Assessment of livelihood adaptation strategies to climate variability in a coastal area of Southwest Bangladesh

Abstract

Climate variability refers to fluctuation of rainfall and temperature around the average without causing the long-term average itself to change. Bangladesh was ranked as the 6th hardest hit country by climate calamities among 180 nations and its coastal zone is the most vulnerable area which is affected more by climate variability than climate change. Climate variability includes unexpected and untimely heavy or less rainfall events, extreme temperature, etc. which ultimately lead to unexpected floods, water-logging, drought, sea-level rise and salinity intrusion in the zone. To understand the recent scenario of rainfall and temperature variability, statistical analysis was performed. The analysis involved the use of Coefficient of Variation (CV), Standardized Precipitation Anomaly (Z) and Precipitation Concentration Index (PCI) from 1978 to 2017 (40 years). Furthermore, Mann-Kendall test is used to detect the time series trend and its significance. The impact of climate variability on livelihood is calculated by a Livelihood Vulnerability Index (LVI). This index is frequently used to identify and compare climate change-specific livelihood vulnerabilities. The extent of the impact of climate change and variability on livelihood depends on the level of vulnerability of farmers and people of other occupations to these impacts. The index consists of six major livelihood components under which there are several sub-components and they contribute equally to the overall index. Impact of climate variability is evaluated by comparing the present LVI with the past. Participatory approaches like transect walk, group discussion, structured questionnaire survey and key informant interview are used to develop LVI and also to assess the current adaptation practices and evaluate potential future adaptation strategies. The variability analysis shows that for both wet and dry seasons, anomaly of temperature has increased significantly with time but rainfall showed little anomaly over the study period. The trend analyses show that except January, the other months, the dry season and the monsoon season have a notably increasing trend in average temperature during the study period. Trend is significant for 99% level of confidence for average temperature of the months May to September, of the dry and the monsoon seasons and also for the annual average temperature. In case of annual temperature variability analysis, highest variation is seen in earlier years-that is 1979-1984. Variation then decreases a little with time, but then increases notably in recent years 1997, 2006, 2009 and 2010, meaning that the variation in temperature has become more visible recently. In case of rainfall, July has increasing trend and December has decreasing trend at 90% confidence, but the other months show non-significant rising or falling trend over time. The analysis of PCI values demonstrates that 1979, 1984, 2006, 2009, 2011, 2014 and 2015 had very high rainfall concentration having the PCI values of 21.50, 21.92, 20.44, 20.08, 21.84, 22.41 and 22.35 respectively. The Z values indicate that 1985, 1989, 1996 and 2010 suffered from severe drought with Z values of -1.54, -1.36, -1.15 and -1.49 respectively, whereas 1992 and 1994 witnessed extreme drought with Z values of -1.91 and -2.17 respectively. The resulting LVIs reveal that livelihood has become more vulnerable due to climate variability in the present decade (0.443) than the past (0.408). Except livelihood strategies, all other major components (socio-demographic profile, health, food, water and natural disaster and climate variability) depict greater vulnerability in the present decade in comparison with the past. All of these are impacting the livelihood aspects like agriculture, health, water, environment etc. Finally, current adaptation practices in response to climate variability are discussed and potential future adaptation strategies to combat upcoming climate threats are suggested.