OPUS Root Disease

Description

Impacts of Armillaria root disease under no forest management and comparison to impacts predicted by the Western Root Disease Model (WRDM)

Introduction: The purpose of this study was to evaluate the short-term predictive capability of the Western Root Disease Model (WRDM) extension of the Forest Vegetation Simulator (FVS) in central Oregon mixed-conifer forests for project planning situations by using data from permanent plots for comparison to predictions derived using the Southern-Oregon Northern-California variant (SORNEC) of the FVS alone and with the WRDM.

The Western Root Disease Model (WRDM) extension of the Forest Vegetation Simulator (FVS) was designed to simulate the impact of Armillaria and other root diseases at the stand or inventory plot level over time (Frankel and others 1998). The WRDM model provides a dynamic representation of the spatial epidemiology of three common root diseases: laminated root rot (caused by Phellinus weirii), annosus root disease (caused by Heterobasidion spp.), and Armillaria root disease (cause by Armillaria spp.). It is a conceptual spatial model based on a simplified characterization of root disease epidemiology including the geometry of spread, root colonization, and impact on susceptible trees. Input information for the model includes the type of root disease, the proportion of the plot/stand area in acres to be modeled as being inside root disease center(s), and the density of infected trees (if the area is to be modeled as one center.) Additional information not collected as a part of standard stand exams (such as the locations of infected stumps and dead trees) can be included to further define the disease simulation environment. This is the first analysis to evaluate the accuracy of the WRDM in central Oregon.

In 1991 and 2002, stand structure and density were measured in 147 variable-radius plots (40 basal area factor) within 12 stands on a 100-acre unmanaged area in south-central Oregon (Winema National Forest, Klamath Ranger District). These stands include some of the most productive sites for white fir (Abies concolor), Shasta red fir (A. magnifica var. shastensis), Douglas-fir (Pseudotsuga menziesii var menziesii), and ponderosa pine (Pinus ponderosa) in south-central Oregon. Stand elevation ranges from 4,900 to 5,300ft. Stands were infected with Armillaria root disease, caused by Armillaria ostoyae, a common pathogen and important agent of disturbance in mixed-conifer forests throughout western North America and central Oregon. In 1991, individual tree measurements within plots included diameter at breast height (dbh), live crown ratio, and tree condition. Each plot was also given a root disease severity rating based on a visual above-ground estimation of the level of canopy reduction caused by root disease. Trees under 5.0 inches dbh were recorded on a fixed 1/100^th acre plot. Plots were re-measured in 2002 for tree dbh and condition. In 1991, thirty-two of the plots had no detectable root disease while the remaining 115 plots had light to severe root disease. The percentage of stand area with visible root disease ranged from 14 to 100%.

Measured changes in the structure and density of these stands were compared to model predictions made using SORNEC alone and with the WRDM, for projections at both the stand and plot level. The purpose of this analysis was to use local permanent plots in order to evaluate the short-term predictive capability of the WRDM in project planning situations, so the model was parameterized using only standard stand examination data likely to be available to project planners.

Results:

Overall, Armillaria root disease caused a decline in stand density and significant changes in structure. Average decline in basal area was 70 ft² per acre. The largest proportional decline in canopy cover (38%) occurred in medium (9.0–20.9 inches dbh) and pole (5.0–8.9 inches dbh) size classes while the least (0%) occurred in large trees (?21.0 inches dbh).

Where Armillaria root disease was present, the WRDM better predicted changes in forest structure than projections using SORNEC alone.  Overall, the WRDM tended to overestimated root disease impacts compared to actual changes, while SORNEC did not predict any impact. On average, plot predictions with the WRDM reduced the unexplained variation by an additional 35% over predictions made only using SORNEC.

The strength of the correlation between WRDM predictions and measured changes depended on both the primary modeling unit (stand or plot) and the size-class of trees under consideration. Plot-level predictions tended to be more highly correlated with the measured data than the stand-level predictions, and the size classes which the WRDM most accurately predicted changes for were those most severely impacted by Armillaria root disease (medium and pole size trees).

The results of this study support the use of the WRDM for making predictions in stands where Armillaria root disease is an important management consideration. However, the authors also noted several characteristics of the WRDM that may act as deterrents to its use in routine project planning, including the level of effort needed to parameterize and troubleshoot the model. Suggestions for reducing such barriers and making the WRDM more attractive to project planners are provided.

More detailed descriptions and results from this project are available in (Maffei et al. 2008).

Literature cited:

Frankel, S.J. 1998. User’s guide to the Western Root Disease Model, Version 3.0. Gen. Tech. Rep. PSW-GTR-165. Albany, CA: U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station. 166 p.

Maffei, H. M.; G. M. Filip; K. L. Chadwick; L. David, 2008: Western Root Disease Model Simulation Versus Plot Remeasurement: 11 Years of Change in Stand Structure and Density Induced by Armillaria Root Disease in Central Oregon.  USDA Forest Service Proceedings RMRS-P-54. Third Forest Vegetation Simulator Conference, Fort Collins, CO, USDA Forest Service, Rocky Mountain Research Station.