http://hdl.handle.net/1957/21686 Sat, 25 May 2013 01:32:38 GMT 2013-05-25T01:32:38Z http://hdl.handle.net/1957/38697 A Comparison of the Spatial Linear Model to Nearest Neighbor (k-NN) Methods for Forestry Applications Ver Hoef, Jay M.; Temesgen, Hailemariam Forest surveys provide critical information for many diverse interests. Data are often collected from samples, and from these samples, maps of resources and estimates of aerial totals or averages are required. In this paper, two approaches for mapping and estimating totals; the spatial linear model (SLM) and k-NN (k-Nearest Neighbor) are compared, theoretically, through simulations, and as applied to real forestry data. While both methods have desirable properties, a review shows that the SLM has prediction optimality properties, and can be quite robust. Simulations of artificial populations and resamplings of real forestry data show that the SLM has smaller empirical root-mean-squared prediction errors (RMSPE) for a wide variety of data types, with generally less bias and better interval coverage than k-NN. These patterns held for both point predictions and for population totals or averages, with the SLM reducing RMSPE from 9% to 67% over some popular k-NN methods, with SLM also more robust to spatially imbalanced sampling. Estimating prediction standard errors remains a problem for k-NN predictors, despite recent attempts using model-based methods. Our conclusions are that the SLM should generally be used rather than k-NN if the goal is accurate mapping or estimation of population totals or averages. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by Public Library of Science and can be found at: http://www.plos.org/. Tue, 19 Mar 2013 00:00:00 GMT http://hdl.handle.net/1957/38697 2013-03-19T00:00:00Z http://hdl.handle.net/1957/38696 Ethanol Attracts Scolytid Beetles to Phytophthora ramorum Cankers on Coast Live Oak Kelsey, Rick G.; Beh, Maia M.; Shaw, David C.; Manter, Daniel K. Ethanol in sapwood was analyzed along vertical transects, through small spot cankers and larger basal cankers, of Phytophthora ramorum-infected stems of Quercus agrifolia at three sites in California. Trees with large basal cankers, known to attract scolytid beetles, had a 4.3 times higher ethanol level than trees with spot cankers that attract fewer beetles. Ethanol concentrations inside cankers, where scolytid beetles preferentially attack, varied by about four orders of magnitude among samples, with a median level of 16.0 mu g.g(-1) fresh mass. This concentration was 4.3 and 15.5 times greater, respectively, than the concentrations at 1 cm or 15-30 cm outside the canker boundaries. In the laboratory, we demonstrated that ethanol escaped through the bark of a Q. garryana log just 3 days after it was added to the sapwood. At the three study sites, traps baited with ethanol captured more Xyleborinus saxesenii, Pseudopityophthorus pubipennis, and Monarthrum dentiger (all Coleoptera: Curculionidae: Scolytinae) than traps baited with ethanol plus (-)-alpha-pinene, or ethanol plus 4-allylanisole (4AA). Logs of Q. agrifolia with a 50 % ethanol solution added to the sapwood were placed at the study sites, with or without additional bark treatments above the ethanol. The number of scolytid beetle gallery holes above the ethanol-infused sapwood was 4.4 times greater than that on the opposite side of the log where no ethanol was added. Attachment of ultra-high release (-)-alpha-pinene pouches to the bark surface above the 50 % ethanol solution reduced scolytid attacks to a density of 19.1 % that of logs without this treatment. We conclude that ethanol in P. ramorum cankers functions as a primary host attractant for scolytid beetles and is an important link in colonization of these cankers and accelerated mortality of Q. agrifolia. The results of this research shed light on the chemical ecology behind the focused scolytid attacks on P. ramorum-infected coast live oaks, and lay the groundwork for future efforts to prolong the survival of individual trees of this keystone species. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by Springer and can be found at: http://www.springer.com/?SGWID=0-102-0-0-0. Sun, 24 Mar 2013 00:00:00 GMT http://hdl.handle.net/1957/38696 2013-03-24T00:00:00Z http://hdl.handle.net/1957/38559 Stand development, fire and growth of old-growth and young forests in southwestern Oregon, USA Sensenig, Thomas; Bailey, John D.; Tappeiner, John C. We studied stand development in three distinct forest types in southwestern Oregon using six stands each in uncut and clear-felled old-growth stands and nearby young stands (18 total). Old-growth stands showed a wide range of tree ages (>300 years) and low tree densities for several centuries; rapid early growth produced trees with large crowns and diameters, as well as low height-to-diameter ratios. In contrast, young stands established much quicker and at higher tree densities; beyond their initial 20 years, trees had smaller diameters at equivalent ages, slower growth rates, smaller crowns and higher H:D than trees in old-growth stands. Low-intensity disturbance, likely dominated by fire, was common in oldgrowth stands during their early development. Fire scars showed these stands burned frequently from 1700 to 1900, and low levels of tree recruitment occurred in a complex relationship with fire during this 200 years. There was no evidence of fire, however, in either old-growth or young stands after 1909, and their densities were well above that of 1900; in old-growth stands, 15–25% of the basal growth occurred from 1950 to 1990, and it appears that they are on a development pathway different from what they experienced from 1700 to 1900. Furthermore, tree recruitment has been limited in both old-growth and young stands since 1950 while biomass and fuels continue to accumulate rapidly. Past stand dynamics can be emulated by prescribed fire and light thinning to reduce risk of loss from severe fire or insects, as well as to partially restore stand conditions that existed prior to fire exclusion. Our results suggest that young stands can be grown to produce high levels of biomass/wood, or their development can be altered to more closely follow that of old-growth stands depending on management objectives. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by Elsevier and can be found at: http://www.elsevier.com/. Sat, 05 Jan 2013 00:00:00 GMT http://hdl.handle.net/1957/38559 2013-01-05T00:00:00Z http://hdl.handle.net/1957/37602 Modeling Relative Humidity in Headwater Forests Using Correlation with Air Temperature Eskelson, Bianca N. I.; Anderson, Paul D.; Temesgen, Hailemariam Microclimate variables such as air temperature and relative humidity influence habitat conditions and ecological processes in riparian forests. The increased relative humidity levels within riparian areas are essential for many plant and wildlife species. Information about relative humidity patterns within riparian areas and adjacent uplands are necessary for the prescription of effective buffer widths. Relative humidity monitoring is more expensive than temperature monitoring due to greater sensor costs, and it is primarily conducted for research purposes. To make relative humidity monitoring in riparian areas more cost effective, we explored modeling relative humidity as a function of air temperature and other covariates using linear fixed and linear mixed effects models applied to two case studies. Localizing predictions for stream reaches using a linear mixed effects model or a linear fixed effects model with correction factor improved model predictions, especially when large variability among stream reaches was present. A minimum of three to five relative humidity measurements per stream reach seem sufficient to estimate the random stream reach effect or correction factor for the linear mixed and linear fixed effects models, respectively. Including covariates that describe distance to stream and canopy cover in addition to air temperature improved model performance. Although further model refinement is probably needed to allow detection of small changes in relative humidity associated with changes in stand structure from partial overstory removal, the models developed provide a means towards decreasing the costs of monitoring microclimates of importance to riparian area function. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by Northwest Scientific Association and can be found at: http://www.northwestscience.org/. Tue, 01 Jan 2013 00:00:00 GMT http://hdl.handle.net/1957/37602 2013-01-01T00:00:00Z