CDRP – Climate Disaster Recovery Process in Chennai Summary Report Zurich, Switzerland, December 2017 University of Madras About the Initiative The Climate Disaster Recovery Process Initiative for Chennai is an umbrella project of the Climate Policy group of ETH Zurich. This initiative builds on earlier work conducted between 2009 and 2012 as part of the Climate Disaster Resilience Initiative. This current research project was developed in cooperation with the University of Madras and with support from the Greater Chennai Corpora- tion, the Revenue Administration of the Government of Tamil Nadu, MLA offices of Mylapore and Velachery constituencies. A full master thesis of this research project is available upon request. Title CDRP – Climate Disaster Recovery Process in Chennai: Final Summary Report Report prepared by Dr. Jonas Joerin1 Prof. Ramasamy R. Krishnamurthy2 Franziska Steinberger1 Dr. Anna Scolobig1 Acknowledgments Dr. K. Satyagopal, IAS, Principal Secretary/Commissioner of the Revenue Administration of the Government of Tamil Nadu Dr. D. Karthikeyan, IAS, Commissioner of the Greater Chennai Corporation Dr. R. Nataraj, IPS, MLA of Mylapore Constituency Mr. Vagai Chandrasekhar, MLA of Velachery Constituency Special thanks to various representatives from residential welfare as- sociations, NGOs and youth organisations to support data collection and validation process. Title picture New settlements in Perungudi, Chennai Layout Sandro Bösch © 2017 ETH Zurich 1 ETH Zurich, Department of Environmental Systems Science, Universitaetstrasse 16, 8092 Zurich, Switzerland [email protected], +41 44 632 32 40 2 University of Madras, Department of Applied Geology, Guindy Campus, Chennai 600 025, India [email protected], +91 9677045519 Summary Report 1 Executive Summary After an unusual period of prolonged in- • The recovery time for socio-economic fac- tensive rainfall caused by a coincidence of tors (physical health, mental health, em- the northeast monsoon (October-December) ployment, household income, education, and the El Niño, Chennai was badly affected household assets, etc.) took longer to be by the 2015 South Indian floods. As Chennai restored compared to the physical items. is built around a former lagoon andis thus lo- For example, mental health took in avera- cated at the end of a series of water bodies, it ge up to four months to be recovered and is vulnerable to excessive periods of rainfall. household income losses more than three Since the catchment areas in Kanchipuram months in both areas. District were already filled-up by the end of • While physical items took longer to be November 2015, the relentless downpour of recovered in the more affected area rainfallon December 1, 2015, triggered the (Velachery), socio-economic factors took floods and forced the city to stand still. On equally long to be restored in Mylapore this day in average and citywide 286mm of and Velachery. rain fell within 24 hours[1] and the catchment areas of Adyar and Cooum received about 490 • The level of satisfaction about the recovery mm of rain. The amount of rain recorded on process did not differ between residents December 1 was the highest in 100 years[2]. living in Mylapore compared to Velachery. Damages were estimated at USD 3 billion and Therefore, the recovery time does not cor- 301 casualties were recorded in Chennai. relate with the level of disaster affected- ness. In response to this disaster, we assessed the recovery process 10 months after through • Residents in the more affected area (Vela- a household survey in two equally exposed chery) became more active to be better (coastal) constituencies, Mylapore and Ve- prepared (stock emergency supply, make lachery. The objective was to understand how their home flood proof, get informed about the physical, social and economic conditions flood mitigation options, etc.) for a future of the city recovered following this disaster flood disaster compared to residents in event. Furthermore, we looked at the inter- the less affected area (Mylapore). play between communities and authorities in • Residents in the more affected area (Vela- the recovery process. Following the house- chery) also become more solidary compa- hold survey, survey validation meetings were red to residents in the less affected area held with residents of Velachery (June 29, (Mylapore). For example, more residents 2017) and Mylapore (June 30, 2017). participate in a volunteer group, talk to Key findings from this study include: other members of their community and request assistance from neighbours. • Residents living in low-land (marsh- land) and newly built areas in Velachery • Residents in both areas showed equally were more affected by damages (houses, high willingness (around 90%) to become household assets, health) compared to part of the recovery planning process. residents living in the older part of Chen- • The absence of a needs assessment and nai (Mylapore). The median damages per limited involvement of residents in the household wereINR 30,000 in Mylapore recovery planning process are key areas and INR 50,000 in Velachery. where the interplay between residents and • The recovery time (average) for basic ser- authorities needs to be improved in the fu- vices (electricity, water, sanitation, solid ture. waste management, roads and communi- • Overall, this study summarises that the cation) took about seven days in Mylapo- efforts by the various governmental agen- re and around 14 to 21 days in Velachery. cies were effective in restoring physical, Housing took in average 22 days in Mylapo- social and economic items. re and 33 days in Velachery to be rebuilt. 2 CDRP – Climate Disaster Recovery Process in Chennai Foreword message Summary Report 3 Introduction After an unusual period of prolonged in- within 24 hours [1]. Damages were estimat- The 2015 South Indian tensive rainfall caused by the coincidence ed at USD 3 billion and 301 casualties were Floods in Chennai of the north-east monsoon (October-De- recorded in Chennai. The airport and basic cember) and the El Niño, Chennai was badly services stopped functioning, e.g. communi- affected by the 2015 South Indian floods. A cation lines and electricity were interrupted. relentless downpour of rainfall forced the Figure 1 highlights how the 2015 South Indi- city to stand still on December 1, 2015 after an floods unfolded and affected Chennai. in average and citywide 286mm of rain fell Figure 1 Timeline of the 2015 South Indian floods in Chennai The severity and unprecedentedness of was to better understand the physical, social Goals of study such an extreme flood event in Chennai, and economic drivers of recovery following a called for a thorough assessment to under- flood disaster event. Furthermore, we looked stand to what extent the city could recover at the interplay between communities and from this disaster. Hence, the key objective authorities in the recovery process. 4 CDRP – Climate Disaster Recovery Process in Chennai Methodology Few studies exist that quantitatively as- sess disaster recovery processes [3, 4]. Hence, this study aims to contribute to the understanding of the factors driving recovery processes. For this purpose, we selected two adjacent constituencies (Figure 2) with equal Mylapore coastal exposure and inclusion of water Bay of Bengal bodies (e.g. Adyar river, canals in Mylapore and basins/lakes in Velachery). However, Mylapore forms part of the old town of Chen- Constituency nai whereas Velachery was just developed during the last two decades and contains Rivers marshlands. The population of Mylapore is Velachery about 300,000 and 600,000 in Velachery. We identified households in both areas 5km N through a stratified random sampling pro- cess. In total, 257 households in Mylapore Figure 2 and 264 households in Velachery were sur- Study area: Mylapore and Velachery veyed during October and November 2016. The content of the survey included questions that are typical to a post-disaster needs as- sessment, such as the provision of basic ser- vices (e.g. electricity, water, sanitation, hous- ing, etc.) and socio-economic factors, such The analysis of the survey focused on as household income, employment, house- comparing the responses between the two hold assets, education, health, nutrition, etc. constituencies through statistical tests, in- For all these factors, we asked households cluding Mann-Whitney-Wilcoxon tests. De- how long they needed to recover and whether scriptive analysis, including boxplot analysis, the services/factors were better, the same, complemented the tools on how to extract or worse than before. Additionally, we asked relevant information from the survey data. residents whether they became more active, Validation meetings with residents from the solidary and engaged in their communities affected constituencies served to verify the following this disaster. analysed results. Summary Report 5 Results Among the surveyed households, the me- proportion (75% of the survey respondents) Damages on dian of the total recorded damages (damages of households in Velachery recorded dam- households on house, household assets and health) was ages up to INR 100,000. This highlights that INR 30,000 in Mylapore and INR 50,000 in households in Velachery were considera- Velachery. As Figure 3 shows, around 75% of bly more affected by the 2015 South Indian the households in Mylapore recorded dam- floods compared to Mylapore. ages below INR 60,000. In contrast, the same Figure 3 Total median damages in Mylapore and Velachery The reason for this is largely because the A somewhat weak correlation was found floods were higher in Velachery (as high as with the Pearson’s product-moment corre- 3m above the ground) compared to Mylapore. lation for the variables ‘annual income’ and A lack of effective drainage systems, particu- ‘total damage costs’ (p = 0.006, r = 0.136), in- larly in Velachery, is one of the key reasons dicating that households with higher annual why the water could not run off. Further- income also suffered from higher damages. more, Velachery is located in a low-lying area Moreover, Kendall’s tau correlation between with marshlands.