Bilgiustam

Landslide Susceptibility Map


Landslides are natural hazards that have a significant impact globally. Compared to other natural disasters, landslides are one of the most costly and deadly geological disasters that threaten and affect the socioeconomic conditions of many countries in the world. A landslide can be triggered by various natural events and people. It occurs when the slope breaks from the saturation of the soil layers and moves downhill, either due to excessive precipitation events or due to external forces, causing loss of life, property, environment and economic damage. For example, in the US alone, landslides cause about $3.5 billion in damage and cost 25 to 50 lives each year. Also, only in 2014 did a landslide occur in Nepal where animal loss was much higher than human loss and infrastructure damage was more expensive compared to the country’s economy. Slope breaks also cause massive sedimentation in streams and lakes, which is a major cause of flooding.
Landslides commonly occur primarily when the slopes of mountainous areas become unstable due to instability or external driving forces. The hazard of a natural disaster can be classified as high, medium and low according to the possible impact in terms of volume, duration, distance, area and speed at which the slope is defeated. Because they can adversely affect human life and property, it is essential to monitor, detect, map and conduct hazard analysis to reduce the impact of their hazards and save human life, property and the environment worldwide. Susceptibility maps for landslide-prone regions can be developed by bringing together all the potential predisposing factors that cause landslides.
Landslide susceptibility depends on local terrain, land use and climatic conditions, which require spatial information. A susceptibility zone includes historical inventory information with an assessment of future landslide-prone areas, but not an assessment of landslide occurrence frequency. Also, the temporal probability of a landslide is not included in the susceptibility models. High-quality landslide inventory maps can be developed using on-site measurements and field research. However, in situ measurements and field research are time consuming, expensive and difficult for local to global scales. On the other hand, using remote sensing techniques and data such as aerial surveys, unmanned aerial vehicles (UAV), light sensing, range (LiDAR) or satellite, it may be possible to perform landslide susceptibility and inventory maps as well as landslide hazard analysis.
Although used continuously for remote sensing, landslide detection, mapping and monitoring, satellite data is available at relatively coarse resolutions. Therefore, it is generally considered to have moderate efficacy and reliability for landslide studies. On the other hand, hazard assessment requires high-resolution data to describe its spatial distribution and activity states from studies at both local and regional and global scales. Also, remotely sensed data is cost-effective, as most global satellite products are freely available and can cover rough and complex terrain that would otherwise not be possible to assess with on-site measurements. Even in the late 1990s, stereoscopic aerial-photo interpretation was the most widely used remote sensing tool applied for mapping and monitoring landslides. Many studies have been carried out on landslide hazard assessment using geographic information systems (GIS) and geographic information techniques. Recently, GIS and remote sensing tools have become powerful tools for integrating spatial data to conduct landslide studies.
Remotely sensed data and techniques are widely used in studies including landslide inventory, detection, monitoring, mapping and hazard analysis. Timely and high-quality information from space-based observations helps manage natural or man-made disasters. Accordingly, landslide risk mapping and management can help reduce disaster risk. Similarly, early landslide forecasts and warnings are important in reducing landslide hazards. The assessment of landslide vulnerability is used to identify which elements are at risk and why, and such information assists in disaster mitigation measures.
Whether it is weather, satellite or ground-based measurements, the use of remote sensing data for landslide studies is basically classified into three main categories, which are as follows:Landslide Susceptibility Map
• Detection and identification,
• Tracing,
• Spatial analysis and hazard estimation,

Sensitivity Map

Landslide prediction is vital to prevent potential damage and save human life. The landslide susceptibility map is important in estimating as it helps to identify potential landslide areas and any susceptible area. Local topography and hydrological conditions play an important role in landslide susceptibility. Although a suitable landslide inventory provides both spatial and temporal information about previous landslides on an area, a landslide susceptibility map provides information about potential future landslides on an area. However, detailed information on historical records of previous landslides, rainfall or earthquakes is vital in determining trigger thresholds. Landslide susceptibility can be measured from stable to very susceptible, and many researchers classify slopes into four landslide susceptibility classes as very sensitive, moderately susceptible, mildly susceptible, and stable.
Some researchers have used slightly different susceptibility classes to develop landslide susceptibility maps, such as unstable, metastable, moderately stable, and stable. Some studies also used sensitivity indices as very high, high, medium, low and very low. For disaster prevention, a landslide susceptibility map can be used in land use planning and decision making. A detailed sensitivity map for land use mapping helps local authorities manage these landscapes for urban or industrial planning and development. However, developing an effective landslide susceptibility map is always a challenge because it requires multiple spatial information about soil, geology, vegetation and hydrology. For example, Stanley and Kirschbaum identified four main issues that need to be addressed for the development of landslide susceptibility maps. These four main topics are as follows:
• Lack of detailed inventory,
• Minimum available input data,
• Regional differences in the importance of causal factors,
• Lack of expertise in landscape processes in large areas,
In order to develop a landslide susceptibility map, there are many methods in the literature using on-site measurements, models, remotely sensed data alone or in combination. For example, inventories and causal factors are used together with the statistical approach in developing a landslide susceptibility model to predict potential landslides. In addition, statistical approaches, physically based models and deterministic approaches have been used in the development of landslide susceptibility maps in many studies. Besides the different methods available for landslide susceptibility analysis and mapping, it is also important to have advanced tools and detailed spatial information to develop an effective susceptibility map.
Digital tools such as GIS and global positioning system (GPS) are mostly used to analyze spatial data and develop landslide susceptibility and hazard maps. In addition, remotely sensed data and technologies are widely used for effective landslide susceptibility mapping, hazard assessment, and risk assessment that further aids awareness, mitigation and management of potential threats. Many researchers have used remotely sensed data to develop landslide susceptibility maps from local to global scale. For example, Ray, Jacobs, Kirschbaum and team used TRMM and global precipitation measurement (GPM) remotely sensed precipitation data along with slope, geology, road networks, fault zones and forest loss to develop a global scale landslide susceptibility map. Ray et al used remotely sensed soil moisture (AMSR-E) with slope, soil and vegetation characteristics to develop dynamic landslide susceptibility maps at regional scale.
As a result, natural disasters such as hurricanes, earthquakes, tsunamis, and landslides continue to increase, causing damage to property and human life, especially in mountainous regions. The main causes of landslides are lithology, relief, geological structure, geomechanical features, conditioning factors such as weathering, triggering factors such as precipitation, seismicity, temperature change, static and dynamic loads. Traditional methods for landslide studies are mainly based on visual interpretation of aerial photographs and field investigation together. However, these methods are time consuming and not cost effective. On the other hand, remotely sensed data and advanced techniques at high spatial and temporal resolutions can be used for landslide studies at various scales.
More robust technologies and high-resolution data are required to reduce the impact of landslide threats on human lives, property and environments. We also need to develop advanced technologies that can improve landslide assessment, prediction and mitigation. According to Singhroy, the primary challenge is to have advanced technology or high-resolution data to recognize and interpret detailed geomorphic features of large and small landslides and determine if failure has occurred. Although the use of high spatial resolution radar and LiDAR data is very helpful in conducting landslide studies, satellite products with high spatial and temporal resolution are still limited. In the future, it would be useful to study landslide dynamics at various scales, especially in remote and hard-to-reach terrains, if real-time remote sensing products with high spatial and temporal resolutions are available.

References:
researchgate.net/publication/Importance_of_Site_for_the_Mitigation_of_Landslide_Hazard
usgs.gov/faqs/why-study-landslides?
core.ac.uk/download/pdf/234698578.pdf
dept.ru.ac.bd/geology/acad/landslide_hazard_1-35.pdf

Writer: Ozlem Guvenc Agaoglu


Leave a Reply

Your email address will not be published. Required fields are marked *