Home Articles Improvement of RS and GIS Techniques Involving Malaria High Risk Regions Determination.

Improvement of RS and GIS Techniques Involving Malaria High Risk Regions Determination.

Abolfazl Ahmadian Marj
MSc Student of Remote Sensing
Faculty of Geodesy & Geomatics Engineering
K. N. Toosi University of Technology (KNTU)
Mail id: [email protected]

Mohammad Reza Mobasheri
Assistant Professor
Faculty of Geodesy & Geomatics Engineering
K. N. Toosi University of Technology (KNTU)

Mohammad javad Valadan Zoej
Associate Professor
Faculty of Geodesy & Geomatics Engineering
K. N. Toosi University of Technology (KNTU)

Yousef Rezaei
Phd Student of Remote Sensing
Faculty of Geodesy & Geomatics Engineering
K. N. Toosi University of Technology (KNTU)

Abstract:
Malaria has been found in the vast areas in different regions of the world. Particularly many people in the tropical and subtropical regions suffer from this disease. 40 percent of the earth's population lives in zones where malaria exists. In Iran, Malaria is one of the main public health concerns mostly in south and southeast regions of the country. Malaria outbreak is profoundly correlates with the environmental and climatic conditions of a region. Due to the vastness of the potential area, Remote Sensing is a useful tool for detection of the conditions appropriate for malaria outbreaks and consequently helping managing it. This could be done through estimation of environmental information and climate parameters using satellite imageries. Thus, it can be used for organizing a controlling system for malaria outbreaks.

In this study, a methodology is suggested in which at the first step, based on the biology of the insect, the minimum requirements of the environmental and climatological parameters for the incidence of this phenomenon will be determined. This study showed that some parameters such as air temperature, relative humidity, vegetation cover and lagoons and basins are the most influential parameters in creation of potential for of epidemy outbreaks.

In the modeling section, different methods in extraction of environmental parameter were thoroughly studied. Comparison of different methods leaded to identification of the most appropriate strategy for each parameter extraction using Landsat images. Then, high risk regions were located for each parameter.

In the next step, the selected regions were imported in to a GIS (Geographical Information System) environment as independent layers. Weighted overlay method was implemented and finally, high risk regions were determined. Also, for model evaluation, some ground truth data has been collected and the work has so far shown good applicability.

Keywords: Hygiene, Malaria Outbreak, Remote sensing, Geographical Information System.

1.Introduction
Malaria disease could be found in the evasive regions of the world. Furthermore, many people of the various parts of the world live in the high risk regions (Fig 1). Every year, numerous amounts of people are being infected and some even die due to this disease. This disease can be spread out by variety of Anopheles insects each in an appropriate natural condition. Despite of many researches in malaria, this disease is still one of the main threats for global health and still there are many unsolved problems in this regards. Nowadays, it has been understood that the most important way to fight against this disease is controlling it by prediction of its outbreaks. In this research we tried to find an answer to this fundamental question by deployment of remote sensing technology as well as auxiliary environmental data (weather parameters).

Malaria was endemic in most parts of Iran around 100 years ago based on the periodical reports from Iran to WHO/EMRO1. As the result of extensive malaria control programs in the last 5 decades, the malaria incidence rate has dropped dramatically. However, malaria is still one of the most common parasitic diseases in Iran and one of the main public health concerns in the southeast of the country that is Sistan and Baloochestan, Hormozgan and southern parts of Kerman provinces [1].

In this research it is tried to review the application of RS and GIS techniques in identification of the regions with the potential of malaria outbreak. The RS data can help in identifying the relation between the environmental condition (climate) and the outbreak of malaria. Based on this knowledge one can buildup a controlling system that is able to forecast the probability of the outbreak of the disease through determination of suitable environmental condition. Images of 7ETM+ sensors are used for the southeast regions of Iran including the cities of Kahnooj and Minab (Fig 2). Also some ground truth in the year 2003 would be used for model evaluation.

Fig 1.The Distribution of Malaria In The World (WHO/TDR 2003) [7]

Fig 2. The Study Area In Iran

2. Effects of Environmental Parameters on Malaria

Climatic conditions directly influence mosquito and parasite development and the duration of the incidence of the disease [2]. Parameters like air temperature, relative humidity, vegetation cover and basin existence are the most influential parameters on epidemy outbreaks.

2.1. Air Temperature
Temperature is important because it governs the rate at which mosquitoes develop into adults, how frequently they need blood feed (and, therefore, acquire parasites) and the incubation time of parasites in the mosquito [3]. Also Temperature has an effect on the survival rate of adult mosquitoes. Considering this, the suitable temperature limit for incidence of Malaria is between 20-35 degrees centigrade[3].

2.2. Relative Humidity
Humidity is one of the factors that have a direct effect on the survival of the mosquitoes [4]. In other words, suitable humidity is one of the major factors for developing anopheles. Different species need different degrees of humidity. If the average relative humidity per month is below 55% and above 80%, the life duration of mosquitoes will be decreased and thus the amount of malaria incidence reduces. The amount of optimized humidity is between 60-65%.

2.3. Vegetation Cover
Vegetation cover has an important but indirect role on the abundance of malaria [5]. Various vegetation covers and the density and species regarding the kind of anopheles can be a good resting place for transfer of the disease. It almost can be said that all vegetation cover are suitable for anopheles.

2.4. Basins existence

Breeding and early prevalence of anopheles as larva is done in water basins and lagoons [6]. Since the fly range of mosquitoes is limited and breeding should be done in water, then the abundance of mosquitoes can be found around the places where there are patches of stilled waters. The dams built by humans, watering plans and developing agricultural projects can produce patches of stilled water and as a result changing ecosystem which in turn can cause the increase in abundance of mosquitoes.

3. Ways of acquiring parameters

In this section, the possible ways of the estimation of different parameters was studied comprehensively. A comparison of different methods leads us to identify the most appropriate strategies for each section. Timeline comparison for Existing process Vs Automated process

3.1. Air Temperature

To recognize the high risk regions in the terms of air temperature, the SEBAL method was chosen [8]. In this method, surface temperature was calculated in the first step and after that, air temperature was derived from the surface temperature. The following formula is used for computing surface temperature.

3.2. Relative Humidity

In the next step and regards to obtain the high risk regions due to relative humidity, a combination of air temperature image and meteorological parameters was utilized. The following formula is used for acquiring relative humidity [9]:

Where e and s e are water vapor pressure and saturated water vapor pressure respectively. The water vapor pressure values are provided by synoptic weather stations in study area. Also, the saturated water vapor pressure can be calculated through traditional equations using air temperature image which calculated in the previous step.

3.3. Vegetation Cover and Water Basins Detection

In the next step and regards to determine the vegetation cover and localize the water basins, NDVI index was used. The structure of this index is as follow [10]:

Where RED is reflectance in red band and NIR is reflectance in near infrared band. The limit between 0.2 to 1 displays the vegetation cover and the negative amounts stands for water surfaces. Also for acquiring better result a combination of NDVI and EVI can be used. The structure of EVI index is as follow [11]:

Where RED is reflectance in red band, BLUE is reflectance in blue band and NIR is reflectance in near infrared band.

4. Results

All of the previous steps were modeled in ENVI software. After calculating the parameters, thresholding method was used and the suitable limits of each parameter for incidence of Malaria in the study area were determined as information layers (Fig 3).

Fig 3. a.The suitable temperature limit (20-35 °C) in study area.b.The suitable relative humidity limit (55%-80%) in study area.c.Vegetation cover in study area.d.Water basins and wet surfaces in study area

In the next step, the selected regions were imported to a GIS work station as independent layers. Weighted overlay was implemented. Then at the last step the final high risk regions was recognized (Fig 4).

Fig 4.Malaria incidence risk map in study area

After recognizing the high risk regions, the statistical results of the suffered area were compared to the derived raster map.

5. Conclusion

As shown in this paper, Malaria incidence depends on the environmental parameters. Air Temperature, Relative Humidity, Vegetation cover and basins are discussed as the most influential parameters. By using satellite data, these parameters could be extracted, assessed and by GIS's analysis, the high risk regions could be recognized. Therefore, a comprehensive model could be developed to produce an output to recognize the high risk regions of malaria incidence. As discussed in this paper, due to 15 degree centigrade temperature difference and 25 percent relative humidity difference, the uncertainties due to the remote sensing technique could be considered acceptable.

6. References

  1. Edrissian GH, Nateghpoor M, Afshar A, Sayedzadeh A, Mohsseni G, Satvat M, Emadi A. Monitoring the Response of Plasmodium falciparum and P. vivax to Antimalarial Drugs in the Malarious Areas in South-East Iran. Arch Iranian Med 1999;2 (2):2-7.
  2. Martens WJ, Niessen LW, Rotmans J, Jetten TH, McMichael AJ. Potential impact of global climate change on malaria risk. Environ Health Perspect 1995;103 (5):458-64.
  3. Muirhead-Thomson RC. Mosquito behaviour in relation to malaria transmission and control in tropics. London: Edward Arnold, 1951.
  4. Sumana , B. , Sharma , C. , Dhiman , R.C. , Mitra , A.P. , 2006. Climate change and malaria in India. CURRENT SCIENCE, VOL. 90, NO. 3, 10 FEBRUARY 2006.
  5. Gemperli, A. , Vounatsou, P. , Anderegg, D. , and Pluschke, G. , 2004. EPIDEMIO: EARTH OBSERVATION IN EPIDEMIOLOGY. Proc. of the 2004 Envisat & ERS Symposium, Salzburg, Austria , 6-10 September 2004 (ESA SP-572, April 2005).
  6. M.A.Faghih, Malariology and Malaria Eradication, Tehran University, Iran, First Edition, 1996.
  7. Haghdoost, A.A. , 2004. Assessment of seasonal and climatic effects on the incidence and species composition of malaria by using GIS methods. London School of Hygiene and Tropical Medicine,Faculity of Medicine, University of London.
  8. Richard Allen, Masahiro Tasumi, Ricardo Trezza, Wim Bastiaanssen, August 2002, SEBAL: Surface Energy Balance Algorithms for Land, Version 1.0, Funded by a NASA EOSDIS/Synergy grant from the Raytheon Company through The Idaho Department of Water Resources.
  9. Baise, Haris, Robert, General Meteorology.
  10. Huggel, C. 1998. Periglaziale Seen im Luft- und Satellitenbild. Diploma thesis, Department of Geography, University of Zurich.
  11. Huete, A.R., H. Liu, K. Batchily, and W. van Leeuwen, 1997. A Comparison of Vegetation Indices Over a Global Set of TM Images for EOS-MODIS. Remote Sensing of Environment 59(3):440-451.