Infrastructure Vulnerability Assessment of Mumbai City to Natural Hazards

Mishra Praveen Kumar , Prasad Rama Shankar and Karmakar Subhankar *

An exhaustive knowledge of risk, hazard, vulnerability and exposure in different spatial locations is essential for developing an effective mitigation strategy to natural hazards. Vulnerability refers to the susceptibility of a region to damage in natural hazard. By identifying the areas of high risk and vulnerability, it is possible to make more informed decisions in proper management of a watershed during natural hazards. The present study has introduced an infrastructure component of vulnerability considering two types of infrastructure vulnerability: (1) general infrastructure and (2) critical infrastructure facilities. The infrastructure vulnerability is mathematically represented by a method of standardization, which provides a value between zero and one. The overall infrastructure vulnerability for different spatial region is determined by ‘averaging’ all indices of infrastructure vulnerability. Geographic Information System (GIS) has been used as a major tool for calculation of attributes and to map the infrastructure vulnerability. The information may be utilized in disaster preparedness and prevention measures. The developed methodology is demonstrated on 22 wards of Mumbai city and the areas sensitive to damage of infrastructures in disaster are identified.

Full Text

Forest Fire Detection and Monitoring using High Temporal MODIS and NOAA AVHRR Satellite Images in Peninsular Malaysia

Pradhan Biswajeet 1* and Assilzadeh Hamid 2

Forest fires cause significant economic damages and hazard to environment all over the world. Apart from preventive measures, early warning and fast extinction of fires are the only chance to avoid major casualties and damage to nature. This paper describes methodology based on remote sensing and GIS for provision of various early warning of forest fire (so called hot spots) danger conditions for regulatory authorities to take actions for mitigation. Hot spot locations were identified through an automated procedure from high temporal satellite images such as MODIS and NOAA AVHRR scenes. Combination of the Daily Hotspot Images coupled with various GIS layers was used to generate Active Forest Fire Map for the study area. Results from the model can support detection and monitoring for wild fires in the forest and enhance alert system function by simulating and visualizing forest fire and helps for contingency planning.

Full Text

Generation of Windstorm in the Eastern Mountainous Coast of Korea

Choi Hyo1* and Choi Soo Min2

Using a three-dimensional non-hydrostatic numerical model, WRF version 2.2, the evolution of windstorm was investigated near Mt. Taegulyang (alt. 896 m) located in 20 km west of a coastal city, Gangneung, Korea from October 27 through October 28, 2003. On October 27, before cold front passing across the Korean peninsula, no windstorm was detected in the study area and moderate southwesterly wind of 4 to 6 m/s prevailed in the study area. On the other hand, at 09 LST, October 28, just after the cold front passed by the coastal city, positive relative vorticity at 500 hPa level induced downward motion of cold air from the upper level toward the ground surface with a decreasing rate of air temperature of - 6 0C/day at 850 hPa level and simultaneously, negative geopotential tendency of - 160 m/day lay on the study area caused the atmospheric depth of 500 hPa level to be much shrunken vertically. These kinds of downward motion of cold air and compression of atmospheric depth resulted in merging of streamlines with higher speed more than 25 kts of isotach at 850 hPa level and produced the formation of strong surface wind speeds. As strong synoptic westerly wind blowing over the mountain toward the downwind side coastal city was associated with mountain-land breeze by both nocturnal cooling of the ground surface and steep mountain terrain, it could be an intensified strong downslope wind like Katabatic wind. Furthermore, as soon as synoptic westerly wind passed over the steep mountain barrier, a strong intensified downslope westerly wind depicted a cyclonic flow pattern to cause cyclogenesis in the downwind side and the wind speed increased, resulting in a good condition for the formation of a wind storm more than 9 m/s on the mountain and coastal surfaces. When this strong wind passed along the eastern slope of the mountain and by the coastal basin, it could be more enforced underneath much shallower nocturnal surface inversion layer than daytime convective boundary layer, showing more increase of its speed. As the centers of maximum positive vorticity and maximum negative geopotential tendency approached the coastal near noon on October 28, surface wind under the compressed atmospheric depth should be more intensified. Even though daytime convective boundary layer was developed with more increased depth than nocturnal surface inversion layer, the shrunken convective boundary layer under the strong downward motion of cold air behind cold front across the study area could still cause the increase of wind speed, resulting in daytime windstorm near noon in the study area. The windstorm in the lee side of the mountain displayed the propagation of internal gravity waves with a hydraulic jump motion bounding up and down over the coastal basin and sea such as blocking.

Full Text

An Approach to Prediction of Wind Load Distribution on Large-span Roofs Using ANN Method

Zhou Xuanyi

The wind pressure coefficient and power spectrum of wind pressure play important roles in computing wind-induced dynamic responses of largespan roofs. In order to obtain more detailed information of wind pressures on roofs, the neural network method is adopted to predict the mean wind pressure coefficients and the power spectra of the fluctuating wind pressures on a real large-span roof based on the limited data from the wind tunnel test. The mean wind pressure coefficients from the wind tunnel test are used as the training data directly. Due to higher level of complexity compared with pressure statistics, the power spectra of wind pressures are first transferred to the fitted curves and then utilized as input data for the artificial neural network. It can be found that the predicted results are in good agreement with the experimental data. This method seems a practical and accurate one to predict the wind pressures on large-span roofs.

Full Text

Cross-Validation of Logistic Regression Model for Landslide Susceptibility Mapping at Ganeoung Areas, Korea

Oh Hyun-Joo and Lee Saro *

The aim of this study was to apply and validate a spatial probabilistic model for landslide susceptibility analysis at Ganeoung areas in Korea, using a Geographic Information System (GIS). Landslide locations within the study areas were identified by interpreting satellite images, aerial photographs and field surveys. Maps of the topography, soil type, forest cover, geology, lineaments and land cover were constructed from the spatial data sets. The 16 factors that influence landslide occurrence were extracted from the database and the logistic regression coefficient of each factor was computed. Landslide susceptibility maps were drawn for these two areas using logistic regression coefficient. For validation, the results of the analyses were compared, in each study area, with actual landslide locations. As the result, the case of Sagimakri showed 87.07% and the case of Samgyori showed 92.18% prediction accuracy. The validation results showed satisfactory agreement between the susceptibility map and the existing data on landslide location. For cross-validation, the result of the landslide susceptibility analysis obtained from Sagimakri area was applied to Samgyori area and reverse. The cross-validation results showed 73.85% and 80.66% prediction accuracy between the susceptibility map and the existing landslide locations.

Full Text

Application of the Object-Oriented Technique to monitor Coastline Changes- Case Study: Caspian Sea

Mohd Din M.A.1* and Rasouli A.A.2

Coastal zone monitoring is an important task in national development and environmental protection, in which, drawing out of shorelines is necessary as a fundamental research. This technique of measurement helps in the accuracy of monitoring the coastline changes. In this case, Caspian sea is taken as the case study due to its unique characteristic. Its dynamic coastlines features pose considerable hazards to human use and development. Rapid reliable techniques are required to monitor and update coastline maps of these areas to explore rates of environmental retreats. Similar method can be applied to measure other coastal zone. In the current study, various semi-automated methods like NDWI, NDSI and Tasseled Cap have been applied accordingly and the results were integrated with some object-oriented classification methods. Landsat MSS, TM and ETM imageries of the past three decades were consequently processed by an object-oriented approach performed with an eCognition software package. By comparing three classified maps of the south Caspian Sea coasts (Babolsar Port to Feridonkenar) in 1977, 1984 and 2002 with a unique region growing image segmentation technique (multi-resolution segmentation), areas of rapid change were progressively identified. The revealed models demonstrate several yearly fluctuations and considerable periodical changes on the study area coastlines particularly during the last decade observed by TOPEX/Jason satellites. These great variations have occurred as the result of 2.6 meters increase in seawater height from 1984 to 1995. This has successively caused coastal lands to diminish about 185 km2 mainly on the Babolsar Port, changing landcover and landuse types by depletion of significant agricultural and residential areas. Implementations of such significant changes signify that the majority of local biotic and a biotic components, all over the surrounding areas, would be in crucial threat in the near future.

Full Text