Evaluation of Landslide Monitoring in the Polish Carpathians

In response to the June 15, 2010 request from the Polish Geological Institute (PGI) to the U.S. Geological Survey (USGS) for assistance and advice regarding real-time landslide monitoring, landslide specialists from the USGS Landslide Hazard Program visited PGI headquarters and field sites in September 2010. During our visit we became familiar with characteristics of landslides in the Polish Carpathians, reviewed PGI monitoring techniques, and assessed needs for monitoring at recently activated landslides.

Visits to several landslides that are monitored by PGI (the Just, Hańczowa, Szymbark, Siercza and Łasńica landslides) revealed that current data collection (monthly GPS and inclinometer surveys, hourly piezometers readings) is generally sufficient for collecting basic information about landslide displacement, depth, and groundwater conditions. Large landslides are typically hydrologically complex, and we would expect such complexity in Carpathian landslides, given the alternating shale and sandstone stratigraphy and complex geologic structures of the flysch bedrock. Consequently groundwater observations could be improved by installing several piezometers that sample the basal shear zone of each landslide being monitored by PGI. These could be supplemented by additional piezometers at shallower depths to help clarify general flow directions and hydraulic gradients. Remedial works at Hańczowa make the landslide unsuitable for monitoring as part of an early warning network. Monitoring there should focus on continued performance of the remedial works.

Our suggestions for new monitoring at recently activated landslides are summarized in table 1. Displacement monitoring using extensometers and (or) GPS is a high priority at Kłodne, Łaśnica, Łazki, and Siedloki. Geomorphologic mapping of active surface features (scarps, cracks, shear zones, folds, and thrusts) in sufficient detail to reveal the kinematics of each landslide would greatly help in planning subsurface exploration and monitoring. Mapping should take advantage of existing and future airborne lidar data sets of specific areas, where available. Borehole inclinometers and piezometers would complete the basic monitoring package for these landslides. The landslide at Kłodne may be well suited for more detailed monitoring for landslide process research, although research opportunities exist at the other landslides as well. The landslide near Siedloki may be a good candidate for terrestrial laser scanning (TLS). Tandem streamflow gages upstream and downstream from the Siedloki landslide, or laser distance meters to monitor advancement of the toe, may be needed to provide warning of stream blockage of Potok Milowski. A real-time warning system specifically for the Łazki landslide might be considered due to potential concerns about catastrophic movement into Międzybrodzie Reservoir.

Challenges associated with the establishment of a complete real-time monitoring and early warning system are far greater than just the technical and logistical aspects of installing remote monitoring systems at a large number of landslides. Long-term maintenance of a landslide monitoring network will involve considerable effort and expense as sensors break-down from exposure to weather, landslide movement, and harsh underground environmental conditions.

Once PGI’s planned pilot network of 10-20 monitored landslides is operating, a period of observation and analysis will be needed to establish appropriate alert levels and criteria for issuing alerts and warnings. Simultaneously, discussions with authorities will be needed to develop action plans for responding to landslide notifications and (or) warnings. Public resistance to landslide warnings and mandated evacuations may be high given the low historical incidence of fatalities and injuries resulting from Carpathian landslides and the small potential for warnings to reduce landslide damage to homes and land. Careful weighing of purpose, advantages, and costs of a large-scale monitoring and early warning program is needed early in the planning process and should be revisited regularly throughout pilot and final implementation.

In this report, we present a generic plan for monitoring of a hypothetical Carpathian landslide that illustrates how our suggestions for each of the specific landslides could be implemented. The plan includes basic pore pressure, displacement, and weather monitoring, along with supplemental monitoring for special conditions at specific landslides. Table 2 summarizes the overall approach and basic equipment and software requirements.