ABSTRACT: Managing Landslide Risk from Forest Practices in British Columbia. Special Investigation.

CGRG Bibliography of Canadian Geomorphology
Search Results

Author :

Date :

Title :

Publication :

Issue :

Page(s) :

Abstract

The frequency of landslides can be increased by forest harvesting and road building on steep slopes. Usually, landslides are triggered by extreme rainfall events; this was certainly the case in the summer of 1997 when record rainfall over a two-month period in the Salmon Arm area triggered a number of natural and forestry-related landslides. One of these events was a massive debris flow down Hummingbird Creek, near Sicamous, which caused over 3 million dollars damage to public and private property. Hummingbird Creek is a prime example of the damage landslides can cause. Landslide reduction was one of the major objectives of the 1995 BC Forest Practices Code (the Code). The Code established procedures for professional landslide hazard mapping, site assessment and road engineering procedures to reduce the incidence of landslides associated with forest practices. There is a general assumption that the Code successfully reduced the number of landslides, and that professional assessments were a key part of this success, but to date there has been no evidence to support these assertions. To address this gap, the Board examined the management of landslide-prone terrain in three areas, two on the Coast and one in the Interior, through evaluation of landslide rates and review of terrain stability mapping and terrain stability assessments. The objectives of this study are to report on: 1. The incidence and trends of forestry-related landslides and the extent of damage to the environment. 2. The adequacy of terrain stability mapping and assessments. 3. The lessons learned in applying this information to the Forest and Range Practices Act (FRPA) environment. How the Study was Done Airphoto and satellite imagery was used to examine landslide occurrence in Code cutblocks and roads from two areas on VancouverIsland (Kyuquot and Gordon River) and one area in the BC interior (Revelstoke). The terrain stability map hazard rating and the terrain stability field assessment (TSFA) for each cutblock were compared to operational planning documents and to actual landslide occurrence. The comprehensiveness of the TSFA reports was assessed. Incidence and Trends A total of 46 Code road and cutblock landslides were counted in the 455 steep slope cutblocks in the three study areas, or an average of one landslide every year for every 14 square kilometres of steep slope cutblocks. This landslide frequency is a significant improvement over pre-Code landslide activity, but is still an increase over natural landslide rates. There is a significant probability that any landslide that does occur will have an environmental effect. Approximately 60 percent of the Code landslides in the study area had a potential “material adverse effect on a forest value,” based on criteria established by the Board, such as defined impacts on a fish stream, loss of soil, or loss of plantation. Code landslides are not, however, a significant soil conservation issue at the landscape scale, disturbing only 0.3 percent of the total cutblock area. Provincially this figure will be much lower, as the study areas are three of the most landslide-prone areas of the province. Code landslides differ from pre-Code landslides. Code landslides are much less frequent in gullies, along stream escarpments and off roads, compared to pre-Code landslides. This probably reflects the retention of reserves and management zones around streams and gullies, as well as better road location, and improved construction practices under the Code. The factors that contributed to the landslides are difficult to determine; in over half the cutblocks examined in this study, the factors could not be determined. Of the remainder, harvesting or road-building on identified high hazard areas, or windthrow in adjacent areas, are the most common contributing factors. Many more landslides (164 over the eight-year study period) initiated from old roads constructed prior to the Code coming into effect. This is in spite of considerable road deactivation work. While deactivation of pre-Code roads has not been particularly effective in restoring hillslope stability, deactivation appears to effectively stabilize roads built under the Code. Adequacy of Terrain Stability Mapping and Assessments Licensees achieved reduced landslide frequencies in the Code era through a number of measures, including better road locations, improved road-building techniques, reserves around stream escarpments and gullies, proactive deactivation, and increased use of professional engineers and geoscientists. The significance of professional terrain mapping and assessments in this equation is the subject of the second part of this reort. Reconnaissance terrain stability mapping, prepared by professional geoscientists, correctly identified most cutblock areas that eventually experienced landslides as potentially unstable or unstable. No landslide cutblocks were mapped as stable. The criteria used for determining landslide hazard class were not always consistent, however, between mappers in the same region. Reconnaissance terrain stability mapping did not appear to be used strategically in planning the location of cutblocks and roas, but rather was used only for determining whether a terrain stability field assessment (TSFA) was needed. A TSFA was carried out on 92 percent of the cutblocks where required. However, the results of the TSFA were not always incorporated into the silviculture prescription, as is required. About 45 percent of silviculture prescriptions did not address the results of the TSFA; nor were the results of the TSFA addressed in the cutting permit or road layout and design documents. This means that while TSFAs were nearly always being completed where required, in nearly half the cases, the results appear to have been ignored. Part of the reason for this may have been the utility of the TSFAs themselves. The comprehensiveness of TSFAs examined in this study was quite variable. All TSFAs provided an estimate of the likelihood of a landslide occurrence for the cutblock. Overall, the TSFAs accurately predicted a moderate or high likelihood of a landslide in those cutblocks where they occurred. However, half did not stratify the block by hazard class, provide hazard assessments for adjacent areas, or estimate the size, the number or the consequence of the landslides. Some TSFAs provided detailed prescriptions for block adjustment or road construction, while others were silent. The more recent TSFAs were generally the most comprehensive. Implications for Landslide Management under FRPA The system for management of unstable terrain established under the Code seems to work—the incidence of landslides has been reduced. More thorough professional assessments and more consistent implementation of their recommendations should reduce landslide incdence even further. The challenge for landslide management under FRPA is to maintain this momentum. The mandatory aspects of landslide management no longer exist under FRPA, and licensees will now have the discretion to decide when and where to conduct mapping and assessments. The shift from a hazard prevention strategy to a risk management strategy is a significant change from what existed under the ode. The consequence of landslides has now entered the equation and forest managers will have to apply landslide risk management tobalance environmental and timber supply objectives. Six recommendations for implementation of a landslide management system under FRPA are made as a result of this study: 1. The landslide management system that was developed under the Code should continue under FRPA. Terrain stability field assessments can be a strong tool for management of unstable terrain. Licensees should continue doing terrain stability field assessments and use the results. 2. The Ministry of Forests and Range regional offices should establish quantitative criteria for classifying landslide hazard in trrain mapping. A landslide hazard class should be defined by the probability or expected frequency of landslides per unit area, rather than by a subjective description. 3. The Ministries of Environment and Forests and Range should establish what “a material adverse effect” from landslides would be for each resource value in FRPA—with public resources it is the role of government, not individual licensees, to define where the threshold is for a material adverse effect on resource values. It is important to develop criteria for acceptable risk for each of the FRPA values separately. 4. The Ministry of Forests and Range should regularly conduct inventories of landslides as part of the Forest and Range EvaluationProgram, as there is no FRPA requirement for licensees to report landslides. 5. Licensees should use a decision-making framework for the management of landslide risk. A framework separates the process of landslide risk management into distinct steps, and clarifies the respective roles of terrain stability professionals and forest resource managers (see the Board’s Reiseter Creek complaint investigation for more discussion of this concept). Licensees should also develop objective criteria for triggering when a TSFA must be conducted, to provide consistency in application across their operations. One example is the criteria currently being developed by a coastal licensee that combines terrain stability mapping, climatic zone mapping and watershed use to trigger when a TSFA is needed. 6. The Joint Practices Board, or the ABCFP, should develop companion guidelines for forestry managers to the APEGBC Guidelines for Terrain Stability Assessments in the Foest Sector. This guidebook should provide advice to the forest manager in applying appropriate levels of landslide risk management. The guidebook should also provide standardized terminology for effective communication, and clarify the responsibilities of forest managers and terrain professionals.

Bibliography of Canadian Geomorphology