Prepared by:
Michelle Howell
Darren Beckstrand
A.G. Flynn

Prepared for:
Environmental Geology, G561
Portland State University
Winter Term 1999



This report summarizes the ground movement study conducted on the Haussler Road area of Kelso, Washington (Figure 1 not available online). The Haussler Road area is a residential neighborhood constructed on an ancient landslide. The purpose of the study is to determine if ground movement is occurring, and if so, to determine whether the movement exhibits a spatial pattern. The study was conducted using quantitative data collected from a ground surface survey and qualitative data collected from resident interviews and questionnaires. As a result of the study, it appears that ground movement is occurring in the Haussler Road area. Of the 46 homes included in the study, approximately 46% are exhibiting signs of movement. Approximately 30% of the homes located above the ancient landslide scarp are exhibiting signs of movement, while approximately 61% of the homes located on the ancient landslide are exhibiting signs of movement.



According to the Geologic Map of Washington (Walsh et al., 1987), approximately 2 to 14 million years ago, the ancestral Columbia River deposited gravels, sands, silts, and clays in portions of what is now the northwestern United States. These deposits are now known as the Troutdale Formation. In certain areas, the Troutdale Formation was deposited upon a much older, near-shore marine sedimentary deposit known as the Cowlitz Formation. In areas where the Troutdale Formation overlies the Cowlitz Formation, landslides are known to occur. One such area exists in Kelso, Washington. The Troutdale Formation overlies the Cowlitz Formation on a northwest-southeast trending ridge located approximately one-half mile east of interstate I-5 (Figure 2 not available online).


In 1973, a slope stability study of this ridge was conducted by the Washington Department of Natural Resources (Fiksdal, 1973). As a result, the ridge was divided into classifications of slope stability ranging from 1 to 5. These classifications are now popularly known as the Fiksdal classifications, after Allen Fiksdal, who was the geologist working on the state project. A Fiksdal classification of 1 indicates a stable slope, and the classifications increasingly progress to a classification of 5, which indicates an area of active or historical landsliding. Figure 3 (not available online) illustrates the Fiksdal classification boundaries in the vicinity of the ridge. Figure 4 (not available online) provides an explanation of each classification.

Over the last 40 years, the north and west slopes of a portion of this ridge have been residentially developed. This portion of the ridge is called Davis Terrace. The north slope development was constructed in an area with a Fiksdal classification of 1. However, it now appears that this area should have been classified as a 5. An ancient scarp and landslide debris exist there (Koloski, 1998). This slide (popularly known as the Aldercrest-Banyon slide) caused millions of dollars in damage and caused many residents to lose their homes. Portions of the slope have even been declared federal disaster areas.

Since approximately December 1998, the residents on the western slopes of Davis Terrace have reported new cracks forming in their homes. This is of great concern to the residents as ground movement occurring in their area may be the beginnings of a catastrophic landslide similar to the Aldercrest-Banyon slide. One concerned resident contacted Portland State University (PSU) to request that a ground movement study be conducted. The purpose of the study is to determine if ground movement is occurring, and if so, to determine whether the movement exhibits a spatial pattern. The study was conducted as the graduate student project for the Environmental Geology course offered by Professor Scott Burns.


Selection of Study Area

An initial site visit was conducted on January 18, 1999. Homes located at 494 Apple Lane and 131 South Vista Way were visited as these were known to be exhibiting signs of ground movement. At this time, it was noted that many homes in the neighborhood are built on an ancient landslide (Fiksdal, 1973). Therefore, the study area was selected to include all homes in the neighborhood, from the top of the ridge (North and South Vista Way) down to Rons Court (Figure 5 not available online). Note that the study area includes homes both above and below the scarp of the ancient landslide. The study area will herein be referred to as the Haussler Road area.


Study Methods

In order to determine if ground movement is occurring in the Haussler Road area, two basic methods were used: (1) a ground surface survey was conducted, and (2) residential properties were evaluated in terms of degree of movement exhibited. Each method is described in detail below.

A) Ground Surface Survey

On February 5, 1999, eleven sites ("stations") on the ground surface in the Haussler Road area were selected for surveying (Figure 6). The station locations were selected in order to represent the ground surface both above and below the scarp of the ancient landslide. The station locations also had to be situated such that they would be visible from our survey equipment, a total station.

Because of the large size of the study area (approx 0.5 square mile), it was necessary to set up the total station ("base") on Rons Court in order to have visual contact with as many stations as possible. This is problematic because the area around Rons Court is showing marked movement, and ideally the base in a survey should be stationary. However, since one station was located above the scarp and was assumed to not be moving, this was used as a control point. Figure 6 illustrates the location of these stations. Please note that station #2 is outside of the northern boundary of the photograph in Figure 6.

The ground movement study consisted of two separate survey events. On February 9, 1999, the locations of each of the eleven stations relative to the base were surveyed. This was accomplished by setting up a reflector on a tripod at each station. Locations were surveyed in terms of distance, horizontal angle, and vertical angle from the base. Subsequent to surveying each station, the height above the ground of each reflector was noted, and the exact station location was marked with masonry nails and spray paint on the ground. Using trigonometric principles, the x, y, and z coordinates of each point were established. Points #10 and 11 (control point) were then utilized to define a line of known azimuthal bearing. The bearing was corrected for true north, and as a result, the northing and easting of each station could be determined using an assumed northing and easting of 100 for the base. The elevation of each station was calculated using an assumed datum of 100 feet for the base.

On March 2, 1999, reflectors were again set up at each station, at the same height above the ground. Station #4 could not be located as the road at that location had been repaved. This was due to the road being torn apart by slumping at that location. The station locations were surveyed, and northing, easting, and elevation were once again calculated. The coordinates calculated from each event are summarized in Table 1.


B) Residential Property Evaluation: Questionnaires and Home Visits

According to Koloski (1998), signs of movement which are exhibited in homes at the beginning of a landslide can include jammed doors and windows, cracks in concrete foundations, and changes in surface water seepage patterns. Other damage noted in the two homes initially visited included:

bulletseparation of external chimney from foundation of home,
bulletrecent propagation of household cracks,
bulletdrywall over the nailheads bulging (likely due to the drywall sheet undergoing stress and strain),
bullettwisting ceiling beams,
bulletwater seeping into basement through new cracks,
bulletbulging walls,
bulletseparation of internal cabinetry from walls, and
bulletcreaking, popping and snapping noises from house.

In addition to the above signs, the evaluation included searching for pistol-butt, tilting retaining walls, and sloping porches.

On February 15, 1999, questionnaires were delivered to approximately 50 homes in the Haussler Road area. The questionnaires included questions about the above mentioned signs of ground movement. The residents were asked to fill out the questionnaires and return them to the PSU geology department in pre-addressed envelopes which were provided. A sample questionnaire is included with this document as Appendix B. In addition, a concerned resident assisted the study by telephoning neighborhood residents and encouraging them to return their questionnaires. By the end of the study, 37 questionnaires were returned. This is an approximately 75% response rate.

On February 20, February 27, and March 2, 1999, the students conducted an exhaustive door-to-door survey on as many homes in the study area as possible. A total of 36 residents allowed the students to inspect their homes. These home inspections included both residences who answered the questionnaire and those who did not. Detailed field notes were collected during home inspections, and it was noted whether the home was exhibiting signs of movement or not. If signs of movement were exhibited, they were noted in detail.

Table 2 summarizes the results of the questionnaire responses. For evaluation purposes, "yes" responses were assigned a value of 1. "No" responses were assigned a value of 0. These responses were then summed up for each home, thus indicating the number of "yes" responses. For example, a home with a sum of 3 responded "yes" to three questions. Each home was then assigned a code which categorizes the number of "yes" responses and as a consequence, degree of movement exhibited in the home. Homes with a sum of 0 were assigned a code of "0" (no movement); homes with a sum of 1 or 2 were assigned a code of "1" (possible signs of movement); homes with a sum greater than 2 were assigned a code of "2" (noticeable signs of movement).

Codes of 0, 1, or 2 were also assigned to each home visited. This code assignment was subjectively based on field observation. These codes are also included on Table 2. As indicated on Table 2, a relatively close correlation exists between coded based on survey responses and codes based on site visits. An overall code was then given to each home. Ultimately, for those homes with both a questionnaire code and a site visit code, the site visit code was used. It should be noted that these questionnaire codes are based only on the responses of the residents. As with any qualitative type study, a range of interpretation exists. For example, many residents may not have noticed signs of movement which are being exhibited in their homes. Some residents may have mistaken older cracks, which could be the results of normal settling, for new signs of movement. In addition, many residents may have been hesitant to report signs of movement in their homes due to understandable concerns about property value.



a) Ground Surface Survey

The total station survey data are included as Appendix A (not available online). The calculated coordinates are summarized in Table 1. As indicated in Table 1, movement at each station was calculated using the change in elevation and vertical coordinates between survey dates. The results for stations #3 and 5 were anomalous and were not considered in the study. As a result of the survey, measurable ground movement appears to have occurred at each location. A maximum movement of 24 millimeters (1 inch) was calculated (station #1).

b) Residential Property Evaluation

The results of the overall code assignment are summarized in Table 2 and spatially presented on Figure 7. The property boundary for a home with a code of 2 is indicated in red, yellow for a 1, and green for a 0. As illustrated on Figure 5, 46% of all participating homes exhibited signs of movement. Of these, 66% are located below the scarp (on the ancient landslide), and 34% are located above the scarp. Movement in homes located above the scarp could possibly be undergoing localized movement associated with sliding on small, local drainage slopes of settling.


Conclusions and Recommendations

Based on the results of this study, it appears as though the ground is experiencing movement in the Haussler Road area in Kelso, Washington. In addition, a spatial pattern appears to exist among those homes exhibiting movement. A greater percentage of these homes are located on the ancient landslide than of those located above the landslide. Of the homes located above the ancient landslide scarp, 30% are exhibiting signs of movement. Of the homes located below the scarp (on the ancient landslide), 61% are exhibiting signs of movement.

Due to the subjective nature of the study, it is recommended that a more detailed study be conducted on the Haussler Road area as soon as possible. The detailed study will need to definitively determine if ground movement is occurring. In addition, the subsurface of the ancient landslide should be investigated to determine hydrology, slip plane location, and nature of the sediments and bedrock. The detailed study should utilize intrusive techniques such as slope inclinometer installation.



Fiksdal, J.A., 1973. Slope Stability of the Longview-Kelso Urban Area, Cowlitz County. Washington Department of Natural Resources Open File Report 73-2, revised 1989.

Koloski, J.W., 1998. Slope Failure Evaluation, Aldercrest-Banyon Landslide, Kelso, Washington. Report, Geotechnical Engineering Services, Geo Engineers, Inc., Redmond, Washington.

McGuffey, V.C., Modeer, V.A, Jr., and Turner, K. A., 1996. Chapter 10 (Subsurface Exploration) in Landslides, Investigation and Mitigation. National Research Board Special Report 247, Transportation Research Council, Washington, D.C.

Walsh, T.J., Korosec, M.A., Phillips, W.M., Logan, R.L., and Schasse, H.W., 1987, Geologic Map of Washington – Southwest Quadrant. Washington Division of Geology and Earth Resources Geologic Map GM-34.

For questions and comments, contact