Download or read book Mapping Post-disaster Need written by Sabine Chandradewi Loos and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: After many disasters, humanitarian organizations, multilateral agencies, and governments must balance damage to the built environment with the disparate needs of diverse populations when planning for recovery. Fortunately, the emergence of nontraditional datasets--such as remote sensing or crowdsourced data--and the increasing focus on incorporating equity into disaster management, present an opportunity to inform effective and equitable planning decisions. The global availability of these nontraditional datasets also makes them particularly appealing for use in data-limited regions in the Global South that will be affected by future disasters. An overwhelming number of post-disaster data is available on building damage, making it difficult for users to synthesize into a single, credible estimate to use for loss estimates. And while there is an overwhelming amount of building damage data, there is a dearth of data that focuses on differences between households in their capacities to recover. This thesis introduces approaches to capture the physical and social needs of a region that has recently been affected by an earthquake, ultimately to support early recovery decisions. Specifically, I introduce two flexible approaches that leverage nontraditional datasets to rapidly estimate regional metrics of need after a disaster. The two measures of "need" I include in this thesis are 1) building damage and 2) non-recovery, a metric I introduce as a corollary to social vulnerability. The approach to estimate regional building damage is called the Geospatial Data Integration Framework, or G-DIF. G-DIF integrates multiple rapidly available datasets on building damage, whether it be from engineering models or remote sensing data, by calibrating them with a relatively small set of locally accurate field surveys. G-DIF was originally developed using damage data from the 2015 Nepal earthquake. Non-recovery is a metric that goes beyond building damage, looking to estimate which households are least likely to recover years after an earthquake. Similar to G-DIF, I develop a model for non-recovery in Nepal, this time associating multiple socioeconomic, geographic, and environmental factors to household inability to reconstruct five years after the 2015 earthquake. For both of these approaches, I evaluate their flexibility, or ability to adapt to new datasets and/or locations, including the Haiti 2010, New Zealand February 2011, and Italy 2016 earthquakes. The future of post-disaster data development not only needs to be statistically rigorous, but also usable and inclusive. Beyond developing clever and complex ways to model the impact from disasters, hazards engineers should thoughtfully consider the end users of our data products, and ultimately what populations we as a community are hoping to support. Overall, the approaches introduced in this thesis to estimate damage and non-recovery start to reconceptualize how we develop post-disaster data for decision-making, so we can start developing holistic and evidenced-based decisions that lead to more effective and just outcomes.