Fuzzy evaluation of air and noise quality.
Planning and design activities of different level i.e. urban, town, rural, transportation etc. is the major channel to control the human activities that pollute urban environment and perform the managing measures to improve the environmental quality. Therefore, urban environmental quality evaluation (UEQE) is at the heart of urban planning and development. Its interpretation and forecast of the urban environmental quality (UEQ) according to the national regulation about the permitted content of contamination is for the sake of protecting human health and subsistence environment. Air pollution occurs in larger cities because of industrialization, urbanization, dense transportation networks, and high population density.
Apart from its severe local effects, urban pollution has deep impact on regional and global scale. To convert these more favorable conditions into actual improvements in air quality, however, there needs to evaluate the air pollution. Noise pollution also effects the urban environment and hereby included in urban environmental quality evaluation. The physical environment represents the external conditions under which human being live. It includes land use, greenness, population density, topography etc. These factors play a vital role in determining the extent of pollution at a particular spatial location. In this study, these factors are also included and have been assigned weights in contributing in UEQE by AHP. There is a kind of ambiguity or fuzziness due to lack of clear defined boundaries between these objects. This approach has gained considerable momentum in Urban Planning specifically in the field of urban remote sensing in classification and evaluation. Because of its ability to deal with uncertain information, it has a lot of potential applicability in urban environmental studies. Much environmental information has the obvious spatial character that can be addressed by geographical information system (GIS). The multiple layers of information of evaluation criterion can be integrated in different combinations in GIS. In AHP, inputs are the comparison matrix and the output is the weights of environmental factors of pollution. Here by AHP, weights are found out of all the indicators by pair wise comparison. The present study undertaken demonstrates the usefulness of remote sensing, GIS, FST & AHP in urban environmental quality evaluation.
Bhopal, the capital city of Madhya Pradesh has started being counted among the fastest growing cities in the country (Bhopal City Development Plan under Jawaharlal Nehru National Urban Renewal Mission). The city which is known as "the city of lakes" is continuously losing its grace and beauty under the growing pressure of up-gradation and densification of activities resulting in increase of service related problems.
Geographical co-ordinates of study area (41 wards) are 77 22 04.71 E--77 26 25.73 E and 23 11 46.59 N--23 17 38.72 N.
Bhopal city nestles in a hilly terrain, which slopes towards north and southeast (map # 1). Most of the study area falls under slope between 0-4 degrees. On relating aspect and wind direction, those areas come under topography whose aspect is NE have more dispersion of pollution. Bhopal is the second largest city in the state with a population of 14, 33,875 in years 2001. The city is distinctly divided into two parts, the old city housing most of the trading and commercial activities and the newly developed area with mainly administrative, institutional and residential activities. The road network in the old city area, with very limited scope of road widening, mainly suffers from very high volume of traffic, heterogeneous traffic mix, and high degree of pedestrian movement and on--street parking. The presence of Bhopal railway station and bus stand in the area adds more problems. One of the critical and most immediate problems faced by rapidly growing cities in developing countries is the impact of urban environmental pollution on health. There is need for major investments in environmental up gradation of the city by the way of urban environmental quality management.
Materials and Methodology
There are 66 wards (as per year 2001) in BMC. In this research, spatial data, socioeconomic statistic data and environmental data are required. Remote sensing data in the form of fused satellite imagery of CARTOSAT- 1 PAN (2.5 m resolution) dated January 2007 and IRS P-6 LISS IV MX (5.8 m resolution) dated January 2007 of Bhopal Municipal Corporation area are used. City guide map (MCB), Wind direction and wind speed from meteorological department, Ward map year 2005 from Bhopal Municipal Corporation, Bhopal Development master plan 2005 from Madhya Pradesh Directorate Town and Country Planning, Road map from Madhya Pradesh Road Development Corporation Ltd are used as reference data. The sample data includes the pollution value of air and noise which is annual mean concentration of year 2006.
The first step in evaluating UEQE is to identify the relevant environmental components (environment pollution, physical environment) and then use these components to establish the relevant evaluation criteria.
Map 2: Location of Sample Stations (for air & noise) in 41 Wards of the Study Area The factors considered are air, noise, greenery, land use, topography and population. Inverse Distance Weight (IDW) method is used on available sample data of air and noise pollution (in 41 wards) to obtain a continuous grid (map # 2). Criteria maps are prepared with final weights. While converting shape file to grid, the cell size is maintained to be 2.5m which is the spatial unit for analysis in this research work.
Fuzzy overlay operation is showed in figure #1. In this process, each factor of the evaluation is represented as a raster data layer. Every bottom indicator of each criterion is overlaid based on fuzzy operation, which is also called intermediate hypothesis. For example, in air pollution criteria, the criterion consists of four indicators (S[O.sub.2], N[O.sub.2], SPM and CO). The first phase evaluation is to overlay these four indicators based on fuzzy operation (Fuzzy Algebraic Sum). Finally, the final hypothesis performing the fuzzy overlay operation of environment pollution and physical environment component to get the final quality map (fuzzy GAMMA).
Data Base Creation
The maps so generated by interpolation are shown as Map#3. As per National ambient air monitoring program (NAAMP), the whole surface is classified into four classes i.e. Low (L), Moderate (M), High (H), Critical (C).
[FIGURE 1 OMITTED]
As per the interviews of twenty eight experts, consisting of scientists of M.P. State Pollution Control Board, Town Planners, Urban Planners and Environmental Planners in Bhopal, the comparison matrices has been prepared by using the AHP scale. Zimmerman (1991) discusses a variety of fuzzy operations. Fuzzy Algebraic Sum is applied to get the final weights of indicators. The final weight is added as attribute to the classified data set and layers with membership values are obtained (Lee S., 2006)
Different varieties of fuzzy operators can be used in the same problem (fig #1). Generally, (TSO and Mather, 2001) Fuzzy Algebraic Sum and Fuzzy OR are used. In this study, Fuzzy Gamma operator has also been used. Thus all the maps can be combined using a suitable operator like Gamma operator. It is to be noted that a single operator would not suffice the combination of maps since it depends on the context of the type of data represented by maps (Kwang,2002). The intermediate hypothesis maps namely air pollution, noise pollution, topography, greenness and population are combined using GAMMA operator (map#4). Gamma operator needs a value that is [gamma], which should be decided judiciously. [gamma] value shows the compromise between Fuzzy Algebraic Product & Fuzzy Algebraic Sum operator. [gamma] value is determined based on the trial and error procedure (Kwang, 2002). Several values of [gamma], were tested and maps were obtained. The a value for which the final map has shown the resemblance of the true picture has been taken. With the a value 0.6 which is the intermediate value between the "increasive" tendency of Fuzzy Algebraic Sum and the "decreasive" effects of the Fuzzy Algebraic Product, the final result is obtained.
[[micro].sub.combination] = exp [[gamma] * log (fuzzy algebraic sum, A ) + ( 1-[gamma]) * log (fuzzy algebraic product, B)]
Final Quality Map = exp [[gamma] * log (A) + (1- [gamma]) * log (B)]
A = 1-[(1-Air Pollution) * (1-Noise Pollution) * (1-Road Buffer) * (1-Topography)* (1-Greenness)*(1-Population)]
B = air pollution * noise pollution * road buffer * topography * greenness * population
The Final Quality Map is reclassified into four different potential categories namely Low, Moderate, High and Critical.
Results and Analysis
The types of "OPERATORS" that have been applied to evaluate the final criteria map using FUZZY LOGIC approach have already been discussed. FUZZY ALGEBRAIC SUM and FUZZY OR are used as intermediate hypothesis. The results so obtained have been classified under four categories of pollution. These classes are Low, Moderate, High and Critical.
Results obtained by Fuzzy Approach
From map (Map-5) using fuzzy approach, it is very clearly shown that the area along the major traffic corridors are in "cyan" color which mean high polluted. This is because that the fuzzy approach employs a series of scores to determine in what degree the criterion belongs to one evaluation class. In fuzzy approach, a continuous value has been assigned with the help of fuzzy weights and scores to the pixels of the generated layer.
The fuzzy approach can deal with the information of evaluation criterion more precisely and prevents the loss of the information (map#6). This is due to the fact that surrounding near by the roads is well vegetated (map # 7). As we know, trees and grass cover act as a medium of decreasing the pollution level. Even with the high vehicular emission, pollution is not in the high zone.
Validation of Map by Fuzzy Approach
To carry out the validation, there should be restriction in using all the sample data of pollution of year 2006 in forming the final quality map. Some point data are used as a reference for the validation of map. These data points remain unbiased since it has not been used in map generation. This ensures consistency of the accuracy assessment, (Congalton and Green, 1999). Since the amount of sample data is not very large so only six points have been removed from the sample data as reference data. This corresponds to the minimum 10% of the data set for validation. Further it ensures the availability of proper amount of sample data for interpolation.
For map validation, six sample data of location P, Q, R, S, T and U in ward are kept unused to get the final quality map. Classes of these points have been identified using expert views and field data and analysis. This procedure is not an exact indicator but with lesser data points high number of test samples would result in erroneous results. However, expert views and some qualitative analysis have been used to identify the classes of the areas which are easily assessable. These areas are identified with GPS coordinates (WGS-84) and located on the registered map. It has been tried to select areas randomly under the constraint of accessibility. Classes have been identified of these specific locations and compared with final quality map generated by fuzzy approach.
Conclusions and Recommendations Conclusions
The ability of fuzzy approach to quantify the ambiguity of complexity of urban environmental quality has been established. Fuzzy approach is able to map areas with all classes pretty well. However, the need of establishing the fuzzy weights at each intermediate stage is central to the map generation. A special attention should be given to identify these intermediate parameters. The study might be restricted to a particular location but technique can be applied for other cities. Obviously, with change in study area parameters are going to change but method of evaluation would be same. The combination of AHP and FST has given a platform to bring environmental as well as physical parameters into same environmental quality evaluation unit. Addition of more criteria like water, soil, solid waste etc are likely to produce a more comprehensive map. A more accurate and balanced quality map of urban environment may benefit the real estate businesses and developers. Future research could explore the possibility of cross-comparisons of aggregate quality of cities or regions within a nation or across countries.
It gives me immense pleasure to express my appreciation of the assistance rendered to me by all those who helped me in completing this project. I am also thankful to the authorities of different organizations namely Bhopal Municipal Corporation, Town and Country Planning, M.P. State Pollution Control Board, M.P. Road Development Corporation, French company Egis BCEOM Pvt. Ltd., Meteorology department of Bhopal, ESRI, EPCO, NRSA for providing necessary information and data.
 Bhopal City Development Plan, Jawaharlal Nehru National Urban Renewal Mission, Bhopal Municipal Corporation, Technical Support, 2006
 Bhopal Development Plan, 005 (Draft), Directorate of Town and Country Planning, Madhya Pradesh.
 Banai, R. (1993). "Fuzziness of geographical information system: contribution from the analysis hierarchy process." International Journal of Geographical Information Science 7(4): 315-329.
 Bonham-Carter, Graeme F., Geographic Information Systems for Geoscientists, Modelling with GIS, Chapter 9, Fuzzy Logic section.
 Central Pollution Control Board, Annual report 2004-2005
 Central Pollution Control Board, Annual report 2005-2006
 CPCB (1995) 'Air pollution and its control'. Parivesh Newsletter 2(1), June 1995. Delhi: Central Pollution Control Board. 20 pp.
 CPCB (2000c) Air Quality status and trends in India, National Ambient Air Quality Monitoring Series: NAAQMS/14/1999-2000. Delhi: Central Pollution Control Board.
 Chinag, C. & Lai C. (2002). A study on the comprehensive indicator of indoor environment assessment for occupants' health in Taiwan.", Building and Environment 37: 387-392.
 Cleveland Cutler J., Environmental effects of urban trees and vegetation, April 2007.
 Choudhry Alok (2002). Morpho- Lithological Analysis and Ground Water Resources Management through GIS Modelling and Digital Image Processing
 Dewaram A. Nagdeve (2002). Environment and Health in India, IUSSP Regional Population Conference on 'Southeast Asia's Population in a Changing Asian Context at Bangkok, Thailand, 10-13.
 Dou Koili (2003). Fuzzy Environmental Quality Evaluation, Wuhang
 Da-Yong Chang, Applications of the extent analysis method on fuzzy AHP, European Journal of Operational Research 95 (1996) 649-655.
 Edwards Ward, How to use multi attribute utility measurement for social decision making, IEEE, Vol. SMC-7, No. 5, May 1977.
 Environmental Status Report of Bhopal City, year 2002-2003, source: Central Pollution Control Board.
 Environmental Systems Research Institute (ESRI): 2002, Using Arc GIS Spatial Analyst, Redlands, CA, Environmental Systems Research Institute, Inc.
 Finan J. S. & Hurleyt, W. J. (1997). The Analytic Hierarchy Process: Does Adjusting a pair wise comparison matrix to improve the consistency ratio helps?, Computers Ops Res. Vol. 24, No. 8, pp. 749--755, @ 1997 Elsevier Science Ltd.
 Green Kass, Congalton, Assessing the Accuracy of remotely Sensed Data: Principal and Practices, edition 1999.
 Gregory McPherson, David J. Nowak, Rowan A. Rowntree (2003). Chicago's Urban Forest Ecosystem: Results of the Chicago, Urban Forest Climate Project, United States Department of Agriculture Forest Service, Northeastern Forest Experiment Station, General Technical Report N E- 186, p 201.
 James J. Buckley, Thomas Feuring, Yoichi Hayashi (2003). Theory and Methodology Fuzzy hierarchical analysis revisited, European Journal of Operational Research 129, pg 48 [+ or -] 64
 Jagannath C. (2003). Development of Fuzzy Logic Architecture to access the sustainability of forest management, ITC
 Jensen, John R., (1986), Introductory Digital Image Processing
 Jiang, B. (1996). Fuzzy overlay analysis and visualization in Geographical Information System. International Institute for Geo-Information Science and Earth Observation. Enschede: 167.
 John M., Martin Hale (2001). Geologically constrained fuzzy mapping of gold mineralization potential, Natural Resource Research, vol. 10, No. 2.
 Kahraman Cengiz, Ufuk Cebeci & Ziya Ulukan (2003). Multi-criteria supplier selection using fuzzy AHP, Logistics Information Management Volume 16 . Number 6 . pg. 382-394.
 Kwang Hoon Chi, No-Wook Park (2002). Fuzzy Logic Integration for Land slide Hazard Mapping using spatial data, Symposium on Geospatial Theory, processing and applications, Ottawa.
 Leung L. C. & Cao D. (2000). Theory and Methodology on consistency and ranking of alternatives in fuzzy AHP, European Journal of Operational Research 124, pg. 102 [+ or -] 113
 Lee (2007). Application and verification of fuzzy algebraic operators to landslide susceptibility mapping, Environ Geol 52:615-623
 Mrinal K. Ghose, Paul R. & Banerjee S. K. (2004). Assessment of the impacts of vehicular emissions on urban air quality and its management in Indian context: the case of Kolkata (Calcutta), Environmental Science & Policy 7, pg. 345-351
 Mohanraj R. & Azeez A. P. (2004). Health effects of airborne particulate matter and the Indian scenario, Current Science, VOL. 87, NO. 6, 25 Sep.
 Martin Raubal and Claus Rinner, Multi criteria decision analysis for location based services, Geoinformatics 2004 Proc. 12th Int. Conf. on Geoinformatics Geospatial Information Research: Bridging the Pacific and Atlantic University of Gavle, Sweden, 7-9 June 2004
 Netzband M., Redman C.L. & Stefanov W.L., Challenges for applied remote sensing in the urban environment. In C. Jurgens and T. Rashed (Eds), Remote Sensing of Urban and Suburban Areas, Dordrecht, Netherlands: Kluwer Academic Publishers.
 Peter A. Burrough, and Rachael A. McDonnell (1998). Principles of Geographical Information Systems, Oxford University Press Inc., New York, 1998.
Anshu Gupta (a),*, Vivek Dey (b) and Dr Adity Goel (c)
(a) Lecturer, Centre for Remote Sensing & GIS, NIT, Bhopal, India
(b) Student, Civil Engineering Department, Indian Institute of Technology, Kanpur, India
(c) Asst Prof, Department of Electronics, MANIT, Bhopal, India
Table 2: Classification of Six Reference Point Data for Validation Point Ward Pollution Final quality map Final quality map No Level of Ward by fizzy approach by Boolean approach P 42 High Critical Critical Q 45 Moderate Moderate Low R 25 Low Low Low S 35 Low Low Moderate T 51 Low Moderate Critical U 52 Low Low Moderate
|Printer friendly Cite/link Email Feedback|
|Author:||Gupta, Anshu; Dey, Vivek; Goel, Adity|
|Publication:||International Journal of Applied Environmental Sciences|
|Date:||Sep 1, 2008|
|Previous Article:||Thermodynamic analysis of solar water heating system of swimming pools.|
|Next Article:||Floristic composition and phenology of a submontane grazingland as affected by scraping and heavy grazing.|