We used state and transition models to integrate natural disturbances and management activities for a 275 000-ha landscape in the central Oregon Cascades. The landscape consists of a diverse mix of land ownerships, land use allocations, and environments. Three different management scenarios were developed from public input: (1) no management except wildfire suppression on federally managed lands, (2) manage Federal lands to increase multistory forests of large and very large trees, and (3) manage Federal lands to move toward historical conditions. All scenarios treated privately owned lands as if they were wildlandurban interface (WUI) areas and all recognized wilderness, reserves, and general forests within federally managed lands. Models were run for 200 years and 30 Monte Carlo simulations to include variability in fire years and other natural disturbances. Passive management on federally managed lands resulted in small increases in single-story and multistory large-tree forests and increases in highseverity wildfire and insect outbreaks. Managing toward multistory large- and very-large-tree forests resulted in minor increases in those forest types and increased wildfire and insect outbreaks. Contrary to intent, this scenario did not generate appreciable increases in multistory large- and very-large-tree forests. Managing toward historical conditions resulted in strong increases in single-story large- and very-large-tree forests and decreases in high-severity wildfire and insect outbreaks. All three scenarios resulted in conversion of most WUI to open grass, shrub, and forest conditions.

As landscape scale assessments and modeling become a more common method for evaluating alternatives in integrated resource management, new techniques are needed to display and evaluate outcomes for large numbers of stands over long periods. In this proof of concept, we evaluate the potential to provide financial support for silvicultural treatments by selling timber harvested during treatments designed to achieve non-timber objectives such as fire hazard reduction or wildlife habitat improvement. We introduce the concept of dimensionless indices to describe and compare physical accessibility, harvest and hauling costs, and potential revenue from wood products. These indices are combined into a composite utilization index that portrays the relative potential of each polygon for wood utilization and associated cost offsets from integrated resource management activities. To illustrate these concepts, we simulate vegetation dynamics, management interventions, and natural disturbances over a 100-year period and summarize results into both tabular outputs and maps for a 178,000ha landscape with more than 50,000 stands.

Breast-high stem sections were sampled from 56 western hemlock (Tsuga heterophylla (Raf.) Sarg.) trees growing in 15 plots representing a wide range of tree and site conditions in northwestern Oregon. Growth and wood density traits of individual rings were measured via X-ray densitometry, and relationships of ring density and its components to age and growth rate were analyzed. Ring density was highest (0.49 g/cm(3)) near the pith, declined to 0.40 g/cm(3) at age 10, remained stable to about age 25, and then increased gradually and remained between 0.43 and 0.44 g/cm(3) from age 38 to 45 and beyond. A negative influence of rapid growth on whole ring density was greatest at young ages and diminished with time, becoming nonsignificant beyond age 30. Earlywood density, latewood density, and latewood proportion were all negatively related to ring width at young ages, but by age 21-25, latewood proportion was the only component of ring density that remained significantly diminished by increased growth rate. Residual differences in wood density (after age and growth rate were considered) did not appear to be related to either stand density or site class. Overall, young-growth hemlock trees are relatively uniform in wood density and likely to be more so if grown in intensively managed stands.

Stem irregularities can influence estimates of tree and stand attributes, efficiency of manufacturing processes, and quality of wood products. Eccentricity and fluting were characterized in young, managed western hemlock stands in the Oregon Coast Range. Sixty-one trees were selected from pure western hemlock stands across a range of age, site, and densities. The trees were felled and disks were removed from breast height, base of live crown, and various percentage-of-height locations along the bole. Indices of out-of roundness (OOR), pith-off-center (POC), and fluting severity (fluting index) were computed, flutes were counted, and depth of deepest flute was measured on each disk. These variables were related to relative tree height, tree diameter, and several stand- and tree-level attributes. The lower portions of stems tended to be more out-of-round and have a higher degree of off-centered piths than upper stem heights. Basal and breast-height disks had a higher number of flutes, deeper flutes, and higher fluting severity compared to disks from upper stem positions. Tree size (diameter at 4.5 ft) was only weakly correlated with OOR and uncorrelated with POC, but strongly correlated with more flutes, deeper flutes, and higher fluting severity. OOR and POC were not significantly correlated with any other tree or stand attributes. The fluting variables were positively correlated with stand age and overall tree growth rate and negatively correlated with trees/ac. Thus, silvicultural practices that result in more rapid growth, wider spacing, and longer rotations are likely to result in more extensive fluting but will have little or no effect on stem eccentricity.

Extractives can account for between 1 to 20% of the oven-dry weight of wood of various tree species and can influence wood density values appreciably. Removing these chemical deposits (extraction) in wood samples can help establish a consistent baseline for comparing wood densities where extractives are expected to differ between sample parameters. Although western hemlock is a very important timber species in the Pacific Northwest, laboratories that determine wood density may or may not remove extractives prior to density assessment. Wood density values were compared before and after extraction for 19 young-growth western hemlock samples. Extraction was performed using 95% ethyl alcohol-toluene solutions. Ring density values averaged 0.045 g/cm 3 lower for extracted samples compared to unextracted samples across rings. Slightly higher amounts of extractives were found at rings near the pith; however, a general consistency in extractive content existed among samples and along the radial profile.

It would be valuable economically to know what are the biological triggers for formation of mature wood (currently of high value) and (or) what maintains production of juvenile wood (currently of low value), to develop silvicultural regimes that control the relative production of the two types of wood. Foresters commonly assume the bole of softwoods produces juvenile wood within the crown and mature wood below. We tested that assumption by comparing growth ring areas and widths and wood density components of the outer three growth rings in disks sampled from different vertical positions of 34-year-old Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) trees. The 18 trees were sampled from one site and had a wide range of heights to live crown. Most of the variance (63-93%) in wood characteristics (growth ring area: total, earlywood, latewood; growth ring width: total, earlywood, latewood; latewood proportion: by area, width; and ring density: total, earlywood, latewood) was due to within-tree differences (related to age of the disk). Stepwise regression analysis gave us equations to estimate wood characteristics, after which we analyzed the residuals with a linear model that included whether a disk was within or below the crown (defined as the lowest node on the stem with less than three live branches). After adjusting for tree and disk position, only 2-10% of the residual variation was associated with whether the disk was in or out of the live crown. There were no statistically significant differences at p = 0.05 between a given disk (by node number) in versus out of the crown for any of the factors studied. Moreover, the wood density characteristics were not statistically significant at p = 0.30. This research suggests that there was no effect of the crown position on the transition from juvenile to mature wood as judged by wood density. Therefore, we found no evidence to support the concept that tree spacing and live-branch pruning have a significant effect on the cambial age of transition from juvenile to mature wood in Douglas-fir trees of this age.

Cross-sectional disks were cut at two stem heights (1.5 m and 3.0 m) from 9-year-old trees of three Populus clones grown in an intensively-cultured plantation in western Washington. At age 1.5 years, when the trees averaged 3.4 m tall, half of the trees were pruned by removing all branches below 1.8 m. Ring width, wood density, and fiber length were measured for each ring. Pruning had no effect on mean ring width or wood properties, averaged over the entire disk or on rings produced during the 2nd through the 4th years. Averaged over all trees, wood density of the 1.5-m sample was 0.37 g cm(-3) during the first 3 years, decreased somewhat at age 4 or 5, and then increased to an average of 0.45 g cm(-3) at age 9. Fiber length increased from 0.57 mm at age I to nearly 1.0 mm at age 9. Averaged over all disks at 1.5 m, clones differed significantly in ring width, wood density, and fiber length. Mean values for the two wood properties at 3.0 m were slightly lower than those at 1.5 m and did not differ significantly among clones. Within clone correlations between ring width and wood density or fiber length or between wood properties were low, and generally nonsignificant or inconsistent.

Pruning trials in young alder stands were sampled to evaluate response to pruning. Effects of pruning (1) live branches on different dates, and (2) dead branches with or without damaging the branch collar were assessed on trees pruned in 3- and 6-year-old plantations, respectively. Six years after pruning, stem sections were collected and dissected in the longitudinal-radial plane to expose the center of the stem and branch stub. Ring counts and linear measurements were made for various boundaries or points, including time of pruning, stub length, defect, and beginning of clear wood formation. Pruning during the growing season and, to a lesser extent, late in the growing season when leaf abscission was beginning, resulted in shorter times and distances to formation of clear wood (2.1 years, 14.5 mm) than pruning in the dormant season or just prior to the beginning of the growing season (2.6 years, 18.6 mm). Cutting the branch collar on dead branches led to shorter times and distances to clear wood (2.8 years, 21.9 mm) than intentionally avoiding such wounding (3.5 years, 24.8 mm); these differences were associated with shorter branch stubs as there were no differences in the amount of defect. Epicormic branching was minimal in the two pruning studies, averaging less than one branch per tree in the date of pruning test and only two branches per tree in the branch collar wounding study. Assessments for comparable unpruned trees indicated that times to form clear wood after branch death would be markedly greater and that epicormic branching was equal to or greater than that determined for pruned trees. Although statistically significant differences occurred among different pruning dates and with branch collar wounding, the decision to prune or not prune is of much greater practical importance, regardless of when (date) or how it is done. Such pruning decisions can be made by using this information on time and distance to clear wood in economic analyses developed with available data on tree growth, log volume, lumber recovery, pruning costs, and price differentials for clear vs. knotty wood.