Analysis of Landslide Movements Using Interferometric Synthetic Aperture Radar: A Case Study in Hunza-Nagar Valley, Pakistan

Mohib Ur Rehman,Zainab Khalid,Filippo Catani,Sajjad Ahmad,Xiaojun Su,Ijaz Ahmad,Dongxia Yue,Gohar Rehman,Yi Zhang,Xingmin Meng

doi:10.3390/rs12122054

Abstract

From a geological standpoint, northern Pakistan is one of the most active and unstable areas in the world. As a consequence, many massive landslides have occurred in the area in historical times that have destroyed infrastructure, blocked the Hunza River, and damaged the Karakoram Highway repeatedly. However, despite the high frequency of large magnitude landslide events, and the consequent damages, the entire area is largely understudied, mainly due to the difficult logistics and the large distances involved. This work is aimed at applying the potential use of Interferometric Synthetic Aperture Radar (InSAR) for landslide identification and investigation for the Hunza-Nagar Region. Sentinel-1 images covering a period of more than two years (February 2017–August 2019) were used and processed by adopting the small baseline subset (SBAS) method. The obtained deformation rate measured along the line of sight (VLOS) varies from −114 to 20 mm/year. The downslope velocity deformation rates (Vslope) range from 0 to −300 mm/year. The Vslope stability threshold for our study area was calculated to be −14 mm/year from the Vslope standard deviation. Four active landslides with Vslope exceeding 14 mm/year were recognizable and have been confirmed by field inspection. The identified landslides listed from the most active to least active are the Humarri, Mayoon, Khai, and Ghulmet landslides, respectively. VLOS exceeding 114 mm/year was observed in the Humarri landslide, which posed a threat of damming a lake on the Hispar River and was also a risk to the Humarri Village located below the landslide. The maximum mean deformation detected in the Ghulmet, and Mayoon landslide was in the order of 30 mm/year and 20 mm/year, respectively. More importantly, it was found that in places, the slope deformation time series showed a patchy correlation with precipitation and seismic events in the area. This may indicate a complex, and possibly uncoupled, relationship between the two controlling agents promoting the deformation. However, the collective impact of the two factors is evident in the form of a continuously descending deformation curve and clearly indicates the ground distortion. The results indicate a potentially critical situation related to the high deformation rates measured at the Humarri landslide. On this specific slope, conditions leading to a possible catastrophic failure cannot be ruled out and should be a priority for the application of mitigation measures.

Highlights

Landslides are the most common and frequent hazards in the mountainous regions of the world which, directly or indirectly, affect the lives of the native dwellers [1]
The landslides identified through the Interferometric Synthetic Aperture Radar (InSAR) technique, and the displacement velocities measured, remarkably correspond to the actual ground situation surveyed on the field, where scars, cracks on the ground and on buildings, and infrastructures were clearly observed
Comparing ground conditions at the studied sites to the known recent events of landslides in the broader North Pakistan region, prompts the conclusion that these unstable slopes may progress towards potential catastrophic failures in the near future

Summary

Introduction

Landslides are the most common and frequent hazards in the mountainous regions of the world which, directly or indirectly, affect the lives of the native dwellers [1]. In the period between 1998 and 2017, landslides, and their associated hazards, affected an estimated 4.8 million people and caused more than 18,000 deaths globally [2]. The northern rugged mountainous regions of Pakistan are prone to landslide hazards [3]. The Hunza-Nagar region in northern Pakistan is well known for its unique geologic and geomorphologic features. The Hunza-Nagar valley is characterized by deeply incised valleys and rugged mountains, with elevations ranging from 1746 m to 7315 m above sea level (Figure 2d). Northern Pakistan is located in a region which is seismically very active and has experienced many disastrous earthquakes in the past. The authors of [8] have genehraaztaerddamlaonnditsolriidneg iinnvtehnetovrayllemya[p14f]o.rAtsheCHPEuCnzwai-lNl baegathrevamllaejyorbdyiruescitntgratdhee rvoiusuteaflrionmterCphrientaattioon of the SEPuOroTp-e5, tshaeteMlliidtedilemEaagset,raynadnAdfrmicaap, tpheedpraotboatballeobflo1c7k2algaenidnstlriaddeesd. uNeutomlearnoduslsidlaenhdaszlairddewscilalrhsacvaen be seengolonbtahl eefvfeacltles.y walls along with mass movement deposits commonly found on the valley floors and the lionwtheTeropreeagnritsouonrf,ettthhheeevsdaaeflteleetcyytioowfnat,hlalesnc(doFmimgmuonureintoi1tr)iie.nsTg, hthoeefoisnludvcehlsatndmdeevsnalitsdstaeintidntehgpeeoafsrfieetacst,saannoddf tlaahsnesdocsoclniiadttienesduiisltayeyosesfretsnraotdifaelvery fine-[g13ra,1i5n]e. dHloawteevseedr,imtoednetteocftvaanrdiambloentihtoicrktnheesssloinpeanmootvheemrwenitsseosftesuepchgaracdoimenptlecxhaarnenaewl pitohinatltaorgweards the fnruemqubeenrtodf aslmopmesinisgaocfhtahleleHnguen[z1a6]R. iver valley in the past [9]

Objectives

Results

Conclusion