Results

2001 Survey

There were 1,164 polygons delineated as dying oak trees on the IKONOS satellite imagery. A sample of 119 polygons, or 10 percent of the total, was randomly selected for ground truthing (Table: Diagnostic results for a sample of tree mortality locations randomly selected from photo-interpretation of satellite imagery of Ft. Hood, TX). Oak wilt was found to be the cause of mortality in 82 (69 percent) of the centers. The major factor, other than oak wilt, delineated as dying or dead trees on the imagery, was brush-clearing operations (19.3 percent), where piles of dead trees resemble the crowns of dead, standing live oaks. With two exceptions, all of the brush piles consisted of Ashe juniper cut and stacked during land-clearing operations. Relatively few other causes of mortality were found, including fire, wind damage, and damage to trees caused by military operations. At eight of the sites, trees identified as oaks were actually some other species, or causes of mortality were not readily identified.

Of the 1,164 polygons, 821 fell within the perimeter of Fort Hood. Of those 821 polygons, 144 or 18 percent were located in designated GCW habitat. Of the total 82 oak wilt centers identified in the survey (including the 1-mile buffer), 60 were located within the post perimeter. Only 7, or 12 percent of the oak wilt centers found within the Fort Hood perimeter, were located in designated GCW habitat.

2003-04 Survey

The highest stand densities, 1298 trees/ha, were found at the nesting sites within GCW habitat (GCW/NS) (Table: Species composition and average diameters of trees located in sample cluster plots for the 2003 - 2004 survey). Stand densities were also relatively high in habitat where there was no oak wilt (GCW/non-OW=886 trees/ha). At oak wilt locations both within and outside of habitat, the stand densities were relatively lower. There were also notable trends in the species among the various categories. The proportion of juniper was far lower outside GCW habitat where oak wilt was present (13 percent) than in the plots located within GCW habitat (62 percent) (Table: Species composition and average diameters of trees located in sample cluster plots for the 2003 - 2004 survey). Live oak density was greater in oak wilt locations, whether they were within (21 percent) or outside of habitat (33 percent), than in the uninfected plots within habitat (4 percent) or nesting sites (2 percent).

The juniper to oak ratios (J:O) varied widely among the four sampling categories. The highest J:O ratio was 6.57:1 in the GCW habitat where there was no oak wilt (Table: Juniper to oak ratios for each of the four sampling categories at Fort Hood). The lowest was 0.24:1 in oak wilt centers outside of GCW habitat.