Enhancing the Performance of Nearest Neighbour Avalanche forecasting Model using Differential Evolution

Kushe Neha Ajit and Magar Ganesh M.

Page No. 1-6

Snow and meteorological data have been predominantly used for snow avalanche forecasting in nearest neighbour algorithm along with Euclidean distance as a distance metric. Method using cosine similarity with nearest neighbour algorithm was further enhanced by using the concept of weighting. The weighting factors were decided by considering the effect of each parameter on the avalanche occurrence situation.

However, deciding the weighting factor can be considered as an optimization problem with the aim to attain high precision in forecasting model. This research study explores the use of differential evolution algorithm for determining weighting factors for the parameters of interest in avalanche forecasting. Heidke skill score is used as an objective function to gauge the accuracy of the model while applying differential evolution algorithm. A rise of 70.03% in the value of Heidke skill Score is observed for the testing dataset after the application of weights obtained by applying differential evolution algorithm.

Full Text

Comparative study on air contaminants discharged from several industrial stacks operated from various forms of fuels

Karthick J. and Samuel Devadoss R.

Page No. 7-14

Chennai, a metropolitan city located at southern part of India, Chennai has a floating population of 1.3 million and has a higher level of vehicular movements. In addition to that Chennai city comprises of many industrial belts which include numerous small-scale industries to large scale industries. these industries emit predominant emissions and they have been discharged into the local environment. this activity highly alters the air quality in the ambient environment and also meteorological factors favour for the pollutants transportation from the point of discharge in to the other areas. point source of discharge with respect to industries are carried out through the stacks, hence stacks serves as a primary source for the pollutant discharge into the atmosphere. current study was carried out in Manali industrial area, Chennai.

The study involves the selection of industries from Manali industrial area and to investigate about the stack profiles, mode of fuel used and the stack emissions from the respective industry. in addition to that meteorological studies were also carried out to find the predominant wind direction and wind speed. three industries (various fuel based) were selected and the stack monitoring was carried out. from the results it is been observed that the coal-based fuel operated at the stacks from a thermal power plant, produces the higher level of particulate matters SOx and NOx when compared to the other forms of fuels like furnace oil and high-speed diesel.

Full Text

Assessment of Drought using SPI at Weekly and Monthly Time Scales in Junagadh (Gujarat-India) Region

Gundalia Manoj

Page No. 15-25

Drought is an insidious natural hazard. Its longevity and widespread spatial extent cause severe impacts on the animal life. It is difficult to define, quantify and monitor the drought because of its complex nature. Several drought indices based on one or more hydro-climatic parameters were reported and used by researchers of hydrology, meteorology and climatology fields. The SPI (Standardized Precipitation Index) is a flexible and powerful index. It is simple to calculate and it required only precipitation data as an input parameter.

In this study, the drought variability in Junagadh region is assessed by using SPI at weekly (SPI-W) and monthly (SPI-M) time scales. 35 years (1984-2018) of rainfall data are considered to compute these indices. Results show that SPI-M explicitly differentiated monsoon season into wet and dry season. The SPI-W indicates severe drought in post-monsoon season while it was characterized as moderate drought based on SPI-M. Severe drought events that occurred in the year 1987 are explicitly detected by SPI values at both time scales which indicate that SPI has capacity to detect an intensification of drought severity. Statistical analysis endorsed that SPI-W provides more reliable and consistent results than that of SPI-M in the region.

Full Text

Comparative Experiments on the Application of Reference Image according to Seasonal Conditions in Multilinear Regression Landsat Image Simulation

Paik Geun Woo, Park Jin Su, Seo Dae Kyo, Kim Hye Jin, Eo Yang Dam and Pyeon Mu Wook

Page No. 26-33

An optical image is subject to many weather conditions, hence obtaining sufficient reference images to simulate image production is difficult. This study analyzes the difference between the results of using four seasonal data for several years and those using only the data for three months within the target period in the production of simulated Landsat image.

The experimental results show that the simulated images in May and October, which include seasonal changes, have high accuracy when all of the four seasonal data are used, whereas the simulated image in August, which is seasonally stable, demonstrates high accuracy when only three-month data are used. Therefore, the application of only a small number of adjacent period images is proven to play a positive role in the simulation results for seasonal stability.

Full Text

Assessment of Flood Risk due to Urbanization in South and Southeast Asian Countries: A Panel Data Approach

Rakshit Satarupa and Irfan Zareena Begum

Page No. 34-41

The climate-related disasters are becoming increasingly common in urban areas exposing millions of people to risky statuses. Asia is home to several fast-growing cities which are witnessing tremendous disaster losses due to urban growth in vulnerable zones which lack proper infrastructure with high income disparity. There is either dearth of proper disaster policies, mismanagement despite policies on paper or absence of awareness in these nations. The objective of this study is to analyze the relation between flood losses due to increase in unchecked urbanization in selective South and Southeast Asian countries which have registered high urban growth rates in past few decades.

The study uses panel data approach to monitor the variables over the period of 1988-2017, three models to represent the total deaths, total affected and total damages caused by flooding incidents spanning across the time frame. The analysis suggests that flood deaths are higher in less developed countries while the economic direct damages are highest in the developed nations. The conclusions and insights from this study can become a reference point for urban centers to take up risk assessment for better targeting of flood risk. The study concludes with policies suggestions to minimize the losses even with the ongoing sprawl and how to pave way for sustainable growth.

Full Text

Disaster management cycles and their application to lifestyle’s all factors

Kyoo-Man Ha

Page No. 42-50

The factors that influence lifestyle can be classified into three factors namely natural disaster, manmade emergency and other factors. A number of disaster management cycles have been used in various regions of the world and have usually been used in dealing with both natural disasters and manmade emergencies. The goal of the study is to delve into how to apply those cycles to lifestyle’s all three factors. While utilizing qualitative content analysis as the main methodology, this study has cross-examined three major cycles (the U.S. four phases, South Africa’s seven phases and Korea’s three phases) with lifestyle three factors.

Namely, major stakeholders must change their application to lifestyle’s disaster (plus emergency) factors to application to lifestyle’s all factors. In doing so, they must enhance disaster awareness, all professionals’ commitment, disaster culture, multidisciplinary approach and education and training. This research has initially applied disaster management cycles to lifestyle’s various factors compared with previous studies.

Full Text

Drought Management Policy of India: An Overview

Sharma Ashish

Page No. 51-62

India aims to achieve the Sendai framework for Disaster Risk Reduction targets along with a few sustainable development goals of the United Nations. For this, the policies towards disaster management should be appropriate and effective at national as well as the sub-national (state) level. Among all-natural disasters, drought is a common and frequent type in India causing significant agricultural and other economic losses. Therefore, the present study overviews the drought management policy of India. The study specifically examines the drought phenomenon (concept), its management (monitoring and declaration) and the post-drought relief policy framework. It also compares the new (2016) drought management guidelines with the previous (2009) policy guidelines to assess its efficacy and usefulness in reducing drought-induced losses.

The overview findings suggest that drought definition, management and subsequently, the drought relief practices significantly vary at the sub-national level in India. Moreover, these policies are also ineffective in reducing the drought losses. The study also observed that the central government transferred its financial obligations to the States for mild and moderate drought incidences. As a result, the States may be financially more stressed in arranging and deploying the additional financial resources. The findings of the overview have a rationale for the governments, policy makers and the relevant stakeholders for strengthening the drought management policy framework at the sub-national level.

Full Text