AHP and GIS-based Risk Zonation of COVID-19 in North East India

On 31st December 2019, a novel virus was reported from Wuhan City of Hubei Province of China, and later it was recognized as SARS-COV-2 (COVID-19). As the virus is highly human to human contagious, it has spread worldwide within a very short time. Since 24th March 2020, after the first reported case in North East India, the total confirmed cases reached up to 4,633 on 11th June 2020. In this work, an attempt has been made to delineate risk zones of COVID-19 in North East India using the Analytic Hierarchy Process (AHP) and overlay analysis in Geographical Information System (GIS). The evaluation is based on 14 criteria that were classified into promoting and controlling factors. The promoting factors include population size, population density, urban population, elderly population, population below the national poverty line, and percentage of marginal workers. In contrast, the controlling factors include available doctors, other health workers, public health facilities, available beds, governance index (composite and health), and testing laboratories. The results were classified into very high, high, moderate, low, and very low risk zones. Most densely populated states with massive pressure on health facilities are likely to have a higher risk of COVID-19. Assam, Tripura, Meghalaya, and Nagaland show a high COVID-19 risk, which constitutes almost 76.93% of the North East India population, covering 48.80% of surface area. The states under a moderate risk zone include 6.92% of the population over 8.52% of the area. Lastly, 16.15% of the people living over 42.69% of the total area belong to the states with a lower risk zone. Current World Environment www.cwejournal.org ISSN: 0973-4929, Vol. 15, No. (3) 2020, Pg. 640-652 CONTACT Gibji Nimasow gibji.nimasow@rgu.ac.in Department of Geography, Rajiv Gandhi University, Rono Hills, Doimukh – 791112, Arunachal Pradesh, India. © 2020 The Author(s). Published by Enviro Research Publishers. This is an Open Access article licensed under a Creative Commons license: Attribution 4.0 International (CC-BY). Doi: http://dx.doi.org/10.12944/CWE.15.3.29 Article History Received: 17 June 2020 Accepted: 17 November 2020


Introduction
The World Health Organization (WHO) country office has been informed about the pneumonia cases of unknown etiology detected in Wuhan City, Hubei Province of China on 31 st December 2019. 1 Unexpectedly, it spread to different regions of China as well as other countries across the world, despite China's considerable efforts to restrain the infection within Hubei. 2 Later, the epidemic was recognized as novel coronavirus of 2019 or SARS-CoV-2 resulting in the disease COVID-19. 3 On 31 st January 2020, the WHO declared coronavirus as a public health emergency of international concern. 4 It is a member of a large family of coronaviruses resulting in Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS). 5 Compared to the SARS-CoV (2002/2003) and the MERS-CoV (2012-2014), COVID-19 has remarkably faster human-to-human transmission as it took only 48 days to infect 1000 people, whereas MERS took around 2½ years and SARS took about four months to reach that figure. 2 Based on the global spread of COVID-19, the WHO declared it as a pandemic. 6 In general, the virus is capable of infecting people of all ages, but the population with above 60 years of age and people with heart disease, asthma, diabetes, chronic lung disease, kidney disease, etc. are at increased risk of severity of COVID- 19. 7 Although health facilities and socio-economic conditions of the people have drastically improved since independence, human development and its growth are destitute in North East India (NEI, hereafter). In fact, it is lower than many underdeveloped nations of the world. 8 Since the initial detection of COVID-19 in Kerala on 30 th January 2020, it had spread to many parts of the country. Presently, there are 276,583 confirmed cases, 7,745 deaths, and 135,205 cured cases of COVID-19 in the country as on 10 th June 2020. 9 The cases have increased tremendously from 519 confirmed cases with ten deaths as of 24 th March 2020 to date. NEI is located in the easternmost part of the country, which is inhabited by 3.88% of the country's total population. The earliest infection of COVID-19 in NEI was reported from Manipur on 24 th March 2020, 10 and it took only 78 days to reach 4,433 confirmed cases as of now. 9 The number of cases has been increasing despite the entire nation been put under lockdown (in different phases) from 25 th March 2020 by the central government. 11 Due to the absence of a vaccine, avoidance of touching the nose, eyes, and mouth, frequently washing hand, the practice of hand sanitizers, covering of face with a proper quality mask, social distancing, and respiratory hygiene are the quotidian measures to stay safe from the virus. 5 The inherent large-scale regional disparities in terms of demography and socio-economic characteristics, along with depressed health conveniences, are likely to exacerbate the pandemic situation in the region. Therefore, an effort has been made to delineate the risk zones of COVID-19 in NEI using the data gathered from various sources of the Government of India applying Analytic Hierarchy Process (AHP) and Geographical Information System (GIS).

Materials and Methods Study Area
The study area constitutes 8 NEI states, namely Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim, and Tripura (Fig. 1). The region is characterized by mountains, hills, and plains with rich culture and biological diversity. NEI shares an international boundary with Nepal, China, Bhutan, Myanmar, and Bangladesh and a state boundary with West Bengal.

Methodology
Based on promoting and controlling factors of COVID-19, fourteen thematic layers (Table 1 & Fig. 2) have been considered to carry out the present study using ArcGIS 10.3 software. AHP was used to assign the weights for each individual reclassified layer (generated after converting ancillary data into raster format) to perform the Weighted Overlay technique to generate the final risk zonation map of the study area.

Analytic Hierarchy Process (AHP)
As per the literature review of available materials and expert opinions, Saaty's fundamental 9 -points scale values were assigned to each thematic layer according to their potentiality on generating risk zones of COVID-19. The weights assigned to different layers were normalized and checked for consistency (consistency ratio) as suggested by Saaty (1980). 23 The consistency ratio reflects the probability that the matrix ratings were randomly generated. The consistency ratio was derived using the following equations: Consistency Index (CI) = (λ max -n)/(n-1) ...(1) Where, λ max is the largest eigenvalue of the pairwise comparison matrix, 'n' represents the total number of parameters.
Consistency Ratio (CR) = (Consistency Index (CI)/ (Random Consistency Index (RI) ... (2) The value of the Random Consistency Index (Table 2) was obtained from Saaty (1980). 23 For consistent weights, the value of CR should lie between 0 and 0.1 (i.e., 10%); otherwise, the corresponding weights should be re-evaluated. In this study, the consistency ratios of pairwise comparison matrix for promoting and controlling factors were 0.044 and 0.025, indicating that the comparisons of evaluation criteria are consistent.

Delineation of Risk Zones
The factors (promoting and controlling) undertaken to carry out the study are considered to have the potentiality to influence risk during any kind of pandemic situation (Annexure 1 & 2). These influencing factors are weighted as per their response to taking risk where the higher value in promoting factors represents a high potential to promote risk, and the lower value in controlling factors represent high potentiality to defeat the risk. A weighted overlay analysis was executed using both the factors (Fig. 2) in the GIS environment to delineate the risk zones using the following formula:

Results and Discussion
All the sub-criteria of selected thematic layers were assigned relative ranks based on their influence in promoting and controlling the situation (Table 4 & 6). The overall potentiality of promoting and controlling the COVID-19 pandemic has been generated through overlay analysis of the layers. Finally, the risk zones were delineated out of the promoting and controlling layers by providing equal importance (Table 7).

Promoting Factors
Among the promoting factors (Annexure-1) that would increase the cases of COVID-19 in NEI, the dominant factors (Table 3) are the concentration of urban population (24.57%), followed by population density (22.18%), population below national poverty line (25.19%), population size (13.89%), percentage of marginal workers (11.09%) and percentage of the elderly population (8.70%). As the nature of the virus is human-to-human contagious, the factors that promote human gathering and make trouble to stay inside the home for a longer time get higher weights of influence. The results based on relative weights (Table 4) and overlay analysis (Fig. 4)    Large population size with high population density was found to promote a high risk of COVID-19 in Assam, followed by Tripura. A moderate risk of promoting the pandemic was found in Mizoram, Manipur, and Nagaland. Sikkim and Meghalaya were found to have low risk, while Arunachal Pradesh has a very low risk of promoting COVID-19 (Fig. 4).

Controlling Factors
Among the controlling factors (Annexure-2), the most influencing factors were availability of doctors (24.01%) and available testing laboratories (24.01%) followed by other health workers (14.84), bed available in public health facilities (10.99%), good governance health index (8.08%), number of public health facilities (7.27%) per capita income (6.40%) and good governance composite index (4.40%) as shown in Table 5. The results based on relative weights (Table 6) and overlay analysis (Fig. 6) 25 At present, most of the reported cases of COVID-19 are from the quarantine centers, and the death rate from COVID-19 is very low (Fig. 7) with only six deaths recorded out of 4,633 confirmed cases as on 11 th May 2020. 9 Among the states, Assam has the highest 3,092 confirmed cases of COVID-19 as on 11 th June 2020 (Fig. 8). The total number of active cases in Assam was 1,893, followed by Tripura with 655 cases.  (Table 7), the potential risk zones of COVID-19 generated thereof has been shown in

Concluding Remarks
The study shows the applicability of AHP and GIS in delineating the risk zones of COVID-19 in North East India. The urban population, population density, population below the national poverty line, population size, the proportion of marginal workers, and the percentage of the elderly population appears to play an essential role in promoting Covid-19 in North East India. While the influence of testing laboratory, availability of doctors, other health workers, bed available in public health facilities, and good governance health index plays an essential role in controlling COVID-19 in the region. Assam and Tripura have a higher risk of promoting COVID-19 transmission in a very short period. On the other hand, Assam, Meghalaya, and Tripura have very weak means to control the severity of COVID-19.
Overall, Assam, Tripura, Meghalaya, and Nagaland have a high risk of COVID-19, while Mizoram, Arunachal Pradesh, and Sikkim have a lower risk. Therefore, the respective state governments need to assess their strengths and weaknesses and develop strategic plans to fight against the pandemic. Lastly, frequent testing of COVID-19, immediate quarantine of suspected people, proper social distancing, and regular practice of face mask and hand sanitizer may decelerate the transmission rate of the disease in the absence of a vaccine of COVID-19.