Spatial Analysis of Natural Hazards Effects in Rural Areas Using Geographically Weighted Principal Component Analysis (GWPCA) (Case study: Alamut in Qazvin)

Document Type : Research Article

Authors

• Hassanali Faraji Sabokbar ¹
• Seyed Ali Badri ¹
• Reza Abbasi Verki ¹
• Abbasi Verki ²

¹ University of Tehran

² University of Shahid Beheshti

10.22067/geo.v3i2.27223

Abstract

1. Introduction
Natural hazards have existed in all periods of human life, but nowadays due to the exponential growth of human and population density in all aspects of life, especially in high risk areas, human has been faced with major disasters such as the Asia tsunami, Hurricane Katrina and the earthquake in Sichuan of China with significant casualties even in developed countries.
These hazards in many cases have caused severe damages on human community in both urban and rural societies and their effects are perceptible in environmental, social, economic and psychological aspects in human settlements for several years. A basic concept in natural hazards analysis on human community has been hidden in pathology of economics and mainly depends on diversity of economic and macroeconomic performance before the occurrence of a natural disaster. For instance, Caribbean is small country that its vulnerable economics mainly depends on tourism, agricultural exports and sales of products.
2. Study Area
Alamut region is adopted for the case study. The mountains native ,the valley location of the Alamut region and poor access of this area to road cause to consider it as a human rural area with 67 settlements (Claye Moalem city and 66 villages of up and down Alamut roud).
Ecological position of Alamut area (in both natural and human), on the one hand, depicts many factors and potentials for the occurrence of natural hazards. On the other hand it indicates the weakness of existing structures, particularly in human ecological aspect when the hazards happen. The appropriate reasons, (for example, the numbers of active faults, weak tectonic bed, existence of tectonic failures, and wide belt of thrust sedimentary sand stones, record of seismicity and also placement of 60 villages in expose of flood and earthquake risk, distribution of village and low levels of literacy in rural areas), take the attentions toward the natural hazards and identifying common risks as the first step of the natural hazards management.
3. Material and Methods
This methods can be used to transform data, such that the transformed data have similar characteristics (e.g., mean or variance) at all locations, are discussed. In other words, the concern is with nonstationary models and with methods that can be used to transform, or otherwise modify, data so that a stationary model can be applied to the transformed data.
A widely-used approach to accounting for spatial variation is a geographical weighting scheme. A distance matrix can be used to assign geographical weights in any standard operation:



Where si is the i th location s, with coordinates x-y, and win is the weight with respect to locations i and n.

3.1. Principal components analysis

Principal components analysis (PCA) is widely used in many contexts for reducing the dimensionality of multivariate data. PCA is based on the variance–covariance matrix or the correlation matrix and, in the present analysis, the latter is used. The variances of the log-ratios differ markedly because the ranges of values on which they are based vary.

When large multivariate datasets are analyzed, it is often desirable to reduce their dimensionality. Principal component analysis is one technique for doing this. It replaces the p original variables by a smaller number, q, of derived variables, the principal components, which are linear combinations of the original variables. Often, it is possible to retain most of the variability in the original variables with q very much smaller than p. Despite its apparent simplicity, principal component analysis has a number of subtleties, and it has many uses and extensions. A number of choices associated with the technique are briefly discussed, namely, covariance or correlation, how many components, and different normalization constraints, as well as confusion with factor analysis. Various uses and extensions are outlined.

3.2. Geographically weighted principal components analysis

Fotheringham et al. (2002) introduce the idea of geographically weighted principal components analysis (GWPCA). With GWPCA, geographically weighted means and GW variances and covariances around the means are obtained (Fotheringham et al., 2002), with the result that there is a set of GW means, variances and covariances for each of the n data locations. Once the geographically weighted variance– covariance matrix is obtained, conducting GWPCA is straightforward.
In this study statical society includes 2080 and 2589 persons in down of and up Alamut area separately that all were selected in 15-64 age group in 1390.
Kokaram sampling with 0.95 confidence factor and accuracy probability (P, Q=0.5) in 610 samples was calculated. Spatial analysis of natural hazards based on the initial findings in seven natural risks such as earthquake, floods, falling, snow break, frostbite and blizzard with risk indicator method such as 11 identified indicator interns of conceptual is performed.
4. Results and Discussion
The output of the GWPCA model illustrates dominance and spread of the seven natural risks such as earthquake, flood, landslide, landfall, Avalanche, frostbite and blizzard using data co-variance matrix in each village with at least 30 neighborhood points.
The index offear among residents indicates the spread of 56% flood risk. The index of disruption of rural transportation shows the spread of 57% landslide risk, which is the most important disruption of rural transportation. Finally, the index of prevalence of disease suggests the spread of 67% frost risk, which impressed the Valley Villages and villages in southern high lands of the Alamut.
5. Conclusion
The results demonstrate that flood and frostbite risks within the seven studied risk has the greatest impact on rural areas at Alamut valley. From the villagers view point, flood risk have greats fear, tend to immigration, destroy the infrastructure of the village and the risk of frostbite has the most villagers despair in the agricultural, interruption of rural economic activities and out breaks of disease among the villagers. Earthquake and landslide risk because the largest forced migration and the most disruption of transportation in villagers separately.

Keywords

• natural hazards
• Geographic weights matrix
• geographically weighted principal component analysis
• Alamut

---

References

پورطاهری، مهدی، سجاسی، حمدالله و صادقلو، طاهره (1390)، ارزیابی تطبیقی روش‌های رتبه بندی مخاطرات طبیعی در مناطق روستایی(مطالعه موردی، استان زنجان)،پژوهش های روستایی،سال دوم، شماره سوم، پائیز 1390،صص 31-54

رکن الدین افتخاری، عبدالرضا، صادقلو، طاهره، احمدآبادی، علی و سجاسی، حمدالله (1388)،ارزیابی پهنه بندی روستاهای در معرض خطر سیلاب با استفاده از مدلHAC_GeoRAS در محیط GIS مطالعه موردی: روستاهای حوزه گرگانرود، مجله توسعه روستایی،دوره اول، شماره 1، پائیز و زمستان1388.

عسکری، علی(1388)، راهنمای استاندارد بین المللی ایزو 31000 مدیریت ریسک: انتشارات نور علم. تهران.

Askari, A., 2009. International standard guide of Risk Manaagment, ISO 31000, Noor Elm publication, Tehran.

Anderson,M.,1990: Analyzing the Costs and Benefits of Natural Disaster Responses in the Context of Development. Environment Working Paper 29. Washington, DC, United States: World Bank

Benson, C., Clay, E.J., 1998: The Impact of Drought on Sub-Saharan Economies: A Preliminary Examination. World Bank Technical Paper No.401. Washington, United States: World Bank.

Charveriat, Celine, 2000: Natural Disasters in Latin America and the Caribbean:An Overview of Risk. Inter-American Development Bank (BID), New York Avenue, N.W.Washington. Latin American Economic Policies Newsletter.

Eftekhari R A,Sadeghloo T.,Ahmadabadi A.,Sojasi H.,(2010) Zoning of Rural Regions in Flood Hazard with Use of HEC-GeoRAS Model in GIS Sphere (Case study: flooded villages of Gorganrud Basin)

Fotheringham, A. S., Brunsdon, C., & Charlton, M., 2002: Geographically weighted regression: The analysis of spatially varying relationships. Chichester: John Wiley&Sons.

Getis and J. Aldstadt, 2004. Constructing the spatial weights matrix usinga local statistic. Geographical Analysis, 36:90-104.

Graham, D. T. ,1995. Social and demographic patterns and trends in Northern Ireland: a multivariate approach. Irish Geography, 28, 35–47

Hilson. D., 2002. Extenging the risk process to mamage opportunities. International Journal of Project management, 20, 251-268

Inter-American Development Bank. 2000: Social Protection for Equity and Growth. Washington, DC, United States: Inter-American Development Bank

Jolliffe, I., 2005: Principal Component Analysis. Encyclopedia of Statistics in Behavioral Science, John Wiley & Sons, Ltd

Lloyd, C. D., 2010. Analysing population characteristics using geographically weighted principal components analysis: A case study of Northern Ireland in 2001. Computers, Environment and Urban Systems 34 , 389–399

Lloyd, C. D., 2011: Local models for spatial analysis. Boca Raton: CRC Press

Mather, P. M., 2004: Computer processing of remotely-sensed images: An introduction. Chichester: Wiley.

Oliver J., 1989: A survey of public interpretation and opinions in Queensland on the presenttropical cyclone warning system, Part 1: Case study of tropical cyclone Winifred and Part 2: Case study of tropical cyclone Charlie.In: Disaster Management Studies Centre, Cumberland College of Health Sciences, East Lidcombe, Sydney, New SouthWales

Pichon, F. and Uquillas, J., 1997. Agricultural Intensification and Poverty Reduction in Latin America’s Risk-Prone Areas: Opportunities and Challenges. Journal of Developing Areas. 31, 479-514.

Pourtaheri M. ,SojasiQidari H , Sadeghloo T (2012).Comparative Assessment of Ranking Methods for Natural Disasters in Rural Regions (Case Study: Zanjan Province), Journal of Rural Research, 2(3), 31-54

Scott, C.D., Litchfield, J.A.,1994: Inequality, Mobility and the Determinants of Income among the Rural Poor in Chile, 1968-1986. London, United Kingdom: London School of Economics

Sivakumar , V.K. , 2005 : Impacts of Natural Disasters in Agriculture,Rangeland and Forestry: an Overview.In: Natural Disasters and Extreme Events in Agriculture,Springer ,1-22.

Smith Keith , 2001:Environmental hazards, third edition by Rutledge England.

Spielman, S. E., Thill, J.-C., 2008. Social area analysis, data mining, and GIS. Computers, Environment and Urban Systems, 32, 110–122

Tobias, M., Bhattacharya, A., White, P., 2008. Cross classification of the New Zealand population by ethnicity and deprivation: Trends from 1996 to 2006. Australian and New Zealand Journal of Public Health, 32, 431–436

Wheeler, D., Tiefelsdorf, M., 2005. Multicollinearity and correlation among local regression coefficients in geographically weighted regression. Journal of Geographical Systems, 7, 161–187.

Wold, S., K. Esbensen, et al., 1987 . "Principal component analysis." Chemometrics and Intelligent Laboratory Systems 2(1–3): 37-5