forest soil type of spruce stands, where one third of all damages occurred, was recorded: fertile and humid Myrtillo-oxalidosum site. Survey of 26 storm damaged trees round wood structure showed, that 19.8% reduction of merchantable volume is due to breakage, and due to that fact was seen significant increase - 2.1 time amount of residuals. Windblown tree round wood quality assessment showed significant difference in timber quality if we compare it with normal conditions... [to full text]

Hintergrund und Zielstellung Im Zuge des Waldumbaus wird die Rotbuche (Fagus sylvatica L.) über Voranbau in Fichtenreinbestände eingebracht. Deren Bewirtschaftung zielt meist auf die Erzeugung von Wertholz ab. In diesem Zusammenhang dient der Fichtenschirm als Instrument der Steuerung der Ressourcenversorgung und damit der Qualifizierung der Buchen („edle Halbschattform“). Die in der Vergangenheit zumeist niederdurchforstungsartige Behandlung bedingt jedoch, dass viele Fichtenbestände hinsichtlich ihrer Stabilität keine geeigneten Ausgangsbedingungen für eine langfristige Schirmstellung bieten. Die Schadbilder infolge immer häufiger auftretender Windwurfereignisse verdeutlichen dies eindrucksvoll und belegen unmissverständlich, dass diesbezüglich Handlungsbedarf besteht. Die vorliegende Untersuchung ging daher den Folgen eines unplanmäßigen, sturmbedingten Schirmverlusts für Wachstum und Qualität vorangebauter Rotbuchen nach. In diesem Zusammenhang wurden auch der Einfluss von Pflanzendichte und Entwicklungsstadium des Voranbaus geprüft. Material und Methoden In den beiden Modellregionen „Thüringer Wald und Hügelland“ sowie „Sächsisches Erzgebirge und Vorland“ wurden Qualitätserhebungen (vgl. BÖRNER ET AL. 2003) auf insgesamt 29 Buchenvoranbauflächen durchgeführt. Auf 17 dieser Flächen wurde der Fichtenschirm im Januar 2007 durch den Orkan KYRILL meist vollständig geworfen. Seitdem entwickelten sich die Rotbuchen ohne Schirmschutz. Alle weiteren Voranbauten stehen nach wie vor unter Fichtenschirm und dienen als Referenz für Wachstum und Qualität. Jedem Voranbau wurde das auf den Freistellungszeitpunkt bezogene Entwicklungsstadium (vor oder nach Dickungsschluss) zugewiesen. Zur repräsentativen und objektiven Festlegung der Aufnahmeeinheiten in den Flächen wurde ein rasterbasiertes Probekreisverfahren gewählt. Das quadratische Gitternetz mit einer Weite von 20 m wurde mittels GPS im Gelände eingemessen. Die Rasterschnittpunkte bildeten zugleich die Zentren der 19,95 m² großen Probekreise (Plotradius: 2,52 m). Die Datenaufnahme erfolgte auf insgesamt 204 repräsentativen Plots. In jedem Probekreis wurde die Pflanzenzahl erhoben und auf Hektarwerte hochgerechnet. Für die Referenzflächen erfolgte hier weiterhin eine Ansprache des Überschirmungsgrads. Einbezogen wurden nur Plots mit locker-lichtem Fichtenschirm. Die Datenerhebung wurde an den maximal sechs höchsten Buchen eines jeden Probekreises vorgenommen und konzentrierte sich damit auf die (vor-)herrschenden und somit vitalsten Bestandesglieder. Daraus resultierte ein Stichprobenumfang von 895 Einzelbäumen. Die Datenaufnahme fand nach Abschluss der Vegetationsperiode 2012 statt, sodass seit der Freistellung sechs Jahre vergangen waren. Mathematisch-statistische Auswertung Aufgrund der räumlich geschachtelten Versuchsanlage kamen bevorzugt gemischte Modelle zur Anwendung. In Abhängigkeit vom Skalenniveau der Response-Variable und dem vorliegenden Verteilungstyp wurden lineare gemischte Modelle (LMM) oder aber generalisierte lineare gemischte Modelle (GLMM) eingesetzt. Multinomialdaten mit einer festen Rangordnung der Response-Kategorien wurden mit einem erweiterten logistischen Regressionsmodell, der Proportional Odds Logistic Regression (POLR), analysiert. Zur Umsetzung aller beschriebenen Methoden wurde die Statistiksoftware R (Version 3.0.1) genutzt. Ergebnisse • Wachstum und Zuwachsreaktion: Sechs Jahre nach dem Sturmereignis besteht kein signifikanter Höhenunterschied zwischen den Buchen auf Sturmflächen und unter Schirm. Allerdings stiegen die Haupttrieblängen ab dem zweiten Jahr der Freistellung an, sodass der Wachstumsgang der freigestellten Individuen signifikant von dem der überschirmten Bäume abweicht. Hohe Stammzahldichten im Voranbau fördern das Höhenwachstum. Das Durchmesserwachstum profitiert im Vergleich dazu deutlich stärker von der Freistellung, wie die signifikant größeren Wurzelhalsdurchmesser der freigestellten Buchen belegen. Entsprechend intensiv fiel auch die Reaktion des Radialzuwachses aus, die ohne Verzögerung bereits im Jahr der Freistellung einsetzte. Der stärkste Grünast zeigt einen im Vergleich zum Stamm synchronen Wachstumsgang, wenn auch auf niedrigerem Niveau. Unabhängig von der Überschirmung dämpfen hohe Bestandesdichten das Durchmesserwachstum. Gemessen am H/D-Verhältnis weisen die Buchen unter Schirm sowie bei hoher Pflanzendichte einen schlankeren Wuchs auf. Zwischen den beiden Entwicklungsstadien, das heißt vor oder nach Dickungsschluss, bestand kein Unterschied. • Astigkeit und Astreinigung: Sechs Jahre nach dem Schirmverlust sind an den freigestellten Buchen signifikant größere Astdurchmesser bei zugleich steileren Astwinkeln zu erkennen. Mit zunehmender Pflanzenzahl im Voranbau werden die Buchen feinastiger, während der Astwinkel unbeeinflusst bleibt. ASIX und Grünastdichte sinken nur auf der Freifläche merklich, erreichen dadurch aber ab Pflanzenzahlen von 8.500 St./ha bzw. 6.000 St/ha niedrigere Werte als Buchen unter Schirm. Zwischen den beiden Entwicklungsstadien sind hinsichtlich der Wirkung des Schirmverlusts keine Unterschiede zu verzeichnen. Totastanzahl und grünastfreie Schaftlänge hingegen werden durch die Freistellung nicht beeinflusst. Auf der Freifläche wurden jedoch stärkere Totäste gefunden. Mit zunehmender Pflanzenzahl im Voranbau steigen Anzahl und Durchmesser der Totäste sowie die grünastfreie Schaftlänge signifikant. Am deutlichsten tritt dieses Phänomen bei Voranbauten nach Dickungsschluss zutage. • Schaft- und Kronenmorphologie sowie Häufigkeit von Z-Baum-Anwärtern: Sechs Jahre nach dem Schirmverlust sind an freigestellten Buchen signifikant häufiger Steiläste und vor allem Zwiesel erkennbar. Im Stadium vor Dickungsschluss treten auf Freiflächen doppelt bis dreifach so häufig Tiefzwiesel auf wie unter Fichtenschirm. Mit zunehmender Pflanzenzahl im Voranbau sinkt die Zwieselanzahl moderat, der Anteil tief gezwieselter Buchen sogar rapide. Dadurch kommen im Stadium nach Dickungsschluss ab 9.000 Buchen pro Hektar unabhängig von der Überschirmung kaum noch Tiefzwiesel vor. Die gutachterliche Qualitätsansprache bestätigt, dass durchschnittlich etwa 60–70 % aller Buchen auf den KYRILL-Flächen den Zwieseltypen zuzurechnen sind, während unter Fichtenschirm mit ca. 90 % die (sehr) guten Formen dominieren. Mit zunehmender Bestandesdichte steigt der Anteil der besseren Qualitäten tendenziell, jedoch nicht signifikant. Die gutachterliche Qualitätsansprache zeigt eine sehr hohe Übereinstimmung mit der anhand von ASIX, Astwinkel und Zwieselzahl einer Buche modellierten Qualitätsklasse. Unter Fichtenschirm ist die Dichte der Z-Baum-Anwärter grundsätzlich sechsmal so hoch wie in freigestellten Voranbauten. Diese steigt aber unabhängig vom Schirm mit der Pflanzendichte des Voranbaus signifikant an. Schlussfolgerung und Handlungsempfehlungen Der Fichtenschirm hat eine herausragende Bedeutung für die Qualitätsdifferenzierung von Buchenvoranbauten. Eine locker-lichte Überschirmung stellt den geeigneten Kompromiss zwischen vitalem Wachstum und Qualitätsentwicklung der Buchen dar. Ungeachtet aller weiteren Faktoren bewirkt ein plötzlicher, unplanmäßiger Schirmverlust gravierende Qualitäts-einbußen der an den Halbschatten adaptierten Buchen. Sehr gute Qualitäten, das heißt wipfelschäftige Buchen mit gering dimensionierten, horizontal ausgerichteten Ästen und zügiger Astreinigung, sind nur durch die Kombination aus langfristig stabiler Überschirmung und Pflanzendichten im Voranbau ab 6.000–8.000 St./ha zu erreichen. Eine Kompensation des Freistellungseffekts ist durch hohe Pflanzenzahlen im Voranbau nur in sehr begrenztem Umfang möglich. Ein hoher Seitendruck kann den fehlenden Schirmdruck demnach nicht ersetzen. Selbst bei höchster Bestandesdichte waren die Formen auf den Freiflächen weitaus schlechter als unter Schirm. Dies gilt unabhängig vom Entwicklungsstadium der Voranbauten. Der Bestandesschluss bewahrt also offenbar nicht vor freistellungsbedingten Qualitätsverlusten. Allerdings gewährleisten Pflanzendichten ab 8.000–9.000 St./ha, dass die Tiefzwieselbildung in geschlossenen Buchendickungen auch bei einer abrupten Freistellung weitgehend unterbleibt und eine hinreichende Anzahl von Z-Baum-Anwärtern zur Verfügung steht. Ist es erst einmal zu einem Schirmverlust über einem Buchenvoranbau gekommen, bestehen kaum Möglichkeiten, aktiv zugunsten der Buchenqualität tätig zu werden. Die einzige Chance besteht im gezielten Erhalt bestehender Z-Baum-Anwärter und deren konsequenter Pflege. Für zukünftige Waldumbaumaßnahmen wird deshalb ein standort- und risikodifferenziertes Vorgehen beim Buchenvoranbau empfohlen. Dieses sollte sich an erster Stelle an der Stabilitätssituation des Fichtenschirms orientieren. Ausschließlich bei gegebener Stabilität ist das Produktionsziel Buchenwertholz als realistisch einzustufen, sodass kostenintensive Voranbauten mit den oben genannten Pflanzendichten angelegt werden sollten (Funktionsumbau). Bei einer offensichtlich (sehr) hohen Prädisposition der Fichtenschirme gegenüber Sturm und Borkenkäferbefall ist stattdessen ein extensiver Voranbau mit reduzierter Pflanzenzahl vorzuziehen. Dieser zielt im Sinn eines Nachhaltsumbaus primär auf eine Stabilisierung der Waldökosysteme ab und trägt zur Erhöhung der Flexibilität der zukünftigen Waldbewirtschaftung bei (WAGNER 2007, 2008). / Backround and Objectives In the course of forest conversion European beech (Fagus sylvatica L.) is introduced in spruce monocultures by advanced planting. Management of beech stands aims mostly on production the of high quality timber. In this context, the shelterwood of spruce officiates as an instrument for controlling resource availability and thus to qualify advanced planted beech. But because of low-thinning treatment in past, most spruce stands do not provide suitable conditions for long-term shelterwood with regard to their missing stability. The situation of forest damage due to periodic windthrow events illustrate this impressively and show need for action. Therefore, the present study analyzed the consequences of an unplanned, storm-related loss of canopy for growth and quality of advanced planted beech. In this context, the influence of stand density and developmental stage of advanced regeneration will be examined as well. Material and Methods In two model regions "Thuringian Forest” (middle of Germany) and "Saxon Oremountains” (East of Germany) quality investigations (cf. BÖRNER ET AL. 2003) were carried out on a total of 29 advanced planted beech stands. In 17 of these stands the spruce canopy was mostly completely thrown by hurricane KYRILL in January of 2007. Thus, the beech trees have grown without shelter since then. The other advanced regeneration stands are still under spruce canopy and serve as a reference for growth and quality. A developmental stage (before or after thicket-stage) was designated to each advanced planting site in reference to the time of overhead canopy release. For representative and objective definition of experimental plots a grid-based plotdesign was used. The square grid with distance of 20 m was calibrated by GPS in the field. The grid inter-sections built the centers of 19.95 m² circular sample areas (plot radius 2.52 m). Data was collected on a total of 204 representative plots. On each plot, the number of beech plants was collected and extrapolated to hectare values. In addition, the canopy closure was also estimated for reference plots within the sample plots. For data collection only plots with moderate canopy closure were included (maximum area of canopy gaps: one crown projection). Data collection was carried out at maximum of six highest trees per plot. Thus it is concentrated on (pre-) dominant and therefore most vital individuals. This resulted in a sample size of 895 trees. The data collection was started after growing season in 2012, six years after canopy release. Mathematical and statistical analysis Due to spatially nested plot design Mixed Models have been applied preferably. Depending on scale level of the response variable and given distribution type, Linear Mixed Models (LMM) or Generalized Linear Mixed Models (GLMM) were used. Multinomial data with fixed order of response categories was analyzed with an extended logistic regression model, the Proportional Odds Logistic Regression (POLR). These calculations were implemented by statistical software R (version 3.0.1). Results • Growth and reaction of annual increment: Six years after the storm event there is no significant difference in height between beech trees on storm areas and such under shelterwood. However the annual increment of terminal shoot rose from the second year after release, so the trend of increment deviates significantly from beech trees under shelterwood. High stand density in advanced planting promotes height growth. Diameter growth benefits significantly more from release compared to height growth. So released beech trees showed significantly larger root collar diameters. Reaction of radial increment, which began immediately after release, was appropriately intensive. The thickest living branch shows synchronous trend of increment, but on a lower level. Regardless of canopy cover high stand densities delimitate diameter growth. The H/D-ratio demonstrated that beech trees under shelterwood and in high density stands showed more slender forms. There was no difference between the two developmental stages. • Branchiness and natural pruning: Six years after the loss of spruce canopy significantly greater branch diameters in combination with steeper branch angles were found on released beech trees. With increasing stand density in advanced planting the beeches showed finer branches while branch angles remain unaffected. ASIX and the density of living branches decreases appreciably for released trees only, so reached lower values than trees under shelterwood by stand densities off 8.500 pcs/ha and 6.000 pcs/ha, respectively. There was no difference of release effect between the developmental stages. Number of dead branches and bole-length without living branches however, were not affected by loss of canopy. However more dead branches were found on beeches in the opening. With increasing stand density in advanced planting, the number and diameter of dead branches and bole-length without living branches increased significantly. This was most distinct for beech stands that reached thicket-stage before release. • Stem- and crown shape, number of crop tree candidates: Six years after canopy-loss, released beeches showed significantly more steep branches and especially forks. Released stands before thicket-stage showed two to three times as likely deep-forks in relation to stands under spruce canopy. With increasing stand density in advanced planting the number of forks per trees decreased moderately, the proportion of deep-forks even rapidly. In consequence, regardless of the canopy cover, deep-forks only occurred marginally in beech stands after thicket-stage with at least 9.000 pcs./ha. The expert quality estimation confirms that on average about 60–70 % of beeches on KYRILL sites belong to a “fork type”, while under spruce shelterwood (very) good shapes dominate with about 90 %. With increasing stand density, the proportion of good shapes tends to increase, however not significantly. The estimated quality class showed a very high correlation with the modeled one, based on ASIX, branch angle and fork number per beech. The density of crop tree candidates under spruce canopy is basically six times as high as in released stands. However, regardless of shelterwood it increases significantly with stand density. Conclusion and silvicultural implication: Spruce canopy is of outstanding importance for quality and differentiation of advance planted beech. A moderate shelterwood is the appropriate compromise between vital growth and good quality development. Regardless of any other factors, a sudden and unplanned loss of canopy effects a loss of quality of shade adapted beech trees. Very good qualities, that means straight to top trees with fine, horizontally oriented branches and timely self-pruning, can only be achieved by combining of long-term shelterwood and stand densities in advanced planting of at least 6.000–8.000 pcs./ha. Compensation of the release effect through high stand density in advanced planting is only possible to a very limited extent. High intraspecific competition cannot replace the lack of shelterwood. Beech shapes were far worse than under shelterwood, even at the highest stand density. This applies regardless of the developmental stage of a beech stand. So closing of advanced planted beech stand (thicket-stage) does not prevent release related quality losses. However, plant densities from 8.000–9.000 pcs./ha ensure that deep fork formation in closed beech thickets is largely suppressed and a sufficient number of crop tree candidates is available, even after abrupt release. Once an abrupt loss of canopy above advanced planted beech has occurred, there are hardly any opportunities to actively engage in favor of beech quality. The only chance is maintenance of existing crop tree candidates and their consistent care. Therefore, for future forest conversion with beech, a site- and risk-differentiated approach is recommended for advanced planting. This should be based firstly on the stability of shelterwood. Only with high stand stability, the target of high-grade beech-timber is realistic, so costly beech plantings with high stand density mentioned above should be applied. For spruce stand with an obvious (very) high risk towards bark beetles and / or storms, an extensive advanced planting with reduced stand density is preferable. In terms of “Sustainability-conversion” this aims primarily on stabilizing forest ecosystems and increasing flexibility for future forest management (WAGNER 2007, 2008).

The thesis was intended to fill the information gap in the ecohydrology of the Pinus rotundata dominated peatbogs in the Třeboň Basin Biosphere Reserve, Czech Republic by studying various aspects of their water régime (i.e., transpiration of the dominant tree species, soil water chemistry, retention ability). The study is based on regular monitoring of the water table, water discharge, soil water chemistry and precipitation (during 1995-2000, study sites Červené blato and Žofinka peatbogs). Additionally, the transpiration of adult Pinus rotundata trees was measured in the field during 1999-2000 at the experimental plot in the Červené blato peatbog (330 ha, 465-475 m a.s.l.). Nowadays, the Pinus rotundata-dominated peatbogs represent almost natural (peaty) forests inside of otherwise human-made forest plantations. Long-term vegetation changes after natural disturbances such as windstorms, insect infestation and fire are only occasionally reported for the central European natural forests. Therefore the evaluation of vegetation changes after disturbances, typical of boreal forests, and their impact on peatbog hydrology has also been included in this thesis (study site Žofinka peatbog, 130 ha, 470-475 m a.s.l.). Transpiration of the central European endemic tree species, Pinus rotundata Link, was...