This study was directed to improve our understanding of the ecology of Swiss needle
cast (SNC) of Douglas-fir, a disease that produces extensive damage to forests and
plantation in the coastal region of Oregon and Washington. A disease prediction model
for the coastal area of Oregon was built by establishing the relationship between the
distribution of disease severity and the environment. Currently available methods of
determining the distribution of SNC were analyzed, and the possibility of mapping the
disease using Landsat TM satellite images was explored.
Two types of regression approaches were used to study the relationship between
disease severity and climate, topography, soil and forest stand characteristics. Although
both types provided useful information and insight, the multiple regression approach was
chosen over the regression tree analysis to build the model, due to its capacity to
produce a continuous prediction response.
Fog occurrence, precipitation, temperature, elevation and slope aspect, were the
variables that contributed to explain most of the disease severity variability. Findings
agree with and formalize our previous understanding of the ecology of SNC: cool and
wet conditions in summer appear to increase disease severity. When the model was
applied to past climate conditions, retrospective predictions suggest that changes in
climate in the last two decades could help to explain the observed recent regional
increase in SNC disease severity.
The resulting model was used to construct a disease prediction map. This map
showed an accuracy equivalent to the currently available SNC aerial survey. The
prediction model, however, is able to produce a continuous prediction surface, more
suitable for testing and appropriate for assisting in disease management and research.
A strong relationship between mature stand canopy defoliation and the Landsat TM
indices greenness and brightness, indicates that it is possible to use satellite imagery to
map SNC. In contrast, young stands showed high variability, most likely due to the
relatively high proportion of exposed understory vegetation.
The possibility of mapping stand defoliation is of great importance because this
symptom can be directly linked to tree growth and forest productivity. Satellite imagery
can be used in future and in retrospective disease mapping. / Graduation date: 2002
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/32528 |
| Date | 17 October 2001 |
| Creators | Rosso, Pablo H. |
| Contributors | Hansen, Everett M. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0016 seconds